PDA

View Full Version : Water Injection, good or bad ???



Mark.A
2nd April 2000, 22:36
When I first mentioned that I was going to install WI, it was suggested that I must have a dodgy Possum map. Because the only reason it would be needed, was to solve a detting problem !!!.

No, my map was great, no det', safe EGT's, and only ever made minor adjustments, + octane booster for the track. This was to stay on the side of caution.

I wanted MORE, but safely.

I'd been running as much boost, and advance as I could, without getting detonation, and if I wanted to go any further, cooling the charge temp was the next step.

Not wanting to go the front mounted intercooler route, and whilst upgrading the turbo to a larger one would help (by getting the same volume of air, with less boost, would/should give lower temps) I still felt that water injection was the way to go. Particuarly after seeing J's engine hold together at Donnington (I know what boost you were running http://bbs.22b.com/ubb/wink.gif).

Bob Rawle helped me fit it (thank god)and after some advise from Richard Lamb at ERL, I re-mapped the Possum, and mapped the controller for the WI.

It's allowed me to run more boost (1.5bar peak, 1.4bar held) and advance the ignition by 1>3 degrees under max load, reduce my EGT, and I'm det' free http://bbs.22b.com/ubb/biggrin.gif.

My car was fully tested at Donnington recently, and ran like a dream. My driving on the other hand !!!!

A proper water tank is needed however, as it can use upto 1/4ltr of water a minute http://bbs.22b.com/ubb/eek.gif

My thanks to Bob Rawle, without whose help I'd still be scratching my head.

I'd recommend anyone running high boost concider WI. Particually if you are using a Possum, or equivelent, and want even more, safely.

Mark. VF22 on order !!!!.

Anders
3rd April 2000, 00:07
Pat is very excited about water injection in fact he sprayed all over me in the car park on one of the restaurants by the 2nd day stage of Catalunya 3!

He had to prime the system first under some pressure and the pipe was pointed at me!

After the water was through he connected stuff back up.

Apparently this was the first time he had used it.

I hoped it would be a "fill and forget" system?

BPM
3rd April 2000, 01:16
What has held most componies from manufacturing a TMIC with water/air is the cost of the pump. The unit is a better option due to water obviously being a superior heat absorber...
Who has one?

What this space :)


Greg Nikolettos www.bpmsports.com (http://www.bpmsports.com)

Mark.A
3rd April 2000, 07:13
Greg,

Are you talking about a "charge cooler", I know PACE tried it, but couldn't get it to work very well. If you can, that would be great.

By water injection, I don't mean "water spray". The ERL system sprays water directly into the intake system, just before the TMIC, mixes with the air, and cools the engine internally.

Anders:

Once the system has been primed, and used, the only time priming would need to be done again, is if you ran out of water, and ran the pump "dry".

I've got a "low level" sender on my water tank, and a LED on the dash. So I shouldn't ever run out.

All I've had to do in the last 2/3 months is top up with water. On the road, you use very little water, as the WI is only used under full load. I've adjusted mine to come on at about .9bar. At Donnington I used about 10/12 ltr of water (this includes the IC water spray, that also uses the tank. I only did about 75mls at Donno' !!!.

If you go for the mappable ERL system (2S) with the MF2 controller, it also has a system failure LED, that will even tell you if the jet is blocked.

Mark.

BPM
3rd April 2000, 07:38
Mark

Communication breakdown , I was referring to a charge cooler.


Greg

Firefox
3rd April 2000, 11:52
Howdy...

Hmm... Water Injection... now let me think...

You ever gonna send me your old intercooler pipe back ? I need it for a customer... http://bbs.22b.com/ubb/smile.gif

I have been running WI for a while.... good results.. no det...even with high boost.... I have the pistons to prove it. http://bbs.22b.com/ubb/smile.gif

I know about water spray on top mount.. I have run it.. (and a kit for it)

I know about water spray on a front mount.. I have run it .. (and a kit for it)

I know about front mount I/C.... I have one... and kinda have a kit (Modified MRT).

I know about charge cooling... My previous car was used to develop the Pace Charge cooler system for the scoob.

A VF22 on order ?

Time for Phase 5 I think.. http://bbs.22b.com/ubb/wink.gif

J.

Mark.A
3rd April 2000, 18:08
Shit J, is there ANYTHING you don't know about http://bbs.22b.com/ubb/biggrin.gif.

I'll call you re pipe.

Mark.

iwatkins
3rd April 2000, 21:45
To answer the original question. Water Injection is good and once fitted doesn't cost anything to run (my favorite bit http://bbs.22b.com/ubb/wink.gif).

I have been running the ERL System1 for about six months now alongside the Unichip. My GT4 (ST165) has a chargecooler as standard also. Running the System1 on a pressure switch trigger set at 8psi with a 0.5mm jet.

I have yet to play with jet sizes/trigger pressure. One of those things I will get around too.

However, we (GT4 owners) have found that the temperature of the water itself does have a slight effect. All owners are currently moving their tanks from the engine bay (very hot) to either the boot or very low down in the front air dam.

On the GT4 we inject post charge/inter cooler as a knock prevention device (mainly) because of traffic heat soak problems.

Scary thing recently discovered about the chargecooler system on the GT4. The chargecooler pump has a computer of its own. It is triggered by going WOT. The pump then runs for about 25 seconds then stops. It will not restart the pump again for at least three minutes no matter what you do. Bonkers.

Needless to say, that computer was in the bin very quickly. The pump now runs full-time (with ignition on) and with a little trickery, the main ECU doesn't know about it http://bbs.22b.com/ubb/wink.gif

J,
Thought about using an air con system to cool the coolant for a chargecooler system ? Apparantly a lot of work has started on this again in the US with good results. I.e. an air con system wrapped around a water/air intercooler. Power loss from running the compressor is made up for in the gains in power. Where is my physics books....

Cheers

Ian

[This message has been edited by iwatkins (edited 03 April 2000).]

Anders
3rd April 2000, 23:36
Ian can you mirror this on the GT4 section as well http://bbs.22b.com/ubb/smile.gif

glenns
4th April 2000, 10:20
A question for those with W.I..

How does the ERL system handle properly measuring the amount of H2O for injection at different revs/loads/boost levels???

Is their set and forget type system workable or is it worth the extra money for the programmable system??

I have looked at getting the ERL system for my "mildly" http://bbs.22b.com/ubb/wink.gif modified WRX...

iwatkins
4th April 2000, 13:45
Glenns

There are basically three systems:

1. Pressure switch driven system (System 1s). This comes on at the preset pressure and stays on until the pressure drops again. E.g. I have mine set to 8psi running a 0.5mm jet. The jet size dictates the amount of actual water injected. The system works well and gives good intake/cylinder cooling. However, further improvements could be made by playing with the jet size and the trigger point. Something I have yet to do.

2. Computer controlled (System 2s). Using the MF2 system can be mapped against boost and rpm hence giving a 3D map. See this page: http://member.newsguy.com/~gtfour/modERL.htm for an end users perspective on this system. This system also uses a pulsed solenoid to give an exact amount of water against boost against rpm.

3. Aftermarket ECU controlled (System 2c). This system is like the System 2s except you don't get the computer. You would hook this up to say a MoTeC or Unichip to control when and how much the water injection runs. I.e. you know the jet size you have fitted so you just set duty cycle for the system at that revs/boost point in a map. It can also do other funky stuff like feedback the system state, i.e. block jet, run out of water etc.

For the full info. see the ERL website at: http://www.aquamist.co.uk/

Cheers

Ian

Firefox
4th April 2000, 18:13
Ian...

Forget the idea of the aircon..

It has been tested.. and believe it or not...found not capable of providing enough cooling ?

Many ideas have been tried... ranging from seperate refrig plants.... dry-ice...etc...

Dry-Ice has proved very good for drags...

I have had a charge cooler fitted to a cav turbo.... and also fitted to my UK turbo (different ones - lol)

The water reservoir and core design of the C/C play an important part.... one problem is still heat soak.... in a way its worse... you have far more metal to cool down.

The way to minimise heat soak to to continously have the pump running (at a low speed to save its life). And then switch to high speed when needed.

If designed right they can be very good. But the choice between charge cooler...top mount... front mount all depends on the use of your car....

if you are dragging go for a front mount....or possibly an uprated top mount. The extra weight of the C/C isnt suitable for dragging... and the system doesnt provide a quick enough cooling effect.

Anyways.. I'm waffling... and havent really got much to say..

tara

iwatkins
4th April 2000, 20:54
J,

Yep, I was joking about the air/con http://bbs.22b.com/ubb/wink.gif

The arguement over air/air and air/water has turned into a long term squabble/joke in the GT4 community as the Mk2 GT4 has a top mount air/air whereas Mk1 and Mk3 both have air/water.

"Mine is better than yours for..." is always boiling away just below the surface http://bbs.22b.com/ubb/smile.gif

The heat soak issue of the air/water cooler on the GT4 is a big one as the unit is top mounted but with the radiator front mounted. Best results have been found by going for a full size radiator that not only increases cooling surface area but also means adding about 55% more coolant to the system. With the pump running full time and a fan (temp controlled) on the radiator gives the best intake charge temps. over the air/air after slow (<35mph) city driving.

On the open road (read decent airflow) both are about the same, but the air/air has the weight advantage.

Drag racing is a good one. While queuing the heat soak really affects both systems but we have yet to check intake charge temps say after queuing (idling) for 10 minutes followed by 15 (or so) seconds of flat out acceleration.

I still like the idea of putting about 2 cubic foot of dry ice under the bonnet for a few minutes before drag racing but this isn't practical for everyday use.

One thing I have heard that works really well is NO2 sprays over air/air intercoolers. I.e. use exactly like a cooler water spray but with NO2 or other very cold gas. Can work a real treat and get charge temps ***below*** ambient. Of course, you have weight/cost disadvantages but could be workable.

Cheers

Ian

Firefox
4th April 2000, 22:38
Howdy...

You will be amazed at some things people have tried... lol

But remember there is a cut of point with how cold you can go... if you go too cold you will have fulling/ignition problems.

People have used broken dried ice on the I/C.. then broken dried ice in the water tank to the I/C spray.... then they run both...

Another trick for dragging is to remove your power steering and alternator belts.... you could try an underdrive pully..

No2 would work... but I'm not sure on the quantities required to cool the outside... Stef mentioned a c02 extinguisher... I suppose any inert compressed gas being quickly released under pressure would work... but then you have to consider the control systems.. weight of the cylinder...etc...

I know you can now get Carbon No2 cylinders... nice and light weight...

I suppose it depends on how serious you are...

J.

shiv
5th April 2000, 08:57
Water injection is a funny thing. And I would argue that it only belongs in system which is experiencing detonation problems. Of course, ideally, a system should be detonation-free. As we all know, this often involves the revision of ignition, boost and fuel maps. If doing this is impossible, water injection is often used-- as a last resort. However, those who state they they never had detonation problems before, and with water injection, are able to raise boost and advance ignition timing, should be aware that it is very likely that they are not gaining any horsepower. All that water mass in the intake charge replaces fuel and air-- you know, the stuff that makes power.

Simply put, I just don't believe water injection belongs on a properly tuned turbo car.

Shiv
EJ25 turbo
USA

iwatkins
5th April 2000, 09:23
Shiv,

Corky Bell (Book: Maximum Boost) also doesn't rate water injection. I don't have the book handy but he really did slag it off. There is a definite anti-water injection bias is the US, I hear it again and again.

Personally I use it as det. prevention. Not that I get det, most of the time with it switched off. However, with varying fuel quality, occasional really hot days/hot car etc. I would rather loose a few bhp everyday than loose a piston on one day. No way I could hear det. on my car at 5000rpm in top gear, especially with the stereo on http://bbs.22b.com/ubb/wink.gif

Water injection does give lowered intake charge temperatures, fact. How much power loss due to the water in cylinder, I don't know. But it doesn't seem to be much.

Drive your car on a dry hot day. Then drive it on a cooler damp day. Which feels the more powerful ? You get my drift.

I would agree though, in an ideal world where 97RON fuel is 97RON fuel, and underbonnet temps, ambient temps, humidity levels and atmospheric air pressure were all constant you shouldn't get det. even on a highly tuned engine. However, we don't live in an ideal world and all these things change on an hourly basis.

Cheers

Ian

Mark.A
5th April 2000, 09:25
Shiv,

So is that why the top rally teams use it, because they are unable to map the cars correctly ????.

WI on used on it's own, will infact reduce power, but if used in conjunction with a re-mappable ECU, not only can this power be regained, but exceeded.

It also allows you to run a leaner mix, that would normally be marginal on a turbo car, and create high EGT's. A lot of tuners run turbo cars artificially rich to reduce temps, this in itself reduces power.

Mark.

iwatkins
5th April 2000, 09:32
Mark,

Good point. After I had my WI fitted, we remapped the Unichip and managed to take quite a bit of fuel out. Toyota's standard maps were rich, in fact, filthy rich http://bbs.22b.com/ubb/smile.gif

Cheers

Ian

shiv
5th April 2000, 09:53
Mark-- Comparing a street car to a race car often isn't the best idea. When you are trying to extract 400+hp out of an intake restricted 2 liter motor that gets torn down and overhauled after every race, tuning requirments are rather unique. If you want to use water injection, that's fine. Just make sure that your seeing actual power gains when all is said and done. Because, in my experience, not only is this unlikely, but relying on a little solenoid and a depletable resevoir of water to keep your engine from pinging to death isn't my idea of good tuning.

Ian-- I would argue that there are better ways to maintain a safety margin in light of an environmental conditions (climate, gasoline, heat soak, etc,). The first method is to use a conservative timing map. Often, tuners will advance ignition timing right to the threshold of audible knock-- not realizing that they can retard timing, depending on rpm, by an additional 4-5 degrees without loosing power. In fact, when coupled with a less rich a/f mixture, more power can be generated, more safely. Second would be to use a user-programmble knock sensor. Set a sonic threshold high enough to ignore typical engine noise while intervening agressively once that noise has been exceeded. Inhibit the sensor above 5500rpm or so (due to engine noise) and use a conservative ignition map from there onwards. The third would be to use an appropriately sized turbocharger that doesn't result in excessive, knock-inducing exhaust back-pressure at higher engine speeds. The stock turbo is undersized for the kind of power people want to generate with these cars. This latter solution would, without a doubt, yield major improvments in not only power, but also knock resistance. And last, the use of an efficient, low pressure loss, bar and plate intercooler would help as well.

Also, drawing an analogy between the cooling effects of water injection and that of a nice cool night isn't very accurate. The latter environment yields a far denser, oxygen-rich charge and generates more power.

shiv

[This message has been edited by shiv (edited 05 April 2000).]

Shaun
5th April 2000, 10:35
Dear all,

Several points are being raised and made by postings on this thread. Some valid, some not so valid.

We all have our opinions, but which are based on physical experience - not hearsay!!

Ian - I know you have ACTUAL experience, so your points are based on factual issues.

The point of saying, that if a engine is mapped correctly then WI is not required is total tosh. This all comes down to the type of engine and how it reacts to certain mod changes. If a re-mapped car was boarder lined to take for e.g. 1.4 bar of boost (anything more would cause det), then one of the ways to get over this limit, would be to use WI to cool inlet temps (but is not the be-all and end-all). So it's a case of horses for courses. Ian is a fine example for this. He has had his ecu re-mapped, but to allow the extra boost (which is some what more than standard), he needed to use WI, to combat det. It all depends, on how far you are pushing your engine. Common denominator is more power/boost = more heat. Just increasing the fueling to combat the heat, will only decrease power.

On the other hand, I have had experience of both charge cooling and WI on previous cars. And I can quite catagorically say while it does improve safety (it was not required for det purposes though, as mods were not immense), the main reason for using it, was to ensure the maximum boost/power was available at all times and in all weather conditions. Now for this purpose it worked very well!!!! The cars were always URGENT all the time, and so much smoother to boot.

There are obivously pluses and minuses for using WI, Charge Cooling and intercooler relocation kits (which I am not going to talk about here!). But for pound for pound value, WI has to be the best way in combating temp related issues, whether its to combat det or to ensure optimum boost/power at all times or just as a extra safety plan.

Shaun.

iwatkins
5th April 2000, 11:45
Shiv,

Yes, my analogy sucked, big time http://bbs.22b.com/ubb/wink.gif But even the same air temp and air pressure but one with 20% humidity and the other 95% humidity, there is a real difference.

You are quite correct. There are many methods of doing the same thing. The main target is to reduce intake charge temps and hence increase the density of the air mass (not volume) entering the cylinder.

You are also quite correct on the mapping front, you can often wind back the advance a bit and not loose any power. All depends on how conservative you want to be.

New turbo and bigger intercooler costs in the UK. Also, larger turbos tend to be more laggy (unless you spend the extra and go for special bearings etc.). Larger intercoolers *can* also cause bigger pressure drops also adding to lag.

It is swings and roundabouts. I for one think that for a mildly modified road car, water injection is a big help (not perfect though). If you start to get serious, then sure, bigger turbos, intercoolers, great.

Water injection also has other benefits such as cylinder cooling, flame front speed control, reduces carbon deposits (another cause of knock) etc. etc.

I do know it is not perfect, but compared to the cost of a new turbo and intercooler plus all the fittings and labour to fit it, it is worth the money (and then some).

Cheers

Ian

Mark.A
5th April 2000, 12:42
This is fun http://bbs.22b.com/ubb/biggrin.gif. Shiv, do I take it that you don't like WI http://bbs.22b.com/ubb/smile.gif.

I know of an STi that was running about 320bhp, this lost about 6bhp with WI installed. It also lost about 30 degrees inlet temp. Had the guy bothered to re map the ecu, the 6bhp would easily have been regained.

As for "solenoids", we're more than happy to let them control our boost !!!, and personally, apart from the Aquamists own warning system, I have a low level sender on my water tank http://bbs.22b.com/ubb/wink.gif.

I admit my knowledge is limited, and I can only relay my own personal findings.

Mark.

Anders
5th April 2000, 14:51
Mark, Shiv is a famous techie from America.
He has built a RS Impreza turbo http://bbs.22b.com/ubb/eek.gif

Shiv reckons his car is faster than a 22B http://bbs.22b.com/ubb/cool.gif

I also value your opinion on the water as I am considering going that route! http://bbs.22b.com/ubb/biggrin.gif

MorayMackenzie
5th April 2000, 16:13
Anders,

Don't worry mate, Americans do quarters, not corners! http://bbs.22b.com/ubb/wink.gif

Moray
(Where has that Anderson shelter gone? http://bbs.22b.com/ubb/wink.gif )

[This message has been edited by MorayMackenzie (edited 05 April 2000).]

Mark.A
5th April 2000, 16:15
Anders,

Shiv is obviously very knowledgeable, and I'm more than open to suggestions. I opted for WI as it seemed like the logical progression (and currently still does).

When the VF22 goes on, this will also help to reduce inlet temps, and also increase power.

Whilst I keep an eye on the Lambda Link when tuning, my main concern is the EGT gauge. This gives a very good picture of what's happening inside the engine. Far more so than those dodgy lambda sensors !!!!.

I can only say, that with my level of mapping experience, diagnostic equipment available to me, and apparent results, it has done what I expected of it.

That's not to say that if Shiv got his paws on it, he couldn't get simular results without the WI, but I doubt it.

When the VF22 is fitted, and I get some time on the RR, I'll try to test it with, and without the WI, and if he's right, I'll probably keep it quiet, and still argue http://bbs.22b.com/ubb/biggrin.gif.

Mark. Always willing to learn, always willing to argue !!!! http://bbs.22b.com/ubb/wink.gif.

ian/555
5th April 2000, 17:40
OK guys I haven't got a clue what half of you are on about http://bbs.22b.com/ubb/confused.gif . But what if you could cool the petrol down before it was injected into the engine to say below 0*C via the A.C. by using some sort of matrix for the petrol to flow through with the A.C. going through/round the matrix, and then some sort of lagging on the fuel line, like what is used on the downpipes to stop heatsoak in the engine bay would it improve things? Does petrol's burn charactistics change alot if it's too cold. I could be way off the mark as my knowledge is very ltd. Awaiting ridicule regards Ian http://bbs.22b.com/ubb/tongue.gif

shiv
5th April 2000, 18:33
<BLOCKQUOTE>quote:</font><HR>Originally posted by Mark.A:

As for "solenoids", we're more than happy to let them control our boost !!!
<HR></BLOCKQUOTE>

Except when a boost control solenoid goes bad, boost reduces to the limits set by your mechanical wastegate. When a water injection solenoid fails, your motor pings to death. I've seen this happen more than once.The only nice thing about water injection is when your motor fails, it's already nicely steam cleaned... heh heh...

shiv

shiv
5th April 2000, 19:08
<BLOCKQUOTE>quote:</font><HR>Originally posted by Shaun:
The point of saying, that if a engine is mapped correctly then WI is not required is total tosh. This all comes down to the type of engine and how it reacts to certain mod changes. If a re-mapped car was boarder lined to take for e.g. 1.4 bar of boost (anything more would cause det), then one of the ways to get over this limit, would be to use WI to cool inlet temps (but is not the be-all and end-all).<HR></BLOCKQUOTE>

Hi Shaun,
When I mentioned that a properly tuned car does not require water injection, I was not referring to mapping alone-- that's only one ingredient of the recipe. Charge control and exhauast back-pressuree managment are the two big others components. From what I can see, the exhaust situation (between the exhaust ports and the turbine wheel) often goes unaccounted for and leads to instances where more boost yields very litte extra power coupled with a greater tendancy for detonation (cause by reversion). This is when people start to use WI as a band-aid. While it doesn't solve the problem, it does address the symptoms. Essentially, one is using water to cool the combustion chamber which is slowly getting filled with exhaust waste during the overlap period. That's like kicking someone in the knee to take their mind off their headache... http://bbs.22b.com/ubb/smile.gif

Shiv

shiv
5th April 2000, 19:42
One more thing. Using a lambalink or any other 02 sensor voltage measuring device to monitor a/f ratios can and will be misleading when using water injection. Even heated o2 sensor elements are influenced by changes in exhaust temperatures. All things equal, the hotter it is, the richer it will read. As we all know, if you dump water into the charge, EGTs will go down significantly. But, all of a sudden, you're reading your lamba monitor and saying, "wow... I'm running a lot leaner with WI and I'm not even knocking. This must be good" Well, chances are, you're not running leaner. You're exhaust is just cooler. This is especially true for those who aren't running catalytic converters in the midpipe.

If you want to run cooler and don't want to run appropriately-sized turbos or intercoolers, might I suggest running on a methenol mix? http://bbs.22b.com/ubb/smile.gif

shiv

iwatkins
5th April 2000, 20:20
Methanol, been there, done that.

We found that any more than 70/30 (water methanol) would cause a increase in problems. The optimum was about 10% methanol in water. This is just enough to stop the mix freezing on most nights. Bargain. http://bbs.22b.com/ubb/smile.gif

Basically I run 90/10 mix in winter and then switch to 100% water once warmed up a bit.

Also methanol has a lower latest heat of evap. if the numbers I remember are right, so any more than the minimum amount required to prevent freezing, will give you less cooling effect.

I once had a nice book with latent heats of various compounds from my chemistry days, cannot find it now though.

Also, some components of water injection systems don't like methanol/ethanol etc., so be warned before you start pouring nitro-methane etc. into you WI tank http://bbs.22b.com/ubb/wink.gif http://bbs.22b.com/ubb/wink.gif http://bbs.22b.com/ubb/wink.gif

Cheers

Ian

shiv
5th April 2000, 20:28
Uh... I meant methanol mix as a fuel (as opposed to straight gasoline)-- Not as a complement to water injection.

shiv

Bob Rawle
5th April 2000, 20:40
Shiv, you comment on temperature making an O2 sensor read rich. IMHO increase temp and the O2 sensor output drops. So say at 400 deg C a given O2% gives 0.99v, at 400 deg C it gives 0.97v, at 750 deg C it gives 0.92v, at 900 deg C it gives 0.88v. Lower output leaner mixture (relatively. Introduction of water into the air/fuel mixture does influence the sensor and it does make it read lean but thats not down to cooling, its down to the fact that there is no (or very little) excess oxy in the exhaust gas whilst the water is running.

Unless I missed something http://bbs.22b.com/ubb/smile.gif

Bob

shiv
5th April 2000, 21:31
<BLOCKQUOTE>quote:</font><HR>Originally posted by Bob Rawle:
Shiv, you comment on temperature making an O2 sensor read rich. IMHO increase temp and the O2 sensor output drops. So say at 400 deg C a given O2% gives 0.99v, at 400 deg C it gives 0.97v, at 750 deg C it gives 0.92v, at 900 deg C it gives 0.88v. Lower output leaner mixture (relatively. Introduction of water into the air/fuel mixture does influence the sensor and it does make it read lean but thats not down to cooling, its down to the fact that there is no (or very little) excess oxy in the exhaust gas whilst the water is running.

Unless I missed something http://bbs.22b.com/ubb/smile.gif

Bob<HR></BLOCKQUOTE>

No you didn't miss anything. Maybe I should have been more clear with my words. Let me restate my comment: I have seen high EGTs yield innacurately high 02 sensor voltages. This doesn't mean that you will not see lean 02 sensor values. It just means that the a/f ratio displayed on your lambda meter (which reads of the o2 sensor) are less lean than they really are. It's all relative. This becomes especially true when operating close to 900C, in which the narrow band 02 sensor becomes partically useless. As it does when also operating too far away from stoich. This has been the fall of many high boosted engines. If you want to play it safer, use a wide band horiba 02 sensor. The $30 sensors that come from the factory are best used to ensure proper cat performance and stoich running conditions-- not to refer to when tuning a car for higher boost.

-shiv

[This message has been edited by shiv (edited 06 April 2000).]

Firefox
6th April 2000, 12:56
Hi Shiv..

Welcome over here... http://bbs.22b.com/ubb/smile.gif

I see u r causing trouble again.... http://bbs.22b.com/ubb/wink.gif

Water injection injection in my case is used as a "last resort".... I can say this.. as I use it... I have it coming in at high boost....if I need it.

You should use other more "permanent" solutions.. larger turbo.. front I/C... etc... where possible

But for most people they cant go down these routes.... W/I is perfect for their needs..

Remember everyone has different needs and requirements. The W/I system is currently what Mark thinks he needs.. and he has shown it can work.

I know you have fitted a turbo kit... but yet you went half way on fitting piggyback units to control timing, etc... and suffered from Checklight errors...its great u r concerned about water running out.. but surely hiding the checklight is a bit of a risky thing to do too?

Mark also has knocklink.. lamba link.. boost gauges.. egt..etc... so he is monitoring things...

Obviously the Apexi kit was fine for your needs at that point.... like WI is fine for Mark.. but your needs changed... and no doubt Marks will.

Everyone has solutions to their own needs...

W/I can work.. if set up correctly. It has been proven that it can. But people who fit it without knowing about it, and whether they actually need it wont work, and then wacking up the boost wont like the results.

Anyways...

I'm off to break some computers.

J.

NITO
6th April 2000, 13:16
Guys,

Where is the best place to install the sender for an EGT gauge on the Impreza...and can I just clarify something which is more of a confession of my ignorance on the matter, but, The hotter the EGT the richer the engine is running or is it the hotter the EGT the leaner the engine is running.

Also, a UK Impreza running 1.15 held boost with a standard ecu...is this cause for concern regarding knocking!

The VF22 Turbo..is this the larger unit used in the Sti's, and is it better than the HKS roller bearing types etc....

Thanks guys,
NITO

shiv
6th April 2000, 16:01
Firefox-- We agree on a lot of thigns. However, I see no reason why "hiding behind a check engine light" is risky. In fact, if that check engine light wasn't triggered under boost, I would have gotten very nervous. As for piggy back computers, thankfully, they are things of the past. They were installed and tested only for magazine testing purposes. They did prove to be marginally suffient at low boost levels. But, as far as I am concerned, they don't belong on cars making twice the output as stock. As I'm sure you know, they simply cannot provide the necessary resolution and tunabilty. Nor do they make good tuning tools as they offer no type of datalogging/monitoring capabilities. For that reason, I am using a stand-alone Electromotive TEC-II programmable engine management system. Crank-triggered and laptop programmable. Tuned appropriately, it has allowed me to run right to my fuel limit-- this corresponds to 10psi of boost on a 9.7:1CR EJ25 running 80psi of peak fuel pressure.

May take on WI is simple: If this motor doesn't need water injection to run 10psi of extra boost, the low compression EJ22 certainly doesn't. It just needs a larger turbo and better intercooler. But I think we agree on that... sorta. Maybe... http://bbs.22b.com/ubb/wink.gif

What's the price of the Aquamist over there? Can't be that much cheaper than a turbo upgrade, can it?

shiv

Mike Rainbird
6th April 2000, 17:20
Dear All,
Water injection is good if done properly. If not done correctly it will lose power at best and kill the engine at the worse!

I like the way some of you are experimenting, but in my mind, the systems you are using are flawed. If you only use a pressure switch to control the water as only boost related, then some days when the atmosperic pressure is high / air particularly cold, then the charge temperature will not get hot enough to warrant the water being sprayed, but because it is on a pressure switch, it will spray anyway. On occasions like this, you will find that you will actually lose power, as the water will not evaporate (the process by which it extracts all the heat from the charge). Accordingly, what you reaaly want is to run the water injection controlled by the charge air temperature and not just the boost. Ahmed did some tests in the late eighties to establish the best temperature for the water to come in at. These figures are now accepted by all the rally teams that run water injection. The water is switched on at 42C and switched off at 38C as the optimum charge temp is 40C. If water is injected below this thresh-hold then it was found to lose power.....

Obviously you cannot just do this as temp related, as it will inject water into the engine when it is stationary because of the heat soak effect. Accordingly you need to connect this into the boost switch as well. That way the water would come in (in Marks case) if the boost was above 0.9 bar AND the charge temperature was above 42C - i.e. ONLY when it was needed.

If anyone is interested, we offer a kit which shows charge air temperature (and samples every 1 second) on a display that can be fitted to work in the manner as described above in conjunction with the ERL water injection. This is fitted for 75 and comes with a full warranty. That way, at least the water is working for you and not against you....
Mike

shiv
6th April 2000, 17:55
Does the Possumlink offer air intake temp compensation in relation to fuel and ignition timing)? The Miatalinks on this side of the ocean don't.

Shiv

iwatkins
6th April 2000, 20:29
Mike,

I doubt the intake charge temps ever get below 40 on my GT4 http://bbs.22b.com/ubb/wink.gif

Seriously though, Mike is dead right. The pressure switch method works, but could be better. But there again, so could everything else on a Toyota.

In the ideal world, the water injection would be used to make sure that the air entering the cylinders = 40 degrees, no matter what the situation is.

This can be done with the MoTeC and the other high-end ECUs.

Mike, I would be interested to hear/see more about this system you describe. Feel free to mail me stuff off-board.

Cheers

Ian

Bob Rawle
6th April 2000, 22:16
Shiv, only Jap spec cars are fitted with an IAT sensor (and not all of those) and yes, the v4 and v5 Possumlink ecu's do take notice of IAT. Miatalink is pretty old technology by comparison I think and not available for the later cars. Subaru seem to have relied on the knock sensor and the temperature sensor in the maf so the facillity was not there to be replicated in the case of European and Australian spec cars.

BTW I think we are in agreement over lambda temperature but ... voltage still goes down as temp increases (as the sensor is a galvanic cell then thats normal) so any lambda meter would indicate low, thats why those of us with the lambda in the headers have to be careful when mapping. BTW spec on wideband is no better than narrow when it comes to withstanding temperature, life span is shorter as well (500 hrs life typical)and most normal ecu's don't have the resolution to make use of it afaik. One way to use the lambda to map is by directly reading the output voltage using a suitably fast dvm or, better, scope, much easier.

Bob

shiv
6th April 2000, 23:47
Bob,
We have different experience regarding exhaust heat and its effect on 02 sensor voltage. I'm drawing my conclusions from watching 700hp turbo rotary engines detonate and fail from lean-run conditions when the 02 sensor read full rich. Their EGTs are typically in the 1700-2000F range, making tuning a tweak-and-pray job for people who like tearing them apart and putting them back together. I've monitored other strange "lean-run" situations in my 350hp turbo Miata which I've traced by to ignition misfire from running too much fuel, sending unburned o2 and fuel down the downpipe which my o2 sensor would pick up and scream "lean!" at me. Problem was solved by leaning out the mixture.

True, the wide band is subject to heat-induced misreading just as much as the norrow band unit. No one said that it wasn't. Instead, its advantage is being able to accurately read on either side of stoich-- something that the stock 02 sensor simply cannot due accurately enough. Also, one does not use the ecu to read its signal. Tuners use wide band units as a temporary tool during the tuning process. The 02 sensor displays its ouput on its own monitor, which you keep on the passenger seat during your runs. Unfortunately, the cost (as much as a MoTeC computer) and limited lifespan keeps wide bands out of the reach of most enthusiasts. Then again, if on is hell bent on making 150+hp/liter, it's a tool that should be taken advantage of. Most good dyno tuning facilities will have one on hand and be able to assist in the tuning process.

As for air intake compensation, may I suggest a programmable ECU that not only adjusts fuel delivery but also ignition timing? The system I've put together constantly monitors manifold temp (just after intercooler), adjusting spark and fuel accordingly. Coldest night, hottest day... doesn't matter. The engine is always 3-5 degrees away from the knock threshold. The programmable knock sensor rarely intervenes-- only when I get a bad tank of gas. But even then, only barely.

Mike Rainbird
7th April 2000, 07:56
Ian,
So how do you know what the intake charge temperature is? Have you got a guage fitted? If not WHY not! http://bbs.22b.com/ubb/wink.gif If you have spent that much time / effort on setting up your engine, surely you would want to know the exact effect the water is having on the charge air?

Ahmed has an old adage that he has instilled in us all: "TEST don't guess....!".

I will bring the gauge with me to Santa Pod on 16th April and you can see if it is something you would want in your car.

As just a charge temp monitor it is 50 fitted, or with the extra control over the water injection, an additional 25 (fitted). Fully back lit LCD gauge.
Mike

Bob Rawle
7th April 2000, 07:59
We have had similar experiences then, excess "media" in the exhaust gas wether fuel or water residues has the same effects. The temperature effect on a Lambda sensor is a physical (and chemical)fact, given the sensor only reports what it sees then its clear that the composition of the exhaust gas in "excess" terms is going to affect the outcome. I guess you monitor the egt to keep an eye on this. The temperatures you mention kill the sensor, even readings in the circa 850 deg C area is going to affect the outcome. Having used wide band lambda then its "easier" to pick the point, the trick with the narrow band is to "callibrate" it against temperature. I did this by connecting a separate dc power supply to the heaters and then winding the voltage up to increase the temperature of the sensor, very interesting, the results showed good stability in the 500 to 850 deg C range but major differences outside that. It was not a scientific experiment but good enough. Since the sensor produces an output based on exhaust gas composition and conditions then away from stoichmetric the voltage has to be what it is, its a matter of reading that voltage. Given unlimited funding, then the Motec or similar has to be the best hardware option. Other ecu's offer much though and, as long as the functionality is understood, can provide superior performance (to standard), improved fuel consumption and much better driveability. As there are far more Subaru Turbo models in the UK then a far wider spectrum of people want to "upgrade" and not all are just after "performance", although its good to have extra.

BTW could discuss this for ever, if you are happy to keep this on the bbs then its going to help the uninitiated understand some of the issues I think.

Bob

shiv
7th April 2000, 09:46
Bob--- Yes, this is good tech info. And yes, I spend a good amount of effort watching EGTs. Especially in light of the meager amount of total ignition advance I have to run with such a high compression motor. Making matter worse, towards the limit of fuel supply, I can see EGTs rise above 850C (with the probe mounted just after the exhaust port on cyl #4). Especially at night when air density increases. Partly caused by reduced thermal efficiency (less advance), party caused by not enough fuel. It is here, as you have noted, that 02 sensor performance gets compromised. I'm not at all surprised with your temperature findings. It corresponds quite nicely with what I've seen first-hand-- even on non-rotaries.

Just curious... what are typical EGT readings (at cruise, WOT, etc,.) from a stock Impreza turbo. And where was the probe mounted?

-shiv

[This message has been edited by shiv (edited 07 April 2000).]

iwatkins
7th April 2000, 10:14
Mike,

Thanks for that, see you at Spod.

I don't have a temparature gauge fitted, that's why I'm interested.

I have however, 'borrowed' units before for setup/testing.

Very hard runs, post charge cooler temps of around 60 degrees C before water injection. I have yet to try the temps with a gauge since fitting the WI.

I basically want to spend some time instrumenting the car so I can see what is going on. According to ERL, I should see approx 15 - 20 degree drop with the WI with the size of jet/ water volume I am currently using.

I want to instrument before I actually start trying to set the WI up to a point where I can basically set the charge temp. And that is why I am so interested in your product.

We will talk next Sunday.....

Cheers

Ian

Firefox
7th April 2000, 19:44
Howdy..

I'm not sure about temps at cruise... my gauge starts at 450.. and at cruise its less than this.

At WOT I would suggest about 850-900 tops... The VF turbos maximum operating temp is 950.. If you are getting this high.... then you need to rethink your options...

J.

shiv
8th April 2000, 06:09
<BLOCKQUOTE>quote:</font><HR>Originally posted by Firefox:
Howdy..

I'm not sure about temps at cruise... my gauge starts at 450.. and at cruise its less than this.

At WOT I would suggest about 850-900 tops... The VF turbos maximum operating temp is 950.. If you are getting this high.... then you need to rethink your options...

J.<HR></BLOCKQUOTE>

Firefox--

Your readings are interesting. The n/a MY99 2.5RS cruises at ~750C. And this is in stock, non-turbo, from-the-factory form. Wierd. Are you sure that you see less than 450C during cruise? Where are you reading from? Perhaps the US cars are tuned to run leaner during cruise than the turbo cars. FWIW, my turbo 2.5 runs as high as 850C under full boost at redline.
shiv

Mark.A
8th April 2000, 10:06
Shiv,

One issue I do have at the moment, is the position of the EGT sensor. I was badly advised to locate it just "after" the turbo, next to the lambda sensor.

This will be addressed when the VF22 turbo is fitted, but do you have any knowledge of the difference between "before", and "after" EGT's.

My car cruises at between 640 &gt; 700c (same as stock) and max EGT is about 810/820c.

Mark.

marcus
8th April 2000, 18:42
Okay let me see what you guys think of my situation. I've got an Evo I with an exhaust and boost controller. I'm running 16psi now which by general consensus out here is a safe upper limit for boost. This is on 90 Octane gas. At this level my a/f ratio meter shows full rich up to redline
(more than 0.9v), but I don't have an EGT gauge to know the exhaust temps I'm reaching so I don't really know how close I am to getting in trouble. This will be my next purchase. Now I want more power, and assuming I am right on the limit for definition, I was considering water injection to allow greater boost to achieve it. It seems like the easiest way to me, even though it does have a "cave man" image and I don't like the idea that the water could run out on me.I am shying away from a programmable ECU because of the expense and the possibility that I will not be able to programme it to even run as well as the standard ECU because of inexperience. What recomendations can you guys give me? Maybe I should have posted this as a seperate post.

Firefox
8th April 2000, 18:42
Oops..

My mistake.... only a slight one... lol

Its been so long since I've been in the damn car... lol

Its idle that is lower than 450.

Mark.. I thought we established there was about a 50 degree difference ?

Shiv.. my EGT is just before the turbo.. in the manifold.

Cheers,

J.

Anders
8th April 2000, 22:03
How hot do exhaust gasses have to be to turn the manifold violet?

shiv
9th April 2000, 04:21
Howdy Mark,

I've seen temp. differentials of 50-100C on either side of the turbo. Post-turbo placement, as one can surely imagine, often buffers/blurs/obsures any transient spikes in EGT. But I have heard arguments against installing the probe before the turbo, suggesting that, if it gets heat damaged, it may fracture and demage the turbine wheel. However, if temps are hot enough to melt the probe, it's easy to argue that there are other, far bigger, problems to worry about and address.

I wish I was better at C -&gt; F conversions... http://bbs.22b.com/ubb/smile.gif

shiv

Richard.L
12th April 2000, 07:20
Lambda sensor questions, has anyone used the Bosch wide band lambda probe? If so, where can I get more information on it - Motec has been using it on their digital A/F meter.

Richard.L

pat
13th April 2000, 11:51
Richard,

I'm still trying to source the Bosch wide band lambda sensor. There's a group of units, belonging to the LSM11 "family" which apparently are heated, and produce a similar output to narrow band units (ie voltage rather than current).

Will let you know how I get on....

Cheers,

Pat.

Richard.L
20th April 2000, 19:08
Thank you Pat, I look forward to hearing from you soon.

Richard Lamb

Richard.L
20th April 2000, 19:19
Question for Shiv,

Just a repeat question posed by Mark:
(posted 05 April 2000 10:25) *** **** ** **
---------------------------------------------Shiv,
So is that why the top rally teams use it, because they are unable to map the cars correctly ????.
---------------------------------------------

I can not make sense of the answers you gave to Mark A. Could you please explain the reasons to me again? are you saying water injection is only good for racing cars and not for road cars?

Richard Lamb

Firefox
20th April 2000, 21:44
Hi Richard...

Welcome aboard.. http://bbs.22b.com/ubb/smile.gif

J. (ERL system installed and has been for quite a while)

iwatkins
20th April 2000, 23:00
Hey, it's Richard Lamb.

Welcome my man. You will be glad to hear that I have had no further problems after fitting the remote mount kit.

Before fitting, pressure switch drift = about 3 psi (up) for every 500 miles. After fitting, pressure switch drift = 0 psi. Top product.

Will call you next week with regards to the tank relocation (to the boot), need to pick your brains.

Cheers

Ian

[This message has been edited by iwatkins (edited 21 April 2000).]

pat
21st April 2000, 19:25
Richard,

unfortunately no news on the Lambda sensor front (yet). The conversation with Bosch's UK base went something like this:

"I'de like to order an LSM11 Lambda sensor, please"
"There's more than one, do you know the part number?"
"No, I just know that I need an LSM11"
"There are many with different lead lengths, outputs etc we would need the part number"
"OK, do you have datasheets, so I can find one which best fits the application, then we'll use that part number?"
"NO."

Very helpful bunch :-( I guess that I'll have to call the German office to see if they have any specs..... the quest goes on....

BTW, the MF-2 uses a PWM period of 75ms when running WI [according to the manual], but of course the HSV will run faster. Is it possible to select a shorter PWM period to use in conjunction with the HSV without getting out the soldering iron?

Cheers,

Pat.

Richard.L
23rd April 2000, 17:57
Pat,

Bosch is getting so big now and the head size of the people working there has grown in similar proportion.

I have contacted them many times and failed to make sense of a simple comversion.

MF2... If the "F" is link is the same time as well as the the "W", the speed of the pulses will be at around 80Hz. or 12.5ms. Lert me know if you wanr me to send you a jumper link.

Richard Lamb

Richard.L
23rd April 2000, 18:12
Ian,

Great news, all future WI system will be supplied with a remote mounting bracket for the pressure switch.

I have got some useless information if anyone is interested, how much water is contained in 1 Kg of air at 25C on a rainly day? A. approx. 33mls.
A typical 2 litre turbo engine consumes over 200cc of water per minute at 6000rpm on a rainly day at 1 bar boost. Just a little below what you inject on the Aquamist system.

I hope the people that doesn't rate WI should not take their car out on a wet day...

Richard Lamb

pat
23rd April 2000, 19:04
Richard,

thanks for the info on the jumpers. Had a feeling that the "F" jumper was used to control the PWM period, but I didn't want to chance it, or reverse engineer the board to be sure :)

With regard to the jumper.... well... if I can't find one lying around here I should be shot! One of the "perks" of being an IT Contractor... loadsa (useless? :) stuff lying around :)

I shall go and fit the jumper posthaste... the way that the water is delivered at the mo has me a bit worried about the "off" period, it can be quite long, perhaps long enough to allow the charge to become less humid than I might otherwise like! [fx: thinks about transfer functions of PWM signals through low pass filters... the shorter the period to smoother it gets :)].

Cheers,

Pat.

Firefox
23rd April 2000, 22:20
sad....oh so sad.... http://bbs.22b.com/ubb/wink.gif

J.

Richard.L
24th April 2000, 09:35
<BLOCKQUOTE>quote:</font><HR>Originally posted by Firefox:
sad....oh so sad.... http://bbs.22b.com/ubb/wink.gif

J.

<HR></BLOCKQUOTE>

Hello Firefox, have I upset you in anyway?

Richard Lamb

Firefox
24th April 2000, 10:13
Hi Richard...

No, not at all.. http://bbs.22b.com/ubb/smile.gif

I was joking about Pat... and his "approach" to car tuning and modifiying... lol

J.

pat
24th April 2000, 18:00
J,

One of the side-effects of being an electronic engineer, with a longstanding interest in switched mode power conversion; you kinda get to "know" about PWM stuff, as this is one of the most fundamental concepts in high efficiency power conversion. The actual converter topology is obviously highly dependant on application and/or budget but in the end most of them boil down to PWM....

Anyway, the point was that if the water injection control solenoid is run at a low frequency, then there will be large "chunks" of charge with water in them, and "chunks" without... if the frequency is increased, then these "chunks" become smaller. Ideally, the "chunks" should be at most half the "size" of one cylinder-filling; that way you will get at least one "chunk" of water-treated charge into each cylinder per revolution [depending on relative timing]. My "worry" was that at 7200RPM each clinder gets filled 60 times per second, or every 16ms. With a maximum pulse width of 75ms (as indicated in the MF-2 controller manual) and running, for the sake of simplicity, at 50% duty cycle, the spray would be on for 37.5ms [just over 2 cylinders' worth] and off for the same time [again, just over 2 cylinders' worth]. Ingoring the fact that it takes finite time for the water to start flowing, and stop flowing [we'll assume it is symmetrical though, ie takes no longer to stop than start] and the fact that there is a finite amount of air between the water spray nozzle and the inlet ports, it would not be beyond the bounds of possibility that two out of four cylinders would not be receiving any water [again, ignoring laminar flow which would "smooth" this phenomenon].

Compare this to the current setup... the max pulse width is now, IIRC, 12.5ms, which isn't quite half of 16ms, but getting closer. Assume again that we're running at 50% duty cycle, then the water will be on for 6.75ms and off for the same period. Best case scenario, you have 9.25ms worth of water per cylinder, worst case you have 6.75ms worth per cylinder. No cylinder can run "dry", but there is a discrepancy of about 37% between the wettest and driest cylinders [again ignoring laminar flow etc].

Just to prove the point, let's go for 8ms [ie the target mentioned above]. Again, 50% duty cycle, so on 4ms off 4ms. Regardless of the relative timing, each cylinder will receive exactly the same amount of water, 8ms worth [this does only hold up at 7200RPM, but it's pretty close even below... think about sampling theorem and bandwidth....].

There are other considerations to take into account [like laminar flow introducing a time differential effectively "smoothing out" the water charge over time] but the closer I can get it without any "real world tricks" the better it should behave even when subjected to weird phenomena [like inlet tract resonance which can play havoc with inlet charge velocity... and hence water / air homogenisation].

The HSV is rated to "250Hz", I'm guessing that this means 250 cycles [open-close] per second, which would make the solenoid reaction time 2ms [if open and close are symmetrical, which they may not be... close may take longer than open]. This is still a LOT shorter than my proposed operating point of 8ms PWM period, so should allow good flow control, while evenly distributing water between cylinders :)

The next step, then, is to get the MF-2 PWM period down to 8ms... I fear that this may involve some surface-mount rework.... Richard, what say you?

Cheers,

(Sad) Pat :)

Richard.L
24th April 2000, 21:51
<BLOCKQUOTE>quote:</font><HR>Originally posted by pat:
J,


The next step, then, is to get the MF-2 PWM period down to 8ms... I fear that this may involve some surface-mount rework.... Richard, what say you?

<HR></BLOCKQUOTE>

Hello Pat,

Unless you are using a HSV, or are you? please remind me what you got.

If you reduct the cycling time to 8ms, you can only control between 2ms to 6ms or 25% to 75% water flow rate. The valve opens at approx. 1.8ms and close at about 1.2ms (pressure assisted, high back pressure at the jet will slow this down, by approximately 0.2-0.3ms.
If you want to reduce the pulse time down further, lower the resistor situated left the letter "F" (200K).

RIchard Lamb

pat
25th April 2000, 18:13
Richard,

I am indeed using the HSV, as this came as part of the System 2c kit. At the time I wasn't sure how I was going to control my water setup, so I opted for the controller-less system; later I got the MF-2 to map the water..... I would still like to get the Possum to control the HSV, but in the meantime, I'm using the MF-2.

Seems that my thoughts on open and close times being different were correct... but back to front :) With regard to only going up to 75% flow due to the short PWM period, I would have thought that it would be possible to achieve 100%, but control between 75% and 100% wouldn't be linear, as would control between 0 and 25% [the valve would start to open, but not fully open before the pulse was removed again, IYSWIM].

Will inspect the resisitor later tonight. Lowering the value to 128k would get the value close. Guess the next closest would be 120k....

Cheers,

Pat.

Nova Turbo
28th April 2000, 16:16
Whats all this water injection, my mate, Dyna, owns a Nova 1.2 Merit Saloon (quite rare), is it possible to fit WI to his set up, he already has a peco backbox, synthetic oil and aftermarket spark plugs and leads. He has also fitted a spoiler off an Essy cossie, which looks pretty rad. He has a set of 13" 3 spokes which look like 17's. He has twin headlight stickers, and splitters.
An ALBA stereo has also been fitted.
THe best bit is that he has a Turbo 16V sticker on the back of it, which tends to scare a lot of scooby owners. CAn anyone point him in the direction of someone who can fit a turbo and WI to his already rapid engine....
P.s I have just fitted Corbeau bucket seats to keep me tucked in when i fly round corners at 60mph in 1st. Coooooooooooooooooool

Anders
28th April 2000, 17:09
Sadly despite pleanty of advice mr Nova and proud of it, has been banned for offensive irrelevant posting.

AndyMc
28th April 2000, 18:44
Mr Nova

0-60 in 3.9 secs
60 mph in first gear
Very impressive!
Are you sure your not getting your car mixed up with a Caterham R500?

[This message has been edited by AndyMc (edited 28 April 2000).]

Richard.L
1st May 2000, 12:48
To whom it may concern ....

Due to so much speculations and hearsays surrounding the water injection myth, we have decided to post some mathematical facts onto a temporary section of the Aquamist site for those who are interested.
Here is the url:
www.aquamist.co.uk/rescr/rescr.html (http://www.aquamist.co.uk/rescr/rescr.html)

Only the "FAQ" section is active.

Email me if you have found any mistakes.

Richard Lamb

Richard.L
1st May 2000, 12:59
Forgot to mention only the topic under "Common scepticisms" in FAQs (far left) is relenvent to the current WI discussions.

Richard Lamb

glenns
3rd May 2000, 09:38
Just a note on something that was mentioned earlier in the topic...The Mounting of the EGT sensor...

To get an accurate reading of the EG then the probe MUST be placed infront of the turbo..The reason is that the turbo utilises both the kinetic and heat energy in the exhaust charge so the post turbo temps will not be an accutrate relection of what is going on in the engine...What was said earlier was right, if you can melt a temp probe designed to measure temps exceeding 1000 degrees celcius then you have bigger problems than destroying a turbo....

Bob Rawle
3rd May 2000, 10:59
Had my egt probe after the turbo, then moved it to the intersect point at the bottom of the up pipe in the headers. Difference recorded between 100 and 150 deg C depending on situation. Gauge response is also much faster here as well. As I have an early Jap spec car then this was where my lambda was installed. It had been seeing temps of up to 900 deg C. I installed a new one after the turbo.

Bob

glenns
3rd May 2000, 11:30
The 100-150 degrees that is seen pre turbo is energy that the turbo uses to spool up and is a much more accurate reading, as shown by the increased response time....

Bob Rawle
3rd May 2000, 12:39
Not all of this is used, the temp drop through the turbo, one side to the other, is closer to 50 deg C. The rest is lost in radiation, conduction etc. The hotter the temp the faster heat is lost.

Bob

shiv
7th May 2000, 18:39
Richard.L asked:

"I can not make sense of the answers you gave to Mark A. Could you please explain the reasons to me again? are you saying water injection is only good for racing cars and not for road cars?"

Sorry, I just saw your post-- I've been off list for a while.

No, I said the driving conditions experienced by a rally engine are drastically different than that of a street motor. For one, these motors are typically inlet restricted. This means that the turbos operate at a much higher pressure ratio. In order to get good turbo response (and 400+ ft-lbs of torque they need in the midrange), they cannot use massive turbos as one would expect given the flow and pressure ratio requirements. Moderatedly sized turbos operating at massive pressure ratios make a ton of heat. Add in the fact that vehicle speeds are often rather low and that cars often travel sideways as often as they do forward , one could argue that intercooler efficiency is often marginal at times. Of course, having dirt and mud caked and splattered over the core doesn't help charge cooling either. Hence, the need for another charge cooling device such as water injection.

As nicely configured road-going turbo system, by comparison, operates with an (hopefully) unrestricted inlet, appropriately-sized turbo, and intercoolers with cooling efficiencys of 80% or higher. All of these go very very far to reduce the messy need for squirting water into the combustion chamber. I still believe that with good EFI tuning/turbo sizing/IC matching, WI is simply not necessary (or desired) for those low compression motors you guys get on that side of the pond. Currently, I'm running 12psi with a T3/T04 turbo and a top-mount Spearco core on a 9.7:1 CR EJ25 engine. At full boost, inlet temps are just 5-12 deg. C above ambient. Ignition timing at the peak torque point is still above 14 deg. BTDC. EGTs stay below 1520C. No water injection. No detonation on pump gas. Full boost by 3500rpm. Over 300hp by 6000rpm. If this system doesn't need water injection, I would argue that a properly configured EJ20-based system doesn't either.

But if we really want to talk about race motors, let's fondly recall those 1300-1400hp 1.4L Formula One engines. Or how about all those 8 and 9 second, 600+hp 1.8 liter Honda drag motors over here. Of what about those 500+hp Porsche turbo GT endurance race cars. Or even about SAAB's up-and-coming variable compression 240hp 1.6 liter supercharged @ 24psi street car motor. The list goes on. None of them use water injection to reliably achieve those specific outputs.

Does water injection work in reducing knock? Sure, it does. But as I've said before, there are other better, far more effective ways to do this before resporting to a band-aids such as WI.

Just my opinions, YMMV.

shiv

Firefox
7th May 2000, 21:10
This should be interesting..

Wasnt Saab one of the first to use WI ?

Anyways.. I await Richards answer http://bbs.22b.com/ubb/smile.gif

J.

Mark.A
7th May 2000, 21:50
Shiv,

Glad to see you're back.

I think cost should be brought into the equation. You have undertaken a big project, and have the luxury of choosing the right components to achieve your goal. We on the other hand, are trying to improve on what we already have.

The Aquamist WI system starts at 300. How much have you spent on your turbo, IC, matched headers, etc', etc'.

If I had the resources of a top race team, where an engine only has to last one race, my approach may be different.

As for the drag Honda's, etc', how many of them are using NO, and as for endurance Porsches, I'll let Richard comment on them !!.

Taken in context, you'd find it hard to get equivalent results for circa 300.

Mark.

shiv
7th May 2000, 22:18
Hi Mark,
In some cases, the stress of one single race is more stress than our wimpy stress engines will ever see during the course of their entire life times. This is especially the case with high output endurance race motors. I would also argue that drag motors are very stress-resistant as well-- just in a very different way. And while many of these motors do use NO, they are often used as a fill gap until boost pressures are acheived.

As for engine upgrades (WI, IC, turbo, etc,.), I agree that there is a cost issue involved. But my only argument has only been that WI is a band-aid, not a solution.

shiv

iwatkins
7th May 2000, 23:00
Shiv,

Buying a new turbo and IC is a lot of money. If I had it I would fit it. (oh yes I would http://bbs.22b.com/ubb/smile.gif)

WI performs as advertised and doesn't cost anywhere as near as much.

As far as I am concerned, it *is* the solution.

Non-confrontational cheers

Ian

[This message has been edited by iwatkins (edited 08 May 2000).]

MorayMackenzie
8th May 2000, 12:20
Ian,

To be completely honest... if you had the money for turbo+intercooler upgrades, wouldn't you save it towards the price of your Evo VI? http://bbs.22b.com/ubb/smile.gif

Moray

Mike Rainbird
8th May 2000, 12:24
Dear Shiv,
I am not a Scooby owner, but a Cossie one and therefore do not have to suffer the limitations of the intercooler location or the boost I can run (as the engine was designed to run big power in the first place).

Accordingly I run a peak of 35psi boost at 3500, backing off to 25psi at 6500 revs, a TO3/4 turbo (0.63 a/r), a 3ft x 4ft front mounted intercooler and have 419bhp at 6500 revs and 401 lb ft of torque at 3500 revs on normal 97 RON pump fuel (bearing in mind that this is only a 2.0 litre engine), using four Bosch 400 injectors (497cm @ 3.5 bar fuel pressure). 7.7:1 c/r medium ported head with longer duration cams.

However, I HAVE to run water injection to keep the charge air temp down. Without it the charge air goes above 60C at which point the ECU backs the ignition timing off and it dets at high revs. With it in operation the water comes in if the charge air exceeds 42C and the boost is above 7psi and switches off again if the charge air drops below 38C or the boost drops below 7psi. This keeps it at 40-45C (weather dependant) at what-ever the revs. Without it the engine would probably overheat at high revs causing head and piston damage.

My question is, if water injection is not the solution, how would YOU stop this from happening (without losing any power OR flexibility). I would advise that this engine was live mapped on a proper engine dyno (as in test bed) and not in the car, so the fuelling and ignition cannot be improved upon, as they are spot on.

Awaiting your repsonse with interest (but please, no compromises!).
Best regards
Mike Rainbird

[This message has been edited by Mike Rainbird (edited 24 May 2000).]

iwatkins
8th May 2000, 13:44
Moray,
You are as always, dead right http://bbs.22b.com/ubb/wink.gif How you getting on with your Mitsubishi ? http://bbs.22b.com/ubb/wink.gif

Mike,
There seems to be a few people recently whose posts seem to indicate that they are actually aiming to get charge temps at or around 40C. Why 40 ? Shouldn't the aim to be to get the charge temp as low as possible ?

Looking forward to Shiv's response to Mike's post. http://bbs.22b.com/ubb/wink.gif

Cheers

Ian

MorayMackenzie
8th May 2000, 14:21
Ian,

It's slow and dulled down... but what's new there! http://bbs.22b.com/ubb/wink.gif

Moray

shiv
8th May 2000, 18:33
Mike,

Four 497cc/min injectors, running 3.5 bar of fuel pressure, cannot supply the fuel necessary to safely support those hp numbers. Assuming you aren't using a rising rate fuel regulator, those injectors should only be used for up to 260-300hp applications. Sounds like you're running lean.

Also, have you looked at the compressor map for the T3/T4 t.63 turbo? Can it efficiently support the airflow and pressure you are asking it to? If not, that would also explain your intake temps. If your boost is falling off in the top end, this too would suggest that the turbo is undersized. Still, 35psi is a lot of boost to run on a street motor sipping pump gas. This may be an application for real race gas if all else fails. No compromise, right?

How did you fit a big 3'x4' front mount IC in there?

Shiv

Richard.L
8th May 2000, 21:49
To Shiv,

Thank you for your answers, I need time to digest them all.

Just a quick observation, your tuning method as I see it - you are using a turbo as a band-aid for a poorly aspirated engine, and adding intercooler (band-aid again) to compensate a less than 70% efficient air pump. Lastly how do to you evolve from your setup to get more power without resorting to bigger intercoolers, bigger turbos and bigger...

It has been so quite in your absence, welome back.

Richard Lamb

Richard.L
8th May 2000, 21:52
excuse my spelling mistake... quite (quiet), may be more to come soon.

Richard Lamb

Firefox
8th May 2000, 23:09
lol..

Richard... http://bbs.22b.com/ubb/wink.gif

Oh Mike.... I told you, you were only running 300bhp http://bbs.22b.com/ubb/wink.gif

lol

J.

Mark.A
8th May 2000, 23:35
I want to see a conversation between Shiv, and Ahmed. Now that would be interesting.

Richards comments about "bandaids", just goes to show how views, and oppinions vary between tuners. It all becomes very subjective.

I'm sure both routes will give simular results, just at very different costs.

Mark.

shiv
9th May 2000, 01:27
<BLOCKQUOTE>quote:</font><HR>Originally posted by Richard.L:
Lastly how do to you evolve from your setup to get more power without resorting to bigger intercoolers, bigger turbos and bigger... <HR></BLOCKQUOTE>

Simple, Richard. I don't. I believe a tuner should build a system with a set of goals in mind. These goals will determine what components need to be used. Once the system is assembled, it is slowly tuned and tuned until the goals are safely acheived. Once this happens, the wrenching and keyboard hacking ends and the driving fun begins. The key is knowing when to stop tweaking. Those who don't know when to stop usually end up with either detonation and holed pistons or water injection.

Mark: I don't believe for a minute that these two different approaches yeild similar results. One is a fully-contained, inherently safe, simple, adaptable and powerful system. The other is an inherently unsafe, less powerful system which requires a depletable aux. system to keep it from catastrophically failing. The only reason, IMHO, that water injection has become popular for comsumer use is often used to hide the effects of penny-pinching and lackluster tuning. But if one has enough $$ to buy a 22B, WRX or EVO, I would expect that they would want to do things the right way. There's a lot to lose otherwise.

I'm sorry if my response to Mike R.'s horsepower claims was blunt. I didn't mean to be rude. It is entirely possible to make more horsepower than simple injector math would dictate. But injectors, run static, often behave stangely and tend to fail prematurely at the worse possible time. With more information about the system, it should be possible to find the source of Mike's detonation problem. But again, 35psi is a lot of boost to run on any engine while operating on pump gas.

Shiv

PS. Who's Ahmed? I get the feeling I'm being humoured here...



[This message has been edited by shiv (edited 09 May 2000).]

Mike Rainbird
9th May 2000, 07:59
Shiv,
Ahmed is the guy who live mapped my engine. For your information the injectors supply more than enough fuel as was proven on the dyno when it was live mapped (I was there and watched it). The way he "tricks" the injectors into supplying the required fuel is to put multipliers in the chip which make them pulse more often between valve openings. The dyno operator sits there with a pair of det cans connected while Ahmed monitors and adjusts everything from ignition timing to the voltage supply to the injector (to confirm it is not running lean or something?) until the virge of det is detected at 500 rev increments, the fuelling / ignition / boost is then adjusted to remove the det and then he moves on to the next increment and does the same until he has a perfect fuel / ignition map for the engine. The good thing about an engine dyno is that it can reproduce identical loads again and again and again. Obviously this allows the engine to be taken to the very edge of every limitation and then backed off slightly to make it safe. However, it is so well set up (or too close to the edge probably in your opinion) that the reduction of octane from 98 RON to 97 RON in this country in October last year required a tweak in the map (more fuel) above 6500 revs to stop it from detting.

I will e-mail you some photos so that you can see where the front mounted intercooler goes. Ahmed maps engines for the likes of Ford (he did the Ford Racing Puma) and many race teams (in conjunction with Pectel the ECU people).

Just to prove that it has 400+bhp, the car does a standing quarter of 12.64s at 115mph (is rwd don't forget, with a stick shift, so is very hard tto get the power down, I am doing the first 60ft in 2.2s!) and weighs 1344Kg (half tank of fuel and driver) as well as being timed at 172mph on a runway (with 700 revs to go, we ran out of room - it was still accelerating!). The mid range torque causes it to break traction in the wet in fifth at 100+mph if you plant the throttle....

As you rightly point out, the turbo IS undersized for the engine if I wanted to go any further, as it is already suffering from "consumption" (as in the engine is consuming the air faster than the turbo can supply it at high revs. This is due to the head and the cams making it flow much more air). However, as I said to you I didn't want to compromise the flexibility of the engine. If I want ANY more power (I am on the very limit of the injectors and the turbo), then I would need a bigger turbo (T4) and higher flow injectors (or eight smaller ones). However, this would be a sacrifice in bottom end power, as the bigger turbo would take longer to spool up (peak torque would move up the rev range by 1000 rpm) and I have already said to you that your "fix" must NOT sacrifice the current drivability / flexibility. So, as I requested, without compromising the current power / flexibility, how do you propose that I cool the charge air temp other than with water? I forgot to say that flat out at 170 mph the EGT is 875C according the peak memory recall of the gauge.

Ian,
Ahmed did some tests in the late 80s and found that 40C is the optimum temperature for a nice clean combustion, the majority of rally teams now use these figures as well....
Best regards
Mike

iwatkins
9th May 2000, 08:48
Mike,
Thanks. I would have thought that the charge temp. should be as low as possible though. I.e. the colder it is, the more oxygen there is. Do you know the details of the cleaner burn at 40c or should I ask Ahmed ?

P.S. Is this going to be a record thread ? http://bbs.22b.com/ubb/wink.gif

Cheers

Ian

Richard.L
9th May 2000, 09:03
For Shiv,

Re: water injection (band-aid) for all...

Rally and Street Engine:
Surprisingly, there are quite a bit of similarity between the above.
First of all, they all travel sideways, intercooler efficiency (air speed) is hampered by either twisty rally courses or traffic congestion experienced by most street cars.
Both turbos runs at high pressure ratios, rally cars have no choice due to the inlet restictors; street cars just inherited small turbos form the manufacturers.
In both cases, water injection is a logical solution.

Formula One engines:
Ferrari used WI during the 80s on their 1.5 litre 126C2B and 126C3 engines, they won the constructors championship that season (1983).
Renault's Chief F1 research engineer Philippe Chasselut presented a press release on 1997, explaining the reason of why they used WI.
references: www.aquamist.co.uk/dc/coollinks3/index/race/ferrarif1/ferrari.html (http://www.aquamist.co.uk/dc/coollinks3/index/race/ferrarif1/ferrari.html) www.aquamist.co.uk/dc/coollinks3/index/race/renault/renault.html (http://www.aquamist.co.uk/dc/coollinks3/index/race/renault/renault.html)

To be fair and objective, Folmula One cars do not suffer from straight line speed, travelling sideways and mud splats (only when they got it wrong) , so why do they use WI ? Simple, they lacked of good engineers and could not find a better way tuning an engine by the ways you suggested. By the way, I got one of their intercoolers, not quite 3' x 4', but not too far off it (16" x 32" x 3.5" thick).

SAAB:
It would be interesting that you make a note of this, SAAB presented a press release: www.aquamist.co.uk/dc/coollinks3/index/rally/saab/press/press.html (http://www.aquamist.co.uk/dc/coollinks3/index/rally/saab/press/press.html)

The article highlighted the effectiveness of WI as well as replacing fuel dumping, a band-air for all those who can't manage or control their in-cylinder temperature surging during WOT condition.

GT endurance racing:
We have just supplied a such a car, twin WI pump etc. The reason they gave was quite logical, in the previous race in pouring rain, their lap time was faster than the their best on a dry day !

Drag racing:
Your compatriot, Adam Saruwatari's new creation Supercharged Honda-NSX (CR-10.5) has recently installed our WI system: featured on SCC- December 1999.

Back to the street cars:
On the other hand, the 80% intercooler you mentioned- is it 80% at all conditions? variables such as core restriction, vehicle speed, ammbient temperature, humidity.
I can named a few independent published articles on WI on road cars:

Nissan SE-R (USA): Mike Kojima for SCC http://www.se-r.net/about/200sx/scc/feb99/water_injection.html
Honda Civic Si (USA): ... made more power than any other modification to date!... by Dave Coleman for SCC http://www.gr8ride.com/articles/009271ah/009271ahb.pdf

I can go on an on, and on. put it this way, band-aid (medical) has been sold for many years and it is more than likely it is go ing to stay that way for many years to come.

Richard Lamb

PS. I will get round to the FUEL- specifics later (Cossie 400BHP?), my view is 2 litre of fuel per minure is plenty good enough for 400+ BHP (band-aid must be used- of course!).

Mike Rainbird
9th May 2000, 09:59
Ian,
You'll have to ask Ahmed or Richard the answer to that one....
LOL
Mike

iwatkins
9th May 2000, 12:26
Mike,

Cheers http://bbs.22b.com/ubb/wink.gif Don't get me wrong, I'm not saying 40C isn't the 'right' temperature, just would like to know why it is better than say 10C.

I'll give Ahmed a call .....

Any input on this Richard L ?

Cheers

Ian

shiv
9th May 2000, 17:20
For Richard,

Rally and Road engines:
I would argue that there is only one car on this list, from what I've seen, that runs pressure ratios similar to that of a rally car. And that is the Cossie. As for my EJ25 turbo, despite traffic and low speeds, I've measured intake temps to stay just around ambient with no form of IC or intake spraying. Although temps spike briefly during prolonged periods of idle, they cool down withing seconds of driving. Again, a function of appropriate IC and turbo choice.

Formula One engines:
From that same web site: "...Water injection was given up when the use of heavier elements in petrol offered a better route to increase power. " Likewise, I believe our puny low output motors, except for some extreme cases, can do just fine pump gas.

SAAB:
Sure, water injection allows a car to run leaner under WOT. I've never argued that point. If emissions is one's first priority, that's great. It's not mine. Nor should it be yours if safe power is the goal.

GT endurance racing:
So everyone is using WI? I know of at least three people who have tried it and found no benefits. Of course, these people have put together a properly designed system with no need of continuous and active knock supression.

Drag racing:
Adam Saruwatari's Supercharged Honda-NSX (CR-10.5) street car has installed WI system because he is NOT using an intercooler. He couldn't fit one in. The better questions is: Is his turbo'd and intercooled 30psi 600+hp 9 second drag RX-7 running WI? Or how about the other drag race boys who extract 500+hp from 1.8 liters? Not all of them use NO and none that I know use water injection.

Back to the street cars:
Nissan SE-R (USA): Mike Kojima for SCC

I had dinner with Mike last week. Mike's (or actually, Serle's car) was pinging without WI. That non-turbo SR20DE has a 9.5CR (!) at was turbocharged to over 1 bar of boost (!). Understandably, 92 octane fuel couldn't cut it. This is an example of insufficient octane and a situation where WI can work to raise the knock threshold. However, I don't think anyone would argue that running race gas would have been the ideal solution. I suspect that I would have to resort to similar means if I were to run my EJ25 at high boost levels (over 1 bar).

Honda Civic Si (USA):
It should be noted that the Project Civic had to be CARB approved. CARB, for those who don't know, is a nastly legislative body in the US set up to get rich while making life difficult for car modifiers. In California, only CARB approved modifications are legal. CARB approval has very little to do with actual emissions. This (and the 10k price cap established in the Civic Challenge competition) prevented the use of a CARB-illegal programmable EFI system or any additional ignition controller (J&S, ITC, etc,.). The car ran 8psi from a Roots blower with no intercooler. Not exactly a optimal system. The car also pinged itself to death during the Si Challenge. With Paul Tracey driving the time trials, the water tank ran dry. Blown headgasket.

"I can go on an on, and on. put it this way, band-aid (medical) has been sold for many years and it is more than likely it is go ing to stay that way for many years to come."
-- Yes, as long as people continue to run insufficient ICs, push their turbos
into surge, and ask for more octane than our gasoline can give them. It's a great band aid for those who insist on hurting themselves during the "tuning" process.

Shiv

Anders
9th May 2000, 19:51
This is my favourite thread of all time both from the old BBS and this new technical one!

Shiv:" USA top guy"!
Richard "mr Aqamist"!
Mike Rainbird(Ahmed):"Ultimate Cossie power"!

Fantastic! http://bbs.22b.com/ubb/biggrin.gif

Richard.L
9th May 2000, 20:42
For Shiv,

Rally and Road engines:
Are you saying that it is OK to use WI on the Cossie because it is running similar pressure ratios to thr WRCs? In that case, why can't anybody on the list evolve with WI ? Do you favour fuel dumping to WI ?


Formula One engines:
I will agree with you water injection have to give way to heavy element fuels, but note that the pump fuel is getting worse, not better. Until then, WI is always an option. I will welcome the 130 octane fuel, but someone will use WI to stretch the PR further, sky is the limit excepts people with fixed views.

SAAB:
Putting emission aside, SAAB has managed to reclaim all the power loss with WI at Stoichiometric. More importantly, the engine will pink itself to death at that ratio without WI. Again, it is not necessary to dump fuel, as you might have suggested that Mike's engine could not possibly produce the power level he claimed. What is your personal preferred AF ratio for Mike at WOT?

GT endurance racing:
Give it time- they will come to the WI way, just like the WRCs. It is still quite new (as far as we are concern) to the GT racing, except the three GT cars you mentioned, they achieved "zero-gain" with WI, how sad!, please let me know who there people are and I like to know how their expert "zero-gain" WI systems worked, may be I can cxhange their minds. I am sure that there are a lot more forward thinking people there that haven't tried. Based it on your method of tuning, each one of GT racers will turn out exactly the same power year after year, it will be more interesting watching the vintage car races.

Drag racing:
Adam Saruwatari used WI becuase he wanted to use it, not becuase he can't afford to fabricate an small and highly efficient intercooler.
In Drag racing, WI is still new... just be patient. Remember the 850BHP, "English Cossie Probe" ? it used CO2 gas to cool half of the intercooler core, 7.9sec. 1/4. Houses for courses. It doesn't use water injection nor did he use any one of your tuning methods, just a lot of good engining, brute pressure and RPM. Julian (engine builder) is only 25 minutes away from me. I believe the probe went to the US but didn't meet any challengers (rule: 4-cylinder engines only), so where were all the US drag boys? Shiv, I really like an update on it, although it has nothing to do with WI.

Back to the street cars:
I am glad that you have confirmed that the SE-R pinked like mad without WI, It has a big turbo and a big intercooler in front, isn't it right? So what has he done wrong? Mike Kojima, shared my WI view, although you and him contribute to the same Magazine SCC, you don't always have to agree.

Honda Civic Si (USA):
Another example of of what will happen without WI, blown head gasket. Si challenge with Roots blower, as you called it - less than optimal, but it exists. There are over 1,800 non-intercooled Eaton supercharged converted vehicles out there, with no chance of an intercooler, wouldn't you agree that WI is the best alternative?


Finally....
Like all new products, you need to give it time to mature, I can imagine what the first turbocharger might be like, highly unreliable piece of band-aid to compensate the low volumetric efficiency of the early engines. The new generation of Aquamist water injection is only the begining, it is normally for people to be highly sceptical. I am surprised at your reluctance to accept WI, unless there are "no other alternative" (is it a form of acceptance on you part? ;)

RL

Anders
9th May 2000, 21:01
Yes the first 100 poster!

I had anticipated water injection as part of my rebuild but I have run out of money! http://bbs.22b.com/ubb/frown.gif

I have a MRT/Scooby mania large top mounted inter-cooler, a BRD VF22 turbo (supplied within 24 hours! http://bbs.22b.com/ubb/cool.gif ) Scooby Sport equal length headers, coated by Firefox Technologies. Due a map by Sir Robert of Rawle with the link.

May be water next year but Bob reckons that I will have fun in the meantime http://bbs.22b.com/ubb/biggrin.gif

Richard.L
9th May 2000, 21:09
For Anders,

I have posted my last reply to a "new topic" headed by " same old story" folder by mistake, could you delete it for me.

I look forward to hear from you soon.

RL

Richard.L
9th May 2000, 21:17
Ian,

40C inlet air temperature is Ahmed's idea, I think it has something to do with the fuel droplet size ideally suited to the intake geometry of the Cossie engine.

Any chance of getting Ahmed to join in on the "fuel dumping" topic or "lack of it" on Mike's engine?

RL

shiv
9th May 2000, 21:48
<BLOCKQUOTE>quote:</font><HR>Originally posted by Mike Rainbird:

As you rightly point out, the turbo IS undersized for the engine if I wanted to go any further, as it is already suffering from "consumption" (as in the engine is consuming the air faster than the turbo can supply it at high revs. This is due to the head and the cams making it flow much more air). However, as I said to you I didn't want to compromise the flexibility of the engine. If I want ANY more power (I am on the very limit of the injectors and the turbo), then I would need a bigger turbo (T4) and higher flow injectors (or eight smaller ones). However, this would be a sacrifice in bottom end power, as the bigger turbo would take longer to spool up (peak torque would move up the rev range by 1000 rpm) and I have already said to you that your "fix" must NOT sacrifice the current drivability / flexibility. So, as I requested, without compromising the current power / flexibility, how do you propose that I cool the charge air temp other than with water?<HR></BLOCKQUOTE>

For one, you could upgrade to a ball bearing center section with a larger turbine a/r. You could also run a slightly larger compressor. Spool up should not be adversely affected. In fact, it may even be improved. The bigger turbine side will alleviate much of the exhaust back-pressure which is currently constricting your motor and leading to knock-inducing exhaust reversion and poor exhaust scavenging. The fact that boost is falling off by 10psi at the top end indicates that you're system is very far away from being what many tuners on this side of the ocean would consider well set up. That, and the fact that your injectors cannot provide enough fuel, is why you need water injection to keep the motor in one piece. I've seen knock thresholds move around by as much as 10 degrees of advance just by swapping turbos.

Shiv

[This message has been edited by shiv (edited 09 May 2000).]

Beef
9th May 2000, 22:23
Well, this lot is waaaay over my head, but I chuck in the fact that some of our Supra guys have now installed WI on their MKIV TT's, and those that have, have said very favourable things about it. One guy I know of has a gigantic FMIC, but still uses WI to prevent detting, and that's using 103RON fuel! Not sure what boost he is running, but I believe it is aroung 1.5bar. Something like that anyway.

Not sure what relevance this may have, but I thought I'd add it as possibly useful info for someone else.

Mike Rainbird
10th May 2000, 13:45
Shiv,
The fuelling IS absolutely spot on. The water injection only comes in on days where the ambient air temperature is high (or low barometric pressure days) totally controlled by the ECU. However, because of the huge front mounted intercooler, in the winter or on really high barometric pressure days it doesn't come in until well into three figure speeds in 4th and 5th gear and only above 5500 revs (I know when it is operating as I have a light that comes on in the dash). Now that it has been re-mapped at the top end, the onlt time it would det at high revs without the water injection is if held flat out for several miles. However, I would notice a reduction in power without the water injection, as the ECU would start to retard the ignition above 60C (which as you know makes matters worse) in an attempt to prevent any possible det. Having the water come in prevents the ECU ever having to retard the ignition due to high intake temperatures.

However, funny you should say that about the roller bearing turbo (combined with higer flow injectors), as I am currently looking down this route, but not because of the thermal properties, but for more gee-gees! http://bbs.22b.com/ubb/smile.gif The water injection WILL stay though, as the best way of keeping the charge air temp down to acceptable levels.

I would also imagine that the majority of your time is spent "cruising" at fairly low speeds where heat is not an issue, whereas we tend to have seen 150+mph seven or eight times per tank full of fuel....
Best regards
Mike

Richard,
I have e-mailed Ahmed this link and it will be up to him if he choses to take part... http://bbs.22b.com/ubb/biggrin.gif
LOL
Mike

Dennis
10th May 2000, 17:42
Hi All,

I'm from the GT-Four list, but couldn't resist throwing in a few points on this debate on water injection ;) A few here will know me.

1. WI is dangerous because if it fails you will destroy your engine. Hmmm....O.K., based on this view it would be time to remove lot's of parts. What happens if your fuel filter get's clogged? Can't use fuel filter, it could be dangerous. What happens if my fuel pump starts to give trouble? Can't use a fuel pump, it might be dangerous. What happens if one or more of my injectors get clogged, and flow decreases? Can't use them there injectors, it could be dangerous. Oops, time to design a completley fail safe fuel system...but how??? Oh no...what a minute..I just realized I gotta design a failsafe coolant system too. I mean what happens if the water pump fails? On second thought, that's what we have gauges and warning lights for. Although the ERL system can also use gauges and warning lights to keep us informed, just like a coolant temp gauge, a fuel pressure gauge, EGT gauge, etc.

Seriuosly, how can anyone who deals with modifying a vehicle, expect a totally failsafe enviroment. You obviously don't want to modify a street car to the ragged edge, but you also can not expect that things can't go wrong. If anyone can tell me that WI failing is the ONLY thing that can destroy an engine if it fails, then they shouldn't be dealing with cars.

2. WI is a band aid, and worst, a band aid that doesn't work. First of all, with no disrespect meant, there are band aids all over the place on performance engines. From compensating timing over compression or forced induction, to intercoolers, to fuel dumping compensation. It is a simple as picking the right band aid for you. As for WI not working, the only thing that needs to be said is that the WRC cars use it. Trust me, these people are NOT going to put any band aid on one of their vehicles that doesn't work. A comment was also made about them not being able to tune the vehicles enough. I somehow seriously doubt that a "works" team won't be able to tune a car properly. There was also a comment, which I was actually found funny, that a rally car is sideways a lot, and mud clogging up the intercooler, hinting at airflow over the intercoler being diminshed. Hmmm....well, there really ain't much mud in the tarmac events, and they aren't sideways that much, so that theory drifts out the window.

3. WI takes up some of the air space, and as such reduces power. If it were only that simple. It's all about density. If you take the same two volumes of air, one has a temp. of 30C, and the other has a temp of 55C, there will be more oxygen in the 30C air than in the 55C air. What does this mean? It means the volumetric effeciency of the engine is higher, and the higher the VE, the more power you can make. So you have a trade off again. You lose a little space, but gain density, but wait, there is more. Water has such a high latent heat that it will also reduce the in cylinder temp, and raise the knock limit. Hmmm...now you can get your timing advance closer the optimum point in areas where the engine is under high load.

4. Another point was made that the timing can be retarded by 4-5, instead of WI. Wow...that's a lot of reduction, especially on a forced induction engine. First you need to understand something about mapping timing. Timing should be advanced until the engine output no longer increases. However, you will find that on a forced induction engine, you will reach the point of knock before you reach the point of optimal timing advance, especially in areas where the engine will be under heavy load. So what you will be doing by retarding the ignition 4-5 is taking the timing even further away from the optimal advance. This will cost you power, and chances are a good set of power, just by that little bit of reduction. If you can retard by 4-5 and not loss any engine output, your timing maps weren't right, and went beyond the optimal point. This is easily potrayed by some modern forced induction vehicles, that advance timing until trace knock. These will slowly find the highest point (not beyond optimal), so that the car can produce more power if higher octane fuel is used, or if the car resides in a cold climate etc. So if you water inject a vehicle like this, it will take advantage of the raised knock limit.

In the end, we must always put things in perspective, and look at them from a logical, and open minded approach. The bottom line is that WI does work, and works well, when properly setup, and is another viable option to intake temp control, and raising the knock limit.

Dennis

shiv
10th May 2000, 18:21
Richard,

Give WI time to gain popularity? You've got to be kidding me. The idea of squirting water into the combustion chamber has been around for decades. Of course, Aquamist has only recently made it affordable and readily available to the public. But what about cost-no-object race teams? To this day, water injection is rarely (if ever) used in CART, IRL, Nascar, F1, etc,... That should speak loader than the results of isolated examples sprinkled here and there. The idea of running water injection on low compression, moderately boost turbo engines is just a silly solution to a silly problem caused by silly engineering.

Let's get something straight before we keep on going. I agree that running water injection is a better than running detonation. And in all instances were WI yeilded power, detonation was present without it. I also believe that for a fuel-limited, non-detonating system to gain hp through the use of WI, athorough ECU re-mapping is necessary. I also firmly stand by the belief that a well-fueled, properly-designed system will not benefit from water injection. I keep on hearing the same funny argument from WI proponents: "Not all systems are well-designed so WI is the only solution."

If that is how one wants to look at it, that's fine. I agree that all those people who poorly/marginal designed turbo systems should run water injection instead of blowing up their motors. I also believe that a tuner's job should be to make a poorly tuned system work well by fixing the underlying problem, not by covering it up with WI. Of course, the former solution requires more effort and money. It is, however, more effective.

Many reputable tuners find WI to be a crutch for an improperly fuel system. You refer to what we do as "fuel dumping" which, to me, is perhaps the greatest misnomer since "military intelligence." Not only does fuel absorb heat, it also burns and makes power. Give me fuel dumping over water squirting any day. All this, of course, assumes the presence of a sufficienct intercooler and turbocharger.

I made a couple phone calls today. One to Russ Collins, president of RC Engineering and another to Charles Cadle, Advanced Product Dev. Engineer at Honeywell (Garret turbo division). I spoke to Russ about WI. And I spoke to Charles about Mike's Cossie.

Russ' arguments against WI were as I would expect from an experienced fuel injection specialist. "It's a crutch." His reasonings were similar to mine (which we all know by now).

Charles at Garret (who, by the way, is also friends with Mike Kojima) offered his opinion about the Cossie's turbo choice. After I told him the turbo configuration and the boost/hp levels, he basically said "yikes... no wonder he's experiencing detonation problems." In other words, that turbo is being used *well* beyond its safe and efficient operation range. As I've said before, the engine is getting killed from the exhaust side NOT only the inlet side. In other words, it's not an issue of intake charge cooling-- it's a matter of cylinder charge cooled caused by exhaust reversion. I asked for his advice on turbo replacement given the caveat that hp or turbo response cannot be sacrificed. After some number crunching and back and forth Q&A, he recommended the following turbo:

HKS GT30/37 (built by Garret, proprietary to HKS)
56 trim compressor
84 trim turbine
.72 A/R

Far more modern than the venerable T3/T4 (which was designed in the 70s), the GT30/37 has a ball bearing center section,~20% less rotational intertia (better spool up), ~7% greater turbine efficiency and larger turbine housing (less back-pressure). It should, by numbers alone, yield a sizable power increases compared to turbo Mike is currently using in his Cossie. It will be able to provide the same airflow with less boost and far less exhaust backpressure. Less boost, more power. Great deal. The net result is safer engine operation and a far higher knock threshold. In fact, this upgrade alone should most definately eliminate the need for water injection at your current water-injected hp levels. Isn't that the goal? Charles also recommended extrude honing of the turbine housing and proper porting of the exhaust manifold (if not done already). There you have it. A quick spooling, safe running, 400+ hp system that doesn't require water squirting or excessive fuel "dumping."

Favourite quotes of the day:

"Water injection makes more horsepower than detonation... but it's still a crutch for an inadequate fuel system"-- R. Collins, RC Engineering

"That turbo [on the Cossie] is waaay overworked, poor motor."-- C. Cadle, Garret

"All things considered, you would be far ahead never to have heard of a water injector."-- Corky Bell, Bell Engineering

On another subject, Russ Collins brought to my attention that Mike's Cossie race motor was probably built to operate at a brake specific fuel consumption (BSFC) of around 0.45. Typical, OEM turbo motors best work with a BSFC of somewhere around 0.6, which means that for each hp it makes, it will consume 0.6lbs of fuel per hour.

Assuming a max duty cycle of 80%, four 500cc/min injectors (48gallons/hour), and a BSFC of .45, we get...

4x((48x0.80)/.45)= 341 horsepower.

Assume 100% (static) fuel injector duty cycle, we get
4x48x1/.45= 425 horsepower.

So, I stand corrected on that issue. It is possible to make 400hp with those injectors on that particular engine, especially if one runs batch. May not be ideal, but it is possible.

It's also interesting to see how different countries resond to different products. Oh well, this will be my last post on the subject. Cheers.

Shiv

Dennis
10th May 2000, 19:43
Shiv,

I never knew that F1, CART, etc. used pump fuel. That's amzing how they got those cars to run on pump fuel without WI. I mean for you to compare them, as an example of where WI is not used, surely they would use pump fuel. BTW, if you do some research, I am sure you will find that Formula One did use water injection. It started in 1982 when Ferrari used it on the 126 C 2 engine (just 1.5 liters and producing 650hp), soon followed by almost all other Formula One cars by 1983. To my knowledge it was later banned, and instead, one hell of a concoction of what is termed "witches brew" fuel was used.

Why is WI a silly solution to any forced induction vehicle? Let's throw a quote back your way. Quoted from "Automotive Fuel Injection Systems, A Technical Guide" by Haynes:

"Water possesses a high amount of latent heat, which makes it suitable for internal cooling of the combustion process.The water improves the engine's knock resistance and cools the exhaust gas, therby permitting higher boost pressures while lowering the thermal stress on the turbine section.

By slowing down the burn rate, water reduces the pressure rise rate and the maximum cycle temperature. That cut's down on NOx emissions. It cools the end gas, therby preventing it from reaching critically high temperatures and causing self ignition. Water and water-alcohal mixures have proved able to prevent, to a great extent, the build up of carbon deposits in an engine. They also protect the valves and valve seats from burning, pitting and distortion. Water injection may permit changes in spark timing to get higher effeciency."

So let's see what exactly is silly. A system that helps to protect valves and seats, helps prevent carbon build up (which can lead to hot spots, and knock), helps keep combustion chamber temps from getting too high, cuts down on NOx, and allows more power through increased boost and/or timing advance....yep..that is soo silly.

Now, I must also inform you that many reputable tuners, openly admit to the benifits of WI, on even highly tuned cars. They have found that there is a point where the only way to keep things working, without knock was WI, with the exception of using very high octane fuels. I am not even going to touch the comment on fuel dumping, as I believe most people know what happens to power output on too rich a mixture, especially if they have seen the results on a dyno.

Ah well...WI is a crutch...then what is an intercooler? An intercooler cools the charge air, WI cools the charge air. An intercooler, does not stop carbon build up (and fuel dumping will only help to promote it), it does not actually cool the in cylinder process. I wonder what does do all that? I also wonder why lot's of reputable engine builders use WI. Ever heard of Motec? I think they know a little about engines...well guess what the head guy at Motec U.K. has installed on cars. ERL water injection systems. Now, unlike you, I am not saying an intercooler is silly. By all means intercoolers work great, but then again, so does water injection.

On last thing to put in perspective. When one type of racing team uses a product, and another does not, it would be illogical to assume that the reason the team that didn't use it, was because it didn't work. It would be more rational, to determine that it does work, and that it either didn't suit the other type of race team, or rules didn't allow it, etc. In short, extremely high budget motorsport teams are NOT going to use something that does not benifit the performance and/or reliability of the car.

Dennis.

shiv
10th May 2000, 21:08
I guess this is something we're not going to see eye to eye on anytime soon. But I'm willing to keep an open mind. Richard L., I don't see your email address anywhere. Could you forward it to me?

Shiv

Richard.L
10th May 2000, 22:18
Shiv,

I agree to disagree, it has been an interesting discussion, we'd better stop before it gets to repetitive.

My email address is erl@aquamist.co.uk

I will probably meet you again on another list, care to join in on the syclone, Audi, few others list? one way to keep the WI on peoples mind ... just joking

Mike Rainbird
11th May 2000, 09:39
Dear All,
Please find below Ahmed's comments on the subject:
QUOTE
Hello Mike,

I would love to get involved in the debate, (and will do if you insist!) only that I've been there and forgotten more than most people know about the subject!

Suffice to say, that water injection as I developed it, (for charge air temperature control, as opposed to ERL's claims about detonation suppression etc...), is a form of band aid for high powered turbo engine's, but NOT because of poor tuning, it's a final Air Charge Temperature control safety buffer should the normal charge cooling mechanism not cope with the demand!

The problem is, that simply injecting water into an engine causes a loss of power - FACT!

The trade off that the engine tuner has to make is! What is the better compromise for that particular engine and it's operating environment!

In my case it was easy, the production Escort Cosworth compound water/air - air/air intercooler was not capable of keeping the ACT's in the range that produced best power and throttle response!

So I developed a water injection system and ECU control strategy that supplemented the production intercooler, and in this particular case the insertion losses were far outweighed by the power gain's.

E.G. In a case where an engine's technical specification demands a high static compression ratio for good throttle response, and the ensuing result is full load detonation, then the introduction of WI can be regarded as a 'band aid'!
This is not poor tuning, it is a considered option by the engine designer and has worked very well on engine's such as the old 'TAG McLaren Porsche' Formula 1 engine! Something which most people in the 'debate' are probably are not aware of!

Road cars; I personally would develop a road car engine to work WITHOUT WI! But I'd fit WI all the same, in this instance I do not regard the WI as a 'band aid', but as a mechanism to stabilise the ACT, which therefore stabilise's the power output from climatic temperature fluctuations.

The engine maintains good power* in all weather conditions, and the system is totally fail safe! (Assuming the thing is mapped properly!) If you lose the WI, you simply might lose some power depending on the climatic conditions, but you won't destroy the engine!

Bear in mind, that when the FIA introduced the WRC formula, (and thereby greater freedom of intercooler size and position) we at Ford, and everybody else (I think? including Subaru!) fitted the largest possible air-air intercooler at the front of the car!

If WI was a complete solution, do you think we would all have gone down this route?

Also, WI controlled by pressure alone is a flawed solution, it must have boost and ACT inputs to it's control algorithm, otherwise you can end up with misfires and other difficult to solve maladies.....

*Within the limit's that WI can help to stabilise the ACT and allowing for Barometric variations!

Hope this has been of some help to you...

Best regards,

Ahmed.
UNQUOTE

I hope Shiv reads this, as I have e-mailed Ahmed with his last comments as we have been attempting to source a suitable turbo for my requirements (thanks everso for taking the time to do this Shiv, I really appreciate it), and if there is any chance you could e-mail or fax me (01603 702 936) the flow characteristics for this turbo for Ahmed to examine, I would owe you a large drink if you were ever in Norfolk!
Best regards
Mike Rainbird
PS. Shiv, although the engine might be to "race spec" it is in a road car that does 12,000 miles a year, numerous track days and and drag events. It has four doors, five seats and you can see pictures of it on our website www.rsnorfolk.co.uk (http://www.rsnorfolk.co.uk)

[This message has been edited by Mike Rainbird (edited 11 May 2000).]

mike_nunan
11th May 2000, 12:58
Hi Shiv,

I'm a beginner in all this stuff, having read about half of one good introductory text on engine internals, but I've been following this discussion with interest as it seems to touch on a lot of important issues. I don't belong to either "camp" in the argument, but there are a couple of points that are nagging me and I'm wondering if you can settle or clarify them.

First let me check that I've got a handle on the basics. AIUI the achievable power output with any turbo motor is essentially limited by the amount of boost you can run before it starts to knock. Also, leaving aside knocking problems, then if we look at what happens when we vary the mixture we find that maximum torque is obtained a little way rich of stoch.

Starting from that point, lets say we maintain the same AFR and increase the boost until we reach the knock point. Now, we can add more fuel and the knock goes away, which appears to be a generally accepted solution that is used broadly by the major car manufacturers. However, your argument seems to imply that this is just as much a band-aid fix as using WI (leaving aside issues such as WI solenoid failure for the moment). Is that an accurate representation of your opinion or have I missed something? That's question (1), anyway.

Personally, I find the idea of using excess fuel to reduce inlet temperatures pretty graceless. Faced with a choice between injecting water (cheap, harmless) and fuel (expensive, potentially leads to damaging deposits and emissions) I'd tend to pick the water, but the confidence factor also plays a big part. When it comes down to it, neither solution seems particularly desirable.

The presence of water vapour *or* excess fuel in the charge will slow down the burn rate (correct?) which will increase the tendency to knock, working against our overall intention. For this reason, I can see that it's better to use different componentry (e.g., bigger turbo and more efficient intercooler) to reduce the charge temp and hit a certain target output.

However, if you're not thinking in terms of a particular target output, but simply want to get as much power as possible, you could always go further by overfuelling or using WI, right? That's question number (2).

As I say, I don't really have a strong opinion either way (except that I don't like wasting fuel at 87p a litre http://bbs.22b.com/ubb/wink.gif) but I would like to improve my understanding. Please excuse the ramble!

Best regards,

-= mike =-

pat
11th May 2000, 17:31
Richard, Mike....

a while back we were looking for affordable wide-band lambda sensors, and I was trying to get some info out of Bosch. Well, I've finally found the elusive LSM11 sensor :)

At the time, I wasn't sure which member of the LSM11 family it was, so I just bought it because it *was* part of the family. It's Bosch part number 0 258 104 002, listed as an "LSM11 OE Spec". Guess it's the cheap & nasty one :) But it does appear to be lead-tolerant and have a life of about 80,000 km on low-lead fuel. God only knows how long it will last on unleaded! They are available from Genesis Motorsport or MoTeC [it's the one MoTeC supply with their wide-band option]. They're happy up to 800 degrees C with the heater connected, or 930 with it disconnected. Oh, it also has two wires for the signal, so you can use a differential amplifier to get rid of nasty interference :)

There are another two members of this family which may be of interest... B 261 209 105 which is a "Standard LSM11" sensor, about twice the price of the one I got. Finally, there is the B 261 209 101 which is listed as a precision part, costs over five times as much as mine, but still only one sixth of the Bosch Lambda meter, the "LA3" which has part number F 00K 000 367 [yeah, oh F OOK! that's what your bank manager will say if you buy one! :)]

Trying to get vaguely back to being on-topic, and just to add to the confusion, I recall reading some time ago that WI will influence Lambda readings. I guess this is due to lowering of charge temperature and perhaps also altering the oxygen free radical content of the exhaust gasses. So what IS the best way to map? My gut feeling at the mo is to map fuel with WI off, then ignition with WI on.... hmmmm... any ideas?

Shiv,

With regard to WI's suitability, consider the engineering challenges faces by the car manufacturers... good fuel economy, low emissions, good bottem end, high power output. A few mutually exclusive things in there! The compromise most use is to keep the compression high-ish, and retard the ignition at higher boost to prevent the engine destroying itself. For not a lot of money you can fit WI, which will allow more advance and get you closer to MBT timing. Used within the manufacturer's design parameters, the turbo should not be operating outside its efficiency map, so a more suitable turbo isn't applicable. A bigger intercooler might be. But the standard one will keep the charge down to about 40-45 degrees so fitting a different intercooler won't help a lot. Given that a turbo is operating efficiently and the charge air density is good, and you still have to back the timing off to counter knock, what can you do? Lower the compression, and you'll lose bottom end torque (not to mention a large wedge of cash). Pour more fuel in and you lose power and increase emissions drastically. Inject water and you do lose a bit of power (at the same advance) but you're not increasing emissions. Increase the ignition advance and you'll have more power and still reasonable emissions. Sounds quite reasonable to me.

Note, this is all about standard engines, BUT the same could be said of any system. *IF* you can run MBT ignition on full boost, you will almost certainly have poor low end torque. You could use another band-aid to fix that (nitrous oxide ALS).... one end of the scale will always achieve a gain at the expense of the other. Let's not forget we're taliking about ROAD CARS here... losing bottom end grunt in a race engine isn't that much of a problem since you won't be using that part of the rev range out on the track.... but you do use the whole range in road cars!

AFAICS, the only way to get good top and bottom end power, without band-aid like nitrous or WI is to use variable compression; the best of both worlds, high compression when you need it, lower compression when you want to keep spark advance at MBT but higher compression would cause knock. Until there is a viable production variable compression engine available, I think I'll just use the environmentally friendly band-aid :)

Cheers,

Pat.

shiv
11th May 2000, 21:47
Mike,

Wow, am I reading this correctly? Do Ahmed and I actually see eye to eye on the big issues here. From all the hype leading up to this point, I was expecting a big US vs. UK tuner brawl http://bbs.22b.com/ubb/smile.gif

Please correct me if I'm wrong, but I'm inferring from his post that he installed water injection only after running out of IC/turbo options. And second, he appearantly prefers to tune road cars without WI, meaning that he is not using WI to actively supress detonation (as many people on the list are doing). Instead, it seems as if he is relying on WI as a auxillary failsafe due to ACT instability. In this situation, if the WI system fails or runs dry, the engine does not go POP. I can agree with his approach. Hell, I think it's a good idea and worth a try.

But I think my basic arguments still stand. That is, I don't agree with the following:

1. Using WI before addressing obvious compressor and IC limitations
2. Using WI as a continuously active form of knock surpression

Neither of which constitute as an example of good tuning, IMHO. The reason I've been such a hate-monger when it comes to water injection is because I've seen, especially on this forum, simply offensive examples of WI implementations where undersized turbos were being spun into surge, simply incapable of sustaining boost pressures to redline. Coupled with stock, flow-restrictive and low efficiency intercoolers, the results were frieghtening-- a car that needs a continuous water supply on boost just to keep from blowing up! To be fair, the WI by itself, is not offensive. In fact, I think it's got a lot of merit IF used correctly. In fact, I've softened my hard head enough to try it on our big turbo'd, big intercooler'd, EFI programmable, high-compression EJ25 turbo Impreza. It will be interesting to quantify all the variables involved.

Shiv

[This message has been edited by shiv (edited 12 May 2000).]

shiv
11th May 2000, 22:20
<BLOCKQUOTE>quote:</font><HR>Originally posted by pat:

...Given that a turbo is operating efficiently and the charge air density is good, and you still have to back the timing off to counter knock, what can you do?... Inject water and you do lose a bit of power (at the same advance) but you're not increasing emissions. Increase the ignition advance and you'll have more power and still reasonable emissions. Sounds quite reasonable to me.<HR></BLOCKQUOTE>

I would argue that for higher-than-stock hp levels, the turbos are not operating efficiently. Measuring intake temp is almost meaningless when it comes to evaluating turbo performance. Try measuring before-the-intercooler turbo outlet temps instead. I suspect you will see outlet temps approaching 300F under race conditions and mid-200F during spirited street driving. While even the most marginal of OEM intercoolers will remove 60-65% of the heat generated by compression, there don't help when it comes to the issue of exhaust back-pressure which, gone unchecked, will dramatically reduces an engine's knock threshold, blows open wastegates, kills power, etc,..

OEMs maintain a margin of safety with their turbo cars by running rich with a conservative map. Unfortunately, we use up that margin of safety when we cranking up the boost. All things equal, as boost goes up so does exhaust-backpressure which, in turn, will eventually lead to exhaust gas reversion and then detonation. This is why a good turbo upgrade can yeild sizeable power increases with no extra boost and only marginally cooler post-IC intake temps while still allowing addition ignition advance with a higher knock threshold. With a little research, it is possible to find upgraded turbos that will move more air, more efficiently, while not adversely affecting spool up. In the past few years, many advances have been made with ceramics, low inertia turbo shafts/wheels and ball-bearing center sections. It costs money, but doesn't everything?

[This message has been edited by shiv (edited 11 May 2000).]

Bob Rawle
11th May 2000, 22:32
I too have been following this debate with interest and also thought that we were getting into a UK v US type scenario, its good to see though that at the end of it all there is some predominance of viewpoint. Having mapped cars with water injection fitted I also refuse to map with water on, only with the fuses pulled!! I think that water is a very good "band aid" whenever a car is knowingly going to be subjected to severe stress for prolonged periods (Nurburgring for example).

Pat, I have been using a NTK wide band four wire sensor which is very good and very cost effective as well. Its the one that Haltech use with their wideband lambda kit. If you do the work I have done I think you will/should be suprised at how sensitive/good the Lambdalink is within its range. I have been running back to back tests in my car using narrow band and lambdalink, wideband with wideband meter, wide band sensor and lambdalink and narrow band sensor and wideband meter. Suffice to say that all combinations have their merits and issues. Once I have completed all the work I might post on the issue but the wide band combination did not lead me to change my fueling at all. (the map is not bad to start with though. As a comment make sure that you have callibrated the lambda offset. I used a dc power supply with 5 digit dvm to do this and also to set up the wideband meter sensitivity. I think early cars are more susceptible to volt drop than later ones. Also I can confirm the need to move the lambda sensor from the headers to the downpipe in the early cars. At 800 deg C max limit you will kill your new sensor very quickly if you leave it there.

As an aside I have also established that EGT, when measured using the select monitor in standard UK and imports, produces values in the same range that I get with my egt probe at the header intersection at the bottom of the "up pipe" and are in the range of 800-850 deg C. This is typically what I would see on fast cruise/hard blat.

Bob (now going to forget all about Impreza's for 10 days of r & r)

Richard.L
11th May 2000, 23:24
Hello all,

I would like to tell a short story, it is not meant to upset or have a dig at anyone, but it will be as simply as I could put it.

I am an engineer at heart and soul, I like solving problems using the simplest and the most cost-effective way possible.

I can see that there are two ways to tuning your car, go for the Aquamist way or go for the non-Aquamist way.
If you go for the Aquamist way, pay your one-off 350 and you get a lot of fun without sacrifying any derivability, of course you need to up your boost a bit, otherwise you loose power, as Ahmed stated.

Going for the Mr XXX's way, it would be very exciting and there will be tens and hundreds of experts ("experts") rushing to help him determining what's best for his engine ... he would naturally go for the one that impressed him most: "... I was an ex-... therefore I am the best"; "I know so and so ... so I know it all". Guess what, before long, he was offered a bigger intercooler and a bigger turbo as the start-up routine.

Bang !!! "Oh dear, you should have upgraded your fuel pump, it went lean, not my fault, I have only sold you an intercooler and a turbo, You needed an engine rebuild!".

A twin fuel pump later.... "WOW, it goes like a bag of S*** !" exclaimed Mr XXX. Bang !!! "your fuel injectors has maxed out and the management could not dump any more fuel for cooling. You should have known better, your engine was not built for track events". The expert told him bluntly, engine out again, more dosh.

Triple fuel-rail with twelve injectors, 4-bar MAP sensor, ECU re-programmed by ... . "Go and race it, totally bullet proof now". The expert said confidently. A few weeks later Mr XXX took his car back to his trusty "expert", "It was not very fast sir and I have to fill up every 50 miles, got stopped by the police a few times on the way, accused of destroying the public highways by setting fire to it ", he continued to explain: "I think it was the two-foot long trailing flame that came out of the exhaust pipe that did it. " No rebuild is necessary, assured by the "expert".

Mr "expert" offered more expert help. "The only way you are ever going to take advantage of those big intercooler, big turbo, big injectors is to lower your compression ratio, 4 to 1 is ideal", he explained. A rebuild was indeed necessary, "throw in a few improvements at the same time, forged pistons with total sealed rings, billet crank and steel con-rods, a pair of high lift cams with special profile of course from 'you know who' " he continued. Mr XXX was very grateful for the advice and decided to proceed.

Mr XXX couldn't wait to collect his car on the day. He was late arriving because his bank manager refused to unlock the interviewing room door until he has signed over his freehold.

Now Mr XXX is really happy, beating every car and motor bike in sight on his favourite stretch of motorway. Just about few yards from home at the stop-lights, he met the "Aquamisted car". When the lights went green ....

Poor Mr XXXX.


Here is the moral of the story, aquamist has robbed thousands of dollars from the "experts", no wonder it is met with sooooo much hostility.


I must to go to bed, it is really getting late, need to make a few more Aquamist tomorrow... Richard Lamb


[This message has been edited by Richard.L (edited 18 May 2000).]

[This message has been edited by Richard.L (edited 14 December 2002).]

shiv
12th May 2000, 00:01
I dunno... I kinda like shooting flames out of the tail pipe http://bbs.22b.com/ubb/wink.gif

Shiv
EJ25 turbo
9.7CR:1
12psi
~320hp
300-350 miles/tank
850 peak EGT
30k miles and still going strong
... with big fat flames at fuel cut!

Anders
12th May 2000, 00:28
I love this thread and I spent 30 mins creating a funny and poignant reply to Richard. Computer crashed! http://bbs.22b.com/ubb/frown.gif

Instead!

As a summary!

I bought snake oil (re map)and it cost me 20 X more to put it right!

I set up this bbs to appraise new techology and I am very pleased at the moment http://bbs.22b.com/ubb/biggrin.gif



[This message has been edited by Anders (edited 12 May 2000).]

Dennis
12th May 2000, 07:28
Hi All,

This is gonna be a long post on the subject, but hopefully interesting and informative.

There are two main approaches to modifying a turbo car. One is the average person who just wants a little more power, and the other is the guy that wants to go all out. Aquamist is ideal, and the best bang for the buck, for the the guy that just wants a little extra from his turbo car. Aquamist is also an excellent complement to a "no holds barred" type modification.

Let's first examine the most common of the two, the guy who doesn't have very deep pockets, but wants a little more fun, with the equation being fun=more power. The first thing he's going to want to do is, you guessed it, raise the boost. He soon discovers that his car isn't running that well. Why? Well, it will usually be caused by some, if not all of the following. The ECU's internal fuel and ignition maps don't extend to the boost he is running, so the mixture is getting a little lean, andd there is bit too much advance. The turbo (adequately sized by the manufacturer for stock boost pressures) is no longer running within it's best effeciency range, hence the compression process is creating unwanted heat. The intercooler (also adequately sized by the manufaturer for stock boost pressures), isn't effecient enough to reduce intake temps to an adequate level. He now must find a way to solve these problems.

There are two ways he can go. The first is to replace/upgrade the turbo, replace the intercooler, and invest in some some sort of add on fuel computer, and while not neccessary in some cases (dependant on the internal characteristics of the engine, and the timing maps), invest in an add on ignition computer. Now remember, this is for the majority of people who just want a little extra boost. O.K., let's look at the bang for the buck, and I will be very conservative with prices (all in U.S. dollars).

Turbo Upgrade (replacement will more than likely be over the $1,000.00 mark) 500.00
Intercooler 500.00
Fuel computer 500.00
Ignition computer 500.00

That's $2,000.00, and I doubt you will be able to get the above for those prices. Well, that's a bit expensive, but wait, these things don't install themselves. So let's throw in another 500.00 for installion and tuning your add on computers, and that is cheap. From the time you install the add on fuel computer, you will need to tune it with a proper wide band A/F ratio meter, so you are most likely going to end up at a dyno facility. So the final conservative figure is 2,500.00.....ouch.....that 5 psi increase from 8 psi, cost you more than you thought. Also, watch out for errors that can be made with these types mods. Don't add that larger turbo and crank it to your desired boost increase pressure on the stock turbo, because the cfm on the larger turbo will be higher than the cfm on the smaller turbo at the same pressure. Fuel computer....ohh boy.....you can really do some damage with those things. My advise, only increase fuel, as running rich won't kill your engine, but unless you have a good A/F ratio meter, don't lean it out.

The second way is water injection, and....well....nothing else. How? Because of the wonderful properties of a virtually free liquid called water. Let's take it step by step.

Your turbo strayed from it's best effeciency, hence more heat, and your intercooler isn't effecient enough to bring the charge air temp back down to accetable levels. How will water injection help that? Simple, because of water's high amount of latent heat, which is a fact. This process is used on World Rally Cars, and has been clearly proven to work. But what about fuel and ignition? Here's more good news, the water injection will also raise the knock limit on the engine. This is where the skeptics will chime in and say, no way. Unfortunately for the skeptics, this has been proven by Saab, when they took a 2.3 litre engine to appox 220 hp,with a curve that only barely dipped at high rpm, on a mixture of 14.7-1, no that's not a typo, and you read it right, 220hp at stoichiometric. The most they could get out of the engine without water injection, at 14.7-1, was around 160hp, and the hp rapidly dropped off at higher rpms. O.K., but what about igniton timing? If your timing is a little to advanced for that boost level, knock might occur. Again water injection will help. This has been proven almost fifty years ago, by J.Jalbert in 1952. He was the director of a company in France called Compagnie des Moteurs a Combustion, and when he did some tests with water injection, he found that with 65 octane gas, he could raise the compression ratio from 6:1 to 10:1. Since increasing boost is like increasing compression, and increasing compression requires the use of higher octane fuel, ignition retard, or fuel dumping, it's clear to see that water injection will allow for extra timing advance. A side note here. Timing advance will not usually be a problem, as most manufacturers use very conservative timing figures. Now for the cost. An Aquamist water injection system will set you back approx 500.00 to 800.00 depending on which system you choose. Can you say bang for the buck?

Here it comes, the big turbo, large intercooler, more injectors, add on electronics people will be warning you how dangerous water injection can be. They will tell you if it fails, or the water runs out, your engine will blow up. First off, the systems are built to high quality, and aren't prone to failure. Mine has been working for over two years, and I have never had a problem. Of course nothing is immune to failure, but almost all modern, and even not so modern turbo cars, have something called a knock sensor. The manufacturers realize that even a bad tank of fuel, amongst many other things, can cause knock, so to safegaurd the engine they install a device called a knock sensor, which informs the ECU of the presence of knock, and the ECU in turn, winds back the ignition advance to save the engine. So if your water injection system does fail, you will realize something is wrong when the car no longer pulls as it should, but the need for a rebuild is highly unlikely.

Now we can move onto the guy with the deep pockets, that wants to go all out. This section should be a lot shorter, I hope.

You have built a monster, with a large turbo, huge intercooler, higher cc/min injectors, upgraded high pressure/volume pump, programmable ECU, cams, strengthened the bottom end, and have ported and polished everything, and are now running 30 psi, and your latest dyno figures show 450hp. However, wouldn't it be great to ensure that 450hp remained consistent when you take it out on the road/track? You find you take it out and sometimes it feels 100%, and sometimes it doesn't quite feel 100%. The likely reason for this, would be a good ECU tuner. Huh? That doesn't make sense. Actually it does. Most good engine management systems will have an "air retard compensation map". The programmer will add more retard, as needed, as the air intake temperature increases. The net result will be a power output drop off, as intake temps rise. But I have a huge intercooler you say, how could my intake temps increase? Lot's of common and not so common reasons. Things like changes in ambient temperature, lots of tight cornering where air flow over the intercooler is slow, to the unlikely, like debris in the intercooler. So how will water injection help? You could purchase the System2C, which can be controlled by most aftermarket engine management systems that have a PWM map. You set it up so that as soon as your ideal air intake temperature is exceeded, the system starts injection, to keep the air intake temp ideal. As simple as that. Ooops, here come the "engine will blow up if water injection fails" people. Remember that air retard map, well, it is still there, and if the water injection does fail, or your tank runs out of water, the computer will start to retard the timing if air intake temps increase.

It is not my intention to insult anyone, but I would just like to point out a few things about Shiv's quotes of the day. None of the quotes had any testing or sound scientific rational within them, to back up there claims. As a matter of fact, it is interesting that each quote actually benifits the person quoting.

"Water injection makes more horsepower than detonation... but it's still a crutch for an inadequate fuel system"-- R. Collins, RC Engineering
It's a crutch for an inadequate fuel system, so buy some larger injectors, does RC Engineering by any chance sell injectors? ;)
My opinion of RC Engineering, excellent. I have had injectors done by them, and I am very confident that these guys know there stuff when it comes to injectors.

"That turbo [on the Cossie] is waaay overworked, poor motor."-- C. Cadle, Garret
Come on give that turbo a break....buy another Garrett turbo. Again, I think highly of Garrett turbos, so no bias here on my part.

"All things considered, you would be far ahead never to have heard of a water injector."-- Corky Bell, Bell Engineering
Corky Bell knows his stuff, that is why that statement really surprised me, but then again, instead of water injection, he can sell you an intercooler....hmmm...interesting.

As for my replies ro Shiv, I enjoyed reading his write up on the Impreza, and by no means wish to convey that his knowledge is limited. Quite the opposite, but I do feel his knowledge on the subject of water injection could do with a little research ;)

I am not quite sure if Anders was associating water injection with snake oil, but being the person I am, I like facts, plain, undisputed, documented facts.Theory and opinions are good to hear,and a great place to start, but left to them, and no real world testing, it would still be a fact that the Earth is flat. I got mainly negative information on water injection when I was interested in using it, yet no one had any real information. That is why I spent over three months researching water injection before I purchased it. From it's first trial in 1900 by an engineer named Benki, to the US Air Force, Royal Air Force and Luftwaffe aircraft during World War II, to Ferrari's use of it in Formula One in the 1980's, right up to the present use on the WRC cars. I have found lot's of actual testing that confirms it's benifits, and while I have also found numerous statements, and opinions against it, I am yet to find one actual test showing these negative aspects claimed. I have gotten to realize that the best way to make decisions, is to be informed, and not just by what Mr.X or Mr.Y says, but by real evidence, as in the results of real world testing.

I'll sign off on this subject now.

Dennis.

ian/555
12th May 2000, 08:09
I think that's the end of it!

shiv
12th May 2000, 08:13
So, let me get this straight... you're implying that proper engine tuning is a conspiracy perpetrated by turbo, IC, and injector retailers? Oh boy.

shiv

Agreed. Let's move on to another thread.

[This message has been edited by shiv (edited 12 May 2000).]

Richard.L
12th May 2000, 09:07
Hello Mike,

Re: Ahmed's ACT Strategy for Ford, using water...

As far as Ahmed/Ford is concerned, the application is sound for Ford.

But... the BIG BUT, other WRCs such as Subaru, SEAT, Skoda, Peugoet, Toyota (ex-WRC), Hyundai has completely different Strategies. I know it becuase I worked with them all - as the competition is current, I can not pass on each one of their strategy, that would be unwise as I would be locked up in a prison for life. I have to sign un-disclosure agreements.

To sum it all up, Ford uses water injection for what they think fit, ERL uses water injection for what the market demands - horses for courses. Whereas for Shiv, he doesn't use it, until then he should keep quiet.

As to Ahmed's comment about WI controlled by pressure alone is valid- ONLY for Ford. Their WI system is ancient and as far as I am concerned and is not very effective. A low pressure pump (not ours - thank God) used in conjunction with our water jet is a mis-match. They have at last listened to me after three years and changed it into two smaller ones. Unforntunately, the old pump stayed. Without high water pressures, the ACT will consume more water then necessary and therefore lead to power loss and even occasional misfire. Ever wonder why Ford hadn't won any championships since the launch of the WRC since 1992? I mean championships, not occasional races.

Detonation control. If Ford doesn't use water for detonation control, where does all the water go after it went pass the throttle body, unless they have a very clever WVRV (water vapour recirculation valve- fancy name- isn't it), I doubt it very much.

Water is the best anti-denoatant to date, as far as WRC is comcerned, no other additive is allow to mix with water, no argument there.

I wonder what will the Ford ACT do when the RH (relatative humidity) of the air is over 50% ? equates to approximately 150cc/minutes of water (at 300BHP), of-course there is a humidity sensior there, or is there? ... I know, they only compete on dry days.


Water injection by pressure alone.... flawed?

Within the narrow confine of the ERL aquamist operating cycle, normally a few psi below maximum boost, introducing a temperature probe is totally unnecessary. For instance, there is very little temperature changes within +/- 3 psi. For Ford, it is vital as ACT is their main strategy, but saying ERL's detonation Strategy is a flawed is totally unfair, I hope he will retrieve that statement. ERL's detonation strategy will also work well with overall ambient tempertures changes, the colder the inlet air, the more power the engine produces, but at the same time, more combustion heat and in turn more likely to detonate.

There is one tuning rule we should all agree to, I hope:

Maximum Power produced by a given size engine is limited by the fuel's knock limit (octane) and compression ratio. No magic intercoolers or fancy turbos will ever change that, eventually water has to be used to increase the knock limit of the system or lower your compression ratio (engine out).

What is the CR of your engine Mike?


Richard Lamb

mike_nunan
12th May 2000, 12:19
Very interesting responses - thanks very much to all. It seems like there is general agreement that WI has its place, although we could probably debate the exact method of application/implementation until the cows come home and then some.

Shiv, all credit to you for keeping with the technical spirit and embarking on your own experimentation. Given that this discussion has been quite adversarial at times, I find it admirable that you are willing to acknowledge a shift in your own position, and that is something the rest of us can learn from.

Richard, your final comment states that CR and RON are the main variables influencing knock, but charge temperature is clearly important. I still haven't decided what my target figures will be for the next phase of tuning, and that topic best belongs in another thread and another lunch hour, but it seems clear that turbo and intercooler changes will play a part.

So might water injection, but first I'll need an answer to one mundane but crucial question: what the heck are you supposed to put in the water reservoir to stop the who system freezing up in the winter? I assume that Bluecol is contra-indicated in this particular application! http://bbs.22b.com/ubb/wink.gif

-= mike =-

PS. My car is an Impreza STi4, btw, and I _think_ the CR is 8.5:1 although I'm sure someone near here will correct that if it's wrong.

Anders
12th May 2000, 13:43
I have amended my post about snake oil! http://bbs.22b.com/ubb/mad.gif
I have know qualms about water injection http://bbs.22b.com/ubb/biggrin.gif

Mike Rainbird
12th May 2000, 15:14
Richard,
7.7 c/r......
Mike

[This message has been edited by Mike Rainbird (edited 12 May 2000).]

mike_nunan
12th May 2000, 15:59
Grief, there're too many Mikes around here! http://bbs.22b.com/ubb/wink.gif

*blush*

-= mike =-

(Belatedly putting down the wrong end of the stick...)

pat
12th May 2000, 17:32
Shiv,

my comments were about components being used in standard state, ie not uprating the boost. Even within the manufacturers' design envelope the engine has to be retarded to control detonation, this being caused primarily due to high temperatures in the endgas as the large volume of air/fuel gets compressed. One way to slow down the rate of temperature rise is to add water; due to the high latent heat it will take longer for the air/fuel mix to reach autoignition temperature, and therefore it is possible to run more ignition advance than possible without WI, on a completely standard engine [no mods bar WI].

We are assuming here, that the original system is as close to optimal, overall, as possible.... good low end torque, sensible top end power and acceptable emissions. If you were going to upgrade your engine, you would no doubt aim for the same criteria; you wouldn't want to sacrifice bottom end torque to gain at the top end unless it was *NOT* a road car (say a track car). But such is not the case in the present discussion. So basically, you've now upgraded various bits (turbo, intercooler, pistons, crank etc) so what you have is what you had, but more. The system will still be subject to the fundamental constraints of the original; namely to have good bottom end (read: off boost) torque you can't lower the compression ratio too much. This is fundamentally at odds with achieving MBT at full boost. You *MUST* back off the ignition in order to prevent destructon of the engine, even though your setup is actually optimal [you could add fuel but this will affect the flamefront velocity and lose you power as well as increasing emissions].

As shown for the standard car, it is possible to advance the ignition simply by injecting water into the intake. This will cool the charge (may not be relevant if your turbo and IC are good) and control the buildup of heat as the fuelmix gets compressed, allowing this extra advance to get you closer to MBT. As a bonus, the cylinder temperature will be lower due to the heat lost to evaporating the water. As the exhaust port opens there is very rapid expansion of the gasses, which can cause some of the water to come back out of the vapour phase, which will quench any glowing carbon deposits (not that you'de be that likely to have any, since the requirement for overfuelling has been alleviated).

Given that any good ECU *WILL* retard the timing if it detects knock, you may get one or two transition knocks, should the WI system fail, then the timing will be back to the safe area for non-WI assisted operation. Given that transition knock is little more than a minor annoyance and very unlikely to cause damage, the odds of an engine blowing up due to WI failure is very slim. Considering that a WI system can be bought for about USD 500, and that the consumable, water, is pretty close to being free, it really is good value for money, IMVHO.

Cheers,

Pat.

shiv
12th May 2000, 18:39
<BLOCKQUOTE>quote:</font><HR>Originally posted by Richard.L:
There is one tuning rule we should all agree to, I hope:

Maximum Power produced by a given size engine is limited by the fuel's knock limit (octane) and compression ratio. No magic intercoolers or fancy turbos will ever change that, eventually water has to be used to increase the knock limit of the system or lower your compression ratio (engine out).
<HR></BLOCKQUOTE>

Richard, but we're not talking about a given engine. We're talking about an entire system. The individual components put together dictate the realistic and safe maximum horsepower of any given system. Of course, CR and octane is the FINAL barrier. A barrier many think they are brushing upon when, in fact, they are not.

CR and octane limitations allow for far more specific power output that most are willing to believe. They reason so many people are brushing against the knock threshol, so early, is almost entirely due to component choice. Running a small turbine-limited turbo into surge will, always, yeild terrible results on both sides of the engine and leading to premature detonation. Does this mean that the compression is too high? Or that the octane is too low? No! It just means that the turbo is too small. Period. End of story. Install a carefully chosen turbo and, all of a sudden, you have a car that is, say 30% more powerful, at less boost, with several degrees more advance, no detonation and with no water injection. This is not to say that WI can't or shouldn't be used. I'm just saying that a lot of people give in to the "CR/octane monster" well before they really need to. I've agreed to keep my mind open to the use of WI, but I'm not going to sit by and let to it assume the role of proper component choice.

Octane Shmocktane. Try boosting a 10:1CR motor on CA 92 octane winter gas and tell me that your 8.5:1CR motors are knock prone.. yeeesh http://bbs.22b.com/ubb/smile.gif

shiv

PS. Pat-- many EFI systems disable their knock sensors at higher engine speeds due their inability to distinguish knock from high speed clatter. I also disagree with your notion that the OEM system is close to "optimal".

[This message has been edited by shiv (edited 12 May 2000).]

Richard.L
12th May 2000, 20:56
Hello Shiv,

Guess what, I am going to agree with you halfway...

I am talking about the final barrier as you put it, some engine such as non-crossflow type with flat piston without squish will knock as early as 8:1 CR, long stroke makes it even worse, on the other extreme, central spark plug with swirl induction pipes and very short stroke- 10mm or so, may never knock at all until 15:1 compression ratio, these engine are normally revving at 15-16K rpm at in the F1 turbo era.

It is almost impossible to talk about the whole system as there are just so many variations. As far as system goes, triming individual components can shift the knock threshold.

What I am trying to say that water can entend the " final barrier", thats all. It may be out of context as the discussion goes, I should have made it clear, thank you for pointing it out before people read it as "in conrtext".

Richard Lamb

Richard.L
12th May 2000, 21:53
For Pat,

Thank you for finding out the WB Lambda probe, I often wonder where you can buy them from. It is so useful for tuning your engine, need to know more about the WI effects, weird- 50% of the engine output is water vapour, another 2-3% (due to WI) shouldn't make such a big differce. Could it be the alcohol they mixed in caused that?

Have you installed your system yet? The sooner you do it the better, I am looking forward to see all your perfectly docuemented reports, being such a precise person as you are.

Richard L

Bob,

thanks for yet another sensor, is the NGK sensor a current driven one as used in the horiba range of instruments ?

Richard L

Richard.L
12th May 2000, 22:57
Anders,

Still waiting for your funny reply to Mr xxx ... may be it is too late

RL

pat
13th May 2000, 00:31
Shiv,

perhaps the big manufacturers are just using the millions of dollars it takes to develop an engine and spend it on pizza [I know I would :)], but seriously, the OEM system setup is as close to optimal as you can get it. Of course this statement does depend somewhat on your definition of optimal! In this context, I am referring to all-round driveability combined with low emissions at a sensible cost, reliability and longevity.

Of course, your definition of optimal may be maximum BHP, in which case, yes, you can get 6000BHP from a small engine, BUT you won't do more than about a mile with it before it needs rebuilding. Not ideal for a road car.

In this context, and given that the manufacturers need to compromise on ignition advance in order to maintain good low end torque, WI could be used to gain more top end power. If you drive a standard (without any mods whatsoever) turbocharged car, you will find it really IS more powerful on a damp day than a dry day.... no coincidence, methinks.

Richard,

WI has been plumbed in since my trip to Catalunya, where I covered Anders with water as I primed the system :) MF-2 seems quite happy to control HSV at the shorter PWM period; all I need to build now is a PWM monitor. I feel a PIC project in the making :)

W.R.T. effect of WI on Lambda readings, WI will cool the charge, thus affecting the oxygen free radical content of the exhaust gas... a bit... errr, it's late :) Suspect you're right about the additives, though, they could upset the readings somewhat.

Not sure about this reporting business.... all that I can say at the mo is that with no WI, my knock monitor detects a lot of engine noise on full chat, with WI on it's virtually "quiet".... I've yet to map the ignition so far advanced as to be det-prone without WI. Methinks I shall wait for my new injectors first; driving any fuel injector beyond 80% duty is not ideal; mine went all the way up to 120% on the rollers, OUCH!

BTW, what's the best way of setting up water pressure? My approach was to connect jet direct to pressure control manifold, then adjust pressure switch so pump was running almost 100%... this should (hopefully) allow maximum flow without initial pressure drop as the HSV opens.... [ie setting the pressure switch higher WILL pressurise the system higher, BUT the pump cannot maintain that pressure once water starts flowing].

Cheers,

Pat

shiv
13th May 2000, 01:31
Pat-- we've been full circle. No need to re-cover any ground.

Soo.... what's the next point of debate? Maybe someone should start a new thread.



[This message has been edited by shiv (edited 13 May 2000).]

Richard.L
13th May 2000, 07:41
Pat,

I was joking about the report, a short note from you normally takes me a few days to read and keep me thinking for months. You rarely leave any stone un-turned- electronically speaking.

A LM3914 plus a bargrapgh display will make a nice simple PWM-visual converter. You need a RC circuit 1M and 1UF, to convert the PWM signal nicely to a pretty linear voltage to feed the LM3914.

At the same time, make another one to monitor the gray wire, this will give you the true flow. The PWM signal from the MF2 output is not the true flow.

OK, lets get back to the WI topic, I heard that a small amount (10-20 PPM) of H2O2 promotes the OH radicals- the start of all hydrocarbon burning process. It will speed up the chemical conversion process of fuel, ie super-fast fuel means less detonation due to less end-gas situations. The H2O2 concentration will have to match the geometry of your combustion chamber. Your local chemist sells a small bottle of 9% by vol for approximate 80 pence.

Have you heard anthing thing about this phenomenon, if it is true, adding a few drops peroxide to WI, changes the ordinary fuel to rocket fuel... just a thought.

Richard

Mike Rainbird
13th May 2000, 20:16
Dear All,
I now see where Shiv is coming from. I originally thought he was dead against water injection - period. However, I now understand he is just dead against using it instead of sourcing the correct "tools" for the job. In this case I am in agreemment with his comments that it is a "band aid" for a poorly developed engine. BUT, if used as a means of Air Charge Temperature control, water is ideally suited for THIS function. By this I mean, not as a det control, but to as a means to sustain the power you have, whatever the climatic conditions. This way you prevent the ECU from ever having to retard the ignition at high ACTs. Accordingly, on a hot day where the intercooler is struggling to keep the charge temperature down to a "comfortable" level, water can be used to do this.

For example, if two cars running identical engines and on the same (hot) day, but one without WI fitted were to be pitted side by side, the one with WI would be faster due to it's ECU not having to retard the ignition. The one that didn't have it fitted could lose as much as 15-20bhp because of the retardation introduced to prevent the engine from pinking due to high ACTs...

My view on this subject (and I'm sure Shiv will agree) is to decide initially (before building the engine) EXACTLY what bhp you wish to achieve and then source all the best bits to achieve that. My engine was built specifically to achieve 410 to 425bhp. Every component was chosen and matched to the engine to achieve that. If I wanted any more than 460bhp, I would have to start all over again, as the head is only a "medium" port type and to get 500+bhp I would need a "big valve" head to flow the air as well as a bigger turbo and higher flow injectors (at the moment the turbo and injectors are FLAT out all the time I use the power.

An engine is VERY simple, all you have to do is think of it as just an air handling unit. The more air you get in (providing it is matched by a corresponding amount of fuel), the more power you get out. This can be by large capacity or forced induction, either way achieves the same aim - to get more air in. Unfortunately with a forced induction engine you HAVE to have trade offs (or compromises). You have to decide what is important to you - throttle response, or top end power - you cannot have both, it has to be one or the other, or you can compromise, making it neither particularly responsive or mentally top endy powerful. My engine is such a compromise, it does not have the bottom end of an almost identical engine (but equipped with a T3/4 with a 0.48 a/r 391bhp@6500) - but mine has better top end (419bhp@6500), which is more noticable on the road than the on paper figures might suggest. Mine doesn't have the top end power though of a T4 equipped car (550bhp@7000!) - but mine has better bottom end than a car thus equipped.... As I said EVERYTHING is a compromise with turbo cars. Fortunately turbo technology is not static and you now have the latest generation of roller bearing turbos that can be a higher flow than the old T3/4 I have on my car (which is VERY restrictive on the exhaust side) AND have a better spool up time so that you don't suffer from as much lag, but they are still a compromise (just less of one http://bbs.22b.com/ubb/wink.gif).

The ideal solution would be a variable rate turbo (such as the one used on the Renault Espace turbo diesals), this has a camera aperture type set up that opens out with revs, making it a perfectly responsive turbo whatever the engine speed. Unfortunately turbo technology has not developed enough to make it transferable to a petrol engine which develops much higher temperatures than the diesals, but I can live in hope that they will one day! http://bbs.22b.com/ubb/biggrin.gif

Anyway, I've digressed - match your components correctly and you won't need to use WI as a "band aid", instead you can use it just as an ACT control.

To end, I'll let Ahmed have the final say:

QUOTE
Hello Mike,

It's interesting to see such a defensive response to my comments about WI...as I clearly stated the comments were based on my findings based on extensive and independent work both on the dyno and in car, using various jet sizes and pump pressure including an ERL supplied pump.

Richards comments assume I dismiss his company's approach to WI.....he clearly hasn't read or understood the text!

In the second paragraph I clearly stated that I was commenting about WI as I developed it for Ford, as opposed to ERL's approach.

In the seventh paragraph I cited a clear example of WI used as an anti-detonant!

But, when I developed the WI system at Ford, the system was NOT used as an anti-detonant, the WI could fail completely with no ill effects to the engine other than a loss of power output. The only functioning 'anti-detonant' on the car at the time was the Spark Retard / ACT function in the ECU!

I fully respect all (Richard included) professional tuners, and their various approaches to resolving the technical difficulties in producing a competitive turbocharger race, rally or road engine.
But I find it very intriguing, when people in this protracted debate make comments without knowing the facts, having had no access to the huge amount of test data generated by a large team of professionals, and the reasons behind making the INFORMED choices we made.

And regarding remarks relating to Ford not winning any championships since 1992, (I presume he means the FIA World Rally Championship, because the National Championships won by Ford Rally cars since 1992 are countless...) well, if that was influenced in anyway, by the WI strategy used on their cars........I'd better retire.........

So, if there is a lesson to be learned from all this! It's this...Test, test, test....believe nothing until you have seen positive results with your own eyes. Anything and everything is possible, development is the key to success...just because someone posts a comment on this or any other BB detracting your approach to tuning something in particular, do not assume they are correct! More often they are NOT!

I will continue to develop my engines and the way I choose to use WI as I wish. I shall continue to win many, many championships and single venues and 'dyno shoot outs' with my engines, including the 400+ BHP Subaru project that has just started. Using my low (6 bar) pressure WI system & Pectel T6 management system, and wonder what all the fuss was all about.

Best regards to all,

Ahmed Bayjoo.
UNQUOTE

So Scooby tuners, watch this space...
LOL
Mike

shiv
13th May 2000, 22:05
okay group hug!!!

http://bbs.22b.com/ubb/smile.gif

Richard.L
13th May 2000, 23:09
Hello Mike and Ahmed,

First of all I must apologise for my comment about Ford not winning any WRC championships, it was totally inapt. It was incited by a particular comment: " ERL's WI stratege is flawed". I strongly believe that too was un-called for. However I have offered my apology. Slagging off each other is not the best way to go forward. Water is there to be used wisely, who ever the provider is.

Water have great potential for inlet air cooling, the Enthalpy is huge. We supply WI to a number of Fuelcell manufacturers around the world (suppose to be the future power plant for the car, I sincerely hope not). Water in that application was solely used for convertor core cooling purpose, during the Hydrogen/Oxygen to electricity conversion process, no "detonation" there for sure.

My comment about water pressure on Ford system is not from hearsay, it is test, test and tested, by a company who employs 150K people around the world, I trusted their results. Droplet size is vital for the ACT-strategy, I hope you will agree with me there, Ahmed.

Ahmed and I have met in the late eights, we were both interested in WI, he has found a particular niche use for WI and he has now advanced so farword using WI as ACT control, I got his data logged chart to prove it- NO argument there, full stop. I owed him a big drink for introducing the concept to Ford's rally programme which ERL have benefited greatly, both in fame and fortune.

We, at ERL developed a range of WI kits for the road going vehicle, whom you are familiar with, but we have also a range of specialised systems for other markets, different jets and different pump (16 bars+) etc, of which is not available to the public.

Rather than drifting off the main WI topic- road cars only... I think ? we are after all in the 22B site. I will continue...

ACT/WI or Knock/WI is the main discussion here, or is it?
Lets be straight, temperature ratio is calculated from absolute zero- that is -273C. A charge air temperature increase from 50C to 100 C is NOT a 100% temperature-ratio change, it is:

[(273+100)-(273+50)/(50+273)] x 100% = 15.5%

The engine is only seeing a 15.5% air temperature ratio change, not 100% !!!

OK, lets look at the pressure. Pressure ratio is calaculated from absolute zero vacuum. Doubling the atmosphere means a 100% pressure gain, no need to work that one out.

This is the reason why ERL's WI is pressure triggered, main aim is to control the onset of knock, induced by pressure ratio increase, hence the final barrier- effective compression ratio increase. Most turbo users tend to increase boost to gain power, very few used the temperature reducing method, there is a limit as to how big an inetrcooler one can fit into the car's engine bay and how much one can afford to buy a bigger turbo just to gain a few % cpmpressor exit "temperaure ratio", it is just not a economical way of tackling a simple problem for a small gain, given the normal road car doesn't have a high flow inlet tract, high lift cams, free flow exhaust system, big valves, etc.

For road car I repeat, is more prone to boost-induced knock rather than air charge temperature rise, the car's (road car) have made provision to detect in-take air temperature, an in-built protection from birth. I add, it is not necessary to use an air temperature probe on the down-stream of the turbo, Hot-wire mass flow sensor (Inherent temperature compensated) produces true air mass; most ECU normally takes a barometric and ambient temperature reading during start-up, selecting the correct map for the day.

I would like to sum up the my argument as fairly as possible, I hope I haven't stepped on somebody toe this time, if I have it is not intended as before.

Ahmed's ACT WI is perfect for the the all-out, no compromised individuals like Mike (Rainbird) and the future Subaru 400+ conversion. We at ERL, will pick up the crumbs and continue to develop our knock/WI kits for the fast-road going cars.

I really hope that from now on, all WI providers will stress their products finer points rather knocking each others product or strategies.


Richard Lamb




[This message has been edited by Richard.L (edited 18 May 2000).]

shiv
14th May 2000, 00:35
There are always going to be people who see things differently as to how water injection should be implemented, if at all. Both Richard, Ahmed and I have had good results with our own tuning methods. I agree that each method having its own strengths and share of compromises.

A few days ago, Richard was gracious enough, at my request, to allow me to install and evaluate the Aquamist system on our project Impreza 2.5RS turbo. I suspect it will be effective as our ultimate hp goals are well beyond the CR/octane barrier we've been talking about. I'm also going to keep my mind open for other WI issues as well. I'll keep everyone posted on the results.

Russ
14th May 2000, 15:49
[QUOTE]Originally posted by mike_nunan:
So might water injection, but first I'll need an answer to one mundane but crucial question: what the heck are you supposed to put in the water reservoir to stop the who system freezing up in the winter? I assume that Bluecol is contra-indicated in this particular application! http://bbs.22b.com/ubb/wink.gif

My water supply is the windscreen (windshield) wash reservoir. I use ww fluid with high methanol content for a little octane boost. For more information about my setup see: http://uk.syty.org/waterinjection.htm

Cheers,

Russ Bland
GMC Syclone

mike_nunan
15th May 2000, 12:36
Thanks Russ,

Over two years and 15,000miles seems like a fair test period. What made you decide that injecting screenwash was safe in the first place tho'? Did you just decide to risk it and see what happened?

-= mike =-

Richard.L
15th May 2000, 18:08
to Mike_nunan,

Methanol sources: Alcohols Ltd on 01279 658464 or Pertolchem on 0181 879 3090.

RL

Russ
15th May 2000, 19:08
<BLOCKQUOTE>quote:</font><HR>Originally posted by mike_nunan:
Thanks Russ,

Over two years and 15,000miles seems like a fair test period. What made you decide that injecting screenwash was safe in the first place tho'? Did you just decide to risk it and see what happened?

-= mike =-<HR></BLOCKQUOTE>

Mike,

Richard Lamb of Aquamist fame advised me that injecting water/screenwash mix is a satisfactory means to a) add methanol to the mix, b) prevent the water freezing in winter & c) avoid the need to install a separate water reservoir in a crowded engine bay.

Screenwash is mostly water plus 20% - 50% methanol or ethanol plus a little detergent. The detergent doesn't do any harm... probably helps remove detonation-causing carbon deposits. I'm only injecting at &gt;10psi boost at the rate of 5ml injectant(?) per second & the detergent won't survive long in the 850 Deg C heat of my combustion chambers at full boost.

Russ.

Russ
15th May 2000, 19:13
<BLOCKQUOTE>quote:</font><HR>Originally posted by Richard.L:
to Mike_nunan,

Methanol sources: Alcohols Ltd on 01279 658464 or Pertolchem on 0181 879 3090.

RL<HR></BLOCKQUOTE>

If I recall correctly, a drum of methanol is 106 for 205 litres = 0.52/Litre.

My water injection system consumes 500ml per week, so a 1:1 water/alcohol mix would use 250ml of methanol per week. 205 litres would last me nearly 16 years!

Also, I don't have room in my garage for a drum that large. Then there's the fire risk. Pre-mixing it with water would make it safe, but would also increase the storage space problem.

Ideally, I need to find someone in South East England that's made such a purchase & is willing to sell me, say, 25 litres. That way I get a 2 year supply of bargain-priced methanol. Anyone?

Russ

Richard.L
15th May 2000, 19:57
Mike_nunan,

Here is some good news regarding Methanol used as an octane booster:
http://www.turbofast.com.au/racefuel13.html

RL

frisby
15th May 2000, 20:03
To get a good supply of methanol, you could try your local 6th form college or university. From my chemistry/biology days, we got through loads of the stuff http://bbs.22b.com/ubb/wink.gif

laters

stu

pat
16th May 2000, 17:11
Richard,

not much to report yet, my Lambda sensor is doing very weird things at the moment, which means mapping is somewhat dubious...

I know that the output of the Lambda sensor will drop with increasing temperature, but on my last mapping session it would not rise again! Curious.... perhaps it really is running leaner as it heats up [yes, I did let the engine warm up before mapping to avoid problems with cold enrichment :)] Yet to figure this one out, perhaps a dodgy fuel pressure regulator, lower density fuel as it heats up, slower impulse response from the injectors, could be many things.

W.R.T. using an LM series LED bargraph driver, it should work, but there will be horrible switching spikes on the HSV drive which could cause inaccuracies in the readings... you are of course right about the grey wire, flow will depend on HSV duty cycle and manifold pressure; the pump control wire will give a better indication of true flow, as the pump tries to maintain water pressure.

Cheers,

Pat.

Richard.L
24th May 2000, 08:07
Hi all methanol users,

I have found a good cheap source of Methanol from Cotswold Chemicals: Price comparison with Alcohols Ltd. All delivery is free to your door. Don't forget to buy some empty drums for dilution purpose.

- Alcohols Ltd on 01279 658464** 25L: 44.00*** 205L: 105.14

- Cotswold Chemicals 01453 825292* 5L: 7.70* 25L: 26.04* 205L: 61.00

RL

johnfelstead
6th June 2000, 14:16
Just read this entire thread for the first time.

Very interesting debate, the things that i am taking from this are that i can get a charge temp guage fitted for 50.
I am having one of those mike, bloody bargain of the centuary.

I too am a big fan of Ahmed, whenever i ask him for advise on any engine related issue he always makes complete sense to me.
He has helped me masively in uprating my cossie powered westfield giving advice on making the fuel system totally safe for track use.

He has also instilled in me the phylosophy of dont guess, test and find the real facts.

He has sold me a superb 350BHP conversion for peanuts money, and he has not tried to have me install a WI system until i have done some real world testing to see if i need it.
Once i have done some proper testing on track i will then have the FACTS to make an informed decision.

If it is proven that my engine would benefit from a WI instalation then he will supply me with a new ECU map for very little cost.

If i dont need it, then it doesnt get fitted, simple realy.

I have no idea how some of you have come to the coclusion that you need WI without doing some charge temp measurements, very odd!

My real world experience of WI to date is in running our group A cosworth rally car in top level UK forest rallies. We led one of the top national rally championships last year using a mountune high compresion spec engine with one of Ahmeds maps.
We ran the car using ELF's TURBO FIA fuel and had a different map available for when we ran on 98RON super.

The WI system makes a huge difference in stage times, we run the car with Ahmeds design of WI systems, it works!

At one service point last year we had a major problem on the car and i didnt have time to refill the WI tank before we ran out of service time.
This made a huge difference to our result, the stage times were 1 second per mile slower than we had been running before the WI ran dry, if that doesn't tell you something nothing will.

so my answer to the original question is simple, it depends. If you do some testing and you see a requirement to gain maximum safe power then its good. If you do some testing and you dont see a requirement then its bad. Simple logic isn't it?

I for one am looking forward to the day Ahmed has his impreza upgrades on the market, if he says it will produce 400BHP then it will produce 400BHP, none of the bullshit figures so popular in some circles.

RETed
9th May 2001, 23:59
Being from both the (big) RX-7 mailing list and the Haltech Support list, this discussion has been a raging debate of recent.&nbsp Hell, I found this due to a link posted in the Haltech Support list. http://bbs.22b.com/ubb/smile.gif

There's a lot of misconception of what water injection can and cannot do.

shiv pointed out barriers due to compression ratio and octane - well, what do you know - this is exactly what water injection was designed for.

Compression ratio and octane are closely related when in reference to pre-ignition and detonation.&nbsp Currently, most tuners throw more fuel in to suppress detonation or up the octane.&nbsp Yes, as shiv has pointed out, there is a limit where no matter how much fuel you throw at an engine, it's not going to suppress detonation - this is where water injection comes in.

To put it succinctly, water injection breaks the barrier which fuel octane holds you back.

Most opponents of water injection need to do more research before opening their mouths - this is the conclusion I'm come to when reading most debates on this very subject.

-Ted

gmonsen
10th May 2001, 04:16
hi, ted, richard... nice to see you over here ted. i notice you're here as well, shiv...

nice post, ted... it seems many in the us and new zealand like to think of water injection as a "band aid". they talk about a "properly tuned" car not needing water injection and proceed to launch into a discussion -- often knowledgeable -- of what a properly tuned engine can do. i just think it may reflect a closed mind to the adaptation of additional and new (at least to them) technology. a long time ago, intercoolers were in exactly the same position as water injection is today. it was a "band aid" to reducing intake temps. over time it became conventional wisdom.

if you "properly tune" any car to the max and run it on the edge you risk detonation. so, you can dial it back a bit for safety. optionally, you can add water injection and add the safety measure to keep running the car at a higher power level.

the rx7 guys have had no problem adding premix synth 2-stroke to the gas to improve engine lubrication, but some of these same guys say that its a big problem to add water... done right (water tank sizing) you add water when you add gas when you add premix ...when you run out of gas...

water does things intercoolers don't do and both can and should be used if you can do it physically and economically. i like to run my single turbo, well-intercooled (properly tuned i might add) rx7 at around 425-450 rwhp on pump gas. water injection will let me run it there more safely and reliably.

i'm fairly sure most others running their cars on the edge over here will be doing this in the next few years and want to thank richard lamb for his help in getting me going.

-gordon

Mark.A
10th May 2001, 20:13
Well bug*er me.....finally a couple of guys with some common sense http://bbs.22b.com/ubb/biggrin.gif

When I first posted the thread, I expected some people to question the merits of water injection, but not just to discount it, based on some mis-conception.

Common sense tells us that we need to fit the appropriate uprated parts, when tuning our cars, and to make sure an adequate ECU is mapped correctly.

As for "bandaids", well, we can only fit "so bigger" IC, size the turbo to our needs, and throw so much fuel (as you guys said) then what ??? Plus, water is much cheaper than fuel, it won't cause bore wash, and excess fuel loses the power we're looking for.

As for Richard Lamb, he must be one of the very few, who will go out of his way to help, and his service/support, is second to none.

Mark.