VR6 hard to start, stalls, lumpy full throttle

[Rpmayne 0]

I've been living with the VR6 stalling occassionally pretty much since I bought it about 7 years ago but now things seem to be getting worse.

 

It always used to start on the button, now it normally (but not always) take several seconds of turning over before it fires. The idle is as its always been, luck more than a constant, hunts, revs seem to drop very quickly, every trip it will stall when the clutch is disengaged at least once. The more annoying thing that its now decided to do is run ok under little load but sound like a Subaru on full throttle. Funny for a while but the power is way down.

 

Another strange thing it how the pump primes. It used to do a approx. 1 second buzz when the ignition it on. It can now either do one very short buzz or sometimes cycle until the car is started i.e. buzz-pause-buzz-pause-buzz....etc... And to top it off, if it does stop priming as it should, turning the indicators on makes the pump prime again. WTF???

 

Has anyone got any tests they have carried out to positively identify the offending components? I've read through many threads on here but the majority are to try changing this. I've never been that lucky to get it right, just spend hard earned Stella money on new shiny bits that make no difference.

[Kevin Bacon 5]

Step 1 - VAG-COM for any codes

Step 2 - Check for intake leaks

Step 3 - Check / replace the fuel pump relay

Step 4 - Check / replace the MAF sensor

Step 5 - Check / clean / replace the idle valve

 

The idle fuelling is very critical. All the idle valve can do is open and close as instructed by the ECU, it doesn't control the idle on it's own. The fuelling has to be 100% c0ck on or it will stall out. The MAF sensor is usually the culprit. Does yours have the wire maf (5 pin) or the film maf (4 pin)?

 

The latter is better and more reliable and cheaper to replace.

[Rpmayne 0]

1. I'll hook up the VAG-COM tonight if I can find the laptop I used to use. Thing is I got very sceptical of the errors it used to throw up. Generally oxgen sensor out of range. Wiring to Lambda redone by VW, on my 3rd sensor, no difference. So took it that it could actually be anything causing it to run rich/lean hence bad burn.

 

2. Top hose had a split, sealed it off. Remember years ago a mechanic spraying WD40 over my 16v Golf intake pipe and found a leak because it stalled it. Tried this, can't find anything else.

 

3. Will try that, is it 109?

 

4. The MAF is the rip-off wire type, is there a procedure to check this? Make up a dummy loom, start the car and check the output to the ECU?

 

5. Cleaned the ISV, opens and shuts (leaves about 1mm gap).

 

Cheers for the help, been away from this forum for a while.

[Rpmayne 0]

Thinking about it, am I right in thinking the MAF is ignored at full throttle?

[fla 9]

3. Will try that, is it 109?

 

No thats the ecu relay, you need 167 (or may be labelled 67)

[Rpmayne 0]

Thanks Fla.

[Kevin Bacon 5]

1. I'll hook up the VAG-COM tonight if I can find the laptop I used to use. Thing is I got very sceptical of the errors it used to throw up. Generally oxgen sensor out of range. Wiring to Lambda redone by VW, on my 3rd sensor, no difference. So took it that it could actually be anything causing it to run rich/lean hence bad burn.

 

Yeah look for cam sensor errors etc. It's an old chestnut but it really does sap a lot of power when it's dead. The lambda out of range doesn't necessarily mean it's faulty, it just means the lambda controller ran out of options to get the fuelling where it needs to be. For example, if the lambda controller is allowed to adjust the fuelling plus / minus 15%, and it needs 20% to meet the target, it will throw the "out of range" error. Hence my thoughts on the MAF. If that's not working properly, it will cause all manner of fuelling related errors and issues.

 

2. Top hose had a split, sealed it off. Remember years ago a mechanic spraying WD40 over my 16v Golf intake pipe and found a leak because it stalled it. Tried this, can't find anything else.

 

Yeah I use lighter fluid or carb cleaner. Any change in engine note points to the leak :D OK, tick that one off the list then!

 

3. Will try that, is it 109?

 

Yeah it's 167 as Fla said. VW have changed this relay at least twice since I've owned mine, so they've clearly had issues with the earlier designs. It's subject to a hefty load and tends to run quite hot. Flip the lid off and check for carbon burning on the contact plates. Sometimes just sanding the pitting off helps. Reminds me of the good old points and condenser days :D

 

4. The MAF is the rip-off wire type, is there a procedure to check this? Make up a dummy loom, start the car and check the output to the ECU?

 

Hmmmm, thought so. Bloody things. I had no end of bother with wire MAFs back in the day. I saw a thread on this forum about converting to the newer film maf - http://www.ecuconnections.com/forum/ I can't find it at the moment though, but it's defo on there.

 

The MAF isn't ignored at full throttle, it's used full time as the main load. It's the lambda that is ignored at full throttle 8)

 

5. Cleaned the ISV, opens and shuts (leaves about 1mm gap).

 

Sounds like it's fine then, good stuff as they're over £250 new now :shock:

[Rpmayne 0]

Thanks for the reply Kev. I couldn't find the VAG-COM at the weekend, been along time since I've used it. The old MAF has been cropping up again and again, might just take the plunge and buy one. Will see what the VAG-COM says though. Cheers.

[Rpmayne 0]

Been a while I know but now being temporarily redundant I've had abit of time to play with the VAG-COM.

 

I have since changed the MAF as a shot in the dark with no improvement.

 

Plugged the VAG-COM in today and no faults except lambda intermittently out of range and engine speed sensor. As it stalled during the test I expect these would come up anyway.

 

A couple of things which I wasn't sure about were:

 

Binary position code reading - sitting at 0010 at idle, 0100 when throttle depressed, goes to 0011 when throttle released then back to 0010. Is this the throttle position as a binary value, and if so shouldn't it count as binary (ie. 0001, 0010, 0011, 0100, etc)?

 

The vehicle speed block always reads 0.

 

Intake temperature is at 60deg.C which seems abit hot to me.

 

Unfortunately no obvious faults to me, I know it well overfuelled, on overrun to idle its popping and spluttering all over the shop, very sooty exhaust. Have changed the plugs, the last ones came out sooty, not the good golden brown.

 

Got lots of ABS faults, differential lock failure, front wheel sensor implausable, etc.. That's another story anyway, used to not having it now.

[Rpmayne 0]

Forgot to mention, plugs have been changed, old ones very sooty, as is the exhaust.

 

When on ignition only I could hear the fuel pump constantly priming. I turned it off then on and this time it didnt prime at all. How is the fuel pressure governed?

 

Also, putting the indicators on does a fuel pump prime... :confused4:

[Rpmayne 0]

I checked the fuel pressure today, borrowed a kit from the local garage where to plum a gauge in-line with the fuel feed between the fuel pump feed and the filter.

 

Without a prime the pressure is 0, when the ignition is turned on there is a small prime (humm from the pump) which barely moves the needle.

 

When starting it turns over for a longer time than normal then once the pressure reaches around 3 bar the engine starts.

 

Once started the pressure sits around 3.6 bar.

 

Now when giving it a quick blip of the throttle, when the revs drop to idle level the pressure rises, sometimes to around 4 bar when the car stalls.

 

Should the pressure be a constant 3.5 bar regardless of what the revs are doing? Seems to me I have two problems, one the pump does not prime to 3.5 bar hence the hard starting, and when revs drop the pressure is not be governed well enough so the engine runs rich and stalls.

 

When doing a prime, what tells the pump the correct pressure is reached? And is it possible the FPR is not reacting quickly enough to limit the fuel pressure when the revs drop?

 

Thanks for any help.

[Timbo 0]

Did you by any chance resolve this? I'm having similar problems at the moment and suspect some combination of fuel pump/pressure regulator. In my case it's possibly self inflicted as I've had the tank off to do a bit of welding, which meant disconnecting the fuel pump for a while. At the same time I cleaned the fuel tank of 16 years of crud and of course the fuel pump was drained as I had to take it out of the tank. Inevitably I introduced some air into my fuel lines but I would have thought this would be self bleeding to an extent?