[DWR]

Originally Posted by TDIwyse

I hope you get your issue sorted out soon. Are you getting any codes?

I'm not entirely sure if the TestO and Torque Lambda/AFR reported values are the actual sensor values, or if it's a calculated value. The way those values snap from 1:1 to something reasonable after ~ 1-2 minutes of driving makes me think it might be a calculated value... However, I've never seen 0's for those.

This is where getting a direct measurement by splicing into the sensor would be beneficial. However, one runs risks by doing this.

CAUTION! DO NOT splice into the wideband O2 sensor lines unless you have the circuitry to do it - looks like that's how iaknown got into this situation. The wideband works in a non-intuitive manner. There is a controller that attempts to match a reference current. Trying to measure voltage is futile and can harm the driver circuits. Interception of the wideband signal takes a special circuit, one that not all piggybacks have, and no multimeters. The actual signal can be measured from TestO. iaknown is getting no signal on all of the loggable O2 parameters.

[TDIwyse]

Originally Posted by DWR

CAUTION! DO NOT splice into the wideband O2 sensor lines unless you have the circuitry to do it - looks like that's how iaknown got into this situation. The wideband works in a non-intuitive manner. There is a controller that attempts to match a reference current. Trying to measure voltage is futile and can harm the driver circuits. Interception of the wideband signal takes a special circuit, one that not all piggybacks have, and no multimeters. The actual signal can be measured from TestO. iaknown is getting no signal on all of the loggable O2 parameters.

Good to know. Thanks to you and iaknown for sharing this with the community.

So do you know if the TestO lambda sensor mentioned previously is the "real" value, or a "processed" value? Is there other values besides this one that can be logged?

[DWR]

Originally Posted by TDIwyse

So do you know if the TestO lambda sensor mentioned previously is the "real" value, or a "processed" value? Is there other values besides this one that can be logged?

That parameter is the OBD PID value. It is processed, but for good reason.

For example, there most likely is a correction for exhaust back pressure. The wideband tries to balance the partial pressures of oxygen in reference cell and 'measurement' cell. Partial pressures increase proportional to total pressure. Note this is different than concentration, which strictly speaking the O2 sensor does not measure. If you are wondering if changing the exhaust system could change the reading, we are wondering the same thing. Some sources say it will, some say it won't. I may have to breakout my wideband meter, at some point.

As far as the list of other parameters that can be logged, there are adaption values, corrections, and other things I can't figure out what they are for. Some appear to be constants, as I have not seen them change over several logs.

[DWR]

Pheno and I continue to work together - as you know, that means he does the work and I run logs and try to look like I am adding value.

I just ran what he calls a "serial" job in Testo. Actually, I have run quite a few of these, but this is the first one that ran without issue. Evidently, communication with our ECU was a little harder to crack. But it is done now. So, what does this mean? How about a sampling rate of 70ms for 4 parameters! It is fast, but that rate also seems to be very stable. Something we did not always see in the other logs.

I'll let Pheno do the official announcement, but I just wanted to say publicly:

Thank you, Pheno!

I'm not allowed to let the cat out of the bag, but I think there are more significant improvements coming. One thing I can say is getting this sampling rate and stability is going to get us closer to a working Road Dyno in Testo.

[TDIwyse]

Excellent.

[iaknown]

Sounds good. When does it get released?

[pheno]

Thanks DWR :-)

Yes, logging just got a little faster than before. I tend to think every millisecond counts in logging and that drives me to seek for better options. Why ? I don't know. I don't even own 335d but I do know you guys are great :-)

With tremendous help from DWR it's now tested and seems to be working very well. Before releasing a new version there's some work left to do to get 'custom jobs' available for this faster 'serial' method. This test was made with fixed amount of 4 parameters.

[DWR]

Originally Posted by iaknown

Sounds good. When does it get released?

The serial jobs are 'hardwired'. To give Pheno a break, I will most likely be the one to put together a selection of these loggable jobs. Please feel free to make suggestions. At this point, I can probably guess 95% of what folks will want to log. I will certainly create a serial version the 'Road Dyno' set of parameters that some have been using to tune their cars.

[TDIwyse]

Originally Posted by pheno

... Before releasing a new version there's some work left to do to get 'custom jobs' available for this faster 'serial' method. ...

How many WTFs/minute are you down to on this new version?

[pheno]

Originally Posted by DWR

The serial jobs are 'hardwired'. To give Pheno a break, I will most likely be the one to put together a selection of these loggable jobs. Please feel free to make suggestions. At this point, I can probably guess 95% of what folks will want to log. I will certainly create a serial version the 'Road Dyno' set of parameters that some have been using to tune their cars.

Originally Posted by TDIwyse

How many WTFs/minute are you down to on this new version?

Ha! Getting better.. only a few per hour. It depends on whether I'm looking at something I did too late and tired.

[pheno]

Here we go.. so TestO works also without Ediabas server as a standalone program. This is possible with ediabaslib project. It means you don't have to install INPA/EDiabas package for TestO. You will need .prg files for your ecu though just copy them into 'ecu' directory.

Since this is a 335d forum I have created a special ready-to-go package for DDE 7.3 US owners. This package contains .prg files for DDE 7.3 and GS19D transmission ecu. All you need is your INPA/OBD2 cable. Before running the program be sure to change com port in 'Ediabaslib.config'

http://phenoboy.kapsi.fi/testo/testo...andalone-1.zip

Enjoy !

[Yozh]

Wow, wow, wow. Thank you.

[iaknown]

Awesome work!

[DWR]

Also, set the com port in Testo.ini. I use COM1, yours may differ. Check in device manager for the correct port.

[Commanderwiggin]

I'm assuming this will work with the X5D as well?
Do you know if a wireless ELM327 will work with Testo as well on Windows Laptops?

[charlie fairmont]

Thank you and congrats

[pheno]

Originally Posted by Commanderwiggin

I'm assuming this will work with the X5D as well?
Do you know if a wireless ELM327 will work with Testo as well on Windows Laptops?

X5D should work also. In the release I've included the ecu file for DDE 7.3 US only. In case X5D uses different ecu file it needs to be copied from ediabas installation. What year model/month is your X5D?

I don't believe using ELM327 will work but you can always try.

[TDIwyse]

Originally Posted by pheno

...Before running the program be sure to change com port in 'Ediabaslib.config'
...

Question regarding this direction... you highlighted the value "COM4" in the screen shot and also make the above comment to be sure to change it. Did you change it from something else to becme "COM4"? The file I downloaded already has it showing "COM4" ... Or are we supposed to make it something besides "COM4"? Or is it supposed to align with the COM port we have set in the testo.ini file? For example, in post 256 where DWR mentions he's using COM1, would this mean he should also have Ediabaslib.config setup to have "COM1" instead of "COM4"?

[TDIwyse]

Well, my tablet has one com port and the system calls in Com 3... Put both of the files discussed above to Com 3 and that seems to make it work.

However, got a couple more questions now.

I still see a whole bunch of ECUs in the list. I see a DDE 7.3 US (Serial) now that wasn't there before. I picked that one.

Question 1: How do I delete all the others so I don't have to scroll down to find the one I want?

Question 2: How many "Jobs" were supposed to be in here? I see 5.

Question 3: The 4 parameters in "Custom set #1" seem to work correctly, but none of the other "Jobs" listed seem to work correctly: "current charge-air pressure", "RPM maf", "RPM maf2", "HP". Are these other ones supposed to function as well?

Question 4: In "Custom set #1" what is "STAT_EINSPRITZMENGE_AKTUELL_WERT"? Is that the total injected amount of all injection events, or the "main" injection event?

[pheno]

Originally Posted by TDIwyse

Well, my tablet has one com port and the system calls in Com 3... Put both of the files discussed above to Com 3 and that seems to make it work.

However, got a couple more questions now.

I still see a whole bunch of ECUs in the list. I see a DDE 7.3 US (Serial) now that wasn't there before. I picked that one.

Question 1: How do I delete all the others so I don't have to scroll down to find the one I want?

Question 2: How many "Jobs" were supposed to be in here? I see 5.

Question 3: The 4 parameters in "Custom set #1" seem to work correctly, but none of the other "Jobs" listed seem to work correctly: "current charge-air pressure", "RPM maf", "RPM maf2", "HP". Are these other ones supposed to function as well?

Question 4: In "Custom set #1" what is "STAT_EINSPRITZMENGE_AKTUELL_WERT"? Is that the total injected amount of all injection events, or the "main" injection event?

There's three way to run jobs: 1) install original Ediabas and use that 2) library that is present in this package that does not require original Ediabas installation. For TestO 1) and 2) look the same only the implemntation of the library is different 3) 'serial' job.. umm.. gotta find a better name for this . That does not use Ediabas at all only plain communication with com port

Main "benefit" of this package is that you don't need Ediabas anymore and you can try a little faster 'serial' job. So pick ecu: 'DDE 7.3 fur N57' - it contains the most interesting jobs. With custom jobs editor (Debug->Custom jobs) it's possible to create/edit new jobs and choose parameters that are requested from ecu. Ok, this is old stuff that you must already know.

(serial) version which is not fully complete yet. Seems there's only one working job I thought they all would work but looks like no. Serial version does not use ediabas interpreter libs at all. I'll let you know when I get serial jobs also working with serial jobs. DWR's tests show that with Ediabas avg speed for 4 params is 100ms per job and with 'serial' it is 70ms per job.

1) edit file: jobs\eculist.txt and leave lines starting with: D73N57C0 and GS19D. .
Also delete all files except D73N57C0_SERIAL.BIV in jobs\serialjobs

2) DDE 7.3 fur N57 has lots of jobs. Custom jobs are on the top of the list. Serial version has only one working jobs 'Custom set #1'.

3) Seems only 'Custom set #1' works.

4) It has to be total amount because there is another parameter for main injection event

[TDIwyse]

Thanks pheno.

[Commanderwiggin]

Originally Posted by pheno

X5D should work also. In the release I've included the ecu file for DDE 7.3 US only. In case X5D uses different ecu file it needs to be copied from ediabas installation. What year model/month is your X5D?

I don't believe using ELM327 will work but you can always try.

I have a 09/2012 X5D.