Author Topic: Updated XDF for Cali 1400 Euro3 HW100  (Read 1230 times)

Offline Meinolf

  • Gosling
  • ***
  • Posts: 243
  • Location: Germany
Updated XDF for Cali 1400 Euro3 HW100
« on: November 25, 2021, 12:29:44 PM »
Hi,

some months ago another deep dive into the code of the 7SM as used in the V85TT and Cali 1400 was begun. Initially with a focus on the V85TT BIN, but a question from this forum about the addresses of the Lambda flags in the "old" XDF let towards the Cali BIN.

The result is a first revised version of the XDF for the Cali 1400 Euro3 HW100.

<Edit 29-12-2021) Updated Link for download: https://drive.google.com/file/d/10UQblBc_U6_H8oIht5CCZCOrYQLDBfYM/view?usp=sharing

The XDF is by no means complete, the tables included are <5% of the total number. But the included tables are the ones where I am sure (>95%) that the meaning and the indices are correct.

The code is much more convoluted than the already difficult 5AM code and the introduction of air mass calcs doesn't make it easier. But, "the code began talking to me", to quote John Th.

Cheers
Meinolf
« Last Edit: December 29, 2021, 09:50:34 AM by Meinolf »

Offline moto-uno

  • Gaggle Hero
  • *****
  • *
  • Posts: 1397
  • Location: Burnaby , B.C
Re: Updated XDF for Cali 1400 Euro3 HW100
« Reply #1 on: November 25, 2021, 01:04:33 PM »
  Hi , for those of us less computer literate , does this mean it's an update for the 1400 ?
  I have a "beetle" map presently in my 2018 Eldorado . Is there anything in particular this
  addresses ?  Peter

Offline Meinolf

  • Gosling
  • ***
  • Posts: 243
  • Location: Germany
Re: Updated XDF for Cali 1400 Euro3 HW100
« Reply #2 on: November 26, 2021, 06:37:36 AM »
Hi Peter,

...does this mean it's an update for the 1400 ?

no, the XDF is a translation tool.

This is the raw code.




This is the same code in a more readable form




The XDF file translates the binary code dump (BIN) to a user-readable format.

Cheers
Meinolf

Offline Meinolf

  • Gosling
  • ***
  • Posts: 243
  • Location: Germany
Re: Updated XDF for Cali 1400 Euro3 HW100
« Reply #3 on: November 26, 2021, 01:06:35 PM »
Hi,

the last days were spent with wrestling with the routines related to reading the air/engine temp sensors and the translation into degree values.

Accordingly the XDF was updated.

The next week I'll be off hiking in Tenerife, afterwards I'll start posting more (graphic) details of the operations.

Cheers, and stay away from -
Meinolf

Wildguzzi.com

Re: Updated XDF for Cali 1400 Euro3 HW100
« Reply #3 on: November 26, 2021, 01:06:35 PM »

Offline lucky phil

  • Gaggle Hero
  • *****
  • Posts: 2047
Re: Updated XDF for Cali 1400 Euro3 HW100
« Reply #4 on: November 26, 2021, 04:00:31 PM »
  Hi , for those of us less computer literate , does this mean it's an update for the 1400 ?
  I have a "beetle" map presently in my 2018 Eldorado . Is there anything in particular this
  addresses ?  Peter

Remember our posts on this Peter. As Meinolf points out this XDF is the programme you use in Tunerpro to open the .bin file (in your case the Beetlemap) with all your ecu mapping information etc from which you can make changes.
It simply translates all the smart "computer people" binary language into a readable and modifiable format in the Tunerpro software for us luddites to use. Without the XDF and Tunerpro you'd be in the world of Binary code and ecu definition file modifying.
 
Ciao
« Last Edit: November 26, 2021, 10:02:28 PM by lucky phil »
If you're not living on the edge you're taking up to much room.

Offline moto-uno

  • Gaggle Hero
  • *****
  • *
  • Posts: 1397
  • Location: Burnaby , B.C
Re: Updated XDF for Cali 1400 Euro3 HW100
« Reply #5 on: November 26, 2021, 07:42:36 PM »
 Thanks again "Phil" for keeping me up on this ! I feel like English is a second language sometimes when trying
to relate to computer speak :) . Peter

Offline Meinolf

  • Gosling
  • ***
  • Posts: 243
  • Location: Germany
Re: Updated XDF for Cali 1400 Euro3 HW100
« Reply #6 on: December 06, 2021, 11:19:36 AM »
Hi,

while there's still a long way to go, here are some current insights using the ignition calc routines as example.

Apart from a sheer endless amount of time loops the, in comparison trivial, jump conditions are currently blocking a complete understanding of which tables are used when. Anyway, the first step is to check the TPS state. If open the jump is to loc_5E054. Here the status of gear/neutral engaged is looked at. If neutral is engaged, IgnIdle_table_958C6 is selected.

If instead of neutral a gear is engaged, then the jump is towards the next branch, which considers the clutch. If if drawn, which equals powertrain not engaged, IgnIdle_table_958C6 is also selected.

If the powertrain is engaged (gear engaged/clutch not drawn) then the drive mode is considered. Ign_Main_DriveMode_ 1_table_94FDC if drive mode 1, Ign_Main_DriveMode_ not_1_table_9521C if drive mode 2 or 3.





The branch conditions here were first formulated based on the 5AM code and then verified by following them in the 7SM code.

The naming of the tables is still preliminary and will be adjusted once all branch conditions are understood.

Offline Meinolf

  • Gosling
  • ***
  • Posts: 243
  • Location: Germany
Re: Updated XDF for Cali 1400 Euro3 HW100
« Reply #7 on: December 06, 2021, 11:53:52 AM »
In case of closed TPS the situation is not so clear.





The first branch condition results from a byte read from the EEPROM, or rather a bit of the byte. Without having a 7SM equipped bike I can't try to find the meaning of it and the usage at 13 other places in the code hasn't yet led to a unambiguous understanding.

If this bit is set then the jump would be towards the idle ignition table, if clear it leads towards the next branch guess_clutch_engage d_byte_AA28. The usage of this is similar to that of the clutch status flag, but the duplication and remaining discrepancies leave the meaning unclear.





Ignoring the above two branches leads towards loc_5E0B4, where the clutch status is evaluated. If drawn, Ign_Table_954AC is selected. If the clutch is not drawn, another condition, number_of_cyl_9A12, is evaluated. As the value is always <> 0, the jump continues towards Ign_Table_9545C.

Ign_table_96096 is, due to the value of number_of_cyl_9A12, not used.

The naming of the ignition tables is, due to the yet to be understood branching conditions, still preliminary.

Cheers
Meinolf



Offline Meinolf

  • Gosling
  • ***
  • Posts: 243
  • Location: Germany
Re: Updated XDF for Cali 1400 Euro3 HW100
« Reply #8 on: December 06, 2021, 12:00:20 PM »
After the basic ignition value process has been completed, the correction based on temperatures and ambient pressure is applied.

First a correction following air temperature. One axis is indexed by air temperature, the other one is yet unknown. The resulting values are added to the current main ignition value.





Then a correction based on engine temperature and TPS status. And once again the two yet unknown branching conditions.





Onwards towards air pressure correction.





And finally a yet to be determined correction using Ign_unknown_correct ion_table_95D52.





All of the corrections values are directly added/subtracted to the main ignition value.

Cheers
Meinolf

Offline Meinolf

  • Gosling
  • ***
  • Posts: 243
  • Location: Germany
Re: Updated XDF for Cali 1400 Euro3 HW100
« Reply #9 on: December 20, 2021, 12:23:01 PM »
Hi,

the last weeks were spent taking a closer look at the closed loop related routines.

While the missing match of ECU external pins to the internal ASICs and CPU leave lots of room for interpretation and in addition the unknown frequency with which subprograms are called makes any time-dependent calculations even more guesswork me thinks that the Lambda flags could be identified.

I've updated the XDF-link in the 1st post of this thread.

Flags 1 and 2 manage the closed loop operations. Flags 3 and 4 seem to take care of lambda sensor voltage and heating management.

It would be helpful if somebody using GuzziDiag or Pads can check and hopefully verify the flags.

As the Cali1400 Euro3/HW100 seems to be a rare animal, this check could also be done on a V85TT with HW320. I synchronize the findings between the respective BINs and could provide an XDF for the V85TT within a day or two.

Cheers
Meinolf

Offline Meinolf

  • Gosling
  • ***
  • Posts: 243
  • Location: Germany
Re: Updated XDF for Cali 1400 Euro3 HW100
« Reply #10 on: December 25, 2021, 12:51:56 PM »
Hi,

and here's the Xmas update. It seems that the speedo correction factor (that's the animal adapting the actual roadspeed to the one shown in the instrument cluster) was found.





Here's the code where the number of milliseconds /h (3.600.000) is divided by the number of pulses from the speed signal generator at the wheel(s). Slightly different from the 5AM another factor (250) is introduced. Which probably can be understood as impulses per 250m, the adapted term would then be 250x4 (1km) / pulses x 4. Which would be 3.600.000ms / pulses per km.

The corresponding address for the V85TT is 92E66h, value (1275dec) being the same as in the Cali BIN.

If somebody would change the value at either address, by say +5%, and note the resulting change in roadspeed (if any) in the instrument cluster this would validate the proposal.

Cheers
Meinolf

Offline Huzo

  • Gaggle Hero
  • *****
  • *
  • *
  • *
  • *
  • Posts: 13202
  • Location: Creswick Australia
Re: Updated XDF for Cali 1400 Euro3 HW100
« Reply #11 on: December 25, 2021, 02:00:44 PM »
Hi,

and here's the Xmas update. It seems that the speedo correction factor (that's the animal adapting the actual roadspeed to the one shown in the instrument cluster) was found.





Here's the code where the number of milliseconds /h (3.600.000) is divided by the number of pulses from the speed signal generator at the wheel(s). Slightly different from the 5AM another factor (250) is introduced. Which probably can be understood as impulses per 250m, the adapted term would then be 250x4 (1km) / pulses x 4. Which would be 3.600.000ms / pulses per km.

The corresponding address for the V85TT is 92E66h, value (1275dec) being the same as in the Cali BIN.

If somebody would change the value at either address, by say +5%, and note the resulting change in roadspeed (if any) in the instrument cluster this would validate the proposal.

Cheers
Meinolf
I usually do not read your posts Meinolf, because I am not even close to understanding what you can do..
But am I to understand you can adjust the speedometer to read correctly on the V85 ?
I tried to do that previously by another method and succeeded, but introduced another anomaly.

What you say is interesting..

Offline Meinolf

  • Gosling
  • ***
  • Posts: 243
  • Location: Germany
Re: Updated XDF for Cali 1400 Euro3 HW100
« Reply #12 on: December 26, 2021, 01:17:28 AM »
Hi,

But am I to understand you can adjust the speedometer to read correctly on the V85 ?

correct.

The procedure is, if we leave aside the actual calculations, quite straightforward. Pulses are generated when the wheel is turning. The pulses are processed and divided by the speedo correction factor. The result is the actual roadspeed, which is used for a multitude of purposes in the code and send via CANBus to the dashboard.

Legal requirements dictate that the roadspeed shown is always larger than the actual one, never smaller. This is typically in the range of up to +5%. A small change of the speedo correction factor to observe the effect on the displayed roadspeed can be done safely.

I strongly recommend not do larger changes as the roadspeed is used too often as guideline/threshold for other procedures. Some such instances are known, many are not. Better to err on the safe side.

Cheers
Meinolf
« Last Edit: December 26, 2021, 01:19:17 AM by Meinolf »

Offline Huzo

  • Gaggle Hero
  • *****
  • *
  • *
  • *
  • *
  • Posts: 13202
  • Location: Creswick Australia
Re: Updated XDF for Cali 1400 Euro3 HW100
« Reply #13 on: December 26, 2021, 05:31:38 AM »
Hi,

correct.

The procedure is, if we leave aside the actual calculations, quite straightforward. Pulses are generated when the wheel is turning. The pulses are processed and divided by the speedo correction factor. The result is the actual roadspeed, which is used for a multitude of purposes in the code and send via CANBus to the dashboard.

Legal requirements dictate that the roadspeed shown is always larger than the actual one, never smaller. This is typically in the range of up to +5%. A small change of the speedo correction factor to observe the effect on the displayed roadspeed can be done safely.

I strongly recommend not do larger changes as the roadspeed is used too often as guideline/threshold for other procedures. Some such instances are known, many are not. Better to err on the safe side.

Cheers
Meinolf
Ok Meinolf, that is good news.
I will have to wait until someone does it and can give the procedure for those like me with little understanding. I will look forward to seeing the progress.
Thank you.
Peter.

Offline lucky phil

  • Gaggle Hero
  • *****
  • Posts: 2047
Re: Updated XDF for Cali 1400 Euro3 HW100
« Reply #14 on: December 27, 2021, 01:48:14 AM »
Ok Meinolf, that is good news.
I will have to wait until someone does it and can give the procedure for those like me with little understanding. I will look forward to seeing the progress.
Thank you.
Peter.

Huzo. Analogue brain in a digital world  :grin:

Ciao
« Last Edit: December 27, 2021, 01:48:37 AM by lucky phil »
If you're not living on the edge you're taking up to much room.

Offline Huzo

  • Gaggle Hero
  • *****
  • *
  • *
  • *
  • *
  • Posts: 13202
  • Location: Creswick Australia
Re: Updated XDF for Cali 1400 Euro3 HW100
« Reply #15 on: December 27, 2021, 02:15:02 AM »
Huzo. Analogue brain in a digital world  :grin:

Ciao
Never a truer word.... :thumb:

Offline Meinolf

  • Gosling
  • ***
  • Posts: 243
  • Location: Germany
Re: Updated XDF for Cali 1400 Euro3 HW100
« Reply #16 on: December 27, 2021, 11:22:48 AM »
Hi Peter,

Ok Meinolf, that is good news.
I will have to wait until someone does it and can give the procedure for those like me with little understanding. I will look forward to seeing the progress.

the process is simple. Load the BIN into Tunerpro, change the value of the correction factor, save it and upload it to the ECU. Check if the actual speed (as measured with GPS for example) gets closer to the dashboard value. If yes, the change is in the right direction. If not, same procedure but in the reverse direction.

Cheers
Meinolf

Offline Meinolf

  • Gosling
  • ***
  • Posts: 243
  • Location: Germany
Re: Updated XDF for Cali 1400 Euro3 HW100
« Reply #17 on: December 27, 2021, 11:29:33 AM »
Hi,

I've updated the XDF (and the link in the 1st post) to include all current findings.

These are
- added comments to the ignition value tables
- changed main fuel (tables) to VE tables. The formerly known main fuel interpretation doesn't work when following the table values thru the code. I now believe that they are the VE values as calibrated by the manufacturer.
- Accordingly changed the fuel correction tables to fuel main tables. The table values correspond nicely to this assumption.
- added the speedo correction factor as parameter.

The last few days a Aprilia RSV4 BIN was dissected based on the current findings. All parameters can be identified and are plausible in a completely different engine. A XDF was created and posted on the Aprilia forum.

Now the V85TT XDF will be brought up to par, then the Cali1400 Euro4 BIN will be looked at.

Cheers
Meinolf

Offline mondtster

  • Hatchling
  • **
  • Posts: 127
Re: Updated XDF for Cali 1400 Euro3 HW100
« Reply #18 on: December 27, 2021, 11:30:34 AM »
Hi,

and here's the Xmas update. It seems that the speedo correction factor (that's the animal adapting the actual roadspeed to the one shown in the instrument cluster) was found.





Here's the code where the number of milliseconds /h (3.600.000) is divided by the number of pulses from the speed signal generator at the wheel(s). Slightly different from the 5AM another factor (250) is introduced. Which probably can be understood as impulses per 250m, the adapted term would then be 250x4 (1km) / pulses x 4. Which would be 3.600.000ms / pulses per km.

The corresponding address for the V85TT is 92E66h, value (1275dec) being the same as in the Cali BIN.

If somebody would change the value at either address, by say +5%, and note the resulting change in roadspeed (if any) in the instrument cluster this would validate the proposal.

Cheers
Meinolf

This is great news! The speedometer error is one of the things that really annoys me about the V85. I just wish it wasn’t the middle of winter here in the US, or else I’d test it for you. Hopefully someone with better weather can verify the functionality.

I’m actually a bit surprised this is something that can be adjusted in the ECU. I assumed it was a calculation made in the instrument cluster.

Offline Huzo

  • Gaggle Hero
  • *****
  • *
  • *
  • *
  • *
  • Posts: 13202
  • Location: Creswick Australia
Re: Updated XDF for Cali 1400 Euro3 HW100
« Reply #19 on: December 27, 2021, 04:31:15 PM »
Hi Peter,

the process is simple. Load the BIN into Tunerpro, change the value of the correction factor, save it and upload it to the ECU. Check if the actual speed (as measured with GPS for example) gets closer to the dashboard value. If yes, the change is in the right direction. If not, same procedure but in the reverse direction.

Cheers
Meinolf
That is a beautiful skill Meinolf.
For me it is difficult, but I now have the advice I need that can be relied upon. The combined knowledge of Mark (Beetle) and yourself, is a treasure.
Thank you.
« Last Edit: December 27, 2021, 04:34:27 PM by Huzo »

Offline Meinolf

  • Gosling
  • ***
  • Posts: 243
  • Location: Germany
Re: Updated XDF for Cali 1400 Euro3 HW100
« Reply #20 on: December 29, 2021, 09:52:33 AM »
Hi,

I've updated the XDF (and the link in the 1st post) again. Some minor cosmetic changes and several additional tables.

The V85TT XDF is now also available, more details in a separate thread.

Cheers
Meinolf

 

Quad Lock - The best GPS / phone mount system for your motorcycles, no damage to your cameras!!
Get a Wildguzzi discount of 10% off your order!
http://quadlock.refr.cc/luapmckeever
Advertise Here