NefMoto

Greetings,

I've been playing around with both ECU and TCU (DSG) tuning on the MED9.1 and CXX DSG platform and I would like to share a few things and also see if what I want to do is attainable.

From what I can find, the TCU uses Momentreduktion to protect the gearbox and avoid clutch slips during the shift transition between clutches, it is what gives the DSG their fart noise as well.  Of course it is necessary to prevent excessive clutch wear during the engagement phase, HOWEVER it can be very annoying when the TCU torque reduction command results in the ECU cutting boost / throttle / opening the N249 (DV) when shifting under peak boost/torque (+- 3500 rpm on stock turbo)

I have raised the torque limiters on the gearbox, but I can clearly hear and feel the DV opening on the shift.  Particulary if I shift 2 -> 3 under peak torque.

What I want to achieve is the application of Ignition-cut (Fast-Path) only Momentreduktion, as far as I understand, the TCU will send a desired torque value CAN signal to the ECU, the ECU will then determine how it is going to cut the torque (Ignition/N75/N249/Throttle).

What I can see on the TCU side is:

- MaximaleMomentReduktion(spelling) or Maximum Torque Reduction (Set to 400 nm stock), I assume this means it can cut up to (Current Torque - 400 nm).  Would reducing this to say 150 nm perhaps fool the ECU into thinking that an ignition-only cut will suffice?  Or will this result in a harder than normal shift/clutch slip during the transition?

- Maximum MomentReduktionDuration, would reducing this also affect the ECUs decision to use fast/slow path torque intervention by reducing the duration that torque reduction is active?  I have also decreased the shift times in Tiptronic, so would a lower torque reduction duration be appropriate as it's not needed for as long?

I just want to achieve the fastest, smoothest shifts even if it needs to drop Ignition to 30 ATDC to achieve it, as long as it isn't cutting boost.  I have lowered the minimum ignition timing during shift operation as well (mostly for the sound).

Any help would be appreciated :) I'm not looking to be spoonfed, I'm just looking for guidance from anyone who has also tried to achieve the same.  

So you are asking: what will happen if i change a single value?

my question is: why dont you test this on your own?

I most certainly will, and when I resolve the issue I will update the thread as I've done on previous threads so that we may all benefit from glorious DSG shifts, but I was just hoping that someone could offer a shove in the right direction, i.e should I be looking at an Action change (TCU) or Reaction change (ECU).

two differend things:

constant protection will be commanded by TCU via sending desired torque, eds request and flag for constant protection. ECU will take over the request by useing DV for reducing

fast protection will be commanded by sending desired torque and eds request. The ECU will take over the torque by decreasing ignition angle as long it is possible.
Depending on ECU application and EGT protection, the ECU will not reduce the troque only by ignition angle, when the request is too high or two long.

So, if your torque request is too deep, ECU can go into fuel-cut-off, witch is mutch slower in torque recovery, then igniton angle.


So, to avoid the behavieur, increase time for synchronisation in TCU or decrease minimum ignition angle on ECU

Thank you for the response, by 'time for synchronisation' I assume you mean the Upshift/Gear change duration?  or do you mean the Duration of Torque reduction?

I have already reduced the minimum ignition angle and increased KFZWOP in the high load areas so that the torque calculation shows that the lower ignition angle has decreased the torque a lot (hopefully this is an effective way).

upshift gearchance is devided in different phases:

- engaging of ongoing gear
- filling
- torquehandover
- synchronisation
- disengaging of offgoing gear

so, synchronisation is the engine speed adaption to ongoing shaft speed after torquehandover is done. This can be done by torquereduction or overpressing

Very good information, having this same issue. Well explained, hopefully this will solve my dropping boost when shifting issue.

Modifying those values will do nothing apart from making sync phase VERY slow.

The normal procedure for ECU to reduce torque is:
1. Reduce ign angle into direction of zwmin
2. If that is not enough or thermal limits reached, apply a fade out mask for X amount of cylinders.
3. If for some reason it is not allowed to fade out cylinders, only then will it cut the throttle, because throttle cut is very slow form of intervention.

If it's cutting throttle on shift then the problem is NOT in the TCU. The TCU only tells the ECU to reduce torque on fast path during shift, the ECU determines how it does it.

I can upload logs later today. Car shifts fine when cold and no boost drop. As soon as trans gets over 50c it likes to drop boost in random gears. As soon as it’s I continue driving it starts to do it in every gear and shifts are sloppy. This is my first time messing with dsg. There’s a big learning curve for me. I’m here to listen. Just tell me what data you need and I can provide.

If possible CAN-Log from AntriebCAN.

As prj said:

DSG will only send a troque reduction request and ECU will deside, how to handle.
This will only chance, when DSG, why ever, send a longtime reduction request during synchronisation phase.

And this, you will see on CAN-bus..

When you’re talking about seeing through the CAN, do you mean watching the measuring block maf-sp? I’m messing with simos 8.5 just so you know. Yes in my logs I watch my maf setpoint drop to 800 when it was at 2300 before it shifted. Then I watch the rfp shoot sky high and my boost bleed away until next gear engagement.

No, that's not what he means.
There is constant slow intervention on can and fast path intervention.

The gearbox knows nothing about maf_sp or anything else, it only asks for X torque from ECU, how ECU manages that is up to the ECU.
The problem is your ECU file, simple.

Got it! Now at least I know what direction to go. I’ll keep reading the fr about shifting sp in the ecu. I didn’t have this problem until I went e40 and added a lot of timing. Will keep you guys updated. Thanks for the insight.

Did you get any further with this?

Yes, was a pain and multiple flashes. There’s a limiter that was stopping me. Also believe maf_sp and supercharger efficiency map had a lot to do with it.

How you to this ?

In MED9.1 I think it is KFZWMNGS

If in doubt read the FR (if available)

Can/will anyone confirm this? ME7.5 doesn't have this, but I'm thinking about using KFZWMNKH
 in it's place IF this is the easiest way to keep ECU in fast-path torque-reduction during shifts.

Looks like etazwmn is involved. I's rather not change ETADZW directly since etazwmn is used in many other places.

Hey
I'm fighting with my DQ500 mqb - not a tuner - but i bought x17 to tune myself dq500

can You please share knowlege - what limiter was stopping You?

Ive got a question regarding the DSG fart via selective fuel cut
Could it be achieved with calibration only at MED17.5.x like the simos do? or asm patch is the way?

Ive trying FREDTYP.5 = 1 (getriebeschalt -> selective cut first) , CWEVAB as the mask, REDHYOC/REDHYUC without luck

In me7 i was able to patch b_mdee and this gives result of using redsol as the torque reduction strategy

*Im asking bc didnt have an access to any MED17 car personally for now unfortunatelly, so no chance to trace this things

 did you ever find out what it was?