Audi 90 with AAN/GTX3071 - Tuning Time

[Marty]

Howdy folks,
I'm Marty; currently installing a VEMS in my B3 Audi 90. A few specs to start:
- AAN motor, Garrett GTX3071 turbo with Siemens Deka 550 cc injectors and 5x LS2 ignition coils.
I believe I've got most of the preliminary items installed and configured but I'm not having any luck firing it up.
I have not yet put a timing light on, and I know that'll be the first recommendation. I should get around to that in the next day or 2.

I was hoping someone could take a look at my config file and let me know if they spot any obvious glaring issues. I'm not extremely clear on some of the help info so I've left a lot of items default.

Here is a link to my Wiki where I have a few shots and a link to my current config file.
http://www.vems.hu/wiki/index.php?page=MembersPage%2FMarty
Any help is appreciated!

[Marty]

I did my timing check, and everything looks good. I'm triggering about 2 or 3 degrees BTDC.
There are several unusual things: Whenever I crank, I get a few seconds and then the laptop->ECU connection is dropped. I'm not sure if this is expected but this seems to make it difficult to get a vemslog with much detail (or so it seems?).
I am also seeing 0 RPMs on most of my attempts. Not sure why that is, since my triggers seem to be appearing correctly?

[Marty]

Direct Link to my config file:
http://www.vems.hu/files/Marty/v3.3_u004574-A-2012.05.24-12.58.57.vemscfg
Direct Link to my log file:
http://www.vems.hu/files/Marty/v3.3_u004574-2012.05.21-23.03.03.vemslog

[Joof]

May help to throw a charger on your battery? I was getting random disconnects while cranking when I was trying to get up and running...voltage drop while cranking was my issue as the battery was quite low from a number of unsuccessful starts.

I don't know how the firing order is on that car but I would expect (if you have a nice clean triggerlog and are confident in your wiring) that it is something with firing order/trigger settings.  Set the cranking ignition advance to zero and crank the car with the timing light on it...if it's not strobing at tdc then you need to adjust your trigger teeth settings.  This should be the first place you start.

Good luck!

[Marty]

That was a good tip prior to last night actually. A close observation of my trigger log may have shown that it was barely cranking at the time; however yesterday I traded my crapped out battery in on warranty and put the new one in. Cranking was significantly improved, but lo and behold: the disconnect continued to happen.
- I'm presuming from your reply that this is not expected behavior.

[Joof]

Nope, sounds like noise in that case, do you have a shield in your dsub connector?

I'm not near vemstune so I have yet to look at your log, but regards to starting- are you confident  of the firing order in your config? I assume you've checked that you have fuel pressure/fuel at the rail? (I.e. the pump is kicking on?)

[Marty]

I ordered the prefabbed harness, so I'm not likely to suspect a shielding issue. Perhaps it's a crappy mini-PCMCIA serial adaptor. I may have an alternative adapter cable around. I'll take a look.

Firing order is 1-2-4-5-3
Fuel Pump is priming
I just tried to understand the ignition Outputs Visual page again and I've made some adjustments.
When my Cylinder 1 is at TDC, my 60-2 wheel is on the ~17th tooth past the missing tooth.
I've changed around the settings now hoping I understand correctly. On the Angle slider, I have pulled the 3 degrees of timing to get it to zero (with timing light). I have yet to verify after this change that I'm triggering at TDC.

I believe I misunderstood these values before, but my wiki link has an attached image of the previous setting.

[Joof]

TDC after trigger of 3 degrees is likely too small to charge the coils.

Reduce your FTT to get that "TDC after trigger" over into the green (to the right)

I would send you mine, for reference but again i'm not near vemstune/computer
http://forums.bimmerforums.com/forum/showthread.php?t=1719609
Here's something that may help explain that a bit better...

Know of anyone running vems on similar engine? Maybe you could find a config and copy their trigger settings over and then confirm with a light.

Were you running 0 degrees crank advance under starting/idle>cranking>ignition advance when you had the light? 

[Marty]

OK, Thank you very much for that link. Helped greatly
Here is my new setting:

[Joof]

Cool.

Now set your cranking advance to zero degrees and verify that the timing light strobes at tdc mark.

(May want to disable the fuel pump relay first)

[Marty]

Looks like I'm getting injector pulses if I'm reading my log correctly and I have a light on my #1 plug so I can see it firing. I haven't re-verified trigger at TDC because I need a second set of hands. Seems to me it should be right on if I thought it out correctly, but I'll handle that as soon as I can
Here's a log file image

[Marty]

A few shots


No decent shots of the turbo, but here I was ading an oil pressure test rig


Fear not, I have some acrylic tornado red just waiting for the day...

[Marty]

Tearing my hair out here, and I'm all outta hair!
I'm hoping someone can take a look at things here and help me diagnose. I have been in my garage every single day for some time now trying to diagnose this most rudimentary of #%&^*$ problems
Here are my symptoms and points of interest:
- I have verified my Spark, and spark timing with a strobe
- I've tried 0, 8, 10, and 12 degrees of cranking advance
- I can smell fuel, at times I've cranked enough that the air is dense with it and my eyes are watering
- I've adjusted based on theory that it's too rich, and I've adjusted based on theory that it's too lean
- I have tweaked my fueling PW based on my own theories, and when that failed I have tried every single value I could find in any vemscfg for any similar cars available on the internet...
- Also tweaked my fueling PW based on every single value I could find in ANY vemscfg for any Other cars available on the internet...
- I am using a harness built by VEMS, including the ECU and the Relay/Fuse Harness so wiring is not (immediately) suspect, However:
- As far as Harness Terminations go (injector and sensor plugs), I have verified every single one of these more times than I can count, using visual, multimeter, test lights, noid light, spark indicator, tug and pull, VEMS Test modes...

Never ONCE in my entire experience (1000 start attempts?) has fuel ignited at the expected time in any cylinder... Not once!!
-Occasionally, I get a very brief (audible) burst of pressure. Last night I noticed that my boost gauge is jumping to 10psi when this happens. I also smelled combustion when this happened yesterday. I should note that this engine was running fine prior to changing the ECU (if that removes any suspicions about broken internals). I do not know what this would be a symptom of, but it is quite occasional in relation to my number of attempts to start.

Now since I have a properly timed spark, and I can smell the fuel, the only remaining possible fail modes I can think of are:
 Fuel quantity
 Fuel Timing

my cylinder firing order
1-2-4-5-3
my corresponding injector order
4-5-3-1-2

Could anyone take a look at my log? I've linked it as an .xls. hopefully someone can find meaning in this and help me.
I'm starting to think if I poured a can of gas on this car and threw a match I couldn't get it lit........
http://www.vems.hu/files/Marty/VEMS%20log.xls

and a link to my current config
http://www.vems.hu/files/Marty/v3.3_u004574-A-2012.07.11-20.24.12.vemscfg

-Also, is there a legend I can use to identify what the column headers in the (csv version) vemslog mean? -For example there is a column called "Engine", and I don't know what it is





 

[GintsK]

Quite often in such cases reason is unpowered system components (coils, pump, injectors) during cranking. Haven't you?
Next reason is no compression. Cam timing way off, valves never closes etc...

Triggerlog file with just 1st plug screwed in would be helpful also.

Do you use Hall sensors? For VR trigger edge must be rising.

Gints

[Marty]

I do still have a strange case where my VEMS says "Disconnected" during cranking, but I can not determine why.
- My voltage drops to ~10.2V perhaps this is the reason. Is there a minimum voltage threshold?
Or it is expected behavior that when generating logs, the UI communication is disconnected to save bandwidth for logging?

Since I'm using the VEMS fuse/relay harness, it seems the power and ground connections would be difficult to confuse. My plugs are sparking, and my injectors test out ok. My trigger log shows what I *believe is correct.
I'm using a 60-2 VR sensor on the crank, and a Hall sensor on the cam.

Perhaps I need to evaluate my alternator wiring.

I will retest my compression, but I have done this since the last time the engine started successfully, and the values were ok.
I'll get a 1 plug vemslog this evening