edc15 tuning... getting into the game! very basic questions

[burpod]

i feel like i'm a bit late in the game, but i'm now trying to understand and attempt some edc15 tuning on my own. i've been googling and reading a lot, and have some very basic questions that i can't seem to figure out while i wrap my head around this. these days, due to having a couple little kids now, i have some time where i can't be working on the other things i need to do, but i can tinker on the computer here and there. i finally bought myself a nice refurb windows 10 laptop - never actually owned my own one myself before, always had one through work and they were always linux. pretty handy now i don't have to steal my wife's laptop to use VCDS!

my cars... AHU mk3 is my daily, my AHU rabbit needs some TLC to get back on the road, but both converted to VNT hopped up with big nozzles, 11mm alh pump, 17/22 and other goodies, vwmikel tuned (great tuner!) and likely stay that way for a while because i have no idea if i will have any success with this . i am planning on swapping both over to edc15 with 121 pin alh ecu. back burner project is building up a new block (in the basement) with rostens for the rabbit with an 1856 from Xman, ported head (did it myself came out quite nice if i say so myself, another thread for that), stage3 cam etc. wife has a BEW golf and i have a dismantled 05 BEW jetta as well. and recently bought a 400$ BEW with semi-broken auto trans lol... thanks tdiclub

thanks to oilhammer who will be sending me a couple stock ALH ecus and connectors - i also have some free samples on order from te.com - believe i got the right part numbers. so i plan to make a bench setup so i can work on this stuff when i can indoors at the kitchen table and supervise kids, wife will be very happy lol i bought a $20 mpps v16 cable from uobdii.com (any good? lol) but don't yet have the ECUs so what i've been looking at is some original ALH files downloaded from ecuconnections and chiptuning.pw as well as the vag_edc15 suite from chiptuning.pw as well, and another version i downloaded from somewhere else, i forget at the moment, but i believe it was the original from dilemma - as far as i can tell, they seem to be almost the same.

ANYWAY... some very basic questions, hopefully some of you guys can help and also steer me in the right direction. i have been trying to get through sbeghan's long thread, but a lot of it is a little over my head at the moment. right now i'd be happy if i can get to the point where i can read an ecu and write it back and the car will start without grenading

a) is there somewhere else other than chiptuning.pw to get the newest version of this software? and can it do edc16 (bew)? i read that it can't, but then reading all these threads it seems like people are in fact using it for PDs... i do also have a couple PD powered mk4s so this interests me. seems winOLS does it all, but i'm obviously in no position to be buying the real stuff. no idea if i'll have the time to even get anywhere with this

b) what exactly do the different mpps versions correspond to? just capability for more ECUs and versions like edc16, edc17 etc? seem v16 should be fine for me with my edc15/16. i'm still a little unsure if this mpps v16 clone will run under windows 10. dumb question, but what makes the genuine MPPS "genuine" ? does AMT-cartech work officially with bosch, siemens, etc or whoever is producing the ECUs and programming? and then there is kess/ktag, galetto, etc. i wish someone could do some sort of wikipedia-like explanation of all of these, it's all a bit confusing. i believe they are all expensive tools to read/write to ECUs, but are they reverse engineered as opposed to having contracts with bosch etc? if that's the case, in a way they aren't much different than what the cheap clones are...?

c) none of the supposedly stock/original ecu files for 90hp alh seem to have any injection duration maps. were these purposely deleted? is my version of the software screwed up? am i doing something wrong? i see 24 fuel maps, 12 limiters, 14 misc, 4 turbo and then a bunch of unnamed ones.

d) x/y-axis - drivers wish for example, has only 8 x-axis columns for throttle position. i gather the ECU interpolates the values between? but that seems incredibly coarse. am i missing something obvious here? why would the TPS be so precise only to then be guided by such a coarse map..

e) can the x/y axis (number of rows/columns) be changed in any/all/some of the maps and to what degree? since these files appear to be a fixed 512kb chunk of memory, does the structure allow for varying sized tables and is there a way to determine by how much you can expand a table?

think that's about it for now.... any help greatly appreciated!

 

[Mozambiquer]

My curiosity wants to follow...

 

[burpod]

in my past life i was a computer programmer (quite rusty now), so i feel like i should be able to figure something out with a lot of help. but my waning knowledge has a lot of gaps when it comes to the low level bits and communication, which i used to be more familiar with back in college, but that was many years ago... wrote compilers, networks, OS and all kinds of stuff for my somewhat now obsolete computer science degree. i'm very curious what the file structure of these ECU files are and how they were figured out. i have an idea of how in general they probably or work, and that it is 8bit, which to me makes it tricky to parse out, since many values must have a scale factor to them (eg, 5150 rpm is actually 245 or so). since i haven't been able to find explanations of this, just thinking out loud. is it more of a linked list of varying sized "maps"? where each map must have a unique identifier and header information for the table size so it knows where the next map starts. or are maps always at a unique address? and different for each ecu? if fixed address, can maps be resized and how do you determine what it's max size could be? manually crunching numbers looking at the address of the maps found in edc15? so many questions....

 

[tdi_my live]

I don't want to say anything about clones or anything like that. I can offer you to understand the software a little. There are ECU's that are flashable from EDC15 or the old ones, which have to be soldered and written and written in a prommer.
Your AHU is such an old one.
Basically only the PD have the duration and the VEP motors have the pump voltage.
I work with a translator. Even with that I don't understand everything you asked.
greetings from Germany

 

[burpod]

Aha! That makes sense about injection duration maps and why I'm not seeing them! Lots of posts I was reading weren't very clear .. PD tuning maps mingled in with edc15 discussions.. thanks, that helps a lot! I understand which ecu are flashable and wanting to swap my 68pin msa15 ecu to 121 pin flashable, which I'm relatively confident in doing aside from a couple possible cluster details.

I have relatives in germany, meppen and heidelberg mostly

 

[Rob Mayercik]

Have you read this thread?

help-me-understand-my-engine-map-1-9l-alh-or-i-intend-to-do-stupid-things

 

[burpod]

Yes. Not all of it but Working my way through it, but much of what I'm trying to understand at the moment is kinda assumed in that thread. That thread got more into how to diy tuning with changing the maps and I'm trying to understand on a more basic level.

 

[tdi_my live]

Times listed chronologically how the development was. I'm only talking about diesel control units.
MSA11 there are DILL28 modules with 64kb memory size in 8bit. The data are divided individually as LOW and HIGH, mostly soldered. From the MSA12 and MSA15, PLCC32 modules with 128kb are installed, already 16bit and as LOW and HIGH, but combined.
From the EDC15 onwards, flash modules PSOP44 with a data size of 512kb are installed. However, the very first one cannot be flashed via OBD or table.
Which control unit number are you looking at or which file?

 

[burpod]

well my willem gq-4x4 arrived today, and amazingly, after a bunch of googling and bumbling around (took a while to figure out how to put the chip in the programmer - turns out the software program tells you that lol), i was able to read some of my AHU vnt chips, merge files in vagsuite and view them! at first it wasn't working, but soon discovered you have to use the HHH/Hi file as file 1, and LLL/Low as file 2 (had initially tried it low first, then high). tried to view them in winols demo, but got the NOREAD error. more googling, and managed to crack into (wasn't the simple replace NOREAD for FFFFF, had to use hexedit and search for some other character sequence and replace in each file) it but of course the demo version doesn't show map names, but i did cross-check a couple maps based on address and appeared to be the same as that showing up in vagsuite.

so the big question is whether i can take this ecu file (for vnt turbo) that was in my AHU JB ecu (Seems to have all the usual maps) and flash it do an alh 121 pin ecu, and have it work. if all the maps are identified in vagsuite, look relatively normal, i would think it would work?

i understand what checksums are but a little confused how they pertain to these files. does the checksum corrector (say from chiptuning.pw) emulate the algorithm that bosch uses, so when the ecu reads and checks everything, it thinks all is good and let's the car start? i'm also still curious how these files are organized internally.

so i feel like i've got my foot in the door

 

[burpod]

looking over things more, googling more, i'm inclined to believe it wouldn't be so simple to clone my msa15 chips to edc15. i noticed i see a SOI limiter map for altering SOI based on temp and only one SOI map, whereas the edc15 files use several SOI maps for different temps. i also see an IQ to airmass map, which i don't recognize from edc15 files. so that leads me to think the maps aren't 100% compatible. i'm thinking now i will need start with the stock map i read from my alh ecu and tweak from there to match what i have.

 

[tdi_my live]

Which software versions are you looking at for the MSA15 and the EDC15? Tell us the ECU numbers.
Copy and paste is never good.

 

[burpod]

the chips i've been reading have been for my JB ecu - 028 906 021 JB. vagsuite says msa15.7 when i load up the file. fwiw they had AMD29F010B chips.

i don't yet have an alh ecu on hand, but it will be the 121pin from a 90hp alh tdi here in the US. i've just been looking through stock 90hp (and other) files i've downloaded to try and get an idea how things are setup

 

[tdi_my live]

The AMD29F010 are not stock and are reflashable epromms.
EDC15 AHU is the 038906012AF

 

[oilhammer]

The ECUs I am sending are the BD suffix. Remember, all the US-spec ECUs are different than what Europe used. Mostly for OBD2 readiness stuff. I think Europe was still using E-OBD during the period we had the ALH.

He had three ALH ECUs (all were US-spec OBD2 compliant): early ones with non-flashable PROMs, no immobiliser, which was 1998 and 1999 only. Second type, flashable PROMs, immobiliser, and the larger connector, and the third type, which is the same as the second except they tweaked the immo and went from the simple ON/OFF 2-channel glow system to the duty-cycled 4-channel system.

 

[burpod]

ah nice, i was wondering if those were reflashable, i had a feeling they were- and correct these are not-stock tuned chips. i have VNT 17/22 on both my AHU cars mk1 rabbit and mk3 jetta.

so since i have my willem eprom read/writer, could i modify these maps in vagsuite and then write to the chip? before writing, run the checksum corrector on the file? have to do more reading on that.. i would like to attempt to do some minor fuel-related tweaking as i swapped in larger injectors a few months ago. my tune originally was for ~26psi and pp764 injectors (.232 from what i've read?) and i put in .260 from Xman. perhaps i would first try something simple like adjusting idle speed. wouldn't mind playing around a little bit even though i'm planning on changing the car over to edc15 and alh ecu...

 

[tdi_my live]

26psi, what pressuresensor you have instaled?
each modification need the new checksumm.
how you want adjust the idle?

 

[burpod]

Mk4 style map sensor (3bar) on my ic piping now, I have a pair wires soldered to the ecu two a 2pin connector. Probably be going to 4bar in the future would like to do a dynamic idle like 1023 when cold and once warm standard 903

 

[gmenounos]

b) what exactly do the different mpps versions correspond to? just capability for more ECUs and versions like edc16, edc17 etc? seem v16 should be fine for me with my edc15/16. i'm still a little unsure if this mpps v16 clone will run under windows 10. dumb question, but what makes the genuine MPPS "genuine" ? does AMT-cartech work officially with bosch, siemens, etc or whoever is producing the ECUs and programming? and then there is kess/ktag, galetto, etc. i wish someone could do some sort of wikipedia-like explanation of all of these, it's all a bit confusing. i believe they are all expensive tools to read/write to ECUs, but are they reverse engineered as opposed to having contracts with bosch etc? if that's the case, in a way they aren't much different than what the cheap clones are...?

I think the "genuine" MPPS is genuine because it's from the company that created it and not any of the countless vendors selling pirated versions. It also costs several thousand dollars as compared to the eBay versions that sell for less than $100. AMT-Cartech does have some release notes explaining what's been added in newer versions but they don't go back very far: https://amtcartech.com/software-updates/

As for what works on Windows 10, many of the pirated versions are so old that they only work on XP. I have a MPPS version 5 cable (not genuine) that I used to use to back up my tunes, but it no longer runs now that I upgraded my basement laptop from Vista to Windows 10. Here's one that the seller claims works on Win10: https://www.ebay.com/itm/MPPS-V18-1...-7-8-10-Remaping-tool-Chiptuning/324333378613

I doubt that AMT-Cartech has any cooperation with Bosch. I suspect that when a new ECU is released, people take them apart, figure out what processor is inside, look up the datasheet, figure out how to download the ROM and then disassemble and reverse engineer the code. I was able to do that with one of the MKIV VDO clusters in order to reverse engineer the challenge/response that is needed to read/write the EEPROM:

kw1281test: A Free VDS-Pro / VagTacho Alternative

2022-09-20 When I started this thread, I didn't realize that it would grow so large. While reading through the 1000+ posts might be interesting, it's not going to get you your SKC or mod your CCM very quickly. To save you some time, I'll include some basic info here before the historical stuff...

forums.tdiclub.com

Another technique that people use is to attach a logic analyzer to the K-Line and record what happens when a factory or aftermarket tool interacts with the ECU.

Check out Mike Naberezny's project on GitHub: https://github.com/mnaberez/vwradio
He's done an amazing job of reverse engineering many different MKIV radios.

Also check out ecu-tool: https://github.com/fjvva/ecu-tool
It's an open-source hardware device based around an Arduino that can read/write the flash on the EDC15 (and possibly EDC16).

 

[burpod]

thanks. i have been meaning to check out that thread you started. i have some cluster things i would like to figure out at some point, like seeing if it is possible to correct my speedo/odo for the larger tire/wheels i use.

if anyone has any suggestions/links to any reflashable chips for the AHU ecu's i'm wanting to buy a few to try some test flashes. i'm not very chip-tech savvy and having a hard time figuring out exactly what i should be getting since the AM29F010 i have are NLA. all the threads i'm finding for chip recommendations are super old. will keep looking...

 

[gmenounos]

thanks. i have been meaning to check out that thread you started. i have some cluster things i would like to figure out at some point, like seeing if it is possible to correct my speedo/odo for the larger tire/wheels i use.

if anyone has any suggestions/links to any reflashable chips for the AHU ecu's i'm wanting to buy a few to try some test flashes. i'm not very chip-tech savvy and having a hard time figuring out exactly what i should be getting since the AM29F010 i have are NLA. all the threads i'm finding for chip recommendations are super old. will keep looking...

Searching through old emails, it looks like I bought 10 AM29F010 chips about 8 years ago for my 99.5 ALH. I think I only used 2 or 4, so you're welcome to the rest if I can find them.

Last year I found this post recommending 39SF010A-70NHE, so I bought some of those and they also work fine:

Reading/writing eprom chips on MSA15 ECU.

Hey guys, I have a question. Been wanting to play around with some maps on my 99.5 so I got a William programmer. I put each chip in the programmer and read each chip seperately. Once I had the two files, I used VAG EDC suite to merge the files. I did the LLL file first and HHH file but it...

forums.tdiclub.com

You (or anyone else that's interested) is also welcome to my MPPS V5 cable/software as I can't get it to work on Windows 10.

 

[burpod]

another basic question that is driving me nuts, i can't seem to find a straight answer for - i know our NA BEWs are edc16, but what exactly is edc15p? i'm gathering a european verion of edc15 that has maps for PD engines? would it be fair to compare them like msa15 is to edc15 as edc15p is to edc16? vagsuite loads the euro PD maps i've downloaded, but doesn't seem to recognize any maps when i load a BEW ecu and it only shows 30 maps total, so something is totally wrong there.

could a NA BEW be converted to an edc15p ecu (immobo delete of course) relatively easily? i think they use different ecu connectors, but do the engine sensors etc work the same? would it be a matter of swapping pins or is it more involved? similar in work to converting an msa15 car to edc15?

 

[oilhammer]

Europe had a different PD engine (they may have been 16v?) starting around 2000, and they may have even had Siemens injectors in them. That might be what you have discovered.

By the time we got a PD, they were already sort of matured some. Of course, we still only got a pretty tiny amount of them.

 

[tdi_my live]

From 2000 to 2003/4 there were only 8V PD engines with EDC15P ECU. Then came the EDC16 with 8V and 16V motors. Siemens injectors with 16V PD motor and PPD Siemens ECU have only been around since 2007 and later.

 

[burpod]

thanks QuickTD, sort of confirms my thoughts.

well i got impatient, so i made a copy of my 28psi tune, tweaked the idle rpms in edc15psuite, ran the chiptuning.pw checksum calc on it, split into two files and erased/wrote the other chips i have with the gq-4x4, popped them in the car and whallah! car fired right up with a higher idle that slowly started coming down after a minute. pretty psyched, that was relatively painless turns out, i did have a dynamic idle but just never really realized it lol.. but it wasn't enough of an increase in rpm at the temps we usually see here so i just attributed it to the ecu just doing what it's doing.

feel like that was a major breakthrough. as long as i can get mpps to run on windows 10, i should hopefully not have any issue with the alh stuff.

looks like soon i could attempt to tweak the tune for the larger injectors (went from pp764 which from what i've read are .232 to .260 from Xman). from what i've been reading, it looks like to make it run optimally, the SOI can be bumped up a bit (perhaps n146 pump voltage reduced a bit?) to account for the injectors injecting more fuel in the same amount of time. more reading on all this stuff before i end up blowing anything up

 

[burpod]

well the tuning fun didn't last long, apparently i angered the TDI gods by getting too close to the sun i must have been nearing some deep secrets that were supposed to remain hidden. what was supposed to be a nice little sunday, perhaps a little chip tweaking in the morning, outing with the family, and organizing my basement ended up with shattered lifters, pulling the head and an angry wife apparently i had a "timing event"



the other day i thought i would try the MAP linearization for 4bar map - seemed pretty straightforward and low risk. used the values suggested in many posts, and the numbers made sense. car started up quickly but hard, fast, very loud and smoky (apparently it was reshaping my lifters). did not sound good at all. tried it again, same thing. didn't think a whole lot of it, figured maybe i somehow screwed something up while viewing other maps (don't believe i touched anything else though). pressure washed the engine to clean up a leak i had a few days ago.

then today tried rewriting the chips again, thought maybe they didn't write correctly, and didn't start immediately, had to crank for a second or two (normally fires up immediately), didn't sound quite right but not banging loud like before, and smoky. checked out with vcds quickly, MAP was off (774 at idle), and a code start of injection control regulation... uh oh.. put back in my untouched chips i was using before, same results... not sounding right at all and smoky.. timing belt looked great. so checked physical timing and lo and behold what i was fearing was true. about 13 teeth off tensioner looked like it was still tight, the alignment mark was in the correct place. maybe a loose nut i dropped somehow managed to find its way through the TB cover? i haven't taken the belt off fully, but looks undamaged, in great shape actually. all the intake lifters have damage, one broke completely. the teeth of the cam gear were much shinier than usual...

the head is off now. i don't see any major signs of valve/piston contact, but probably some kissing. was dark and freezing out so i called it a night. i'm thinking this was just a freak mishap. the belt and tensioner are old (maybe from 2011?) but only 35k on them, the car sat for 7 yrs or so. i redid the timing belt a few months ago when i put in the 11mm pump, put in a new roller, but everything else looked mint, i figured it would be fine for a little while longer.

good times!

 

[burpod]

i do have a block and (ported by me) head at the machine shop, but not happy about trashing a head. although i did want to change this stock head out of for something better. this head will also have stiffer valve springs and colt stage3 cam aside from my port job, which did come out quite nice.

cylinder walls look great, although i do see a few superficial vertical scratch marks, perhaps caused during the "timing event" ? unless the rings are recently cracked (205k on the original block). car previously ran awesome, tons of power and no problem getting 50+ mpg on non-winter fuel. always starts on the first crank. hoping i can get away with a quick hone and upgrade the rods/pistons - going to try the cheap maxpeeding forged rods... i just upgraded the clutch a few months ago. don't really want to pull the whole thing. i'm supposed to be making progress on my home renovations, not putzing with my cars - sore point with the wife

 

[oilhammer]

So... PodTune Stage 1 is.... MINUS 100hp?

Your wife and my wife probably have similar concerns.



Wife: "New Sprinter really is quite peppy, I like it!"

Me: "You know the Rocketchip guy knows I got a Sprinter, he told me he could remove the 83 MPH speed limiter."

Wife: [stern look]

Me: [puppy dog eyes]

Wife: "No"

 

[burpod]

rofl, my stage1 tune is more of a valve recalibration tune. it's only the first step towards getting your car ready for a stage 2

 

[tdi_my live]

the other day i thought i would try the MAP linearization for 4bar map - seemed pretty straightforward and low risk. used the values suggested in many posts, and the numbers made sense. car started up quickly but hard, fast, very loud and smoky (apparently it was reshaping my lifters). did not sound good at all. tried it again, same thing. didn't think a whole lot of it, figured maybe i somehow screwed something up while viewing other maps (don't believe i touched anything else though). pressure washed the engine to clean up a leak i had a few days ago.

Can you post a pic about the values that you progged!?

 

[burpod]

MAP linearization -

previously they were:
4500 - 2900
1250 - 780

in my other set of chips they were:
4648 - 3000
400 - 200

not sure why the difference x/y axis values, as the line/function is the same? i changed it to:

4746 - 4160
0 - 100