Supercharged Stg2 VR6 Stalling & Overfuelling when cold

[antera309 0]

Sorry if this topic has already been covered - I tried looking for it in the "Forced Induction VR6" thread, but my eyes started to wear out at around page 90.... :hic:

 

At the last remapping session, Vince @ Stealth was able to get 280BHP from my Stage 2 supercharged 2.8 cabrio. Once warmed up, the car runs lovely and pulls like a train, but it's still overfuelling badly when cold - really badly - it absolutely reeks of fuel and floods very readily.

 

I've spoken to Vince about the problem and what's happening is that the ECU is upping the fuelling by a fixed percentage whenever it gets a low reading from the coolant temperature sensor. This is by design (it's effectively the "choke" function). The trouble is, because I have bigger (red top) injectors in there, the set percentage equates to a lot more fuel than it does with the standard injectors.

 

Vince says the "% increase when cold" ECU parameter is fixed and cannot be changed on the rolling road, so the answer lies with frigging the output from the coolant temperature sensor somehow.

 

Have any of you with forced induction VRs come across the same problem? I'd be very interested to hear about what solutions have been tried and what does and doesn't work...

[CorradoVR6-Turbo 0]

Was your mapped with the Map or without? Sounds like it was done mafless.Mine was done without the Maf and it was rich from cold start up,so much so my eyes hurt if i ran it indoors,i would have to hold the throttle for about 30+ seconds before it would idle normal ish...

 

Mabey changing the temp switch for a new one to see what happens?

[antera309 0]

The MAF is still connected on my setup. With the MAF disconnected, the cold idle problems were worse. That's why we kept it.

 

I guess it might be worth trying a new coolant temp sensor, but according to VAG-COM the current one is giving the expected readings.

[CorradoVR6-Turbo 0]

Is it giving correct readings when cold? They are only about 12 quid so its worth trying?

Vince dosent know how to do the cold start/idle settings yet,this is something he needs to figure out as most supercharged VR's suffer from this fault.

That why im not using the OEM ecu for the turbo as im wanting MAP sensor for better running all round.

[eatthis 0]

im running obd2 with the maf and dont have any problems is urs obd1 or 2?

[H8RRA 0]

mines fine on a cp ecu as well - apart from idle on startup for a very short time but i've got no MAF, ISV or throttle body heating and it still copes...

[KADVR6 0]

i have never been 100% happy with mine since the re-map. my car really hunts for the revs for about 1-2 minutes, and you can really smell the fuel. but as soon as its warm its perfect. but park up for about 5-10 minutes, engine is still warm, but it still hunts for the revs.

 

when i have a spare day i will have to go and see vince.

 

karl

[antera309 0]

im running obd2 with the maf and dont have any problems is urs obd1 or 2?

 

Mine is OBD1 and the MAF is stil connected.

[eatthis 0]

im running obd2 with the maf and dont have any problems is urs obd1 or 2?

 

Mine is OBD1 and the MAF is stil connected.

 

i think obd2 is better for chargers as obd1 doesnt give as much info so vonce has to guess a bit but ask him hell tell you properly (i may be wrong but i thought thats what i heard)

[antera309 0]

I replaces the coolant temp sensor and it didn't make any difference. I ran up VAG-COM and this bears out Vince's assessment - when the engine is cold, injector times are running at ~5ms at idle. When the engine warms up, they fall to ~2.7ms.

 

I'm very surprised that no-one else has noticed these problems, even those with MAF-less OBD1 setups. Not sure about OBD2 - that might work differently (although I doubt it).

 

The temp sensor has a resistance of about 300 ohms at normal running temperature. I think the answer might be to hook up a resistor that will fool the ECU into thinking the engine is warmer than it is and hence get it to put less fuel in (just like those ripoff resistor chips on eBay, except in reverse!)

[Kevin Bacon 5]

Most people live with it, I did. Mine was OBD2 and it ran stinking rich for a few seconds, but OBD2 manages the fuelling a lot better and didn't stink for as long as OBD1 does.

 

VAG-COM and Mapping software etc deciphers the streams of raw hex data into meaningful english as you probably know. The problem with the software Vince has got is it only deciphers the main fuel and ignition load sites. Things like cold start fuel and timing, air and water temp compensation etc are not translated. And unless you understand raw data, data which changes every millisecond, it's impossible to pick out the cold running parameters.

 

I know from experience that Vince has spent days and days on this cold starting issue that so many complain of (unpaid) trying to figure it out. Give him a chance. He's still working on it inbetween day-to-day jobs and he will definitely sort it out!! You have to remember how badly these kits ran with the supplied GIAC software in the first place and how far forward Vince has taken the drivability.

 

We've discussed it so often and at great length. We've played with MAP sensors unsuccessfully and various other tweaks failed. The bottom line is the stock ECUs aren't up to it and the red tops are too big.

 

Cynical marketing from VF Engineering fills people with false hope that they run like stock. They don't.

 

If a little morning roughness is bugging you that much, I'd say investigate a standalone. DTA S60 or the new Emerald will be perfect. Or, wait until Vince comes up with a good fix. You can also do a bit of invesitgation yourself and see how folk in America (Vortex) get round it?

[Walesy 0]

Vince @ Stealth was able to get 280BHP from my Stage 2 supercharged 2.8 cabrio.

 

:shock: - I bet the scuttle shake is fun!

[Guest]

I'm also not running a MAF, and don't get any cold start problems, and it idles spot on! It does drive a bit lumpy for the first 5 mins though...

When warmed up, I do sometimes get the idle hunting a little when slowing down to juctions, but as soon as i stop, the hunting stops!

Still not entirely happy with not running a MAF, as the car is a bit unpredictable and unresponsive. I've been meaning to have a play with the MAF connected, and mounted in different locations, but just haven't had the time...

[Kevin Bacon 5]

The MAF is the engine's load sensor, without it it relies on the lambda and TPS for part throttle fuelling and fixed tables for full throttle, so throttle response will be a little flatter. In many cases, the benefits of not having a MAF outweigh the negatives, such as short life span and stalling when they operate out of range.

[antera309 0]

I know from experience that Vince has spent days and days on this cold starting issue that so many complain of (unpaid) trying to figure it out. Give him a chance. He's still working on it inbetween day-to-day jobs and he will definitely sort it out!! You have to remember how badly these kits ran with the supplied GIAC software in the first place and how far forward Vince has taken the drivability.

 

We've discussed it so often and at great length. We've played with MAP sensors unsuccessfully and various other tweaks failed. The bottom line is the stock ECUs aren't up to it and the red tops are too big.

 

Cynical marketing from VF Engineering fills people with false hope that they run like stock. They don't.

 

 

I've been in the modding game for years and I am WELL aware that there is no such thing as "plug & play" modding, or at least not when you're talking about 100BHP+ increases.

 

Vince has been very open and honest from the start that he is still learning when it comes to mapping these forced induction installs. The fact that there are only two niggles remaining (the stalling & the cold overfuelling) is actually not bad going. But I do think we should be looking into how to resolve these problems rather than just putting up with them. Any problem can be engineered around - I refer you to the chorus of people who told me that my flap-less VSR manifold was nothing but an expensive paperweight...

 

Vince agrees with you that the redtop injectors are too big for the application, chosen by VF-E simply because they are cheap. There are plenty of intermediate size injectors available, but testing these would be a very expensive game of trial & error that no-one is gonna want to play.

 

I remember the old Courtenay Turbo kits for Vauxhalls used to get around the problem by retaining the stock injectors and adding a 5th injector that only cuts in at high revs when the extra fuel is needed. Wet Nitrous kits work on a similar principle. I'm not sure whether the VR6 ECU could handle a 7th injector setup (possibly using an external control box a-la Schrick manifold), but it's worth looking into.

 

The subject of the MAF comes up in every discussion I've read on s/c VR6s and my feeling is that people have been willing to ditch their MAFs (and endure all the problems inherent in doing this) too readily. At the end of the day, engines with FI from the factory (such as the 1.8T) run MAF sensors without any problems at all. The 1.8T MAF doesn't start sending screwy values to the ECU whenever the recirc valve operates, so why should a supercharged VR6 be any different? Are we just talking about some re-engineering of the pipework here? Could a MAF from a different car (e.g. 1.8T) be used? and if not, why not?

 

Coming onto the cold running problem, this seems to just be a matter of manipulating the output from the coolant temperature sensor. Vince agrees with this approach. He suggested a switch on the dash to switch the ECU input between 300 Ohms and the true sensor resistance, but I'm gonna go one step further and put together a circuit that handles switchover automatically. I've got draft 1 of the circuit diagram in front of me right now! Wish me luck!

[Kevin Bacon 5]

At the end of the day, engines with FI from the factory (such as the 1.8T) run MAF sensors without any problems at all.

 

The 1.8T MAF doesn't start sending screwy values to the ECU whenever the recirc valve operates, so why should a supercharged VR6 be any different? Are we just talking about some re-engineering of the pipework here? Could a MAF from a different car (e.g. 1.8T) be used? and if not, why not?

 

Well not quite. Early 150 1.8T engines can be easily destroyed through a MAF operating out of range. Ask Vince, he knows only too well about that, as do I ;-)

 

225 1.8T engines have MAP, MAF *and* EGT, so plenty of contingency options, but MAFs on their own have always been Schitt, regardless of engine or application. The only exception is Hitachi MAF sensors, they never go wrong and it's a common mod for Audi S4 owners (2.7 biturbo).

So if you're looking for blame, blame Bosch.

 

What screwy values are coming out of the MAF when the recirc opens? I thought we agreed that was down to the spring tension in the Recirc? Never had that problem with mine when an appropriate spring was fitted.

Besides, the values aren't screwy, the MAF tells the ECU exactly what it sees, and the VR6 ECU was never designed to deal with forced induction.

 

1.8Ts were, simple as. They have boost compensation maps among other things, which in FI, is absolutely KEY to how the engine performs. VR6s obviously have no way at all to compensate for boost as it's not expecting to see boost.

 

Yes a larger maf body will solve a few issues, but the MAF tables need remapping, which Vince says you can't do on OBD1, but I think you can.....ask IBIVR on the VR6OC.

 

Vince is but just one man with other work to do. Bosch pay teams of 100s to develop their ECUs. GIAC couldn't help. AMD couldn't help. So I personally think what Vince is doing is commendable. He didn't want to know originally....until Raj got involved.

[antera309 0]

What screwy values are coming out of the MAF when the recirc opens? I thought we agreed that was down to the spring tension in the Recirc? Never had that problem with mine when an appropriate spring was fitted.

 

Vince explained to me that this is what causes the stalling - hot air being dumped onto the MAF by the recirc valve, which in turn messes up the MAF's signal to the ECU. Is this not correct?

 

I'd be VERY interested to hear about the spring tension mod you mention - I was not aware of this.

[Kevin Bacon 5]

Ah.... could be in your application if the MAF is downstream of the charger?

 

With the Vortechs, the MAF is upstream of the charger, well away from turbulence and heat and the recirced air is just a loop between the charger discharge pipe and the intake before the impeller, so the MAF never sees the discharged hot air - in theory.

 

The theory of the springs was that the supplied spring in the Recirc was a bit too weak, and stiffening it up solved some of the air turbulence, it helped on mine..... but the main problem was just the increased air flow through the MAF confusing the ECU. A larger MAF housing (4") and remapped MAF tables is the way to go imo. Should smoothen everything right out, but I wouldn't know how to do it on Vince's software. Not sure if he does either tbh.

[antera309 0]

Cheesewire, my setup is a standard VF-E stage 2, with the MAF upstream of the charger.

 

Your comments make perfect sense, but they paint a very black picture of forced induction on VR6 engines, at least in applications where the stock engine managment is retained.

 

I totally appreciate that the stock VR ECU isn't designed for forced induction, but both of the problems I've described occur at idle, when the charger is not producing any boost!!! On-boost it actually behaves very well!

 

The question I put to the forum is this - does anyone have an supercharged VR6 which:

1) is running stock OBD1 management

2) has the MAF connected

3) doesn't stall at every roundabout & T-junction.

 

?

[buttles 0]

Antera, on the -AG ECU that doesn't work. Tryed it and the car is worse. Only thing that resolved the problem is upping the Throttle setting to 15.4 with a MAF fitted. My car now runs as STd. Lucky me :)

[buttles 0]

Antera, just confirm which ECU you are running. Answer to 1,2,3.. Yes, Yes, Yes

[antera309 0]

Antera, on the -AG ECU that doesn't work. Tryed it and the car is worse. Only thing that resolved the problem is upping the Throttle setting to 15.4 with a MAF fitted. My car now runs as STd. Lucky me :)

 

Sorry, what was it that didn't work? Rigging the output of the temperature sensor to fix the cold overfuelling?

 

I have been experimenting with adjusting the throttle stop to up the idle speed in an attempt to make the engine less willing to stall. No luck so far, but I haven't gone as far as 15.4 degrees yet.

 

I have the 021 906 258 AF ECU - it's very similar to the Corrado AG one.

[Kevin Bacon 5]

Where have you put your MAF? Directly under the charger intake? If so, there's your problem. Get a length of flexi hose and move the maf as far away from the intake as possible and also make sure the MAF sits as it would in the engine bay, i.e. horizontal with the plug pointing at 270 degrees.

 

And how do you know the MAF is actually any good? We've proven time and time again on here that duff MAFs cause stalling.

[antera309 0]

The MAF is mounted vertically fairly low down behind the bumper, between the plastic charger inlet pipe and the air filter. Here is a diagram showing how the pipework is set up:

 

 

I already have a new MAF on order - it should be with me in a couple of days. When I fit it, I'll see if I can mount it horizontally somehow. Can anyone remember off-hand what diameter of hose the MAF outlet takes?

[buttles 0]

Antera, yes, using resistors to fool the ecu of warmer temp didn't work at all. Very strange as telling it the actual temp with a resistor didn't work either. Seems not to like the setup. http://www.revotec.com/flexibleducting/flexducting.htm I think it's 80mm stuff you need but not 100% sure just now. My setup looks like the 'U' bend of a loo now :oops: