DME 413 (silver) knock correction

Discuss Bosch (Porsche, BMW, Volvo, etc) tuning topics here. Request definitions, discuss parameters, etc.

Moderators: robertisaar, dex

Post Reply
farjeman
Posts: 11
Joined: Sun Mar 26, 2017 10:13 am

DME 413 (silver) knock correction

Post by farjeman »

Hi,

Dose any one know if the 413 dme is actively correcting the ignition timing in case of knock or the logic of it?
Limp mode/scaled table/nothing...

Also, if so, can we pull out the correction data of the dme?
Thanks.

Yaniv
olafu
Posts: 186
Joined: Tue Jul 26, 2016 12:35 pm
Location: Finland

Re: DME 413 (silver) knock correction

Post by olafu »

Correction data location wasn't found yet (it has individual load x rpm table in RAM for each cylinder), but i will send you XDF for silver label knock part.
farjeman
Posts: 11
Joined: Sun Mar 26, 2017 10:13 am

Re: DME 413 (silver) knock correction

Post by farjeman »

Thank you!

You think it's stored elsewhere then the eprom or the dme can write over it during operation?
I am now intrigued to compare the last flash file what I'm using now to see if the dme changed it.

If i can do something to help with finding that illusive table I'm up for the task!

Faragy9@gmail.com
olafu
Posts: 186
Joined: Tue Jul 26, 2016 12:35 pm
Location: Finland

Re: DME 413 (silver) knock correction

Post by olafu »

https://www.dropbox.com/s/ljl0nz0325ix8 ... k.xdf?dl=0 <- knock parts for silver label. Tools -> Custom tools -> New TunerPro Instance ... copy paste.

This ecu has 8kB SRAM chip on underside of the board. It's used for storing self-learning adaptation values (including knock-correction), fault codes, freeze frames etc. And as general RAM.

EPROM writing is possible, but this function is never used in normal operation.
farjeman
Posts: 11
Joined: Sun Mar 26, 2017 10:13 am

Re: DME 413 (silver) knock correction

Post by farjeman »

Awesome, thanks!

I'm guessing reading that part of memory is only by OBD if so, because it is volatile without voltage.
I tried to get live data using INPA/ADS but no luck /:

Not be able to rewrite the eprom make sense..
olafu
Posts: 186
Joined: Tue Jul 26, 2016 12:35 pm
Location: Finland

Re: DME 413 (silver) knock correction

Post by olafu »

You can read ram with INPA, but i can't say where those knock adaptation values are stored. And i'm pretty sure they are in map form, no idea what kind. And own map for every cylinder.
farjeman
Posts: 11
Joined: Sun Mar 26, 2017 10:13 am

Re: DME 413 (silver) knock correction

Post by farjeman »

For some reason i'am unable to connect to the dme (yet...), maybe my installation is wrong..
You someone who did? (:
olafu
Posts: 186
Joined: Tue Jul 26, 2016 12:35 pm
Location: Finland

Re: DME 413 (silver) knock correction

Post by olafu »

My tests with INPA was with -506 DME's -965 bin, because it's not working with 413 DME's -623 bin. I have no idea why, but if you have emulator, try to run your DME with -965 bin. Hardware is cross-compatible. If it works, the problem is propably in the INPA software. You can cheat it with using -965 identification data, so inpa thinks it is -506 / 965 DME and allows communication. It will work, but don't trust anything what it tells you.

I didn't test with -413 silver label bin.
Hairyscreech
Posts: 196
Joined: Tue Jun 20, 2017 3:19 am

Re: DME 413 (silver) knock correction

Post by Hairyscreech »

The biggest advantage to the INPA output that I can see is the ability to put out the data in the registers contained in the ECU processor, there are a few in the low numbered addresses that are not on the RAM chip but are in the MCU.

If you are trying to connect to a Red label 413 ECU then you will need the M3.3.1 fix for INPA that is in the dropbox we have been using.

The folder is : Dropbox\ECU and Binary\INPA M3.3.1 Fix

Paste those 2 files in to replace the INPA config files and it should work with a warning about version mismatch. From what I remember it was created by "FAL" on bimerforums to solve exactly this problem, I am not 100% on what was done but from the RAM dumps I have taken with it it seems ok.
Obviously use at your own risk.

Its a big problem for the E34.

Edit - Or alternatively ignore this as I just saw the silver label bit in the title.
farjeman
Posts: 11
Joined: Sun Mar 26, 2017 10:13 am

Re: DME 413 (silver) knock correction

Post by farjeman »

olafu wrote: Sun Feb 17, 2019 10:55 am My tests with INPA was with -506 DME's -965 bin, because it's not working with 413 DME's -623 bin. I have no idea why, but if you have emulator, try to run your DME with -965 bin. Hardware is cross-compatible. If it works, the problem is propably in the INPA software. You can cheat it with using -965 identification data, so inpa thinks it is -506 / 965 DME and allows communication. It will work, but don't trust anything what it tells you.

I didn't test with -413 silver label bin.
LOL, thanks, that is interesting... so basically i'll to copy all of my data into an 506 bin\xdf? or you think that maybe the basic settings will "fit" (m52b28)?
Hairyscreech wrote: Sun Feb 17, 2019 12:06 pm The biggest advantage to the INPA output that I can see is the ability to put out the data in the registers contained in the ECU processor, there are a few in the low numbered addresses that are not on the RAM chip but are in the MCU.

If you are trying to connect to a Red label 413 ECU then you will need the M3.3.1 fix for INPA that is in the dropbox we have been using.

The folder is : Dropbox\ECU and Binary\INPA M3.3.1 Fix

Paste those 2 files in to replace the INPA config files and it should work with a warning about version mismatch. From what I remember it was created by "FAL" on bimerforums to solve exactly this problem, I am not 100% on what was done but from the RAM dumps I have taken with it it seems ok.
Obviously use at your own risk.

Its a big problem for the E34.

Edit - Or alternatively ignore this as I just saw the silver label bit in the title.
I think I'll give it a shoot anyway ;) I am using a red label ROM so it might work.
How do I get to that magical folder?
olafu
Posts: 186
Joined: Tue Jul 26, 2016 12:35 pm
Location: Finland

Re: DME 413 (silver) knock correction

Post by olafu »

What bin you are actually using? Almost every bins has different offset. This knock-part is for silver label "1 267 358 715" bin.
Red label has usually "1 267 357 623" or "1 267 357 609" bin.
this is a link to hairyscreech's dropbox: https://www.dropbox.com/sh/9bhg4mjk7nxy ... GCoWa?dl=0
farjeman
Posts: 11
Joined: Sun Mar 26, 2017 10:13 am

Re: DME 413 (silver) knock correction

Post by farjeman »

Great info here, will look at it today.

I am in a long trial and error process, most of the time using the 623 base and slightly changing the ignition angle and fuel (this is mainly why i'm keen to find the knock table).

For reference ignition timings I used the ms41 tables’ (with the VANOS corrections) and interpolate the rpm\load headers and values according to what we have in the XDF’s, Hoping not to blow a piston out…

Before I started having fun with the dme, the stock 715 didn’t work well for me, it felt and smelt way to reach (no WB yet).
m52b28, m50 mani, pink inj
olafu
Posts: 186
Joined: Tue Jul 26, 2016 12:35 pm
Location: Finland

Re: DME 413 (silver) knock correction

Post by olafu »

knock xdf in my dropbox is for -715 bin.
You need -623 XDF, -715 bin has different offset. I will upload new one…
M52B28 alu block has also different "running noise" and iron block's knock settings can cause problems. I don't know anyone who are re-tuned those knock threshold maps for alu block.

And i think this ecu has programmable band-pass filter in knock control IC. Maybe those filter settings needs to be re-tuned, but i don't know where it is.

EDIT: https://www.dropbox.com/s/aqaacqo4viqqq ... k.xdf?dl=0 <- knock control parts for -623 bin.
Post Reply