NefMoto

I got an Audi TT 2007 VR6 8J (PQ35) running on EcuMaster Black and DQ250.

The car is still being built. I want to go around 800nm.

While reading the forum I found out about CAN-BUS limitations on PQ35:

What I don't understand is whether it applies to my case (EcuMaster Black instead of ME7.1) or not?

In EcuMaster settings there is a setting called "VAG MDI torque". I found this explanation on EcuMaster's forum:

So here it says EcuMaster sends a percentage of torque, not an absolute number. Thus, a CAN-BUS limit of 630nm should not apply in my case?

Here's my guess how things might work:

• You set max torque to 800nm in the DQ250 tune
• You set VAG MDI torque to 800nm in EcuMaster
• When you're driving full-throttle at 800nm, EcuMaster sends "100%" to DQ250, it applies pressure required for 800nm
• When you're driving part-throttle at 400nm, EcuMaster sends "50%" to DQ250, it applies pressure required for 400nm

Is my understanding correct or not?

Your guess is completely wrong, it will work nothing like that.

Also you will snap the input shaft on the gearbox over 600nm. And the gearbox needs code patches to even calculate clutch torque over 500nm.
Anyway no matter what you do the gearbox will never see over 630nm on CAN, nor is that important because the hardware can't handle more than that anyway.

If you need to run that much torque you need a DQ500.

Thank you for your answer.

I currently have a TVS Stage2+ tune which has max torque at 500nm. My engine is gonna be tuned by an experienced EcuMaster engine tuner but since he specialises on Japanese cars, he's never set up a car with a DSG. So I'm researching the way Ecumaster talks to DQ250 and what needs to be done. Do I need to set VAG MDI torque to 500nm (current max torque in the TCU)?

You just said you want 800nm.
In this case get a DQ500, it will do 800nm by only changing the torque limiter.

How to setup ecumaster - ask whoever sold it to you how to do it. Dealer is obligated to provide support.

Torque gets sent as 6 bit value in Motor 2 in the multiplex byte.
The multiplex is 2 bits, the torque is 6 bits, so 63 is the maximum value that it can carry, that is why 630nm is a hard limit.
You might as well use that.

It is possible to double this value using an additional flag, but the DQ250 firmware does not support this. DQ500 does.

DQ250 will shit itself at 630Nm

Which is why dozens of people run them at 600+ and even 700Nm and none of them has shit itself yet lol

I'm doing upgrades in stages, first goal is 500nm, ultimate goal is 800nm.

I got reinforced DQ250 gears from Monster Performance capable of 1000whp (multiple cars doing that), I'm not worried about the hardware.

He sold me EcuMaster for my engine, says he doesn't tune DSG cars.

Thank you for the explanation. I also found some "5120 hack but for dsg" which I assume is patching DQ250 code which actually allows 750nm+ (like TVS Stage 4 tune). Are there people on this forum that can do it for money?

So buy the TVS Stage 4 if it does everything you want.
Though I guarantee you it does not see more than 630nm from engine or more than 655 Nm clutch torque.

You did not read what I wrote. The CAN matrix on the DQ250 does not allow to read more than 630 Nm from the engine EVER. No matter what you do.
You can always treat the torque as garbage in/garbage out variable though. Easily done by decalibrating a bunch of stuff in the gearbox.

Also the problem is not the "gears" but the input shaft. That said, it takes special kind of stupid to mess with the internals of a DQ250 when the DQ500 and DQ381 are freely available.
If you have any "800nm goal" then sell the current transmission and buy a DQ500.

But hey, you already know exactly what to do, so why are you even asking on here?

I may say - do what you planed and share outcome. Points to look:

Hardware:

Clutch basket - it snaps input shaft from flywheel side - it was mentioned in trail already
Input shafts inside gearbox also teds to bend/twist - rare case but seen happen in drag followed by broken in half
Gears itself - it brakes tooths and burns synchro, even main pair seen broken tooths
Differential - better replace with something more solid - as in stock main axis for satellites tend play up eating material in differential bell with drama followed

Software:

Can it deal with 500nm - yes it can without ASW mods messing up with microslip and suffering bit chaos with improper torque/pressures management. Not ideal but possible.
Patching it - variuos options mentioned in trail. If you investigate bit more you may meet few opensource patches for DQ TQ management for personal use only.
Just buy solution - there no ideal ones - there always will be cons/pross.
External controller also can be option. There quite few in market.

Keep going! :)



 

It doesn't. I want to set my own shift points on a whim, etc. The only thing TVS Stage 4 has (according to their advertising) is 750nm+.

I will reach out to TVS and ask this specifically. If max they can go is 630nm, then their 750nm+ claim is false advertising and should be removed. I will post their reply here.

I also found this message by you in another thread:
I don't understand if absolute numbers don't matter, why can't I underreport torque in a linear fashion from EcuMaster? Is it because other blocks like ABS will pick up the same "wrong" torque and it will cause problems?


I already got it installed anyways. If it breaks, I'll have to switch to DQ500. Until then, I'm gonna try to tune my DQ250.

I'm here to collect information.

Thanks! I'm pretty sure most of DQ250 failures happened during a launch on a turbo car (torque rises very fast and peaks at 3500RPM). My car has a centrifugal supercharger, torque is growing gradually and peaks at high RPMs. I hope this will save my DQ250, haha.

I am sure the gearbox software will hold 750nm. But the gearbox will never see more than 630nm and the clutch torque will never be over 655Nm in the gearbox.
Try to actually read what is written...

Stage 4 is for an upgraded clutch pack, not for stock clutch. So you can just decalibrate everything where when gearbox sees 600nm and it is actually 800nm for example, and the clutch pressure can be decalibrated by faking the clutch coarseness constant.
It won't shift the best that way, but it's possible to do. And 99.99% that's exactly how it's done.
You certainly won't be able to do it yourself, but you can pay TVS for the software and clutch upgrade.

Btw the price for that software, clutch and/or calibration is probably higher than buying a used DQ500 and killing immo on it (if you get one that even has immo).
And the only thing the DQ500 needs is a torque limiter increase to hold 800nm. Zero hardware changes and zero chance of shitting itself.

Clutch torque is calculated from clutch coarseness constant and clutch pressure, and that's limited to 500 Nm without a patch (or skewing adaptations, which makes it drive bad). So underreporting it will give you nothing, the gearbox will calculate it exactly the same way as before. The software you have now won't magically do 800Nm.
I have open software on my site (and patch) that will do 600Nm on the standard clutchpacks, but 800Nm is just unreasonable on a DQ250.

You're not tuning anything. You have no knowledge whatsoever of any tuning process. But you have very strong (completely wrong) opinions about subjects that you do not understand.

Ok, so now I understand there are 2 limits in DQ250 software:
1) Engine torque received by DQ250 from ECU is capped at 630nm (cannot be ever changed even with a patch).
2) Calculated clutch torque is capped at 500nm in stock but can be increased to 655nm (but not higher) with your patch.
Is it correct?

I already got a Stage 2 clutch (+4 friction disks). It is claimed to have +40% friction force than stock clutch.

Here's how I assume (with my very limited knowledge) things could work:
1) Real engine torque is 800nm. We downscale it to 571nm and send it to DQ250 from EcuMaster.
2) DQ250 receives 571nm torque (well within its 630nm limit)
3) DQ250 applies enough clutch pressure for 571nm (let's pick any number like 16 bar for simplicity).
4)
a) If it's a stock clutch, it slips (because 16 bar is enough for 571nm but our real engine torque is 800nm)
b) If it's a Stage 2 clutch, it doesn't slip (because 16 bar together with +40% friction force from Stage 2 clutch is enough for 571nm*1.4=800nm)

Is my logic correct?

Going over 630 is retarded on a DQ250.  That's assuming you are not a dynojet 630, but a proper European loaded 630.  The gearbox will break.  I've been there, done it lots of times and the customer has picked up the bill for my wasted time and cleaning the oil from the dyno.  Sure you might get a few pulls out of it, but it won't be any kind of reliable.  Just buy a DQ500. 

Yes.

It's wrong and everything after that.
You will need a custom software for the gearbox which you for sure aren't going to be capable of making. The transmission does not calculate pressure purely based off of engine torque, that's the nr1 mistake every one does. There's no engine torque to pressure map that exists in the transmission, but it does exist dynamically in the TCU. Anyway, irrelevant.

And as I said before, doing clutch upgrade or any other work on DQ250 is stupid, the gearbox will snap the shaft eventually. Not during launch control as you incorrectly assume.
For the wasted money you dumped in this box you would have had a DQ500 already running reliably. It's still not too late, before you pay a shitton of money to tune it and it breaks soon after.

I was not talking about a stock software in that example but rather a custom software written by a professional tuner. In such a case, my logic was correct?

I'm not gonna tune it myself. As I said, I'm collecting information so that I can ask potential tuners some questions and weed out unprofessional ones. For example, one tuner has already told me he'll simply increase max torque to 800nm and it will work, heh. If it wasn't for this forum, I'd have believed him.

Your assumption is that there's this sea of tuners who know how to properly tune the DQ250, when in reality it's possible to count them on the fingers of one hand.
Everyone else will just buy the tune from one of these or fail spectacularly.

If you are already working with TVS, then go to TVS and tell them what you need and pay your money instead of screwing around.

99.99% of "tuners" do not understand this box whatsoever.

Actually, I don't trust most of them, that's why I'm here trying to get the basics so that I can weed out non-professional tuners. Thank you for answering my questions.

The only downside to TVS is that you can't add small tweaks later (different shift points, etc). So there is no room for experimentation.

By the way, looks like there is a yet another limit of 410nm in diagnostics. VCDS is not displaying anything higher than 410nm (ECU -> TCU torque).

I was offered another approach by a reputable tuner - you set the limit at 410nm (so you can monitor torque in VCDS) and then downscale real torque in EcuMaster to a number where 100% equals 410nm (so if your real max torque is 820nm, you divide everything by 2, etc). Given this is done by an experienced tuner and all calculations/calibrations are done correctly, are there any downsides to this approach?

Just another one of those who does not understand the difference between engine and clutch torque...

Anyway, I'm done with this thread, there's nothing related to tuning here, and you can't (or don't want to) read.