New PITA

jeepguy553 · Post by **jeepguy553** » Fri Apr 16, 2004 8:08 pm

I did a good bit of datalogging that gave me a good BLM table and a good INT table. I made an Excel spreadsheet that recalculates my values for my VE table. I went into TunerPro and entered the values by hand and saved the changes. Then, I burned the chip and verified it with the buffer. I put it in the ECM and ran it. It ran like crap. Sooo...I pulled the chip and read it with TunerPro RT. I saved the buffer to a file and opened it. The VE cells that I changed were not even close to what I programmed into it...by a LOT.
What did I do wrong?

Post by **Mangus** » Mon Apr 19, 2004 3:59 pm

Didn't you say you can no longer reproduce this issue?

Also, note taht you can copy a table from Excel into a TunerPro table and vise-versa (right click in the tunerpro table, you'll see what I mean).

jeepguy553 · Post by **jeepguy553** » Wed Apr 21, 2004 11:43 am

Mark, I am starting to believe that the issue is related to my Windows ME OS. It is so random that I have just started burning the chips with my XP machine and doing datalogging with my Win ME laptop until I can get Win 98SE loaded onto it. Everything works just fine under XP with regards to the USB AutoPROM and TunerPro RT.
I did notice that the "Save BIN As" command will often cause an error in USER.EXE and shut TunerPro down. This is a recent development and I believe it is probably related to the degrading of my crappy OS.
I am wondering if the datalogging troubles I was seeing are also related to the Win ME OS on the laptop. WinALDL works great, but it just won't hook up right in TunerPro RT...again, probably due to the Win ME setup. I'll let you know how it works with Win 98SE.

Post by **Mangus** » Wed Apr 21, 2004 11:58 am

a crash in user.exe is outside of the control of TunerPro, and is VERY likely to be an OS issue.

jeepguy553 · Post by **jeepguy553** » Wed Apr 21, 2004 4:08 pm

That is exactly what I thought. I am almost certain that the Win ME OS is crashing. I have a little time to fool with it tonight. I may go ahead and wipe the HDD and put Win 98SE on the thing. I am sure it will work better. Refresh my memory. I need that riched32.dll file for TunerPro RT to work properly under Win 98SE, right?
Thanks, Mark!

Post by **Mangus** » Wed Apr 21, 2004 4:10 pm

Yeah, you may need to place the .dll in the system path (windows directory or windows/system directory). Alternately you can place it in the TUnerPro root folder.

jeepguy553 · Post by **jeepguy553** » Wed Apr 21, 2004 5:30 pm

Well, I am in the process of swapping over to the Win 98SE OS right now, but my laptop isn't cooperating. I had to kinda fool the setup program by opening it up in a pseudo DOS mode and actually formatting the HDD. It wouldn't let me install Win 98SE over Win ME. It found one bad cluster on the HDD during the format process, so that my very well have been the problem. Isn't that kinda fitting...the ONE cluster that hold important system data HAS to be the one that gets smoked...
Once I get it finished, I'll let you know the outcome. I bet it works like a champ. Your stuff has worked VERY well up to now. I think that Win ME should be outlawed as an OS. IT SUCKS!!!

jeepguy553 · Post by **jeepguy553** » Fri Apr 23, 2004 3:35 pm

Mangus,
Well it worked after I played with the port settings in Win 98. WinALDL worked fine, but I keep getting nothing but packet errors from the TunerPro RT ALDL module. I had to find the right combination of port settings for it to work with WinALDL. I got a couple of pretty good datalogs off it.
Is there ANYTHING else I can do to get TunerPro RT to hook up to this 8746 ECM? At this point, it is likely that the definition is to blame. Right?

Post by **Mangus** » Sat Apr 24, 2004 1:10 pm

Yes, its the definition. Email Jonas and see if you can get the ECM ALDL information for that setup from him. That will allow you to create a definition with certainty.

jeepguy553 · Post by **jeepguy553** » Mon Apr 26, 2004 3:38 am

Well, I heard back from Jonas and he told me that he used A41, A83, and A90 for the 8746 definition. I am still trying to figure out what parts of each he used.

Post by **Mangus** » Mon Apr 26, 2004 9:13 am

Well, the good news is that each of the streams is 20 byte length. This means you should get real data if you set the packet size to 20. =)

Then its just a matter of interpretting the data correctly. From the looks of it, the first many bytes of the stream are identical between these!!!

You shouldn't have any issues creating a def now that you know this.