GTi engine not running crisply

[937carrera]

I've just acquired another GTi (black AVY) which had a few issues which have been sorted out and having taxed it today have driven it for the first time and it's running great. What I particularly noted was the quick pick up / throttle response compared to my (silver ARC) GTi.

To be honest, I have always felt that although the car was quick enough, it never seemed particularly crisp and put it down to being a 16v engine rather than 8v. Now I can do a side by side comparison it's clear that there is some other issue.

Symptoms:
Car lacks power when setting off - almost feels as though it is going to stall (and sometimes does)
Tickover feels a little rough with a deeper note
Car seems to run broadly the same from 3000-6000 rpm, after that the black one seems to stay on cam more
The black car is far more responsive to throttle blips whan stationary

There are no engine error codes save for 17911 load signal from alternator which I have been unable to get rid of.

I fitted new plugs around 5000 miles ago

I am thinking it could be cam timing, but checked that when I bought the car. Certainly the timing of the exhaust / inlet cams seemed correct by using 8mm drill bits in the alignment holes but the view down to the crankshaft marks wasn't great. I think it was right, but am probably only 80% sure.

Any other thoughts or information needed to help with the remote diagnosis ?

[6n2stan]

Is the exhaust on your silver car ok? If the cat or silencer is breaking up - as I'm sure you'll know - it'll stop the car breathing properly and cause a drop in performance. I've had 2 AVY's and as a back to back test one (the quieter car) was better than the other at all the things you mention.

[937carrera]

Thanks Stan,

I had a broken cat problem on my Saab 9-5 Aero and the transformation when I replaced it was unreal, so I certainly do understand the adverse effect a damaged cat can have. In fact solving the performance problem on that car was a collaboration between me identifying a failure of one way valves in the turbo system, followed by a very experienced Saab mechanic saying he thought would be worthwhile trying a cat.

Is there any way to diagnose this better, (choose whether it's cat or back box) without having to resort to the Kwik-Fit school of engineering (just keep replacing bits until the problems gone) - you could well be right though

[6n2stan]

Short of taking the exhaust off and shaking it, it's hard to diagnose. The baffles rattle in the back box but the cat iirc has a gauze over it so you can't see into it. They do get very hot if they are blocking up but its hard to distinguish between very hot and f*cking hot with your hand for example. I think (unfortunately) kwik Fit can probably tell you if the cat is blocking/overheating.

[937carrera]

I have been doing some more thinking and now I have two cars to compare started comparing VCDS values.

I started off looking at display groups 002 and 004, working on the theory that if there was an exhaust restriction then VCDS may indicate a higher load or more fuel going in. On idle there was no difference between the engines with an injection time of 1.03ms and timing around 4 degrees BTDC. On fast idle (2000 RPM) the injection time remained the same with timing moving to around 10 degrees BTDC

There was a noticeable difference when I blipped the throttle. The black car immediately moved to 10.3 ms, with the silver one barely increasing, though if I gave it a longer blip I could get it to peak at 9-10 ms.

After a bit of a wander around Elsawin I decided to look at the cam timing. Using group 91 I identified that the camshaft adjustment was active and then using area 3 of group 93 that the adjustment timing was 1.9 degrees BTDC on the silver car. This is outside the specified range. I then compared to the black car which shows 4.4 degrees ATDC under the same test condition

My understanding of the test results is that the opening of the inlet cam is 6.5 degrees different between the cars. I'm getting beyond my understanding of valve timing now, but Is it possible that this difference means that when the throttle is opened there is insufficient duration (or too much) on the inlet valves to allow an increase in pressure in the intake manifold, which means the MAP sensor does not see any need to deliver more fuel ?

I'm going to go off and do some reading, but any opinions on my theory, or the operation of the VVT timing would be much appreciated.

[6n2stan]

It sounds like you may be on to something with the inlet timing. Like you, I'm unsure as to how VVT would have an affect on a potentially mistimed pulley (or mistiming via the ECU). I've had the same fight in the past with a VVC engine on an MG but that works on a different principle to the Polo. If the duration of the inlet is so different then surely the potential of valve overlap needs to be considered? This would have a bearing on pressure or back pressure too, or is this getting too deep? Lol.

[937carrera]

I have found something called "Volkswagen Self Study" guides, which provide some useful information. I'm still working my way through them but there is one that covers the 16V engine, but not VVT and another that does VVT, but for later more complex systems but explains what technology approaches can be used:

http://www.volkspage.net/technik/ssp/ssp/SSP_196.pdf

http://www.volkspage.net/technik/ssp/ssp/SSP_246.pdf

Once I understand what it's supposed to be doing, the next challenge is going to be checking / adjusting the timing without the plethora of special tools that seem to be required - though it's really the T10075 that's the problem

[6n2stan]

Good read that lot.....

It sounds like a reliable system in the sense that all but the crank sensor can be lost and it'll still run with pre-set values in the ECU. Assuming you have no fault codes, all your sensors must be functioning. Short of the cambelt fitted incorrectly, it could only be an oil supply problem to the cam fluted variators or the actuator control valve that ports the oil to advance or retard the cams.

[937carrera]

Thanks Stan, yeah it was a good read, and the spiral graphics were very good to illustrate how the valve timing / phase adjustment works.

Our 6n2 only has adjustment on the intake and I believe it uses a much simpler system than in document 246 to achieve the change in valve timing in that it creates the change by simply deflecting the camshaft drive belt.

I have been doing some more studying and found the following links also useful

http://www.enginebuildermag.com/2014/01 ... ve-timing/

http://www.hotrod.com/how-to/engine/121 ... advantages

http://www.vwforum.com/forums/f15/v-v-t-12044/

What I understand from this is that at low engine speeds you need to advance the intake cam to enhance torque, while at higher engine speeds you need to retard the intake cam to maximise power output. From the measuring block data information for 091 to 095 it seems that the VVT provides up to 40 degrees of timing change.EDIT: THIS MAY BE INCORRECT


I then decided to take the car out for a run and see what was happening through logging the data. From this I worked out that the "default" position for the cam is the advanced position for idle and that the cam is retarded when the engine is under load and I created a chart which shows the cam timing observed together with a trendline to give an indication of what type of adjustments are made when the car is under load. At idle the advance is 0 degrees and I think you can see that the left hand part of the red trendline will be move higher as you exclude the lower engine load values

EDIT: THIS GRAPH MAY SHOW IGNITION TIMING AND NOT CAM TIMING DUE TO DIFFERENCE BETWEEN DESCRIPTION OF MEaSURING BLOCKS IN VAGCOM AND LOGS



So where does this leave me ?

• Well, the car is not running crisply, but looking at the data the 0-60 times are about where they should be
• By comparison It is pretty gutless until ~ 3000 rpm
  I decided to examine the timing on my spare ARC engine which is on an engine stand. The main cam timing was spot on, but I could not remove the camshaft cover plate at the gearbox end to evaluate how to carry out the measurements of the cam adjustment (at least not without completely destroying the cam cap)
• VAG-Com tells me that the actuation system is working correctly
• Still unsure of the actual problem

What I still don't understand is that according to the links above you need the advance to give good torque at low engine speeds and to provide a smooth idle yet on the 6N2 polo there is zero advance at idle yet there is full inlet advance at 1800 rpm when the car is under load, this advance then decays as the revs rise.

I feel like I have a jigsaw, with quite a few of the pieces in the correct place but I am still unable to see what the picture on the box is

It could be a cam timing problem, but it could equally as well be an exhaust blockage issue as you suggested initially. On the old adage of measure twice, cut once, more thinking time is needed, or someone who knows their onions on this stuff.

[alexperkins]

Interesting issues here. I'll have a read through the PDFs later today, but I once worked on a car that had a similar issue to this and it turned out to be the fuel filter - strange I know, but fuel starvation was causing the timing to throw as there wasnt enough fuel available for the timing to advance successfully

[6n2stan]

This is an educated guess but perhaps the 1.6 has enough low down torque without the need for inlet advanced timing. I say this as the difference between the 1.4 and our 1.6 is the longer stroke length as the bore diameter is the same. A longer stroke lends itself to more torque (Diesels have a longer stroke). Without boring you with the Rover/MG thing again, the 1.6 k series had better torque than a shorter stroke 1.4 and 1.8cc engines. The VVC was only on the 1.8's with bigger valves too. Perhaps VVT in this case is to gain greater BHP as the revs climb?

[937carrera]

Both of the GTi's seems to have zero cam advance on idle. I haven't got the other one out yet to do a similar data logging exercise, but suspect that the deltas will remain the same.......... and remember my identification that there is a problem with my silver car was because the black one was so much sharper........ and my gf noticed without being prompted.

I don't know much about the MG K series engines, but do know that one of the prohibited mods in Porsche racing some years was the fitting of the 2.4E cams to the 2.7RS engine. The profile difference was such that although there wasn't much loss in peak power the torque and area under the curve was so much better. I agree that longer stroke generally equates to greater torque, and lower peak revs.

I don't suppose you have ever removed the camshaft end plates have you - it seems they are made of chocolate and fracture at the slightest amount of leverage.

[6n2stan]

I can't help with the end plates, it's not something I've done on this engine

[937carrera]

Thanks again Stan

I've been out in the black car today (230 faultless miles) and my logging abilities have improved and chose to log three groups at the same time.

I'm now suspecting that area 4 of group 091 is measuring ignition timing and not cam timing, which agrees with the vagcom description, but not Elsawin measuring blocks. It seems that block 3 of group 093 is what measures the cam timing adjustment

If anyone can confirm my understanding, or correct me, it would be much appreciated. I'm planning do do some more logging in the silver car in the morning and see if I can improve my understanding

[alexperkins]

Elaswin measuring blocks are accurate for the VAS5051. They don't always tally with vcds for some unknown reason - worth noting.