View Full Version : Big turbo upgrades

13th April 2000, 20:59
I have had a IHI VF22 Turbo installed today.
My last turbo a IHI VF23 suffered a leaky oil seal as a consequence of a very hot engine prior to the demise of my car! http://bbs.22b.com/ubb/mad.gif

I have noticed a minimal amount of lag and the turbo now kicks in 400-500 revs higher.

Does this mean that I can cruise at a slightly higher speed without the turbo cutting in, or is it all throttle position related?

Any way I have been told by BRdevelopments who sold me the turbo with a days notice that only when I get a mapped link will I really enjoy the benefits! Can't wait http://bbs.22b.com/ubb/biggrin.gif

13th April 2000, 23:25

the VF22 is a large-ish unit so I'de expect a bit more lag from it. Fortunately it will also flow a lot of air :)

With regard to being on boost while cruising... depends on your ECU. I have found a "bug" with the PossumLink whereby it suddenly builds up turbo pressure.... I never had this "problem" with the JECS.

The Link will try to achieve the boost targets programmed when the throttle position is at about 60% open. This happens to be about the figure needed for a 200km/h cruise... so while sat on the motorway, a minuscule movement of the loud pedal to compensate for a shallow incline actually causes the manifold pressure to rise from about 0.8 bar to about 1.4 bar [absolute, so 0.4 bar boost], with a resultant substantial increase in acceleration.

For normal cruise, you should be running manifold vacuum, unless it is a very high speed cruise, say 145MPH plus, then you'll need some boost to overcome the drag (which a a cubic function of velocity, IIRC).

It's then just down to the ECU to make an intelligent decision on whether to try to achieve the boost target or not.... the Link seems to employ an "on or off" strategy, but a graduated setup would be smoother and reduce the risk of det. Let me explain....

With the throttle cracked open to 60%, at 5500 RPM, it's about right for cruise. So there's no real need to try to achieve the boost target of 1.3 bar boost, since the extra air ain't needed. It tries to bring the boost pressure up to full whack, but of course the manifold pressure will be lower than that in the intercooler, for no other reason than the throtle butterfly causing a flow restriction.

Because these cars don't have two MAP sensors [one for fuelling and one for boost control], it would be easy for the boost pressure to be higher than desired. If you now hit the loud pedal, you will get momentary overboost, something which could be controlled by a momentary ignition retardation, but I can find no facility for this in the Link.

Another side-effect is increased turbo wear, since it is trying to produce boost, but the target it unachievable. The wastegate will be shut to try to meet the boost target, so the turbo will be going like the clappers, generating a lot of heat, and putting the bearing under quite some strain as it tries to push the air into the engine, against the restriction of the butterfly.

Of course, this could all be complete and utter bollox, and I could have just gone into "funny" mode again, high on cast iron vapour from my gas-flowing attempts on the Subaru manifolds earlier this evening.



14th April 2000, 09:29
In a nutshell, Yes Pat ! I too have noticed that it is very difficult to hold a reasonable quick cruise. My MPG went down a fair bit which I put down to this.

Bob Rawle
14th April 2000, 10:12
Sounds as though you guys are still struggling a bit, I can cruise at 5500 rpm at zero to minus 0.2 bar boost with no problem at all. Economy, well on a fast motorway run (I mean fast) I get circa 22 mpg. At 4500 rpm I can cruise at minus 0.3/0.4 bar vac. (all in 5th gear)

Perhaps you forgot that TPS proportions boost ? If the boost response settings are adrift then you get the results you describe.

I gather you've finally refitted the ecu now Lee ?

Pat, since your car and ecu are the same as mine then you will/should be able to achieve the result in the end. Bearing a floating sleeve bearing turbo the TD05H comes up more slowly so the temptation is to try and quicken it, however it then whips up and overshoots. Try and set it up so your torque peaks at circa 4000 rpm, that seems to give reasonable part throttle boost.

Lee a MY99 is even better in terms of part throttle boost control once the parameters are there.


14th April 2000, 23:04

You could say that I am "struggling" a bit :) Especially now that my lambda sensor has gone AWOL... must have damaged it while swapping it over to my new manifold..... they are usually a royal pain to do! Anyway, for the timebeing, shall source a replacement unit, until I can get hold of that nice Bosch wide band lambda sensor... will then fit that into the downpipe and close up the hole in the manifold.

I too can cruise at about 0.8 bar absolute, ie minus 0.2 bar to use your convention. It's just when the road starts to climb a bit, it needs more power and only very gently applying the loud pedal causes a massive surge, up to about 1.4 bar (ie 0.4 bar boost). As long as things stay steady, it's OK though....

With regard to fuel economy, it was about the same as yours. At the mo it is terrible, I have had to increase fuelling to compensate for increased volumetric efficiency [headers flow more freely, but are Subaru ones]. When the lambda was still working, I had to up MASTER FUEL by about 5% to compensate! This caused idle to go rich, but crusie / boost remained lean. Have increased a little further; it now "feels" right again, but fuel economy is terrible! When I get an iterim Lambda sensor, I'll adjust the fuelmix to get it "right" again :)

My unit does not *appear* to use any correlation between boost and TPS, other than at 60% it goes into closed loop mode and tries to achieve the boost target for 5500-6000RPM, which IIRC is about 230kPa. You may have newer software which has "smoothed" the transition out. Then again, I muy just be naff at mapping :)

Our cars are similar... mine "still" has the FSE on it, awaiting larger injectors. I just cannot get the fuel in with less pressure :-( I also have a set of gas-flowed headers which have made the car *quieter*.... but I did replace the heatshields with heat-wrap, so that probably drowns out some of the noise :) What does emanate from the back doesn't sound like any normal Scoob... but still very satifying :)

With regard to trying to "speed up" the 05H, yes I think we all would like it to spin up a bit faster, but I wasn't pressing the loud pedal all the way to the floor when I observed this discontinuity in power delivery... a mere 1 or 2mm movement was sufficient. But then I can hear my turbo start to spool up at about 2000RPM, achieving between 0.6 and 0.8 bar boost at about 3000RPM. From there on up it just "explodes" as it very quickly builds boost.... IIRC it was still boosting 1.3 bar at 7200RPM, but this was causing a very very lean mix, so we had to curtail it a bit :-( Still trying to control overshoot but I've yet to find a good set of paramters which achieve this.

Last measured, torque peak was at 4800RPM, IIRC, but the whole curve was quite smooth, no "lunar landscape" :) I'll try to move it down, but 4500-5000 RPM seems to be very prone to detting :-( Guess this is the engine's VE peak, so it's cramming in a lot of charge.... hoping that the addition of heat-wrap will cool under-bonnet temp, allowing a cooler charge, helping control this problem. Water injection does seem to help, also :)

More fun things to do, methinks.... and I've got to get that oil cooler plumbed in.... unfortunately the take-off plate won't easily sit on the current location :-( Perhaps with the "new shape" manifold it may clear now :) Failing that, a small sandwich should do the trick..... famous last words.....



15th April 2000, 15:54
I can cruise at 5,500rpm at circa 150mph http://bbs.22b.com/ubb/smile.gif

15th April 2000, 18:16
Of course, you could always try and find a more up to date design of turbo that allows for a reduction in lag AND more air flowing ability at the same time... Wouldn't that be nice! http://bbs.22b.com/ubb/wink.gif


15th April 2000, 20:19
Or a VF22-24 hybrid.
With my type R gears it does not effect driveability too much.

When the link is installed it will be a rocket ship http://bbs.22b.com/ubb/biggrin.gif

16th April 2000, 12:51

A VF22/24 hybrid, would you be refering to the VF23 there? The only differences between the 22,23 and 24 units is the size of the compressor inducer.

You can get hold of hybrid units with bigger inducers that are of a more recent blade design, so they work more efficiently than the older style kit. Imagine full boost by 3000rpm _on_completely_standard_management_ and flow as good as, if not better than a vf22. Hmmm... Nice!


16th April 2000, 12:54
PS: There may be _slight_ design differences from the 22/23 inducers and the 24 unit, but the 24 is definitly smaller and flows less.

16th April 2000, 17:54
Moray time to spill your rebuilt beans http://bbs.22b.com/ubb/tongue.gif

[This message has been edited by Anders (edited 16 April 2000).]

17th April 2000, 11:02
I'm sorry, I simply don't know what you could be talking about! http://bbs.22b.com/ubb/wink.gif

Turbo was latest casualty of the Power Engineering Phase One Conversion... bringing total cost of that "upgrade" to 3k+! http://bbs.22b.com/ubb/eek.gif Still, the car has been so much happier since I got the conversion removed, which is nice, I suppose... http://bbs.22b.com/ubb/rolleyes.gif

Of course, having to replace the turbo potentially gives you the chance to let the car catch up with modern turbo design techniques... oh well...


17th April 2000, 19:15
My upgrade experiment is costing over 5 figures and I still haven't finished shouting! http://bbs.22b.com/ubb/mad.gif

20th April 2000, 13:53
If you are spending 5 figures on a rebuild/upgrade, which ECU replacement are you considering? I suspect a Possum link may be a little out of it's depth here.

20th April 2000, 15:25
Anders, Moray is aboslutely right especially in view of the fact that he has not (as far as I know) had a 5-figure engine rebuild on his car and has, nevertheless, chosen to buy the only managements system that no one will ever be able to tell how deep can really be, and it is called MoTeC. I wonder which new genius will claim that he/she can map it any better than their rally Group N and A programs which are already too well known as unsuitable.

Look forward to new revelations!


PS. Perhaps Nicky Grist did wrong without knowing any of this. What do you think, Doc?

Bob Rawle
20th April 2000, 22:13
Ok, I'll bite

Moray, I have some first hand experience of Motec versus Link ... engine dyno work. I'll gladly share it with you .. mail off line. May help when you have yours done ..


20th April 2000, 22:43

Fallen for it again... http://bbs.22b.com/ubb/smile.gif

Dont do it.. Moray will cry.. lol


21st April 2000, 13:13

21st April 2000, 14:53
Ok, I seem to have upset or irritated Branko with my "out of it's depth" comment.

I was not intending to rubbish the Possum Link, I was simply suggesting that it may not be able to get the absolute best from Anders expensive new engine. Every ECU has it's limitations (Jecs, Link, MoTeC, Pectel, all of them), but some are more limited than others.

My point is that there are options other than the Link that have more flexibility/facilities which could squeeze more performance/reliability from Anders engine... for instance, you can individually trim the fuelling and timing on each cylinder with MoTeC, so you can put less stress on the known "weak" cylinder in an engine... http://bbs.22b.com/ubb/smile.gif

I assume that Nicky Grist has had a Possum fitted to his (roadcar) 22b and is happy with it. Good. As yet, I haven't heard of any of the WRC teams switching to Link ECUs for their cars, although I'm sure you will tell us when they do! http://bbs.22b.com/ubb/smile.gif

I didn't mean to caus offence, and I apologise if I have done this.


21st April 2000, 19:13
I have spoken many times with Bob about the link.

My friend Theodore had a Motec fitted to his 22B and had all sorts of problems.

An ECU is only as good as the mapper.

Bob is "the best link tuner" we have, I do not know who could tune a Motec as well.

I am fed up with my car being experimented on! http://bbs.22b.com/ubb/mad.gif

That is why it is now in the hands of tried and tested people! It will recover, I am not happy at the moment! http://bbs.22b.com/ubb/frown.gif Full storey in 22B chronicles, hopefully sooner rather than later http://bbs.22b.com/ubb/rolleyes.gif

24th April 2000, 18:21
Just out of interest, has anyone had a Pectel unit fitted to their Scoob, and if so, who mapped it / what benefit was gained?



25th April 2000, 19:47
Just got back from my Easter sabbatical.

As Moray and a few relevant others already know, no one was irritated and everyhing has been made crystal clear on the day when 'critical' posts took place.

We shall all look forward to hearing from Moray as and when he has some report to offer and the whole issue was totally unrelated to the product which seemed as if it could have been made look less capable than it is.

All good and well and very productive after all.


25th April 2000, 22:35
Welcome back, Quattro! :)

With regard to capabilities, I don't think that there's any doubt about the PossumLink's ability... it is by all counts an extremely capable ECU! I think that you know that I'm not one to take the cheapest approach, and the fact that I went for a Possum instead of the MoTeC or Pectel says more about my faith in the former than than my ability to buy the latter.

None of that alters the fact that there are certain features which the more expensive units posess which the Possum does not. You can, for example, map cylinders individually. There would be little point in mapping cylinders individually unless there was some inherent problem with, say induction or fuel injection systems; would it then not be more sensible to spend the extra money to correct inherent design problems rather than fudge them with ECU tricks? [I'm not saying that there are, it's just a hypothetical question].

AFAICS, the only time that you would need this functionality is where it would be impossible, for either economic, engineering or racing regluatory reasons to correct such design issues. A good example would be a maximum effort engine, such as the WRC engine. I seriously doubt that there are any fuelling problems with this, or that there are any other induction restrictions [bar the 32mm job] which might affect the engine in such a way as to need individual mapping. Looking deeper, there could be issues with flow in the water jacket round certain cylinders / areas in the heads, which would require a block/head redesign. The way round this would be to map cylinders individually [richen up the one with cooling probs, for example]. All very important when you're trying to shave 0.5 seconds off a 20 minute stage time, not so important for your trip down to Asda :)

The big question is where in the spectrum does Anders' engine fall? It has certainly been built with very high spec components and should take some very serious boost. Can any extra safety be gained by deploying a unit which can map cylinders individually?.... or perhaps more importantly, even if this were to be the case, could a tuner be found who could actually map such a unit in such a way as to provide a definitive advantage over a well setup Link [and I think we're all agreed that Bob is very good at doing that! :)]

Perhaps it is also worth considering safety margins... if you are mapping cylinders individually to get round non-induction/fuelling issues, the only reason you would be doing this is to get the engine as close to the edge of its performance envelope as possible while still keeping it intact.... but for how long? I seriously doubt that a WRC engine would last over 100,000 miles :) So if we're not going to push the envelope that hard, it could be argued that the tools which allow same are not strictly necessary.

In the end it's up to Anders to decide which way he wants to play this one.

Back on topic, this was a turbo thread originally... has anyone tried one of the HKS / Garrett GT series roller bearing / water cooled turbochargers? There are some interesting claims being made about these which it would be nice to verify..... if they really will provide full boost at 2000RPM and continue to deliver high boost all the way up to the redline, then that is certainly worth a look!



27th April 2000, 06:17
Check our website we have been offering the Garret G series for quite sometime.

greg www.bpmsports.com (http://www.bpmsports.com)

28th April 2000, 07:48
An email wrx buddy of mine has the 450G and he reckons it doesn't start pulling until well after 4000rpm. Who knows, maybe he doesn't have a good boost controller, so he is falling victim to 'wastegate creep'. My VF22 is seriously accelerating the car BY 4000rpm, but it is pretty hard to put an exact figure on the boost threshhold rpm (i.e. when positive pressure is seen in the manifold if the the throttle is nailed at, say, 1500rpm in fourth gear). Then again, I'm not that happy with my post 6300rpm top end pull, but I think that has more to do with camshafts than turbocharger sizing. Maybe my 450G buddy feels hard acceleration right up to 8000rpm (or whatever his redline is.) It's a complicated juggling act.


28th April 2000, 11:29
The hybrid unit I have mentioned achieves full boost by 3000rpm and pulls as strongly as the standard ecu will allow right to the rev-limiter (8250rpm). As mentioned before, this is on Standard ECU management.

Jan Shim
1st May 2000, 14:37
Moray, what size downpipe is your car running ?

2nd May 2000, 10:08
Twin dump(short) 3", high-flow cat in 3" mid-section.

3rd May 2000, 10:11
Similar to Anders, I have had a VF23 (Apexi unit..) and a VF22 fitted to my car...With a stock 2.0lt engine the major benefit in the VF22 is that with good tuning I had 1.2bar of boost at 3500rpm, as opposed to 3250rpm with the VF23, and it would pull all the way to red line and hold 1.5bar of boost...

After an extensive rebuild to a 2.2lts including a new set of Japanese Zerosports extractors, the VF22 is now on full boost at 3000rpm and feels very strong in the lower/mid rev range...Downside is that it is starting to operate outside it's compressor map at high rpm and I am losing boost at the top end as it can't flow enough air...The VF22 will only efficently flow about 320-350hp...

The HKS/Garrett units are good but the IHI sourced units are more efficent and smaller in physical size to boot...Apexi has an IHI based turbo for the WRX based on the AX series of turbos...These are rated in excess of 450hp and are a bolt on type of arrangement...

As for the VF22/24 hybrid, I would be very careful of these..They are not produced by IHI out of Japan so their operating parameters/compressor maps are totally unknown...They are made by fitting the smaller VF24 exhaust housing on to the VF22 compressor and core... The maximum rpm limit of the VF22 is lower than that of a VF24 so by placing a smaller exhaust housing onto a VF22 you could effectively "over rev" the VF22 shaft bearings and compressor wheel...Increased wear and possibly serious turbo malfunction could be the end result...Just a though....

3rd May 2000, 17:13

Which VF22/24 hybrid are you refering to?