NefMoto

Guys hello! I need your help.

I have Passat B6 2010 DSG7 with CDAB 1.8 TSI MED17.5 06J906026ET, 504587, 0261S05549
Gearbox is DSG7 0AM300048M 9033

Trying to reach 1Bar at middle revs and around 300nm. But some limiters are working and I cannot reach more than 250nm according to the engine measurement.
Stock and modded bin in attachment.

The problem of jerks at WOT is killing me... trying to overcome it for a month, but cannot find what can limit it

Logs of load and turbo.
(https://imagizer.imageshack.com/v2/1024x768q90/924/obBLH7.jpg)
I see that K03 is not enough to give request air at 5300-6000revs... so I already decrease request.
Logs of fuel
(https://imagizer.imageshack.com/v2/1024x768q90/924/5wWnSe.jpg)
Logs of lambda
(https://imagizer.imageshack.com/v2/1024x768q90/922/AW9dvB.jpg)
Logs of torque
(https://imagizer.imageshack.com/v2/1024x768q90/924/0PMc8l.jpg)
Logs of retards and gerbox intervetion with torque
(https://imagizer.imageshack.com/v2/1024x768q90/923/IGN35F.jpg)

Jerks!:
(https://imagizer.imageshack.com/v2/1024x768q90/921/YMo2lb.jpg)
(https://imagizer.imageshack.com/v2/1024x768q90/924/b7PZtv.jpg)
(https://imagizer.imageshack.com/v2/1024x768q90/922/foWuYC.jpg)
(https://imagizer.imageshack.com/v2/1024x768q90/923/YCsPGK.jpg)

To summarize: I cannot see any reason for these jerks, but they are awful.

Four last graphs you can see serious jerks, it feels at the weel as hard deceleration and then again acceleration.

Moreover, you can see the logs of torque by engine - it is around 250. I think there is should be some limiter which limit torque at this level and that is why jerks has come!

I check gearbox intervention logs - no intervention from TCU side according to the measurement...


So guys any suggestions? Seems just some map limiting me.

What are you using to log? If the jerks are occuring when the throttle is closing, it's because your actual boost is exceeding requested and the ECU is intervening by closing the throttle.

I looked at your bin, it's basic stuff covered in the S4 Tuning Wiki that you've missed with regards to KFLDHBN, LDPBN which is limiting your requested boost. Since you are raising the boost request, you may need to fine tune the PID since it's already overshooting your requested boost, MED17 is very different compared to ME7 and MED9 which are covered in the S4 Wiki. The load request is similar to whats in the S4 Wiki, go and read and Understand, then you will know what you are missing and how to mak changes to get your desired boost request.

Thank you for the attention to my topic...
In fact I have ODIS and VCP... so I just put their logs into the excel to work with.

Concerning the KFLDHBN and KDPBN - they are higher that load I request.
I request maximum 167% load at 3000-4000... it means something around 2000mbar with 20 IAT (which is shown in logs as requested in fact)
So it seems I am not reaching the limits of HBN and PBN as they are:
(https://imagizer.imageshack.com/v2/1024x768q90/923/hDoFcD.png)
As I understood PBN is used when engine is too hot, so 2000 is OK in this condition.

What do you think about Com_trqMaxNorm_C and MDNORM maps? Maybe it is limiting the torque? I cannot find it in my bin... trying for an hour without result.
(https://imagizer.imageshack.com/v2/1024x768q90/923/IpoltA.png)

Hi,


On your BIn i found  1)engine-Moment of inertia relative On MDNORM(JMOTOR) @ 5C118 (16bit - offset 0,000015258789 )
                            2) Maximum Indexed engine moment for Moment-Standardization (MDNORM) @ 5EA54 (8bit - offset 10 )

try to lower MDNORM(JMOTOR) in order the car to see lower torque and stop limitng you  . 

Really stupid idea to touch mdnorm on a car with DSG gearbox.
Gearbox needs torque limits lifted as well, but there is a limit to how much the dry clutches take and the 1st clutch takes about 80nm less than the 2nd one before it starts slipping.

Either way, looking at the approach and what has been done, the OP is SO over his head that I don't see anything ever coming out of this :(

Understand and agree about necessity for DSG the right torque to be given, but now I have only around 250nm output according to calculations and measurements of ECU, where DSG DQ250 TCU in stock has limit of 300nm , so it will not interviene till 300nm.
Am I right, that in case of slipping of DSG disks I will feel it as at usual manual gearbox (revs going higher without accel) or it will immediately interviene to torque and reduce it?

I am experiencing limits now just around 250nm... and they have to be in ECU.

Didn't understand about OP and SO...
Any ideas what can be the limit and how to overcome it?

TCU 250nm limit

I thought the same at the beginning.
We already modify the firmware for TCU. In attachment stock 300nm and modified 400nm limit.
No meaning of course.

find gearbox protecion maps and some other torque limiting maps.....this the solution in your problem.....also your file need more work in general....

Yes, I can understand this - but the problem that I cannot find them either any information about them in general... googling gives no results.
So if you can show me the path - i will appreciate.

The aim not just correct the file - for me is to understand the causes and consequences and the processes... that is why I came to this forum and I hope community will help me as I cannot find related information on the WEB.

As I haven't got FR; what do you think about MDASGPH: DEGS map... I have value 255 here.
Zeitliche Begrenzung Reduktionsmoment Getriebe  (Time limit reduction torque transmission)

Torque limiters are not touched in your file. They are all stock.
I already told you, you are in over your head.

Can you suggest which trq limiters they are?
I cannot find anything related through WEB.

No, I will not tune your car for you for free.

I don't know why you are even attempting this... Your knowledge is insufficient to tune even ME7, but you are diving right into MED17.
Start with something simpler, this stuff was difficult even for me (SIMOS12 + DQ200) when I did it the first time, and I do this stuff every day AND I have a dyno where I could verify things in steady state, on the road it's even more difficult.

You will also find out that clutch 1 will slip at a different torque than clutch 2, and the answer as to why you will see when removing the box. The gearbox is a POS.
Also might be that it's happening already if the clutches are worn - easy to tell, if it happens in 3rd but not in 4th, then clutch 1 is slipping.

But I am not asking to tune my car for free.
I just asking for help and maybe some suggestion where to dig to find the answers.

I started with med17 because my car has it and I can make some changes and see result without problem. I do not have a second one with me7 or so...

I do not feel any slippings now at clutches neither at 1st neither at 2nd.

The point is you've gotten absolutely nowhere.
Not with modifying the box nor modifying the engine.

You have -16 deg timing on WOT, throttle cut, your lambda target is all over the place and your gearbox torque limits are untouched.

You are trying to do in a month what took many of us years and years.
You will not succeed. It is more reasonable from both the view of the longevity of your engine and gearbox as well as time spent to pay a professional to tune your car.

If you are just getting into this, I'd try to understand engine basics and get a project car with a MUCH simpler ecu. For example ME7 with a manual gearbox. Get super comfortable with it (that'll take a few years at least), and then move on to MED17.
You are trying to run before you can walk, and there is nothing I can do to help you, because no matter what I give you, you will ask for more, as you have not managed to set a single parameter right on your own.
You don't need "help" - you need a full tune from scratch :/

I don't want to sound condescending but no one will learn this stuff in a month, NO ONE.

are you joking?change one value does not meaning dsg tuning.....

You know how to demotivate;
However, in a month or so I managed to get to know basics of OLS and the bins structure as well as how to work with DAMOS files and what the process in general. I don't think it is nothing at all?

I fully understand that it is not the easiest stuff, but it is feasible and as my car is working and giving more boost now I'm on the right way. I understand that the file need works, but guys - you are telling me that it takes for you years to understand all things and as I am learning it only about 1-1.5 months and this is my first file - it is already something.

On the WEB I find guides only for ME7/MED9, but nothing for MED17 that is why I came to this forum to ask for your help to guide me what to read or where is a mistakes. This is one of the aim of specialized forums I think.
I do not have opportunity to get a project car with simplier ECU, so I'm working with that I have.

I can afford to buy a tuning, but it is a consumer approach; where i'm very interested in the process itself and would like to understand it in details and make it by my own.

More boost with -15 and lower degrees timing. Do you even understand what you are doing?
Do you know what your EGT is doing? And as I said - there is not a single thing done right in your file as evidenced by the logs of your engine parameters.
Even worse, you don't seem to understand engine basics. Best case it'll never work, worst case you'll need a replacement engine if you continue.

You don't have the "possibility" to get a project car, but you have the money to buy a new engine for your passat? Or you think "this can't happen to me"?

OK, I think that I've had a bit of an ephiphany about this forum here...

This thread is a picture perfect example of what happens in 99% of the threads here;

*Noob* "I have X problem"

*Forum Gurus* "You don't have a clue what you're doing, stop, learn German, memorize funktionramen, smear yourself in paste and do a voodoo dance around a bricked ECU" (dispenses cryptic bit of clipped advice)

*Noob* "oh, OK.... I guess...?

END OF THREAD

My point is, where I think this forum falls short is that many of those who actually have answers to questions will hint that they know the answers, but often stop short of explaining it under the premise of "I don't want to spoon-feed someone".

The reality is, while it may take years for us to learn something, if you've truly mastered it, you ought to be able to articulate it in such a way that others can comprehend it. Maybe it's a skillset that I naturally possess and take for granted. However, I'd like to believe that the ability to efficiently pass on knowledge to others is not a skill that's unique to me.

While I can appreciate that the forum gurus don't want to waste their time answering inane /repetitive/easily searched questions, I'm sure I speak for many when I state "I would really appreciate more straightforward answers from those who can provide them".


When I look at this thread, I see an opportunity to start a discussion about DSG tuning, a subject about which I know little, but would love to learn about. However, absent a project car, logging software and definition files, I can only rely on other members here who DO have access to these things to learn anything about the subject.

While I realize that passing on this knowledge is a gratuity on the part of those in the know, I counter with this question, "if you choose to post on a thread, why not do so as a contributing member of this community?"

I guess I just don't see the point of taking the time to tell someone that there is an answer to their question, but they're not going to tell them. If that makes any sense. When this happens, all we end up with are a bunch of stunted threads that don't go anywhere and little fractured fragments of knowledge here and there, but no cohesiveness in it all.

Finally, I DO want to acknowledge that a lot of the answers I refer to have commercial value and I can see why there may be some reluctance to share that knowledge. However I'm certain that there must be a way to pass on knowledge in such a way that answers the question without letting the cat out of the bag so to speak...

PS: PRJ, this isn't meant as an attack on you personally, it's just something I've really began noticing lately.  

I feels the same.
When we are talking, for example, about any computer/mobile phone/electronic vehicle components (multimedia, etc) - first of all the information available, guides everywhere and people at specialized forumes are willing to help and make noobs understand the issue and help them to overcome it.
When I fall into ECU tuning - I feel so much reluctant community, not only in my thread - also in many forums, where I tried to find needed info.
I understand that this knowledge is valuable; but guys, common, we are at the specialized forum, where we are to help each other and share info we have.
I will never be you competitor as I do it just for own interest; and anyway there are millions of people who are not willing to do this stuff and will buy not just tuning files - the whole service.

Look, since I can't tell you how to fix your file, I CAN at least help you articulate your questions in such a way as to possibly get some answers from those who can help.

Tuners are a lot like lawyers, one of the biggest turn-offs is when someone comes in and gives a really vague description of what they want to do or what their problem is and expect a comprehensive answer.

Consider that it takes a LOT of time to write out a comprehensive reply to a question and that time is money.

It's a lot like preparing to meet with a lawyer. Lawyers love it when a client comes in prepared and they've clearly outlined their issues. That means that the lawyer can then quickly assess the situation and provide succinct, clear and concise answers.

When I looked at this thread, although you've done well with posting up pictures of your logs and such, I don't really have a clear sense of;

1) what are you trying to accomplish

2) what are the issues that you are having?

3) HOW are those issues manifesting (i.e. how is the car behaving?)

4) what ideas do you have to solve your issues or accomplish what you're trying to do?

Anyway, I've got to get some shut eye here. But I'll leave you with this. PRJ does have a point that you're trying to learn to swim by jumping in the deep end when you start tuning with an MED 17.5 file. However, from what I've seen of your posts, you're at least willing to try and out some effort in rather than just asking for a free tuning file.

If PRJ is going to give you his time, you need to carefully consider the 4 points I laid out above. Otherwise this thread will go nowhere like so many others.

Good luck

Thank you for time and explanations;

In fact for the four points you mention I gave answers in first post:
1) what are you trying to accomplish. Trying to achieve lets say Stage1 tuning; around 300nm and 200hp. Now I already have 200hp but stuck with 250nm.

2) what are the issues that you are having? Issue of throttle shut and boost cut due to torque limits of 250nm

3) HOW are those issues manifesting (i.e. how is the car behaving?)  When you go at part throttle (under 50% pedal lets say) - everything goes smooth, but once you kick pedal hard - the car accelerate hard and after 0.5sec start jerks (at logs visible as throttle shut and boost cut) - the load is decreasing and car continue to accelerate but ligher; at high revs (5000+ no such problems as trq is under limit)

4) what ideas do you have to solve your issues or accomplish what you're trying to do?  Trying to find those torque limiters (Maps and logic), but cannot find them neither at Damos (via just seraching by "trq", "geitribe", 'schutz" etc) neither at searching via google my different combination (Med17 torque limit; Med17 trq; Med17 LDRXN; DQ200 torque limiters; DSG7 torque limiters and sooooo on)

You have 200hp with -15 deg ignition timing? Really?
Do you even understand what ignition timing is?

blah blah blah, since when is this forum supposed to be a place for those who know to teach from scratch those who don't, for free and during time off or worse even, whilst in the middle of doing their professional activity, be it tuning or something else? Or am I missing something here?

one more thing: you claimed that you can find online "guides" on how to "tune" ME7/MED9 but not MED17. If you had UNDERSTOOD at least ONE bit of how ME7 works, instead of blindly reapplying someone else's logic (which might be flawed), then you might finally start to realize how STUPID what you've written sounds. MED17 uses the same base logic as ME7/MED9, but with a few extra bells and whistles.

On the logs is just showing as negative
In fact in maps it is as follows:
(https://imagizer.imageshack.com/v2/1024x768q90/923/Mi5AAl.png)
Moreover, as I am expriencing jerks - ignition also floats somewhere... before I increase load more than 250nm it was stable.

The problem I'am experincing related to MED17 with the pair of DQ200, especially torque limitations of 250nm. When the same engine with same MED17 used with mechanical gearbox there is no such problem as I have.
So I cannot find the solution in ME7 guides in how to overcome this trq limit

"Just showing as negative". Well that's the actual timing you are running. And not the timing that you have in the maps.
So again, you really think you have 200hp with -15 degrees timing?

It is hard to make such a big screenshot - so please check attachment.
That is actual timing I have.
Yes they are strange at 2000-3000 because of trq limit and jerks. As I said - they are normal if I decrease load via LDRXN maps and trq intervention is not happening.
The 200hp I get over 5000revs - there I have timings according to the map and yes I have 200hp as 100-200 I did in around 17seconds and air flow is between 150-160.

BTW; I forget to answer about EGT - it is normal, with moderate driving around 500-600, maximum agressive not more than 850, usually around 800. It is also visible in log as CAT1.

Your car doesn't have an EGT sensor.

Lol, Jesus Christ buddy, slow your roll...

And yes, that is PRECISELY what this forum is for. Members like PRJ and others in the know are the ones who MAKE this place with their gracious assistance.

Since you like to hear yourself talk, I'll ask you, what do you think this forum is supposed to be about?

If you don't like the idea behind this forum, then go hang out at MHH Auto where you can do all your little secret back-room handshakes to your hearts content.  ::)

OK, first of all, i have no idea where you ever heard me say that I found guides online to tune ME7 and MED9

Second, I understand ME7 pretty damn well, and I've gotten pretty damn good at tuning with it for your information...

Same goes for MED9 (just made a kick ass 4.2 V8 RS4 file) and MED17, and not sure where you get off saying f you had UNDERSTOOD at least ONE bit of how ME7 works, instead of blindly reapplying someone else's logic

Every tune I've sold to someone, I've made 100% myself. I taught myself how to tune by making well over 500 file revisions on my 2 s4's and analyzing Gigabytes worth of data to see what works and what doesn't.

Way to reply to something that was not even aimed at you.

Guys,
lets back to the issue of the topic.
I see only the way to make it step by step. Before I will find a way how to increase trq limit - I cannot work on the timings and all other parametres as there is all going crazy when torque intervention occurs.
So the idea first of all find how to overcome trq limit - then polish the bin file to achieve best result.

I'm digging in FR for the whole day to find torque limit related parametres;
I found:
ENGTRQPTD - calculation of torque to be given; interesting, but MoX_trqNorm_C = 460nm - so seems no limits here.
GBXECU_INTV - seems to be interventions when TCU found torque too high. But in my case I can see that torque is limited by ECU, because in logs no one message from TCU about intervention at high load.
MDASG - rely on MDIMX (99%) at the end as I see
MDASGPH - seems all to be very high values and cannot limit in my case.
TRA_PRT - transimission protection, seems not to be a limit; all values as Tra_dtrqCANErr_C are very high.
TRA_TRQINC - torque increasing interventions
MDBGRMOT - specific limitations due to some failures (rail press, EGR, EGT) - not related to the general TRQ limit
TRQ_MAX = 1000nm
Many maps linked with Com_trqMaxNorm_C - which I cannot find in my BIN, in DAMOS the value is 400.
Com_trqENG5MuxInfo0_C was 250 - set 400, no changes for my problem.
MOFTRQPTD; interesting values, but I cannot find them in my bin... this place looks so different from damos.

Moreover, as I see in FR - when the ECU calculates needed load/torque (mimax_w) it use many limiting values - if any of this value will be a limit - ECU simply will not request so high load.
In my case ECU do request high loads which is in LDRXN maps, but once the engine achieve them - some limiter reached and cut torque very hard.
So i see as a consequence - that the limit I have to find is not used at all during load calculations

Gurus, any suggestions? Am I right concerning the parameters above?

It's the gearbox, I already told you.

Hm, I just rely on several cases when people do not touch the TCU at all at the same engine and dq200 and have no torque limit problems; they achieve around 300nm at dynotests without TCU tuning.
It means they switch off the GBX intervention maps in ECU?

Moreover, in attachment one more version of DQ200 tuning - could you please check whether the limit are risen there or not?

No I am not gonna check every binary file you upload.
If you want me to make you a file with the limits removed, you can pay me to do it. That's the way the cookie crumbles.

Maybe someone is more generous, I am not.

Ok, but then why in measurements, group 122 if I remember right - we have GBX intervention status window and when I go at WOT - there is written no intervention?

Because there is different type of intervention, and you are only logging fast active intervention.

Just for my understanding;
When ECU reach 250+nm and immediately reduce load via angles, throttle and then pressure - it is not fast active intervention?
And there is no way to log all types of intervention?

Learn how torque control works on all ECU's from ME7 on and you will have the answer to your question.

Well when someone quotes you in the previous post and sends a subsequent post beginning: "and one more thing"... you MIIIGHT be able to see why I thought he was talking to me.

Anyway, neither here nor there. Thanks for helping the kid out though *thumbup*

PRJ, what might be helpful to us all is if you could give us a brief description of the types of torque intervention, though what modules are they occurring and a brief idea of where to look and what to change.

Doesn't have to be a treatise, but it would at least keep this thread alive and informative for others looking through it.

I wonder how many of you (I bet prj did, and not one time ;)) read this:

http://nyet.org/cars/info/ME7sw.pdf

It's even on nyet's site now, when I first searched for ME information I found it elsewhere. Was never an easy read on the first couple of attempts (I was too much in the world of old type ECUs with cable throttles), but it eventually got through. Not that it explains any maps, but the concept that you should be familiar with before you start.

Thank you for the link.
I never know that this kind of document is existed.
I just have read its and I was very inspired by the logic and idea... it was developed in 1990s'!

Have you ever seen somewhere such type of document for modern controllers? Just interesting what have been changed since that time

I was once interested in knowing how the throttle-less valve based multi-air technology engines from Fiat are controlled, but could not find much and later lost interest. But do a search, in some cases it does not take 1 minute and you have to try a couple of queries before google decides to give you what you want ;)

You must be very young to be surprised at the age of these things. They take time to develop before they hit the market, and ME7-s in mass go at least back to 2000. You would be even more surprised if you knew how far back are the roots of the OS of your Android phone (if you have one) ;)

Indeed,
Just entering the word of ECUs...
Yes, of course I understand how much time needed for developing... /working for one of the biggest OEM manufacturer, but not as engineer/
That is why it was so interesting, that in 1998 they have ready or almost ready product lets say with this type of logic.. it means BOSCH start to develop M*7 much earlier in the early 1990s'
But for sure, they even write in this doc, that ME7 is a progressive product from old types, not revolutionary one.

ME7 is the first modern ECU. Everything else, even in this day and age from Bosch in gasoline control is very similar.
1998 were already cars on the market and for sale with ME7 (Audi S4 2.7TT).

Most of your questions are all relevant for ME7. You don't know the basics of how it works and it has been more than 20 years.
So that is why I told you to forget what you're doing now and start somewhere a *little* bit simpler.

Really shit idea to play tricks without knowledge on a possibly leased car where you can't afford to buy a new engine or gearbox.

In Russia leasing of car is not common... So it is fully my and price of it is compareble to the something from 2000s in good condition with ME7/MED9 and so on.
Only rubbish can cost something around 2000$ and I will need 2000$ more to make engine and GB works as should in stock. So economy is very doubtful.

Ниче, в рашке тоже кредиты берут, правда на тачки поновее.

Anyway, good luck.

Finally,
Thanks for teobolo
I used MDNORM 5EA54 (8bit - offset 10 ), lower it by 10% from 400 to 360 and now I have no issues.
Gearbox works fine - no slippage, no kicks - everything as it should.
I didn't increased load too much - to not overload GB, the aim was to reach around 280nm, which more or less I did.
Here is the log:

(https://imageshack.com/a/img924/346/Q6KK1J.jpg)