[moved] Supercharged VR trackday issues

[Conky 0]

Hi chaps, just thought I would run a problem by you lot before the car goes in for a check up at stealth racing. Since I purchased my supercharged golf vr it has struggled to complete a full track session, it tends to loose total power after around the 3rd lap, if I coast around for a lap and let it cool down total power comes back, oil temps never exceeded 96c and water temps are low, and my aquamist system is working correctly. I've been told it could be an air intake temp issue,

I've checked pretty much every sensor on the car with my dmm and had diagnostics on it which throws no errors up at all. Coil pack, plugs, leads and maf have been changed, but made no difference. Fuel pump and injectors have been upgraded and ecu mapped accordingly.

This car was built purely for track use and I was assured it would take all that a track day could throw at it. Before the car was sold to me it was converted to manual (previous driver was disabled) so maybe the changeover to manual doesn’t bode well for engine loads and doesn’t suit the current ecu map.

Here is a rundown of the main engine specs:

-Engine fully rebuilt and balanced by Vince @ Stealth Racing

-ZR2 supercharger running a 6psi pulley

-JE Forged Pistons and bored to 2896cc

-ARP fasteners used throughout

-polished head, schrick 268 cams, schrick variable inlet manifold

-custom remapped ecu, larger injectors, 4 bar rail.

-6 branch and racing exhaust system.

-Aquamist setup

would you chaps maybe suggest I run an intercooler or chargecooler to reduce IT temps?

I’d appreciate your input on this one.

Cheers

[apd 0]

never been on the track with my sc vr,but the previous owner did-with no performance issues.running a v2 with similar mods.i would suggest heatsoak/or like you say inlet temps.does the z sc have a restricting air filter underneath the airbox area?i have fabricated 3 cold air feeds piped to the sc air filter.the air filter bolts straight on to the v2-and just fits due to limited room!its quite a large k& n unit designed for competion use.vince commented that getting more airto the sc is a plus.hope this helps!

[Conky 0]

I have a large Ramair filter fitted which is located down on the bottom right side behind the bumper, it is completely out of the engine bay and has a direct supply of cold are from a vent opening in the bumper, so it sucks in the coolest supply of air possible I would imagine, probably could get a wee bit dirty but the car is a trailer queen and only ever gets used on the track so not a real issue.

I'm about to fit a Mocal oil cooler even though oil temps imo are very good anyway just as a preventative measure and remove the nasty standard oil cooler.

Also looking into a compact charge cooler setup to try and reduce intake air temps if they really are an issue.

tbh I'm hoping Vince will find the problem when I take it up there next month.

[SankysVR6 1]

Might get a better response here

[FishWick 21]

I can only see the ECU cutting power if the IATs get stupidly hot, say 90 degs or something. I think the OBD2 ECU only adjusts advance against air temp and Vince said the settings don't make sense, so he never touches it. Not sure what OBD1 gets up to when the motor gets hot under the collar, but Vince will know.

Physics dictates that there is only one air temp compensation map for all petrol engines, and that is for every 10 degree increase in air temp, oxygen content reduces 3%, so you reduce the fuelling by 3% accordingly.

In theory if you hit 90 deg intakes (which I have seen once or twice on my engine) the above theory suggests you need a 21% reduction in fuelling, which is a lot. Knocking off some advance would be wise too. I have tried fuel and advance compensations based on this theory in my DTA and it does work well.

How about the water temp sender? Could be faulty and over reading? I can't think what else could cause a power cut after 3 laps! Very weird.

You need to monitor the IATs on the track really, but bear in mind the air coming out of the charger will be at approximately 70 degrees. VAG com can monitor it but the VR6 air sensor is rubbish and takes 15 seconds to register a change in temp. Fitting the Audi 1.8T sensor as I have is a much better solution as it's instant and more accurate.

I use Schimmel's compact chargecooler system, which you can see in my Turbo fitting thread, or at www.spturbo.com

www.chargecooler.co.uk sell an alternative as the SPTurbo one is very expensive, but very effective. Intake temps average 4 - 8 deg above atmosphere on a cruise (and on boost - turbo in my case) and only ever getting as hot as 45 deg intakes when sat in traffic for hours (compared to 80+ on stock engine) with no airflow over the prerad. It's an impressive peice of kit and has 0.25psi boost loss, so throttle response won't suffer.

It is however designed for Schimmels log manifold though and the pipework length would increase a lot using the stock manifold.

Vince has good experience of the Schimmel stuff now and can probably source and install it for you.

[rodney 17]

I note you have a 6 branch, Is it wrapped or the likes????? If the engine bay temp gets too high then you will start evaporating the fuel before you get to ignite it. The coasting lap allows temps to drop enough then all will work ok.

The issue may be your under bonet temps with all those heat sources. Run a lap or 4 with the engine lid off and see if that cures it.

Also is your air filter sectioned off from your heat sources??

Hope they help

[Conky 0]

Hi kev, thanks very much for your reply! Sorry I forgot to mention it’s an obd2 engine.

I wouldn’t have thought iats would be to much of an issue with this low boost supercharger setup, just can’t see them rising to 90c. I’ve tested the ect sensor, and its output resistance at different temperatures is within tolerance of the technical data I’ve been referring to.

not sure what you mean about switching the air temp sensors, surely the poor resolution of the sensor is down to the standard ecu, also the location of the sensor doesn’t seem to be in the best place on the Schrick manifold.

Hi RCF, that’s correct and part of the plan. The throttle body seems to get silly hot from the heat radiated from the exhaust manifold, this can’t be good so I plan to wrap it up with some nice thick exhaust wrap.

also as mentioned I plan to fit a large 19 row Mocal cooler with the specific Mocal vr6 thermostatic sandwich plate, and try out some Silkolene race oil which people say tends to run a good few degree lower under hard track conditions.

Cheers chaps

[FishWick 21]

Yeah that's a good point made by RCF, I didn't read your spec list to notice the 6 branch. I've not heard of any other track VR6 S/Cs getting that prob though. The manifold does need wrapping tbh. OBD2 is the best VR6 ECU, so I'm surprised it's giving you grief. I'm sure there was someone else on here recently with a similar power cut issue and it was something to do with the MAF signal wire shorting out?! I thought he was you at first....very similar style of typing.

I use Silkolene Pro S 10W/50 and it's a great oil, the VR6 seems to love it. Mocal works great too, as does Water Wetter.

No mate, the IAT sensor's slow response is because of the sensor itself, not the ECU. It's an old fashioned enclosed design and just heat soaks like mad. The 1.8T sensor is an exposed thermistor and sits right in the air flow and responds instantly. I have seen the differences by trying both types on my standalone ECU and the air temp compensation map works a lot better with a faster IAT sensor.

Not sure, best person to try would be Vince as he seems to have a hand in the car's build and may know more.

[Conky 0]

Pretty sure that thread was me chap, that was the 1st problem that had to be sorted. When I bought the car the loom that ran through the bulk head was fried and the +P, neg and signal wires to the maf were shorting pretty much their full length in the nest of wires. tbh this car has been a real pain to sort out, it rarely run right, I seem to have zero luck with coil packs and this power loss on track is pretty confusing.

My old obd1 rado vr I had a good few year ago ran wonderfully, never missing a beat even on the track it held up well, I rember getting a recommendation to use water wetter and that really helped cool the engine better. this is the 1st pure track car I’ve had and it’s a real nightmare 90% of the time, the other 10% showed me what it could do on the track and that’s keeping me going, will persist!

I run Pro S 10W/50 when I track my 200sx road car, but will switch back to 5W-40 soon when the weather cools down. temps went way cooler when I switch from mobile 1, so I'm hoping it will do the same for the vr golf.

With regard to the 1.8t IAT sensor, is it restive and will it work with the standard Motronic ecu?

Cheers chap!

[Conky 0]

Quick update:

Over the weekend I wrapped the exhaust manifold up tighter than an egyptian mummy with about 200ft of heat wrap! and shielded the inlet manifold.

Today we took the car to Stealth, the car ran pretty poorly on the way up to Warwickshire, granted it was pretty hot weather, but the car was certainly not behaving itself.

Vince popped the VR on the rollers and plugged the ecu into his pc and logged the output values on a full throttle run. 1st problem we noticed was the engine was rocking pretty excessively on its mounts, this could be felt from in car but visually looked bad. Typically they would be expensive and fairly new Vibra technic mounts!

Anyway he continued to dyno the car, it seemed to pull well but the torque and power plot it produced was quite ugly with a huge dip at around the 4k mark, after that it seemed to pull well giving a corrected power figure just under 300bhp. Vince felt the dip was due to the Schrick inlet manifold switching its flap over, I forgot to mention a while ago I disconnected the vacuum line to the Schrick actuator forcing the flap to stay shut just to test manifolds effect at this rpm region, but this made no difference to the loss of power at 4k, I still believe its coil pack even though we’ve tried 3 of them, I just don’t trust these Beru ones.

Vince inspected the log file VAG.com had produced and he said everything looked good, fuelling and ignition was spot on and didn’t need tweaking. After asking, he showed me the logged values VAG.com got from the knock sensors, the site regions around peak rpm indicated a little bit of det, Vince said as long as these don’t exceed a value of 12 there not to worry about. Is it normal for boosted VRs to see a tinny tinny bit of det at around peak rpm??

Also AITs were around 60c on the dyno.

Anyways, before Combes September event I plan to have a nice new Temic coil pack fitted, Mocal oil cooler, low temp stat, better oil, water wetter and maybe a suitable bonnet vent plus maybe try it without the drivers side light.

Hope it runs a full session this time around!

[xyber 1]

Ive just fitted a low temp stat, 50/50 mix of water/g12+, and fresh oil and was well impressed on how much that kept temps down, even from when it was standard, well worth doing for the few quid it costs to do! Oil cooler goes on this weekend.

I was also considering a lancia integrale style bonnet vent as it was nice and simple looking

[FishWick 21]

Any more luck Conky?

The 1.8T sensor - Unfortunately that doesn't work on Motronic as it's too fast for it. A mate tested it last weekend and on VAG-COM it intermittantly throws up a fault. That's a shame as it works a treat on a standalone ECU.

60 deg IATs.....that's a smidge high, needs to be lower than 45 if possible to get the most out of the engine, but as you're only low boost it should be OK.

VAG-COM's det retard is a little confusing, it's 12 'units' of retard, not degrees. I have no idea what that is in actual degrees!, but some folk reckon you lose about 10hp with every degree of retard on force fed engines, but that's probably a little extreme in your case as you're not running huge boost. My S/C install (also OBD2) was hitting only 6 units of knock with 12psi on a cold run, but after several runs it was heading toward 12 also because of the heat!

Believe me mate, the TEMIC packs aren't much better. I've got a Vauxhall Omega coilpack (Bosch) at home and the high tension side looks almost identical to the beru one, so I'm going to try and fudge it into the Temic amplifier casing and let you know how that goes. The Bosch packs rarely go wrong.

Sounds like it's making the right numbers though, 300hp is good going.

[Conky 0]

Hi Kev, I’m sure your right, I recently read that both the Temic and Beru’s use the same Telefunken internals anyway. If only I could use a Bosch motorsport coil pack! the 6 cylinder version is very reasonably priced.

Yesterday I got hold of a copy of VAG-COM and a compatible cable and went out to do some data logging.

I manly concentrating on IAT's, Ignition angle and Knock I did a few full throttle runs through gears.

The logged data indicates IAT temperature when moving doesnt really go much higher than 40c, Low speeds see around 46c. Worst temps are when static, but as soon as you move they reduce.

My only concern is the values VAG-COM is coming out with for Knock, like you say its a shame the values from VAG-COM for knock isn’t more meaningful. On the stealth dyno the car stayed around the range Vince thought was safe, but after yesterdays tests they were already exciding this value.

Vince feels the tune is very good on the car, but I can’t help but feel its running slightly to aggressive ignition timing forcing it to be a little det limited on hot days.

As far as det is concerned I always thought that non is an acceptable amount, well its acceptable for damage, but not maximum power, how much of a negligible difference it makes in the case of my supercharged vr is another thing though, might just be how it works.

I don't know how much the stock knock sensors can be trusted, whether their actually detecting det or just some random harmonic at high rpm's, but I don't feel it is the later.

xyber, thats good news! I have a list as long as my arm for bits to aid in the cooling of this motor lol just taking to long for the bits to arrive

Sorry my posts are so long and no doubt boring! thanks for taking the time to read them though.

Cheers

Ade

[rodney 17]

I must admit I will be adding the extra oil cooler myself to help things. I already have the pink water to help me!

I do have a recollection that as the 300BHP starts to arrive intercooling was thoght to be a good idea. Any Thoughts on that? whilst you loose a little power it may be a benefit/help. Bit of a major step tho'

One thing that came to me in a moment of sleep, was have you flushed your system and if so how much fluid etc did it take back in?? Wondering if there was a restiction possibly or if the Radiator was on its' way out ?!?!?!?!?!?

Keep plugging away

[Conky 0]

Thanks for the reply, tbh the coolant system has not been touched yet, I’ve visually check it and thats about it. I was told it had been flushed out and coolant changed not so long before I bought it.

With any look next weekend I'll have a new water pump and uprated auxiliary electric pump which I plan to put on a switch (normally only comes on at 107c) but having it come on at lower temps may help up the water pressure a wee bit and aid with flow and alleviate any hot spots. I’ll flush out the rad when the coolant is removed, visually it looks new.

Defiantly a method of cooling the intake charge is on the cards, but yes its a big step. Something like the setup Kev runs would be perfect. Im just a little concerned about the possibility of a pressure drop the cooler and additional pipe work will create. No doubt fine with turbos, just up the boost a little, but I’m concerned I will lose what little boost the charger produces and consistently loose power. with any luck lowering the charge temps closer to ambient will increase air density and make up for the lost boost. Non of this I'm certain about, just assuming.

Cheers

Ade

[Conky 0]

Hi all, yet another update

Well it was a bit of a rush to get the car ready for Castle Combe but we managed it, mainly waiting on a Corrado vr6 metal heat exchanger end cap so we could complete the oil cooler installation.

so basically what we got done is as follows:

-heat wrapped the manifold

-ran a mix of water and water wetter, no anti freeze

-fitted a low temperature thermostat

-installed a huge 9 series Setrab oil cooler with Mocal thermostatic sandwich plate

-put together some hinge plates to raise the rear of the bonnet to add with cooling (fitted just before I went on track)

-Switched from Castrol RS to Silkolene Pro S 10W-50 engine oil

-fitted loads of reflective aluminium sheets around intake pipe work.

-the removal of the drivers side head light just before we went on track.

-new coil pack, spark leads, plugs.

The night before the event we went for a little test run to check all was well, temps seemed very promising, constant heavy acceleration after heavy breaking didn’t even budge temps.

Next morning one the way up to Combe we realised the engine was running a little to cool! Motorway cruising provided an engine oil temperature of just 68C! a little confusing as the Mocal thermostatic sandwich plate should only let the oil cooler kick in at 80C, I pulled over and covered the oil cooler, which helped to rise the temps to around 82c.

First track session, the car ran well initially (had my laptop hooked up logging data from the ecu) 1st lap power was strong! Second lap was the same, what’s this? not again!! yup it started to loose power big time!! nova fly’s past, struggles to hit a ton down the pit straight, out of quarry corner the car spluttered and simply wouldn’t pull.

The oil temp reads just 91C, I switch the heaters on full blast just to see if this helps and coast for a lap, then back on the power, for the last few laps the car improves vastly managing to hold power for the remaining 4 laps, still not great but definitely better.

Second session we go out with the heater fan on full blast, the car seems to run significantly better, almost lasting the whole session, only loosing power on the last 2 laps.

Once over I ran through the logged data, water temperature never went over 76C at any point on the track!! oil temp peaked at 92C and air intake temps once moving never went over 37C even though it was a mighty hot day! Knock values were low, much lower than previously.

I couldn’t work out why running the heater fan helped so much if the temps were so good in the 1st place, it’s worth noting my auxiliary electric pump needs replacing and has been known to stop sometimes. The outlet of the heater matrix runs directly into the head, maybe the top end is running to hot? hotspot issue?!?

It looks like I still haven’t cracked this loss of power problem, but it did run better on the track yesterday.

Cheers

[rodney 17]

Great to see some progress being made

Just sort of thinking out loud here so bear with me.

When the heater is turned on it takes heat from the water system which would indicate that the water is running hot. That said the numbers on the PC sound great.

The fuel goes in at the front and the heat source is at the left (AS you look from the front) and the rear.... The water is taken into the heater matrix on the right rear (AS you look from the front)

so this would still hint at temperature still being the cause. The running of the heater would drop the engine temperature enough to allow detonation to occur correctly again..... Are the plug gaps correct????

the thinking is that heat basically will cause 2 main things.... there be too hot a mixture so the mixture vapourises on its way rather than be mixed correctly inside the piston (cold air more dense and better bigger bang....) heat will cause metal to expand so other than the plug gap altering what else can move and then not create a spark???? Sorry I forget are you OB1 or 2??? Dissy contact/rotar arm is another thought and where is your fuel line route???? past these heat sources?????

Sorry for it being a little bit rambling but this is quirky one

[Conky 0]

Hi rcf, Im running BKR6E plugs with standard gap, Ive been lead to believe the gap only needs changing when running well over 300bhp.

I think its about time to wip the head off and check the head gasket, might be holes unpunched in the gasket or something ridiculously obvious like that.

[rodney 17]

rather than whip the head off -- check your compression and if you can get the use of one of those sniffer things you use on the expansion tank (checks for CO2) then that may save you some work or at least confirm... Are you using water at all???

Speaking of water i just reviewed a little.......What is your mixture in the aquamist and does it run always or only come on at certain pressure??? Also how retarded is the engine running?? (could be on the edge and is detonating premature if you see what I mean...)

WHen the engine looses power does it kinda die or is it more like a misfire when you accelerate?? (Relates to engine timing......)

I dont think the timing is wrong just grabbing at things which effect the firing...Also what did it show as knock count I seem to recall if the knock gets too high it goes into a default setting until all works again.... Might be very wrong on that... I know other makes do it

I assume it was going well on the other laps then

[Conky 0]

Compression is very good, a week or two ago I check all 6 cylinders and they all read a healthy 13 bar.

The Aquamist pressure sensor is set to switch on at 4bar which works out at roughly 4500rpm, Im using the blue jet and running a 20% Methanol/water mixture.

I’ve got the logs on my laptop and will see what the ecu has been doing with ignition timings.

The car doesn’t misfire, it just loses power and refuses to accelerate, the engine still feels smooth, but just gutless.

[rodney 17]

MMmmmmmm that all sounds as it should be. (except the last bit ) With it still being smooth it does sound like it is being controlled correctly ..

Can anyone else confirm for us if the engine has a kinda safe mode when knock gets too high??? If it does what does it do???

[Conky 0]

This is what it says im my vr6 workshop manual...

*** Knock Sensor I (G61) And Knock Sensor II (G66) ***

Two knock sensors are used. A knock sensor works like a microphone to

"listen" for spark knock or detonation.

When knocking occurs, the ignition timing is retarded until the knocking

is eliminated. Since the knock limit differs from cylinder to cylinder

and changes within the operating range, knock regulation is done

cylinder

selectively.

* Signal usage: *

Knock regulation does not occur until the engine coolant temperature of

40° C (104° F) is reached. Knock sensor I (G61) monitors cylinders 1,

2,

and 3. Knock sensor II (G66) monitors cylinder s 4, 5 and 6.

With the aid of the Hall sender signal, the ECU can determine which

cylinder is knocking. The ignition angle of the knocking cylinder is

retarded in steps until the knocking stops up to a maximum of 12°.

If spark knock is still detected, the ECU will retard the ignition

timing 11° for all cylinders and record a fault.

* Substitute function: *

If a knock sensor fails, the ignition timing angle of its assigned

cylinders is retarded.

* Self-diagnosis: *

The ECU recognized an open circuit if no signal from knock sensor I

(G61)

or knock sensor II (G66) is received by the ECU at an engine coolant

temperature above 40° C (104° F).

My ecu has never once recorded a fault specific to knock.

[rodney 17]

Interesting..... If I read it correctly then you only get a fault code generated if all are retarded. So you can have an individual cylinder retarded and not know about it. Well if we assume that the 2 cylinders next to the charger (heat source) are retarded then that would still run and just have the power (kinda) of 4 not 6 correct?? until the knock count comes down then all would be back to normal. I dont suppose you have a running log do you to see if a couple or more cylinders were altered...????

Do you use this car only as a track day car??? I was wondering if you get the issue if you drive it around for an hour?? can you induce the same fault if you unplug the fans and let it run on your drive/front of the house... Obviously watchin the temps etc.. If you gotta second seat in the car a m8 could be in in your next track day and watch the specific codes etc.... (good excuse to go again )

[Conky 0]

So it seems yes, vag-com is pretty useless at displaying ignition advance, the refresh rate is just to slow, half the time im not sure what to make of the values it produces for ignition timing.

power loss only ever happens on the track, thats the only time I can really push the car.

[Conky 0]

Just fitted a new auxiliary pump, its quite a bit more beefy than the old worn out vr6 one. The Bosch part number is 0 392 022 003, normally fitted to an AMG SL55 coolant system. Pretty easy to fit, you just have to put together a relay board to supply the pump power as the standard wiring isn’t up to the current requirements this new pump needs. outlet sizes are the same.

Got a meeting at Donington park next month so that’s the next test