Rough idle, missfire, strong smell help please

[L30NR]

Hi everyone,

i'm hoping someone can help me please. i have a 2018 ex RAC DSG T6 euro6. i purchased slightly cheaper as when cold it had a rough idle and misfire when driving. also when cold it really stinks of unburnt fuel kind of smell. its not in the cabin its from the exhaust and only smells outside but its very strong. once its warm it runs perfect with no loss of power, missfires or rough idling or smell.
i had a fault on injector 2 when i first scanned it so had this replaced and coded. this has made no difference.
I've now been playing with VCDS and looking at some live data.
only thing that's really jumping out to me is when comparing the cylinder bank internal moment readings across all 4 cylinders. At idle it's pretty steady, all very close to 500nm. as soon as i hold the revs over 1000rpm cylinder 1,3 and 4 go to roughly 6000nm but cylinder 2 drops to around -10nm and stays while i hold the revs. fuel injection time deviation has no pattern and all 4 injectors wander between -80µs and +80µs at idle
i can post a video on youtube if it helps or if there is something else anyone can recommend checking?

i know there has been lots on the net with similar issues, I've tried the VW injector cleaner and another industrial one recommended by the garage. i know the previous owner had this diagnosed as DMF so he had this done a few months ago but again made no difference. I've heard of rods being bent by hydrolocking, i'm not losing coolant, however i do not know when it was an RAC van if it had issues and the EGR cooler replaced. it looks original to me and no obvious signs it's been done. I've read all the threads but i'm a little stuck now.
Thanks in advance

 

[L30NR]

Sorry just to add, its done 140k miles.
Thanks

 

[Dellmassive]

@mmi style vcds scan and analysis?

 

[L30NR]

Thanks Dellmassive, I've seen some posts with MMI that go way over my head so would be great to have their input. I'll try work out how to do the scan download in the meantime. When the injector was replaced and coded should it also of had a quantity adaption done does anyone know?

 

[Dellmassive]

Can you do a video clip and put it on YouTube?

Then post the links here...

Maybe a crank and start..... Showing the problem?

 

[mmi]

i had a fault on injector 2 when i first scanned

What was the fault code?

f it had issues and the EGR cooler replaced. it looks original to me and no obvious signs it's been done.

On the cooler there should be a sticker with date stamp.

When the injector was replaced and coded should it also of had a quantity adaption done does anyone know?

Just the coding - actually an adaptation - the calibration value engraved on the injector stored into the ECU.

 

[L30NR]

Thanks @mmi
The fault was -
Fuel injector 2 (N31)
P026600 contribution / balance fault
This cleared after the injector was replaced but no difference in symptoms


Can't believe I didn't spot the sticker with date on the intercooler but it looks original as the date is 02.02.18

@Dellmassive i will get some videos tomorrow afternoon and put on YouTube. I've got a few from yesterday but I started recording after the van was already running so no cranking over just idle and revs

 

[mmi]

Thanks.

Well, checked a couple of logs (had only few) and didn't see such an imbalance between cylinder's moment.

To better understand the nature of rough idle could you please make a VCDS log

For genuine VCDS HEX-V2 or HET-NET​

Turn IGN on -

In VCDS tick the following under Engine > Advanced Measuring Values

IDE00021 Engine RPM​

IDE00083 Throttle valve position (absolute)​

IDE00347 Air mass: actual value​

IDE00351 Main injection: start of activation​

IDE00352 Main injection: duration of activation​

IDE00401 Intake manifold Runner Control (IMRC): acknowledgment​

IDE00589 Fuel pressure​

IDE04269 Intake manifold runner actuator: activation​

IDE07738 Charge air cooler inlet: pressure sensor 1 bank 1: raw value​

IDE07742 Cylinder head: pressure sensor 3 bank 1: raw value​

IDE07744 Particle filter: air pressure sensor 1 bank 1: raw value​

IDE07757 Exhaust recirc.valve 1 bank 1: posit.feedback - Actual value​

Tick Group UDS requests - by 7 (IMPORTANT!!!)

Log - Start

Start the engine - let the engine idle for 2 minutes.

Stop logging.

The log file will be stored in folder C:\Ross-Tech\VCDS\Logs

For other VCDS versions​

Tick only the following

IDE00021 Engine RPM​

IDE00352 Main injection: duration of activation​

IDE07744 Particle filter: air pressure sensor 1 bank 1: raw value​

IDE07757 Exhaust recirc.valve 1 bank 1: posit.feedback - Actual value​

---

Similar measurements were done here -

Lumpy T6 - Triggers Broom! Misfire & Vibration under 1500rpm.

Hi All, I have a 2018 T6, that I'm running out of patience with... and garages... So I have purchased VCDS to try and get to the bottom of it... First issue was it started to use coolant, I swapped a known good charge cooler to eliminate that, Max to min in around 100 miles, So took it to VW...

www.t6forum.com

 

[L30NR]

It is HEX v2 but it's a Chinese copy. I'm very new to it, basically the first time I used it was this week. Is there a specific way after stopping logging that I get the file with it being a copied hex v2? And upload onto here? I'll have a search on that forum for how to's. Thank you for all your help

 

[mmi]

Is there a specific way after stopping logging that I get the file with it being a copied hex v2?

Oops... forgot to mention that the log file will be stored in folder C:\Ross-Tech\VCDS\Logs

It is HEX v2 but it's a Chinese copy.

I'm afraid it's not actually HEX-V2 but a copy of much older interface cable - thus use the "other VCDS versions" setup - because the recording is too slow to capture full set of data at meaningful rate.

And upload onto here?

Yes, please

 

[L30NR]

Thanks MMI,
Hopefully this has worked OK. Only thing I could do is the 'by 7' in uds requests. See pic below -

 

[L30NR]

@mmi I've done a video of the full data you mentioned that would be too slow on my VCDS copy. Please let me know if you want me to send it or put on YouTube if it will help.
Thanks again for all your help

 

[mmi]

If you made a log with full data please post it, too.

 

[Bav]

If you made a log with full data please post it, too.

@mmi doing what he does best - being a VCDS superstar.

 

[mmi]

Below the data from post #11. A snippet of 30 seconds from the middle.



Something is defnitely wrong. Probably the same can be seen also in

Those internal moments should be fairly close to each others. So suggests something wrong with cylinder 2. The wild variations in injection duration are because the ECU tries to stabilize the RPM.

Also the plot looks very similar what @Blind_Spark had in below.

Lumpy T6 - Triggers Broom! Misfire & Vibration under 1500rpm.

Hi All, I have a 2018 T6, that I'm running out of patience with... and garages... So I have purchased VCDS to try and get to the bottom of it... First issue was it started to use coolant, I swapped a known good charge cooler to eliminate that, Max to min in around 100 miles, So took it to VW...

www.t6forum.com

---

For reference how the same data looks on healthy engine

 

[L30NR]

Thanks @mmi slightly worrying after seeing the outcome in the post of bent rods! Mazing work though thank you. Attached is the full data. Please note the van was almost at temp as I had moved it around so idle is much steadier now. Also not quite 2 mins as my laptop battery was about to die. Do you think the reading on cylinder 2 could be anything to do with the new injector if not coded correctly?

 

[L30NR]

Not sure if this will work but here's a video I did with scope to show cylinder 2 when rpm goes to around 1200

https://imgur.com/a/vcds

 

[L30NR]

https://imgur.com/a/vcds

 

[mmi]

Thanks for the full log. Unfortunately the measurement has slowed down - it takes almost 2½ seconds between samples so the quick variations are lost. But yes - RPM looks more stable - maybe.

Do you think the reading on cylinder 2 could be anything to do with the new injector if not coded correctly?

Certainly worth checking - the injector calibration value can be found in VCDS > Engine > Adaptations > IDE00264-Injector 2 correction value
The same value is engraved on the injector.

Then the next steps obviously would be
a) to swap injector 2 with e.g. injector 3 and see if the deviating internal moment follows the injector.
b) compression test on all cylinders

 

[L30NR]

Thanks @mmi ill try those ASAP.
Realised the video links above didn't work so hopefully this YouTube one will work