riouxc

Hello all,

I have an issue with the LR M-Tune kit that is quite odd. At idle there is no issues. When running, the data logger shows erratic data. See vidéo.
I changed all caps and replaced chimps with new ones on the digital part of the ECU/DME except the S700 (odd chip that is a LM1815 with different pinouts...)

New caps and chips

I isolated the analogue side from any possible noise on the 12V power in with a 12v to 12v stabilizer :


Changed the 5V regulator with a switching regulator (TPSM84205) = lower amps more stable 5v output and grounded pin 17 directly to the battery. Yet I still get these odd readings and the car reacts like a fuel cutoff...

Small video:

Any idea ???
Thanks.

markl951

You're probably not the only one who sees those data glitches. My car runs fine but I still see the same thing in the data log.

What year car is your DME from? What are your PC specs. Also what are you seeing for telemetry rate in TunerPro. While it's logging, there is a little status window down at the bottom of the screen that shows real time telemetry rate. When the rate gets low the data glitches start showing up. I have a high end laptop, two different models in fact and I see the same thing with both.

I have gone so far as to embed the DME Logger inside the DME connected directly to the 8051 TX output pin bypassing all the +12V signaling issues of going across the analog board and out the DME connector and it still does it. I don't believe it's a signal integrity issue. I know the serial data rate is quite high. There must be little margin in the serial to USB converter chip inside the DME Logger. Maybe there's a range of data rates across all the DMEs out there. Joshua go it to work on a few of them and called it good? Don't know. I only have a sample of one. I'm just not willing to spend more money on the problem.

Tom M'Guinn

Is your car actually cutting out like it has a fuel shut-off, or does the car run fine other than the odd data log glitches? Can't help but notice the voltage drops erratically in the video -- maybe as simple as a connection or data transfer issue to the logger?

That S100 chip (not 700) performs a similar function as an LM1815, but there's not much info about it in the public domain. It seems to be a custom chip...

riouxc

Thanks for the info. I do get the engine running like it was a fuel cutoff state...

I cannot reach Joshua as he gave everything to Dave and his board has now a very look-a-like from ftech9 ...

This is my major issue, engine cutoff symptoms... I will try without the datalogger, maybe this is the root cause...

riouxc

Yes it does shows a fuel shut-off symptoms.

Thanks for the correction S100... I did not have the ECU in front of me while writing this...

Tom M'Guinn

Can you post a data log (spreadsheet) of the car when it is happening? Looks like the indicated RPMs jumped around at the end too. Did it do this before you changed all the parts on the DME? Not sure yours is a look-alike with the new DME, since they used SMD parts, etc. Sure all the parts you replaced are truly the same as the old ones?

riouxc

I'm not comparing my ECU but Joshua's built to ftech9. I can upload the .csv file latter. Currently busy with than damn pool pump issue...

Will also check with my scope the voltage check on analogue and digital side:


I can also run a digital analyser on this but this would take some time...

Tom M'Guinn

Neither here nor there, but the ftech9 DME uses a single board with SMD, etc. It appears you just socketed existing chips and replaced them with new. If you have any doubt whether that introduced the issue, you might try another known-good DME. Did you do that trying to fix something? Do post up the spreadsheet. When the car shuts down, does it backfire? KLR blink codes? Rattle speed/ref harness and injector harness to check for bad connections. Any protections or overboost settings in the tune? DME relay fresh? Does the boost gauge die when the motor shuts down? Does the motor spring back to life on its own or do you have to crank the key to get it going again?

Auto_Werks 3.6

Mine generates glitchy data, but runs perfectly. In my mind I attribute it to 35 year old wiring harnesses doing what they do.

riouxc

Neither here nor there, but the ftech9 DME uses a single board with SMD, etc. It appears you just socketed existing chips and replaced them with new.: Yes, I know this. I used latest Texas Instruments chips with the same values. Bosch used cheap chimps to save money on manufacturing.
If you have any doubt whether that introduced the issue, you might try another known-good DME. Did you do that trying to fix something? I had an old ECU that was worst. This is a replace one that was tested on a 951 without any mods by my Porsche expert. No issues with this ECU. Refreshing all solders, capacitors and chips was to fix the issue that I still have.

Do post up the spreadsheet. : Will come latter.

Does the boost gauge die when the motor shuts down? The engine does not die on it's own. Pressure is negative on idle and moves toward 0 Bar when I rev. I did not test on boost.

When the car shuts down, does it backfire? Yes. When I shut it down.

KLR blink codes? Will look at this.

Rattle speed/ref harness and injector harness to check for bad connections. Any protections or overboost settings in the tune? DME relay fresh? Harness is brand new, ordered at LR. DME relay bypass to exclude from the source of the issue. I need to check for possible overboost protection.

Does the motor spring back to life on its own or do you have to crank the key to get it going again? Again, the engine does not die.

The harness was changed on all possible sensors and all. The big harness from LR. Installed by my Porsche expert.

Tom M'Guinn

So are you saying you had the exact same problems with the first DME?

The boost gauge is a proxy for power to the DME/KLR. If it goes fully dead, then chances are your DME/KLR is losing power. If the motor is just suddenly losing power, but not dying, then that's probably not it. I'd look into what overboot protections or other protections you have enabled, and check blink codes for clues. To start anyway....

riouxc

Will do a test run later, I was searching a bit and found this op code from Joshua:
Start_overboost_pseudocode:

load = Airflow / RPM
overboost_threshold = overboost_array[rpm]
if (load >= overboost_threshold)
then fuel_cut = yes

End_overboost_pseudocode

Will look into this on the datalogger... Will upload the csv and other useful data.

Thanks a lot for your time!

ttAmerica RoadsterAWD

Hmmm, I had similar sudden drop in pressure/fuel cut-off with that death stutter and car running like shiitt. Turned out to be the MAP (manifold absolute pressure) switch that is very often mounted above the headers on the firewall. It gets too damn hot and stops working correctly. Just for ****s and giggles, rig it so it has longer vacuum hoses and move it to the back of the battery, closer to the glove compartment. Just far away enough from that heat source. You may be surprised like i was. I believe LR is now recommending this placement to correct some of this behavior in their MAP sensors..

Jaime

riouxc

Here is a spreadsheet of a small test run. I did not post all data(some are useless,ie AC Compressor... it is deleted...).

Look at battery voltage that goes to 8.7 V and all others don't make sense...

I will make a KLR led and post latter today. Will also film the dans board to see if I get any echos from ECU or KLR on this (the pressure display comes from the KLR)

Hope you can find something that would cause this...

riouxc

Will try this. Thanks.