Try on OE chipset. Those modified ones might be too lean. You could have a fuel supply problem though - checked fuel pressure yet ?
-
Paul Nugent
Webmaster http://S2central.net
Administrator http://S2forum.com
1994 S2 Coupe ABY - aka Project Lazarus
2001 A6 allroad 2.5TDi - family tank
2003 S4 Avant 4.2 V8 - daily burble
Purveyor of HomeFries and Exclusive agent for Samco hose kits (S2/RS2)
There are only 10 kinds of people that understand binary - those that do, and those that don't
-
Do you have a subsitute air temp sensor to try? I guess you could measure the resistance at what ever temperature Vag com is registering to see if its plausible. Or actually remove it to measure ambient temp/resistance. Perhaps its over reading and I guess that wouldn't be flagged as a fault.
PRJ developed sw specifically for the S6 so that might be something to consider. How superior the S2 charge cooling is over the S6 only he knows I guess!
I notice that a company that mods Porsche has an add on module to datalog Bosch DME with its own SW. I know PRJ does the same but no idea of the costs or whats involved in terms of sw and ecu mods. It is an area that is somewhat neglected and having the data in itself is not necessarily useful unless we know how to interpret it and have a data bank of logs for comparsion.Comment
-
Yeh, im going to swap to the stock ecu but completely forgot to take it into work with me today
Fuel pressure; yes i have checked it numerous times!
If i get nowhere with the stock ecu, is there anyone on here with another standard car who REALLY knows their stuff with data logging?? Its ok me reading the data but i have no bench mark so to speak!Comment
-
Might also be worth checking the maf in Vag com. Presumably the figure for the AAN should be the same as ABY:
Other than that as others have said, trouble shoot with stock ecu. Leak test it as recommended i.e through the Maf hose with the pcv pipe removed and hose blanked off. Measure any important sensors (particularly maf and inlet temp).
As said previously perhaps this sw is not suited to the dynamics of the S6, especially if it was derived from an S2 which is likely.
I think I remember this vehicle as being particularly troublesome, almost if it was something on the vehicle was effecting things. Unfortunately all you can do is test and eliminate things. I know its nice to brag about power output but sometimes its better to have 95% available all the time than 100% once!Comment
-
Porsche DME logging was done from factory via breakout box, at least all the older ones. That won't work on AAN as there are too many signals really...
I wrote fast-mode logging for a 944 turbo recently, put a RS4 maf on it and so on, no breakoutbox, all via serial from the ecu on 93750 baud. There is some peek&poke functionality from factory in the ROM, but it was never used, because the RxD line is grounded on the MCU on the PCB.
I have software for 2.2T with on-demand datalogging. Basically it's triggered by holding your foot to the floor when switching the ignition on. If you do this, you will have normal diagnostics and can diagnose all modules in the car, if not the ECU goes into a fast streaming mode where it outputs about 25 different parameters 13 timers per second. That includes knock pull for each cylinder, N75 duty cycle and so on.Comment
-
-
Cool, I wrote the same thing last weekend. Dyno tuned the thing on tuesday using the logging.
It's a LOT easier than doing it on AAN though. That ECU is ten times more basic.
And it's ****. I would not do it again, hell megasquirt is more advanced than old ML3.1 - it does not even have a cam trigger. Batch injectors, Batch spark etc.
220hp stock. 337hp tuned - turbo had a bigger comp wheel, bigger injectors were fitted and a MAF from a RS4.
Last edited by prj; 31 October 2014, 19:18.Comment
-
If its that primitive perhaps it will work on the 3B!
Seriously though, can you explain what would be required to have access to the data logging apart from money!Comment
-
Ok, so you need to check for leaks BEFORE the turbo and in the PCV system so try pressurising from the MAF hose with the PCV still connected. You have to just take the big hose off at the MAF, fit a big bung with a valve in it and pressurise. Only go to 1 Bar because you're checking the MAF hose it'self and the PCV system and the crankcase which normally only sees vacuum.Comment
-
And
Maf connector thread
Did you check the DV for leaky diaphragm?Comment
-
No interest in 3B, screw that. And 3B is not really that primitive. It's still M2.3, which is three generations ahead.
M2.3 boost chip runs at 8mhz, fuel chip at 12mhz.
M2.3.2 boost chip runs at 12mhz, fuel chip at 12mhz.
Possible fast baudrates are Fosc/32 and Fosc/64.
That means for the 3B boost chip you have 125000 and 250000, while for everything else you have 187500 and 375000.
If you drop to slow baudrates via onboard generator you get Fosc/2500 and Fosc/1250, which will be 3200, 6400 and 4800, 9600 respectively.
The way I have done it on M2.3.2 one CPU starts sending data first and when it's done with it's packet sends a marker, the other CPU intercepts said marker, and then sends it's data rapidly afterwards. Rinse and repeat. Everything happens at 187500 baud - the two CPU's can see what the other one sends due to the way the ECU is built.
All that is left to do is listen in on K-line and sync to the first processor's end packet marker, et voila.
For 3B one would need totally different code, because there are no common bitrates for the two CPU's and they can not see what each one sends.
So you need to build listening logic into both cpu's, and have the receiver on the computer as an orchestrator, who sends a magic marker, to which the first processor responds, after packet is received, reconfigure the baud rate, send another magic marker to which the second processor responds, then after packet received again set a compatible rate for the first processor and so on.
Lots of ballache, and for what? I almost never see 3B engined cars. Not worth it.Comment
-
Perhaps you misunderstood what I was trying to determine.
I would have thought there was a market for your data logging even for people with stock cars. So I was trying to establish what components were necessary. Is it just SW, a WBO2 a VCDS lead and a computer with a program?
I appreciate for the ordinary person the data might not be of much use but if they could for arguments sake post the data up, it would allow other more experienced members to comment. In fact perhaps even a comparsion between a log taken when healthy and unhealthy might be useful given the somewhat restricted inormation provided my VCDS.
Basically I am asking how it works in practical rather than technical terms !Comment
-
1. Custom software in ECU.
2. Hardware interface.
3. Software for computer.
It is possible to create a patcher, that will patch any current version of software.
But this means extracting the software from the ECU, patching both boost and fuel chip, writing it back on chips, re-fitting it into the ECU.Comment
-
So can the SW be added to either stock or modified maps?
Is the hardware interface what we use for VCDS or something different?Comment
-
Comment
Comment