Pages: [1]
Author Topic: Reverse engineer to find what's causing a DSG error code?  (Read 3921 times)
Reverant
Newbie
*

Karma: +0/-0
Offline Offline

Posts: 6


« on: April 07, 2023, 03:04:02 AM »

Hi everyone. I have made a custom ECU to replace the Simos18 on an MQB car (2018 Seat Leon Cupra). Everything works fine, except for the DQ250, which seemingly works but has the following issues: Will manually shift with a delay of 1 second after you push the shift lever or the paddle, and will not autoshift at rev limit. As soon as you pass over 500rpm (so just after cranking), it throws code P0726 - RPM Signal from ECU - Implausible signal. No other codes. Go to advanced measuring values and the Engine RPM shows 100% correct at idle. Clear the fault and it immediately returns. Have obviously tested with a can bus analyzer to make sure that there is no stray message once in a while as the RPM is broadcasted at 10ms and VCDS may not catch it.

So my only option it seems, is to reverse engineer the DSG firmware to see what exactly the firmware examines and thinks that the RPM signal is wrong.

Is this a valid way of doing it or am I missing a simpler option?

Thanks!
Logged
IamwhoIam
Hero Member
*****

Karma: +52/-115
Online Online

Posts: 1070


« Reply #1 on: April 08, 2023, 08:44:52 AM »

What exactly is a "custom ECU to replace the Simos18"?
Logged

I have no logs because I have a boost gauge (makes things easier)
Reverant
Newbie
*

Karma: +0/-0
Offline Offline

Posts: 6


« Reply #2 on: April 10, 2023, 08:02:31 AM »

Its a standalone ECU. It has everything required so that everything on the engine and on the car works well.
Logged
IamwhoIam
Hero Member
*****

Karma: +52/-115
Online Online

Posts: 1070


« Reply #3 on: April 10, 2023, 08:31:04 AM »

Its a standalone ECU. It has everything required so that everything on the engine and on the car works well.

 P0726 - RPM Signal from ECU - Implausible signal

Obviously not?
Logged

I have no logs because I have a boost gauge (makes things easier)
Reverant
Newbie
*

Karma: +0/-0
Offline Offline

Posts: 6


« Reply #4 on: April 10, 2023, 08:39:39 AM »

Obviously not?

Everything on the car works absolutely perfectly (including AVS and the N493 on the CJX engine), the gearbox works as well, it holds the torque without clutch slipping, downshift revmatching works, hill hold works, the virtual cockpit works perfectly, the klima works, the cruise control, there are no faults anywhere except for this one code in the DSG. Thanks for the sarcasm but you probably don't have any useful contribution to make.
Logged
EanDem
Full Member
***

Karma: +8/-40
Online Online

Posts: 78


« Reply #5 on: April 10, 2023, 10:51:01 AM »

Hi everyone. I have made a custom ECU to replace the Simos18 on an MQB car (2018 Seat Leon Cupra). Everything works fine, except for the DQ250, which seemingly works but has the following issues: Will manually shift with a delay of 1 second after you push the shift lever or the paddle, and will not autoshift at rev limit. As soon as you pass over 500rpm (so just after cranking), it throws code P0726 - RPM Signal from ECU - Implausible signal. No other codes. Go to advanced measuring values and the Engine RPM shows 100% correct at idle. Clear the fault and it immediately returns. Have obviously tested with a can bus analyzer to make sure that there is no stray message once in a while as the RPM is broadcasted at 10ms and VCDS may not catch it.

So my only option it seems, is to reverse engineer the DSG firmware to see what exactly the firmware examines and thinks that the RPM signal is wrong.

Is this a valid way of doing it or am I missing a simpler option?

Thanks!


Sounds like mistakes in DSG CAL..
Logged
Reverant
Newbie
*

Karma: +0/-0
Offline Offline

Posts: 6


« Reply #6 on: April 10, 2023, 10:52:53 AM »


Sounds like mistakes in DSG CAL..

I would agree, but when I reconnect the stock ECU, the error is not retuning after clearing.
Logged
prj
Hero Member
*****

Karma: +1072/-481
Offline Offline

Posts: 6037


« Reply #7 on: April 12, 2023, 02:00:48 AM »

How strict is your CAN raster?
Are you sure you don't have any jitter there?
Logged

PM's will not be answered, so don't even try.
Log your car properly - WinOLS database - Tools/patches
Reverant
Newbie
*

Karma: +0/-0
Offline Offline

Posts: 6


« Reply #8 on: April 12, 2023, 08:35:17 AM »

How strict is your CAN raster?
Are you sure you don't have any jitter there?

Not 100% 10.00ms as is the stock if that's what you are asking. Generally 9.5-12ms.
Logged
prj
Hero Member
*****

Karma: +1072/-481
Offline Offline

Posts: 6037


« Reply #9 on: April 12, 2023, 01:19:24 PM »

Not 100% 10.00ms as is the stock if that's what you are asking. Generally 9.5-12ms.
Yeah 12ms is way too much. These things have to run in real time in a 10ms raster.
The stock ECU manages <0.2ms jitter if I recall correctly.
Also make sure your rpm filtering is up to snuff. The trans is calculating gradients.
If your trigger pickup or handling isn't 100% on point, then you're also gonna have issues.

Most likely what is happening is you're filling the CAN structure in one place roughly at the same time (or maybe even that has jitter, I don't know), and then after that fact you're introducing arbitrary jitter with your sloppy CAN handling.
After that the calculated gradient is all over the place.

25% jitter on PTCAN is unacceptable performance.
« Last Edit: April 12, 2023, 01:25:51 PM by prj » Logged

PM's will not be answered, so don't even try.
Log your car properly - WinOLS database - Tools/patches
Pages: [1]
  Print  
 
Jump to:  

Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Page created in 0.027 seconds with 16 queries. (Pretty URLs adds 0.006s, 0q)