3.0tdi CAPA limp mode

[desertstorm]

Looking at previous logs I have made on my A4 MAF readings at Idle with no boost are around 420-450mg/stroke dependent on the idle rpm whether it's hot or cold.
Normal hot idle speed is around 650rpm.
If you have 1 bar boost then you would end up with around twice that reading. As the air intake isn't throttled there is usually a good amount of excess air.
If you disconnect the MAF you should get a fault code and the engine would switch to a get you home strategy . On petrol cars if the MAF is failing quite often it runs better when it's disconnected as the ECU just ignores the incorrect info and calculates it's own value based on other sensor readings.

 

[bhodgkiss]

Thanks Karl

Yes the cylinders are all approx 500cc so under normal 1 bar absoltue its ~500mg/stroke in a perfect, world, but some volumetric efficiency losses so more like 450mg/stroke.

I guess I'm now back to a blocked DPF, with the remap having turned off DPF fault codes.....

But I will try a replacement MAF first, just in case

 

[bhodgkiss]

Ok todays update:
I tried the (stationary) DPF regen in vcds basic settings, but it won’t do it (it just fails when I put foot on brake and then accelerator)
I have a soot value of 62g (apparently this is high) and can’t clear the value in vcds (I just thought trying this night help)
The last regen is reportedly 120,000km ago, although I’m pretty sure it’s been doing it at intervals since I’ve owned it
What’s are people thoughts on 62g soot please?

 

[The Jargen]

The DPF usually initiates regeneration at 25g soot loading (which is around 40-45% loading). According to your readings its blocked and you will not be able to perform a forced regen as you are over 75% soot loading. Pull the pre DPF pressure sensor and stick a boroscope down the boss, check if the DPF is intact. If not you're at the point of needing another remap.

 

[pelhamsdetailing]

Are any of your EGT temperature readings stuck or abnormally different vs the others? If any are stuck or wildly different in reading, then it won't regen. The sooner it is in a professional's hands, the better. Bobby will take care of you.

 

[desertstorm]

Hi Ben, If the DPF was very blocked I wouldn't have thought the intermittent nature of your issues would be showing and the EGT readings would be a lot higher. It seems to me that you may not have a DPF present if the DPF pressure sensor is showing 0 and the last regen was 120K km ago.

 

[bhodgkiss]

Thanks Both
Maf swap being tried today.
As you say Karl, intermittent (although now most of the time) limp mode you’d think is electrical
Possibly the DPF has been gutted? I presume I should see extra weld lines?

When I looked at egt temps the other day at idle, I was seeing them all come up from 50c to 100c ish, so they were certainly reacting

 

[pelhamsdetailing]

If you can't see weld lines, which may be underneath for visual inspection passing, you could take out the G448 sensor from the exhaust, get a metal straw (or something similar in shape and diameter with a 45* curve), and then poke it through half way, and then rotate it towards where the DPF should be. If it's there, you'll feel it stop against a hard mesh of metal. If it doesn't stop, you've got no DPF in there. Did this myself to answer it once and for all whether or not my DPF was intact. It was, and very sooty.

Don't drop it

EDIT: G448 sensor is the one on the top of the exhaust, with a wire going over the turbo heat shield. It's technically the second sensor after the O2 sensor, but 1 of 2 sensors mounted on the top. It should be spot on before the DPF itself, but just after the catalytic chamber.

 

[bhodgkiss]

Thanks
So this sensor at the top left?

The image shows I had the lambda sensor on the right removed the other day.

Possibly the weld quality around the g448 shows it been gutted?

 

[The Jargen]

No not that sensor, the one to left. The one you currently have removed will only allow you to see the CAT.

 

[desertstorm]

As stated above a bent piece of wire or plastic straw inserted through the pre DPF EGT sensor hole should enable you to determine if the DPF is till intact. Or borrow a small Endoscope camera from work, I am sure they have plenty.
If the DPF has been gutted the weld lines are usually on the back side against the fire wall and difficult to see. Mine looked like this

 

[bhodgkiss]

Thanks All
Ok so interesting developments tonight….
Swapping the MAF still gave limp mode
I then disconnected this G448 sensor (the one top left on my previous image) and full power.
I then proceeded to reconnect and disconnect about 5 times
Each time, connected gave limp mode and disconnected gave full power
So I presume it’s this sensor that’s not happy and needs changing, and the default strategy when disconnected allows all/most of the power
I see it gives a temperature output, as it went to 1000c when I disconnected, but is it oxygen or pressure as well? I can then see if it’s giving a silly value when connected


“

 

[desertstorm]

Thanks All
Ok so interesting developments tonight….
Swapping the MAF still gave limp mode
I then disconnected this G448 sensor (the one top left on my previous image) and full power.
I then proceeded to reconnect and disconnect about 5 times
Each time, connected gave limp mode and disconnected gave full power
So I presume it’s this sensor that’s not happy and needs changing, and the default strategy when disconnected allows all/most of the power
I see it gives a temperature output, as it went to 1000c when I disconnected, but is it oxygen or pressure as well? I can then see if it’s giving a silly value when connected


“

It's just a platinum RTD thermocouple and gives you temperature post cat / Pre DPF . I wonder if the issue is related to the DPF regen and because that sensor is specific for that process it's disabling the limp mode some how. Try disconnecting the DPF pressure sensor electrically and see if that does the same.
IIRC the platinum RTD thermocouples that are used are 200 ohm types. Which means they measure 200 ohms at 0 degrees C , and increase as temperature increases.
They should be around 215 ohms at 20 degrees C. 275 ohms at 100 degrees C.
Have you checked to see if the DPF is intact ?

 

[bhodgkiss]

Thanks Karl
I’ll look more into this after the weekend!

 

[bhodgkiss]

Where is the best place to buy a G448 sensor please gents?
And which website would you recommend to look up the part number? I believe it’s 059906088m

 

[The Jargen]

Get a genuine sensor from TPS or Audi. Have you checked your DPF yet?

 

[bhodgkiss]

Thanks
Audi is £193
Darkside have them with 1 yr warranty at £36.... probably worth a go?
I'll try and check the dpf later!
Running 'fine' so far without the sensor connected

 

[desertstorm]

Did you try measuring the resistance across sensor. Good news that the car is driving again. Had a look on Fleabay, you coulkd pick this up from Birmingham.

WALKER Exhaust Gas Temperature Sensor 273-20058 FOR A5 A4 Q5 Allroad Genuine Top | eBay

Find many great new & used options and get the best deals for WALKER Exhaust Gas Temperature Sensor 273-20058 FOR A5 A4 Q5 Allroad Genuine Top at the best online prices at eBay! Free delivery for many products.

www.ebay.co.uk

 

[bhodgkiss]

Ok so I took out the g448 and with a wire it appears the DPF is there
I bought a dark side sensor but it does look different?
Have I got the right one?
I mean it’s a temperature sensor but is the original measuring oxygen or pressure etc too?

 

[bhodgkiss]

The replacement G448 sensor made no difference
It still has full power when I unplug it (with the temp showing 1000c) and limp mode when plugged in
The fan continues to run now when I stop the engine as I presume it thinks it’s hot
What i don’t understand is that the temp is reading 100c+ when plugged in so why is that causing limp Mode please?
I don’t see how a genuine DPF blockage is causing the limp mode - unless a pressure sensor is showing high back pressure somewhere, but why does unplugging an egt sensor help?
I’m wondering if it’s something else like EGR…..

 

[bhodgkiss]

(If the egr is linked with the egt sensor somehow, and disabling the egt sensor also turns off an egr error, which isn’t showing in the fault codes…..)
Painful!

 

[pelhamsdetailing]

There's the potential that it is running on older ECU software, which allows it to not throw an engine light or hissy fit when the sensor is unplugged. This sensor does not do anything but temperature. You have a dedicated oxygen sensor and a dedicated pressure sensor. Your pressure sensor is effectively directly below the G448 sensor, and your oxygen sensor as already established is the first one before the catalytic component.

If your soot reading is high, like 62g as you previously stated, it will not regen. You put your temperature sensor in, which activates the DPF software loop, it sees the high soot number and will not regen, as there is the potential for an out of control fire if regens at that high of a soot reading. Removing the temperature sensor in question does not allow the DPF software loop to complete, and in this case, the car does not go into limp mode and runs at full power. You would think it would be upset by this, but there's a chance your ECU software is an older version (or tampered with). As far as I am aware you need special software to clear or adjust the soot number manually - to my knowledge: remapping software.

 

[bhodgkiss]

Many thanks
It’s a shame if vcds can’t reset the soot level, otherwise I could take it off to clean and just reset?

 

[pelhamsdetailing]

If it is as blocked as much as it says it is - it's not regenning + how black your G448 temperature sensor is - the DPF needs removing, cleaned professionally, put back in, soot level cleared manually via software, and then if required, configured/corrected so that it actually regens by itself - or at least allows a forced regen via software.

Have you spoken with bobby? Even if it's a couple hours drive, if he's willing to have you there and get to the route of things and give a far more insightful, first hand solution, then absolutely go for it. I regret never properly sorting out my A5's blocked DPF, which was also a 3.0TDI CAPA. Tampered software and EGT readings was the culprit, but it's all £££, and if it's a good condition car, you want to keep it and enjoy it in the long term without this problem you're having, then it's a solid option.

 

[bhodgkiss]

Many thanks
Yes I’d happily spend some £ with Bobby for a dpf delete and remap; just need to look at the timings of getting it down to him!

 

[desertstorm]

The Darkside sensor looks different but I suspect does exactly the same job. As I said earlier they are resistance sensors and a quick check with a multimeter will usually identify if it is faulty. If you can disconnect that sensor and it drives reliably I would get it down to Bobby and get some stock software on it. You would hopefully quickly be able to work out what the real issue is then. If the DPF was really blocked then I would expect the pressure sensor to pick that up. You could check the sensor is working by disconnecting the hose that feeds pressure from the dpf and connecting another one you can blow into . With the engine running and VCDS looking at the measuring blocks you should be able to see the value changing as you blow ito the hose. The pre turbo EGT sensor should also be showing much higher readings if the DPF was blocked.

 

[bhodgkiss]

So yesterday it was in limp mode even with the G448 disconnected
It had an oxygen sensor fault on vcds, but I cleared this and no improvement
Thought I might as well post these lambda related readings when at idle, some don’t look right to me - quite a few zeros, and is lambda 6.4 normal at idle?

 

[desertstorm]

How's it going Ben.

 

[bobby singh]

Hey Karl I had Ben's car in last week found the following:

1) DPF was still in place but off in the ECU file (Ben had no idea about this and was creating excessive back pressure)
2) Bunch of sensors flat lined
3) DTC's killed
4) Injector #5 was blowing via the seal and created this tar over the exhaust manifold
5) Swirl flap DTC's killed but again still in place with a rubber O missing causing a boost leak

Sorted the DPF out, blocked the egr, put on a base tune and went for a test drive.
Within a few metres I started getting loud diesel knock sound, misfiring and plenty of white smoke.
I check the injector deviation and noticed injector 5 (one that had been leaking ) was in a bad way so Ben ordered a set from ebay.
Theory was the injector body has been exposed to excessive heat and the spray patten was no longer healthy.
Anyway installed the replacement injectors and went for a test drive -

No more white smoke, however every time you would give it WOT then it would have a bad idle like a misfire almost and again deviation on 5 was high.

At this point my suggestion was some internal engine issue is causing this and more than likely the leaky injector has been spraying in the wrong direction and now has bore damaged. Again guessing but maybe some oil is passing the rings during the wot run giving #5 a hard time.
On a side note if you drive it easy or leave it for 10mins then it would idle okish again.

This engine has done 235k and appears the inj seal has been leaking sometime so Ben is considering another car at this point as the next step will cost more than the car's value.

Here's some pics

 

[bhodgkiss]

Many thanks for providing the update Bobby, and thanks for your time and great service. It was good to meet you, and your own 600bhp build looks fantastic (what a turbo!)
The frustrating thing is that otherwise the car now pulls like a train - its very smooth and impressive above 1400rpm - and its only the lumpy idle after pushing hard, and general lumpiness below 1400rpm on part throttle thats annoying. Sometimes I do still get white smoke on start up which soon clears. Injector #5 is still shows a +5mg deviation, with injector #2 showing up to -4.5mg (I believe to make up for injector #5?)

Just for my own interest im currently looking into a compression test and borescope to try and see whats going on with that cylinder.

Im probably dreaming if I'm hoping that a clogged exhaust valve (coated in thick carbon) will magically clear?

@bobby singh I presume its not worth me swapping in another injector again, just to be sure? I still have my original set, so could try one of the better looking ones?

On the plus side, its worth as much in bits as it was when it ran properly. I have found another CAPA with 150k miles that looks tidy, so I might go and take a look if this one does have something more major wrong with it

Any replacement would then need to go down to Bobby of course.....

I also saw a nice looking CCWA, but would be worried about the pump failing on those.

Thanks again, I'll update if I do find anything after further inspections. But I think its worth everyone with these 3.0tdi's changing their copper injector seals if theyve done over 150k miles.

 

[bobby singh]

Many thanks for providing the update Bobby, and thanks for your time and great service. It was good to meet you, and your own 600bhp build looks fantastic (what a turbo!)
The frustrating thing is that otherwise the car now pulls like a train - its very smooth and impressive above 1400rpm - and its only the lumpy idle after pushing hard, and general lumpiness below 1400rpm on part throttle thats annoying. Sometimes I do still get white smoke on start up which soon clears. Injector #5 is still shows a +5mg deviation, with injector #2 showing up to -4.5mg (I believe to make up for injector #5?)

Just for my own interest im currently looking into a compression test and borescope to try and see whats going on with that cylinder.

Im probably dreaming if I'm hoping that a clogged exhaust valve (coated in thick carbon) will magically clear?

@bobby singh I presume its not worth me swapping in another injector again, just to be sure? I still have my original set, so could try one of the better looking ones?

On the plus side, its worth as much in bits as it was when it ran properly. I have found another CAPA with 150k miles that looks tidy, so I might go and take a look if this one does have something more major wrong with it

Any replacement would then need to go down to Bobby of course.....

I also saw a nice looking CCWA, but would be worried about the pump failing on those.

Thanks again, I'll update if I do find anything after further inspections. But I think its worth everyone with these 3.0tdi's changing their copper injector seals if theyve done over 150k miles.

Hey Ben thanks.
You could swap the injector but I already tried this - I was thinking maybe the injector seat is in a bad way but again when I pulled the injector I could see markings to confirm it was seating all the way around.
If you do remove the injector it might be worth seeing if the seal has any blow by if so a seat cut might do it.
CCWA has nicer injectors but terrible fuel pumps so a CP3 swap will resolve that.

Cheers