Golf 118 TSI Engine Failures and Service Campaign 24S4

[mr gee]

Prise, this problem had been reported on cars from Singapore.

[logger]

So much for my hunch. I thought, overly aggressive Exhaust Gas Regulation (EGR) was causing some sort of thermal shock to the cold top end during cold starts. Along the lines that a warm engine with plently of gas flow can handle hi temps, but a stone cold one, at low revs could be damaged by similarly hi temps, even if exposed for a brief period.

Initially it really looked like new ECU software had curtailed the cold start EGR to reduce these hi EGTs.
I logged the start of my stone cold engine this morning after the sc24S4 update and it behaved totally differently to what I had seen in the past. Specifically the very high EGTs and rough running were no longer present.

Normally when you cold start, the tsi118 it recirculates exhaust gas by way of valve overlap. Engine idles at 1080rpm and EGT skyrockets to 665 degrees in 20 seconds to get the CAT to operating temp ASAP. This allows early Lambda Control to be initiated. (This is the diesel like temporary rough running people have observed.) As soon as the temp is reached the revs drop back to 680, EGT drops to a cool 250 degrees and lambda commences. (I posted a log of this process a few months back in this old http://www.vwwatercooled.org.au/f112...tml#post478532 thread)

Anyhow now my car is updated with the new sc24S4. It is stone cold, parked outside overnight with coolant at 12 degrees. The start showed the EGR process occuring. 1080rpm as expected, slight but lesser rough running from EGR and climbing EGT. Much less IGN advance than normal. Instead of the EGT and CAT temps rocketing to 665C as per normal though, they peaked at a far more conservative 435 degrees. Then dropped back to 250 degrees and 680rpm.

At this point I thought – Yes they have changed the way it does the EGR to keep the temps down when the engine is cold. I know it can handle over 1000 degrees in the exhaust, when warmed up and at higher RPMs. but 665 degrees in a stone cold engine strikes me as one hell of a temp rise. Especially seeing the revs are only at 1080 and not 4 times that at 4-5000revs when you see higher temps around 800+ when flogging the engine.

I thought I had found something sc24S4 is doing differently, I had better double check.
So I waited about 5 hours until my car was cold again. (never got above idle the first time). Ran the same test again and bugger me, it was back to its old ways of EGR and 665 degrees in 20 seconds. I have NFI why on one COLD start behaved differently to the other.

So this disproves my very speculative theory.

Back to the drawing boards.

If anyone want to have a look through some logs I made you can grab them from here. You may be able to see a difference in the way it is doing lambda before and after? I cannot, but as there was so much data I only looked at the peaks.

SC24S4LOGs.zip

The zip contains 2 pre and 4 post logs with about an hours worth of quite varied driving in them.

Here are the parameters I recorded. The files in the zip are in xlsx format so you will need Office 2007.

Engine Speed
Engine Load
Injection Timing
Intake Manifold - Pressure (G71)
Throttle Valve Angle
Ignition - Timing Angle
Coolant - Temperature (G62)
Lambda Control - Bank 1 (actual)
Lambda Control - Bank 1 (specified)
Catalytic Converter - Bank 1 Temp.
Time since - Engine Start
Exhaust Gas - Temperature Bank 1

I apologise for my previous "Eureka" post. Got a little over excited.

[Maverick]

I would bet money that this is the issue. Normally turbo cars since their inception have pumped out black smoke under full throtle to varying degrees as a result of moving from a lean cruise ratio (lambda 1) to a much richer full boost ratio.

I'm pretty sure that VW engineers know what they're doing after all the engines they're built across so many brands and models.

So, running lambda 1 at all ratio's seems crazy dangerous to me as you are never allowing the fuel to cool the turbo engine as almost all other turbo cars do.

How many years have you spent as an engineer working in a team that builds engines for the largest manufacturer in the world?

You will also note that the GT and GT SPORT earlier cars do not have this failure issue and they use more fuel than the 118TSI.

The new GTI engine uses less fuel and outputs more power than the outgoing GTI but there's no signs of engine failures there. Just about every manufacturer lowers fuel usage and increases power on subsequent models.

[parso_rex]

I think he meant in general or rather traditionally. They still run the GTI's round .7 lambda (10:1 AFRs) but obviously with these TSI's they can keep it close to 1 pretty much all the time. I find that amazing myself and it certainly shows they know what they're doing. All manufacturers try and keep things in closed loop for as long as possible for many reasons. If you look at logger's logs it rarely goes below lambda 0.9 and even then only at almost peak boost - impressive.

[prise]

The good news is that 'brisk' acceleration doesn't ruin the fuel economy (provided it doesn't then cause me to use the brakes immediately afterwards) unlike previous turbo cars I've owned.

[nicandlance]

Unless you've neglected to read the past 12 pages of this thread, its all down to speculation at this point. The only one truly "in the know" would be VW themselves. As for your obvious mistrust for the manufacturer - I struggle to understand why an anonymous individual on an internet forum deserves any more? Although there's clearly an issue here, VW have both acknowledged and addressed it in a reasonable amount of time. We'll just have to wait and see if the update has the desired effect. So lets just get back OT shall we?

Thanks for your sarcastic answer....
We are all concerned about our cars and merely want to know why there have been instances where engines have blown up within months of ownership. The female in question that prompted the article in last weeks Cars Guide is known to my husband and I. It took an article to the Herald Sun to get things moving - hence my 'obvious mistrust' of the manufacturer. VW have a lot to lose if this goes pear shaped - they are going to want to protect their reputation, and their 'car of the year'.
I have read through the thread - and there are a lot of knowledgeable people on this forum. Finding out opinions of people here can often be more helpful than any information VWA are prepared to give - particularly if there is a problem.

Keeping OT is a matter of answering a persons query and being civil. It's when you deviate from that to be rude and sarcastic that it goes 'off topic'.
I love my new Golf and I am hoping it will provide me with years of reliable driving pleasure. I'm just hoping that VW's update will give me that.

[Maverick]

Thanks for your sarcastic answer....
We are all concerned about our cars and merely want to know why there have been instances where engines have blown up within months of ownership. The female in question that prompted the article in last weeks Cars Guide is known to my husband and I. It took an article to the Herald Sun to get things moving - hence my 'obvious mistrust' of the manufacturer. VW have a lot to lose if this goes pear shaped - they are going to want to protect their reputation, and their 'car of the year'.

What did the article have to do with the fix? Volkswagen can't just pluck a fix for a car out of the air when someone complains to the media. They would have been working on this problem since the first report came in, your lack of faith in Volkswagen is unfounded as you're dealing with a company who are pro-active in identifying problems through analysis of warranty claims and this takes time.

I have read through the thread - and there are a lot of knowledgeable people on this forum. Finding out opinions of people here can often be more helpful than any information VWA are prepared to give - particularly if there is a problem.

VWA don't have to give the customer any information other than a "it's fixed". Too many people demand too much, there is no obligation on VW's side to provide it's customers with an indepth report on the problem, how it was fixed, how it now works especially as much of this information would be sensitive information they wouldn't want to provide their competitors.

If you really have no faith in Volkswagen I'd sell the car and find something else before you warranty is up or earlier if it makes you feel better not having it. Cars are so much more reliable than they were years ago but you also have to realise they are so many times more complicated and things can and do go wrong. The difference is that some manufacturers hide these problems (Ford, Holden, Mitsubishi, Hyundai spring to mind) whereas others are a lot more proactive and open.

[prise]

OK so this is what we know so far:

1) The problem affected a small percentage of engines in Australia (and anecdotally Singapore) and the percentage of engines affected is probably less than 1%
2) A software update has been released which VW claims fixes the problem and so far we have no reason to believe otherwise.
3) Based on logged data before and after and anecdotal feedback from owners the update does not appear to reduce the economy or performance of the vehicle (hooray)
4) VW is quoted in the carsguide article as saying that the change involved recalibration of the knock sensor and so far we have no reason to believe otherwise.

Although it is possible the timing of the announcement was made in response to adverse publicity, they must have been working on the fix for a while. Damaged engines and componenes would need to be shipped back for inspection, tests would need to be conducted to replicate the problem, software modified and tests repeated to verify that the problem had been fixed and then the service campaign would need to rolled out. It is possible that in the absence of adverse publicity they would have rolled out the update with the normal service schedule, but we will never know.

My interest in this is motivated by professional curiousity rather than any real concern as VW have produced over 500,000 dual charger engines since the first model was released in 2005. When you start delving into the details of the engine you appreciate the features they have engineered into it for durability.

[logger]

My theory might not be all BS. Cold started my car again this morning. Again it only implemented the mild EGR with low 440c peak, that I had not seen before 24S4. This time I kept logging it for 10 minutes. It took almost 5 minutes of driving around before the CAT reached operating temp and lambda was implemented! This is not what the worlds best engine is supposed to do. It is supposed to be so green that is has emission control running within seconds. This cannot be good for emissions, but surely it is better for my engine, than it getting belted with 665c in the first 30 seconds of a cold start.

I am now wondering if the reason I saw the old, aggressive EGR on the subsequent cold start yesterday, is that it can record the fact that lambda was not implemented during previous run cycle. Therefore on subsequent starts it applied aggressive EGR to force rapid lambda. ( I never drove the car yesterday and as a result it did not get warm enough to get the CAT up to temp to trigger lambda control. )

Anyway this could be something bundled with the update. Implement mild EGR unless previous start did not achieve lambda. OR to do with standard emission control, readiness and unrelated.

Be interesting to see what sort of Start EGTs others are seeing in block 112 on cold starts. The dead giveway that EGR is happening on a cold start is the idle popping up to 1080 rpm for about 30 seconds before dropping back to 680. But unless you look at the EGT in block 112,0 or CAT in 34,1 you wont be able to ascertain which regime it is using. You can tell the greater roughness in the idle when aggressive is used, but it is subjective. Of course you cannot readily re-test as the car will be warm for subsequent starts.

I am far happier with my car doing the mild EGR than the aggressive one.
Here are 3 start logs. SC24S4Starts.zip The latter two also show Lambda Control in row E.

[nicandlance]

Maverick,

It's not that I don't have faith in VW in particular... I am well aware that all cars have issues regardless of their standing. I'm just hoping that my new Golf will give me the years of enjoyment I'm hoping for - just like the Mazda Astina I have beforehand.
So far I am very happy - I have no complaints about my own vehicle - at least at this stage.
I just happen to believe that VW should provide us with some insight as to why this has happened.
It's a pretty major issue to have your engine blow up before you even reach 1500 k's, and in regard to the article.... I was basically stating that it took a Cars Guide article to help this particular person get the ball rolling in regards to a replacement engine. I never said that the article was what spurred VW to 'pluck a quick fix'.

Everyone is entitled to an opinion on this subject without being chastised for it.