Misfire/stutter under load/boost. AMK Engine

[classy]

I've got a 20v AMK engine fitted into a mk2 golf. It's been in a running fine with no problems for the past month or so.

Last night I was having a spirited drive on the way back home when the car started to sputter/heavily misfire when under boost/load, when building early stages to about 0.5 bar it'll happen. If say already at 3k in 3rd no load, if you were to put your foot down engine sputters, holds back. You can drive the car 'normally' with no load any quick build of boost, you probably wouldn't realise there is any fault with the car. Doesn't seem to be speed dependent either, just when the engines loaded.

At standstill, revving the engine - no problems.

I have VCDS - I've had certain faults codes always in there since the conversion, they are second lambda heating/no signal fault. CAN bus faults from other controllers not connected & intermittently a EGT impausible signal fault (keep in mind, this fault has been here the whole time the car was running fine) & N80 charcoal canister purge valve that isn't connected in.

Other than that, there are no new fault codes relating to the most recent problem.

I've been out done some logs, watched the data. Nothing occurs on the misfire counters, after a quick look I cannot see anything in particular to be flagging up out of the ordinary. I've disconnect the air mass, n249 & n75 valve at different times & no difference has been made to the fault. One thing I did notice is when I disconnect the n249 valve, the ecu thinks that the n75 valve is disconnect & vice versa - though this is possibly my fault from the conversion mixing up the wiring even when the plugs were swapped over the fault remained the same. I've taken note of the wire colours & pins & will check them tomorrow at both ends anyhow.

I realise after a bit of digging around for similar sounding faults from other members the EGT sensor could be at fault, though temperature reads fine & doesn't ever seem to be implausible when being logged, the fault (EGT fault code) only seems to pop up after the cars up to temperature. Brake light switch is also new (1 month old -after reading it could be at fault). The coilpacks are also 4 + years old so tempted to start there. Turbo was new/recon when fitted after a months use along with the actuator.

I'm struggling to find a way to upload logs without taking a screenshot of a bit of it, is there a way to upload the whole log? So far logs taken are: 001/031/112 & 002/115/043

Any thoughts/input welcome.

 

[<tuffty/>]

Host the pics somewhere like photobucket and link the full size img using the image button...

Normally use windows 7 snip tool to just cut the section you need rather than display all the irrelevant data inbetween runs...

<tuffty/>

 

[<tuffty/>]

I assume this is running the std map?

<tuffty/>

 

[classy]

It's OE management but apparently it's remapped from awesome GTI back in the day, I took it down to bill @ badger 5 for a smoke test a few months back & he was looking into who's software it was on the ECU but I can't remember what the outcome of it was, or if there was one.

Thanks for the tip, I was trying to hose the excel file, just copied it into paint. Sorry about all the excess in the logs.

 

[JonnyA3]

i recently had a similar problem with my a3 AJQ engine code. was driving fine till around 3k then shuddering/missfiring. about 3 weeks previous to this i had replaced the MAF (from unipart i think) and it was faulty, so got refund and bought myself a genuine bosch MAF and the problem is completely gone, runs perfect now. hope this helps

 

[classy]

Unpluggin the MAF made no difference, readings look ok on the logs also? How did you diagnose yours?

Cheers.

 

[badger5]

103g/s max? very low.
boost request of 1.2bar, and you look to be in actuator only pressure, soft limp.
N80 disconnected = no lambda adaption normally, although block 001 shows adjustments.
Clearly N75 and N249 being reversed is going to mess it up... Whats controlling your boost now?

looks lean on the logs. possibly fuel pressure etc.

I dont remember the map I read tho... looks like a 1ML ecu on VCDS, but was'nt this an 8L0 ecu? APR map? Were you here on Sept 11th?
no egt sensor would also (should) be flagging 1100'c on 112, and fuel dumping.

 

[classy]

Bill, it wouldn't of been far off that date, I bought it down for a smoke test - had a few leaks from the breather pipes underneath the manifold & also purchased a v3 tip off you & a bit of pipe to bypass the leaks under the mani - it's a red mk2.

The ECU is from a seat leon cupra r, the original audi ecu got fried somehow before it was ever used.

The n75 & n249 being mixed up is something I can't get my head around, I've not had chance to look up the wiring yet but for the past month the car's been running spot on without it being an issue - unless this now a fault derived from the mix up. The n249 had been bypassed to.

The EGT fault is intermittent & only happens when the cars warm (pulls a code, I've not logged it when it's actually not been reading data), this fault happens from cold & the EGT as per the logs reads fine as far as I'm aware. It's also a £220 I found out today so won't be replacing it until i'm 100% sure!

Have resistors for the n80 now, on the to do list but can't see it causing the above fault as it's been there all the time.

The air mass I've just compared to another log, though to be perfectly honest I'm still learning & I'm not sure what values I should be seeing with certain sensors - air mass being one of them, i need to clue myself up on lambda a bit better - I can never remember which side of 1 is lean/rich & same for the voltage, think it's reversed?!

It looks as if it's definitely holding back the boost, I have a fuel pressure tester I'll be running tonight to test & I've also got 4 coilpacks to replace the originals as they're old as gods dog anyhow, will be checking the condition of the plugs whilst I'm in there - it's already had the wiring harness to coils replaced.

Here is an older log taken when the car was in good health:

 

[badger5]

a lot more airflow back then obviously..
turbo looks slow to get going tho, 95% duty cycle and 1.1bar boost at 4600rpm

fueling is on block 001 for adaption live lambda adjust, and 031 request and actual fueling.

 

[classy]

Fitted the 4 brand new coil packs & had a look at the plugs & all is well again - pulling like a train.

Need to find out which coil(s) are the weak ones, as the coil obviously had not conked out completely - I'm assuming it has just weakened & didn't have enough in it to produce a strong enough spark under higher cylinder pressures when the boost started building.

Thanks for the input, I'll be in touch soon regarding a custom map once I've pulled my finger out & sorted out the teething problems!