[rjahl]

With help from Pheno, some perseverance and a little luck we have found the settings to make a huge improvement in the Testo logging speed. The last log I took was running about 10 frames per second, this is about three times faster then before.

[Taskmaster]

Originally Posted by rjahl

With help from Pheno, some perseverance and a little luck we have found the settings to make a huge improvement in the Testo logging speed. The last log I took was running about 10 frames per second, this is about three times faster then before.

So...tell me again how I can make this work? I'm here with my laptop and cable.

[rjahl]

On the computer or VM that you use for INPA download in Testo, unzip and run Testo.exe.

I"ll try to write detailed instructions later today. At a soccer game right now.

Edit: Pheno is making changes to TestO that will make things easer and faster. I think the new version should run fine without INPA and be really easy to set up. It's not actually "installed" you just run an .exe file.

I'll hold up with instructions until I can create them using the new program.

[rjahl]

As promised, the following is a quick start

There are two public versions of TestO. One relies on a working installation of Ediabas and the other is stand alone.


Stand Alone Version
http://www.e90post.com/forums/showpo...&postcount=319


Regular Version
Main Link and help is on another forum. Google is your friend.


The stand alone version is an executable file with libraries and folder and does not "install" on your PC.

Step 1,
Just download and unzip where you want. It does not matter.

Step 2,
Modify the Ediabaslib.config file with a text editor to tell the system where your ODB cable can be found Com1, Com2 etc. Com3 on my PC

<add key="ObdComPort" value="COM3"/>

Step 3, Copy the ECU files from your INPA installation into the TestO/ECU folder. For us that will be MSV70.prg and MSV80.prg. I'll attached the files to this thread if allowed.

Step 4, Optional, copy the my File Customjob.xml, into the TestO/Config folder. I'll attach this file as well. I have created two custom jobs for the N52 to get you started. You can build these on your own when you understand how things work.

Step 5, Run Testo.

[rjahl]

Once you have TestO Running.

Scroll down through the ECU list until you get to the MSV70 or MSV80, type enter

Now the list of available jobs will appear under "Jobs". Select "E90 Post" and the Job will appear on a pop up screen to the right

Right click anywhere in that new window and enter "select all" and all of those parameters/jobs will become active.

Now back to the "file" menu and pull up "file logger" window.

Hit the start button when you are ready to record and stop when finished.

Pheno is still actively making improvements to the program and I think we will hit a bug or two along the way. If the program act up, Just shut it down and restart TestO.

By the way, Pheno has put a lot of time into this program and he provides this free of charge so a donation might help keep his interest up.

[rjahl]

Once you have created a log, TestO will unceremoniously drop two files into it's root directory. You will be interested in the CSV version. They will be automatically dated and labeled.

If you keep the logs short the trail version of Magalogviewer can easily read these logs. If you are good with Excell, it's also a good tool. Up to you.

[hassmaschine]

nice, now I just have to drag my laptop to work with me.. lol

I didn't drive the E90 today, but those days are coming to an end.. the rainy season is upon us!

[Taskmaster]

Get my tune today, so we'll see.

[rjahl]

Originally Posted by hassmaschine

nice, now I just have to drag my laptop to work with me.. lol

I didn't drive the E90 today, but those days are coming to an end.. the rainy season is upon us!

I'm really interested to see which "jobs" you find the most useful. Now that we have some real data rates, I want to look at the list again.

I tried simultaneously pulling data from the DME and the transmission module thinking the two ECUs would not conflict with each other but the data rates fell to an unacceptable rate, something like 2 frames per second. I guess that was just being too greedy. LOL

I'm going to look at the torque outputs again. Maybe I can find a way to develop a calculated flywheel torque from the data.

[rjahl]

I did my first run with the Updated version of TestO this morning. It's great!

Lots of good data and faster data rate caught things I could not see before. I'll build graphs later today, time permitting.

Car is not running too bad, Looks like it pulled 0-60 in just over 6 sec and that's me being easy for the first 1/2 second and the Auto pulling power for two shifts totaling about 1 second. Also the car has 150,000 miles on it.

[pheno]

I'd say your car is running very well !

So MSV70 has wide-band lambda(s) ?

Edit: one thing about logging, if you choose 'time from start' in File Logger window for time format I remember that MLV showed the time axis correctly (actual time).

[rjahl]

Yes, we have wide band sensors. I don't have these in the custom job. But I plan on adding them. I've only found a voltage output so I will need to find a way to translate voltage to some usable measurement. Today's log had a bunch of torque jobs and I need to sort through them first. One piece of the puzzle at a time.

I'll try the different time setting on my next run.

[hassmaschine]

Originally Posted by rjahl

Yes, we have wide band sensors. I don't have these in the custom job. But I plan on adding them. I've only found a voltage output so I will need to find a way to translate voltage to some usable measurement. Today's log had a bunch of torque jobs and I need to sort through them first. One piece of the puzzle at a time.

I'll try the different time setting on my next run.

I'm pretty sure I posted the conversion factor dump from Tool32 in this thread or another one.. hmm.

[rjahl]

Now that you mention it. I still have your file on my desktop.

I'm going to pull it up when I get home.

Back to school for me.

[rjahl]

Originally Posted by hassmaschine

I'm pretty sure I posted the conversion factor dump from Tool32 in this thread or another one.. hmm.

OK, I found the link to the wideband sensors. I probably should have shut down the traction control for the run. All of the first gear data is messed up, cleans up through second. MAF data has a little scatter in it today, not sure if was the heat or if it's getting dirty.

I'm now thinking that the function "STAT_MOTORMOMENT_IST_WERT" is the best torque value to use as a baseline for changes.

"STAT_MOTORMOMENT_AKTUELL_WERT" is normally close but reports the exact value of a torque reduction request from the transmission.

STAT_MOMENTANFORDERUNG_KUPPLUNG_WERT, does not record a a toque reductions at all, seems to be a watch of sorts.

STAT_MOTORMOMENT_SOLL_WERT, is the only value that I can pull back into the Parameters section of the tune. These values seem close to the "drivers wish" maps, if I can call them that.

[hassmaschine]

yeah, "ist wert" means actual value (kinda misleading when you look at "aktuell" and think it means literally the same thing!).

soll wert is the target value. your hunch would be correct.

[rjahl]

Originally Posted by hassmaschine

yeah, "ist wert" means actual value (kinda misleading when you look at "aktuell" and think it means literally the same thing!).

soll wert is the target value. your hunch would be correct.

Thanks,

Not much progress from me this week end. I wanted to move up the ignition maps a small amount but I had underestimated the complexity and got lost in research instead.

Discounting the Maps designed for quick heating the Cats or the emergency map, I count three reference ignition maps ( VVT, VVT Safe, GD). I assume these are the optimum ignition angle based on the fuel quality.

VVT Reference has two compliment maps, VVT vollast Ron 98 for full load and VVT Ron for throttled.

VVT Safe (Sicher) reference has only a basic map compliment.

GD (throttled) reference has only a basic map compliment

Is it safe to say that all of these maps must be treated as one complex web and the DME will interpolate between the VVT and VVT safe modes. Or does the DME switch between them based on the fuel "switch".

I am also thinking Sicher or Safe is really the flag for good fuel and higher advance.

[hassmaschine]

I have answers for you but I'm sort of slammed at the moment. the long and the short of it is, you don't need to worry about all those maps. Mainly the VL (vollast = full load) maps and a couple of the part throttle maps.

[rjahl]

Originally Posted by hassmaschine

I have answers for you but I'm sort of slammed at the moment. the long and the short of it is, you don't need to worry about all those maps. Mainly the VL (vollast = full load) maps and a couple of the part throttle maps.

No worries, when you have time.

I'm up to page 270 plus with the M3 group. Funny, I remember reading some of that material years ago but lost interest telling myself that I would never "get in" that deep. LOL

[rjahl]

For anyone trying to follow this.

Sorry, I don't have anything to add. Too busy at work to make any progress. Work load has been horrible.

In my free time I have still been trying to sort out the ignition maps. Maybe, I'm trying to hard to rationalize the maps used for Torque calculation KF_ZWOPT_XX against the regular ignition maps.

I am also a little worried that my ZF gear box is about to drop out. I've picked up the infamous "Seal bark" on my 4>3 down shifts. Research tells me that a fluid change should be the fix. Next week end will be a transmission fluid change, oil change, trip the orchestra and a dinner party.

If the "seal bark" does not go away I will be car shopping.

[hassmaschine]

Same boat (heh). On top of that, my wife hit a deer - so i haven't driven it in 2 weeks. Thankfully the damage is really minor, basically just a broken light. She did a good job though, there were three deer and she dodged two of them and just clipped the third. It could have easily been totaled!

[rjahl]

I'm back on the road again. Did the transmission fluid change with new filter and ZF fluid. Gearbox has not barked or made funny noises since. Old fluid did not look horrible but seemed to lack surface tension of the fresh fluid. Gear shift are certainly firmer but not great. This gearbox has always had soft shifts.

To anyone wanting advice; forget the lifetime sealed transmission story, change the auto trams fluid every 50,000 miles. Late is better than never.