Pages: 1 ... 93 94 [95] 96 97 ... 173
Author Topic: The Volvo ME7 thread:  (Read 933973 times)
thedrill
Full Member
***

Karma: +12/-3
Offline Offline

Posts: 139



« Reply #1410 on: April 19, 2021, 01:51:57 PM »

Hope you didn't forget the rods.. Car should be decently quick.
Logged
prometey1982
Sr. Member
****

Karma: +48/-58
Offline Offline

Posts: 301



WWW
« Reply #1411 on: April 20, 2021, 10:02:28 AM »

I've added possibility for lambda > 1.0 in my BIN. I'm running on 1.1 lambda on low load.
Logged

Россия - Великая страна!
https://youtu.be/fup5GzIFdXk
mxbike
Newbie
*

Karma: +0/-0
Offline Offline

Posts: 1


« Reply #1412 on: April 27, 2021, 11:31:57 PM »

I try to find NMAXDV (engine speed limit at fault of the vehicle speed signal) on this Volvo S60 2.5T 2007.
Anyone who can help me find adress ?
Want it to go down to idle (900 rpm) in the event of a speed signal failure.
Logged
acoffinship
Newbie
*

Karma: +0/-0
Offline Offline

Posts: 9



« Reply #1413 on: May 11, 2021, 09:30:22 AM »

Hello. I have read this thread a few times (along with several other VAG related topics and S4wiki), each time understanding a bit more.
Is anyone still checking out this particular topic? Seems like only tumbleweeds are rolling here.

Anyway, I have read my ECU using Galletto 1260 clone in boot mode. I also read another ECU (purchased from a local junkyard by matching ECU number).

My car is 2001 2.4LPT EU FWD manual stock.
Junkyard car seems to have been 2000-2001 2.4LPT US FWD automatic.

I'm sharing my files in case someone finds them useful.

My car feels like it needs some juice so I'm trying to gather information for a safe Stage 1 remap.

For now I'm interested in cloning my ECU. I've managed to write flash file to my spare ECU.
Anyone care to help with cloning immo using ME7eeprom tool? CSpin is 4.7, not sure about Chip select, since reading using 95p08 produces 1KB file which is half empty starting 0x200. The chip itself on the board is 5p08c3 on both ECUs.

UPDATE: The files attached to this post are incorrect. Bad reads due to incorrect boot mode procedure.
« Last Edit: May 25, 2021, 05:46:47 AM by acoffinship » Logged
vwdenisvw
Jr. Member
**

Karma: +2/-0
Offline Offline

Posts: 26


« Reply #1414 on: May 12, 2021, 04:25:40 AM »

My car feels like it needs some juice so I'm trying to gather information for a safe Stage 1 remap.

The boost can be increased to 0.8 bar with standard injectors 831. For a start, it's enough to simply increase the ldrxn values. And adjust a little afr to 0.87-0.9 in lamfa.
For all stock, this will be enough.
Logged

Volvo S60 2001 2.0T=>2.3T5
vwdenisvw
Jr. Member
**

Karma: +2/-0
Offline Offline

Posts: 26


« Reply #1415 on: May 12, 2021, 04:30:06 AM »

I've added possibility for lambda > 1.0 in my BIN. I'm running on 1.1 lambda on low load.

Есть изменения в расходе?
Logged

Volvo S60 2001 2.0T=>2.3T5
jahko
Jr. Member
**

Karma: +2/-0
Offline Offline

Posts: 47


« Reply #1416 on: May 13, 2021, 07:58:52 AM »

CSpins are different on different ecu's. Has to be identical as far as I've worked out but not done much on it. Let me know how you get on as I haven't got a working clone atm.
jpeg should save you some time, map locations for an 20fwhj I've been working on.


Hello. I have read this thread a few times (along with several other VAG related topics and S4wiki), each time understanding a bit more.
Is anyone still checking out this particular topic? Seems like only tumbleweeds are rolling here.

Anyway, I have read my ECU using Galletto 1260 clone in boot mode. I also read another ECU (purchased from a local junkyard by matching ECU number).

My car is 2001 2.4LPT EU FWD manual stock.
Junkyard car seems to have been 2000-2001 2.4LPT US FWD automatic.

I'm sharing my files in case someone finds them useful.

My car feels like it needs some juice so I'm trying to gather information for a safe Stage 1 remap.

For now I'm interested in cloning my ECU. I've managed to write flash file to my spare ECU.
Anyone care to help with cloning immo using ME7eeprom tool? CSpin is 4.7, not sure about Chip select, since reading using 95p08 produces 1KB file which is half empty starting 0x200. The chip itself on the board is 5p08c3 on both ECUs.
Logged
acoffinship
Newbie
*

Karma: +0/-0
Offline Offline

Posts: 9



« Reply #1417 on: May 13, 2021, 02:48:23 PM »

The boost can be increased to 0.8 bar with standard injectors 831. For a start, it's enough to simply increase the ldrxn values. And adjust a little afr to 0.87-0.9 in lamfa.
For all stock, this will be enough.

Thanks vwdenisvw! Seems to be similar to one of the ways discussed in Community Audi 2.7T stage 1 thread.
I guess I will start from this when I'm confident enough.
Logged
acoffinship
Newbie
*

Karma: +0/-0
Offline Offline

Posts: 9



« Reply #1418 on: May 13, 2021, 03:07:49 PM »

CSpins are different on different ecu's. Has to be identical as far as I've worked out but not done much on it. Let me know how you get on as I haven't got a working clone atm.
jpeg should save you some time, map locations for an 20fwhj I've been working on.
Thank you jahko. This is very helpful.

I flashed the spare ECU with the file from my car's ECU and also flashed immo eeprom (CSpin 4.7 and Chip 95P08 for both ECUs).
Then I read the spare ECU and compared read files with the files from my car's ECU. The reads are identical.

Unfortunately, with the cloned ECU the engine turns but does not start. No immo notifications on the dashboard. Tomorrow I will put in the clone, connect Dice and check the issues.
Logged
keichi
Full Member
***

Karma: +10/-2
Offline Offline

Posts: 67


« Reply #1419 on: May 15, 2021, 12:50:16 PM »

I've added possibility for lambda > 1.0 in my BIN. I'm running on 1.1 lambda on low load.

Hello.

This is my first post so greetings everyone Smiley

I am also working on lean lambda "mod" on my 2002 S60 2.3 T5.
I'v already decompiled my bin in IDA and i am comparing it with FR and V70R 99 damos.

Can I ask what variables and maps did you change to achieve lean lambda?

I am struggling to get it above 1.0. My bin version has only LAMFA and LAMBTS controlling it. But it is not working (or i am missing something).
I know that later volvo bins have also LAMKR, LAMWL and LAMRLMIN functions.
I will eventually figure it out how to force this ECU to drive lean but maybe you have some "tips" that will help?
Logged
acoffinship
Newbie
*

Karma: +0/-0
Offline Offline

Posts: 9



« Reply #1420 on: May 15, 2021, 02:14:42 PM »

Hello and welcome keichi. Smiley
May I ask how did you learn IDA specifics? I managed to load Volvo bins to IDA properly. But if I'm talking sense, I'm struggling converting data bytes to functions/code sections so I could analyse them.
Good luck finding the info you need. There are several topics in NefMoto regarding lambda>1. Try googling "nefmoto lambda above 1" if you haven't already.


Coming back to my ECU clone topic - I think the spare ECU is dead/corrupted. I've put it back into the car and then ran Vida to check the issues.
CEM does not see the ECU at all and therefore reports CEM-1A62 csc. Too bad that I haven't checked it before flashing so I can't really know whether I bricked it or it was already dead when I bought it. Boots fine though - was able to read/write flash without problems. I guess I need a new spare ECU.  Undecided
Logged
keichi
Full Member
***

Karma: +10/-2
Offline Offline

Posts: 67


« Reply #1421 on: May 16, 2021, 03:18:47 AM »

May I ask how did you learn IDA specifics? I managed to load Volvo bins to IDA properly. But if I'm talking sense, I'm struggling converting data bytes to functions/code sections so I could analyse them.

Loading BIN to IDA and then disasembling is the easy part Smiley
Understanding whats going on in the code is much worse - very time consuming.

Make sure you load it with proper DPPs and mem:
DPP0 - 0x4
DPP1 - 0x5
DPP2 - 0xC0
DPP3 - 0x3
RAM - 0x300000

To decompile you basically just press P then CTRL+U on every line with code.
P decompiles and CTRL-U jumps to next not resolved code.
I did a script in AutoIt that just press P and CTRL-U in loop Smiley

Code:
WinWaitActive("IDA - ")

While True
  Send("p^u")
  Sleep(1)
WEnd

But it must be done only on parts with code, not maps and data!
On 99V70R bin i got some code at the begining, then second part 0x8200-0xBFFF and main huge part at 0x22000
But on my 02S60T5 bin second part is at 0x9000-0xBFFF and main at 0x28000. So i assume it may be different on every version of software.

Then you need to have at least basic knowledge of assembly language. You need to understand how data is manipulated in registers, division, multiplication, conditional jumps (C166 Instruction Set
Manual is must have) and so on. But the most important is understanding memory addressing. This processor use 2 types of addressing. With Data Page Pointers (DPPs) and extended overrided mode (with EXT comand). You need to understand BIT operations (ORs, ANDs, shifting etc.) because its everywhere in the code and in the addressing.

How to start finding maps? Find some basic obvious maps like ignition or LDRX and then search for their addresses in the code. Then analyze the code and compare with FR to figure out unnamed variables and start naming them. If you identify variable, press X on it and see references to it. Then go to some reference and figure out what is going on and try to match function with FR. As soon as you notice a pattern you will reveal more variables in that function (by looking to FR). The more maps and variables you identify the further you go into code. Its like puzzles Smiley

The most annoying for me was translating addresses. For example. My LAMFA map address is 0x22140. But in code its addressed as page=0x8 and offset=0x2140. Its extended addressing mode. To translate it to absolute address you have to make bitwise operation (P SHL 14) OR (O AND 0x3FFF). I did a simple calculator app in delphi to convert those addresses back and forth. Maybe i will post it here if anybody wants. Some addresses may be also refereed in DPP mode (like my KFMIOP 0x2214 as DPP -> translates to 0x12214) and some just directly as absolute binary location.

I hope it helps to start, but i think there's no other way than taking a lot of time to practice by yourself.
« Last Edit: May 16, 2021, 03:25:45 AM by keichi » Logged
prometey1982
Sr. Member
****

Karma: +48/-58
Offline Offline

Posts: 301



WWW
« Reply #1422 on: May 16, 2021, 06:31:34 PM »

Есть изменения в расходе?
Ну, как минимум расход из Минска в Новосибирск с Пежо на фаркопе получился 12.1 лира на 100 км. Думаю, без мода, он был бы выше.
Logged

Россия - Великая страна!
https://youtu.be/fup5GzIFdXk
keichi
Full Member
***

Karma: +10/-2
Offline Offline

Posts: 67


« Reply #1423 on: May 25, 2021, 03:11:01 AM »

Ok. I figured out how to force lean drive.
One word constant in the LAMDSK function must be changed (stock 0x1000 = lambda 1.0).
Its used as base lambda when no other lambda modifiers are used and as upper limit when other lambda modifiers are active (algorithm gets the lowest off all modifiers as target).
When idling LAMFA and LAMBTS are not used so target lambda will be hard coded. I set it on 15,7 AFR (~1,07 lambda = 0x1117).
But as soon as you slightly step on the accelerator ECU switch off idle condition flag and LAMFA and LAMBTS comes into play (on my ECU stock TABGBTS is set to -47,50 so LAMBTS is always used).
I also switched off catalytic converter by setting CWKONABG to zero. As a result i don't have cat diagnosis (AFR jumps on idle from time to time) and whats the most important - cat heating.
As it turned out cat heating was retarding timming and enriching lambda a lot from time to time for example in traffic when there was no much load (it was smth like DPF burning on diesels). That was causing increased fuel consumption.

Summing up, this is the most valuable mod in my opinion Smiley After test drive i actually couldn't believe how my fuel economy improved. In the city where i was struggling to get below 13l now i am getting around 10,5. On the highway it was hard do get to 9l and now i achieve 7,5l !!! This is just insane how much fuel stock car is wasting on low loads! All that because of saint catalytic and saint ecology Smiley Probably my car will not have the best emissions now but flashing stock LAMDSK before MOT is no problem Smiley
Logged
acoffinship
Newbie
*

Karma: +0/-0
Offline Offline

Posts: 9



« Reply #1424 on: May 25, 2021, 06:09:23 AM »

Happy to hear about your successful progress on lambda > 1 Keichi. This is some really useful info you've shared, thanks. I haven't had time to sit down and dig into IDA yet.

I have finally cloned my ECU and immo, spare ECU is currently in my car. I've adjusted LDRXN map slightly (~0.65) bar boost and flashed the file. Driving my car feels a little different now, I could compare it to 2.5T. I will try to add some more boost in the future but for now it's enough. Logging in VIDA sucks.

The files I've attached in my earlier post are corrupted due to incorrect boot mode procedure. Being all happy and silly I didn't even think to try and run the checksum test. After a little brainstorming I've noticed that all the flash files start with the same code at the beginning, independent from software/year so I've compared those with my files. There was a visible pattern of zeroes in the code and that caught my attention. I re-read my ecu properly, flashed spare ECU, put it in the car and voila, success!

And since spare ecu can not be re-read since it was overwritten, I'm sharing just one file now.
Logged
Pages: 1 ... 93 94 [95] 96 97 ... 173
  Print  
 
Jump to:  

Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Page created in 0.068 seconds with 17 queries. (Pretty URLs adds 0.001s, 0q)