TCU tuning

[Schnell]

Since HPT hasn't given us access, what software can I use to read/write the TCU? Curious to check out the tables in there.

Thanks

[outlaw_50]

Tactrix to read/write

Winols for checking tables

[Schnell]

Thank you

[Schnell]

Not yet but still planning to dig into it.

[dsptuned]

Since HPT hasn't given us access, what software can I use to read/write the TCU? Curious to check out the tables in there.

Thanks

I can remotely tune your transmission

[Schnell]

I can remotely tune your transmission

Thanks but I like the challenge of figuring it out 🙂

[dsptuned]

Thanks but I like the challenge of figuring it out ��

Verstand or Bflash then.

[Norris Motorsports]

Tactrix to read/write

Winols for checking tables

Whats the software you use to read them? I haven't found anything out there that works without spending $1000s

[Schnell]

Whats the software you use to read them? I haven't found anything out there that works without spending $1000s

Bump for this question... Already have a Tactrix.

[Norris Motorsports]

Tried with Bflash today, they don't offer tuning for the VGS3 in bench.

Verstand offers tunes. It's not an "open source" pass thru software. Also tried.

[chevota]

I had mine done and I regret it. Knowing what I know now, if you offered to pay me $3000 to put it in my car, I'd say no. I lost .2 sec on my 0-60 time and never got it back, and there is only one shift in a 0-60 run. I've yet to check 1/4 mile, half because I'm not sure I want to know, but there are three shifts in those so I can imagine what will happen.

Stock shifts take forever, and the entire point of the tune was to make it shift faster. Instead, it takes longer, sometimes twice as long. There are zero bennies to the tune. Zero...

My guess is the programming is too complicated? It seems to me that it could be made simpler by deleting most everything and leaving the core functions, or starting from scratch, but I can only speculate. It's amazing considering how simple and easy old school trannys are, and to tune. There's an old saying that to really screw something up requires a computer.

[Schnell]

I had mine done and I regret it. Knowing what I know now, if you offered to pay me $3000 to put it in my car, I'd say no. I lost .2 sec on my 0-60 time and never got it back, and there is only one shift in a 0-60 run. I've yet to check 1/4 mile, half because I'm not sure I want to know, but there are three shifts in those so I can imagine what will happen.

Stock shifts take forever, and the entire point of the tune was to make it shift faster. Instead, it takes longer, sometimes twice as long. There are zero bennies to the tune. Zero...

My guess is the programming is too complicated? It seems to me that it could be made simpler by deleting most everything and leaving the core functions, or starting from scratch, but I can only speculate. It's amazing considering how simple and easy old school trannys are, and to tune. There's an old saying that to really screw something up requires a computer.

Which tune did you get? Something is wrong there.

[chevota]

It was a remote tune using Tactrix. I'd assume they're all pretty much the same but that would be a big fat guess.

My theory is maybe they only adjust torque limits and oil pressure? If you were torque limited I guess that would make a huge difference and make people happy, but I didn't seem to have limits before, or I bypassed them in HP, so no gain there. It's never slipped so pressure is, apparently, not an issue. So my goal, and request, was faster shifts because they are just stupid long. 100ms shifts is what I asked for, if possible. Why it takes longer to shift with the tune I'll never understand. What could possibly cause that?

After that fail the tuner adjusted it, which made virtually no difference, but it looks like shifts were actually a bit longer. Then adjusted it again, but same deal. So it seems there's no point in trying again and I basically gave up. I asked to have it returned to stock, but the lack of a reply on that tells me that data is gone.

The oem shift time varied a lot, which is weird but it's computer controlled so of course it is. I hoped a tune would fix that by locking it to the quickest setting, but no. Same variation, but over a longer time range now. The variation of shift time I was well aware of before the tune because longer shifts hurt my 0-60 time. I'd study everything about the run to see what is holding me up, and shifts were certainly a factor, and a much bigger factor than I would've assumed. It seems 100ms longer shift adds more than to the time. I guess the car not only loses acceleration, but then also has to recover from it before it can proceed, it makes perfect sense. If that follows with faster shifts, and why wouldn't it, then shaving 350 to 750ms off my 1-2 shift would be amazing! This was the main factor in my getting the tune. Instead I ended up with the opposite of what I wanted. Reminds me of when I was a kid and people would buy hi-po parts for their cars but many ended up slower than it was. Except with those you could fix the problem.

The quickest oem shift I recorded was 380ms, the longest was 880ms. Tuned it's 440 to I think 1600? After the third tune attempt failed I stopped checking, which was like 6 months ago. Fyi I'm not really measuring the shift time fully, if I did they would be even longer. I start my count when the rpms drop and end where they rise (in HP obviously), because it's easy. I should start the count when the clutches start slipping. What times do you see?


I also wondered, since day one; maybe it's physical fluid flow limitations? That would make far more sense to me, but I've never heard/read a word about that, and everyone says a tune is how you fix it. Then there's the EDOK tune people rave about, which is also tune only. If the separator plate needs some mods then I would've assumed that would be all over the net, and/or places offering replacement plates, but I've never seen it. So if the EDOK makes people happy then how is it different? What is preventing another tuner from basically duplicating it? It really sux being in the dark about all this...

[Schnell]

I've heard that our RWD cars don't have as low of tq limits as 4MATIC cars, but there should still be room for improvement.

Did you try reverting to a different engine tune?

I've never checked shift times... My 1-2 shift seems the fastest of all gears, but is irrelevant on the street, unfortunately. The RPM does "flare" just a little in high tq shifts, so I think a TCU tune would help. I just want to figure it out myself. (It's fun..sort of).

[chevota]

I think my tune is pretty much maxed out, unless I want to push boost more. I have tinkered since the tranz tune, and managed to eek out a little more power the last couple months which shows up between shifts. Like 60-70, 70-80, 90-100mph etc are the best I've ever done, but 80-90 where it shifts falls short compared to previous #'s. If I lock it in 4th so it won't shift 80-90 then again I get the best #'s ever. So the power is clearly there, but...

Yes, I would love to dig in there and tinker. If I had more motivation I'd buy a used/trashed trans and figure out where to physically mod it, then tweak mine. Seems to me if it had more fluid flow it would simply work. With older trans that's exactly how you do it. Our trans use solenoids so at first I thought they're just On/Off? But maybe they're fed a duty cycle, which would explain how shifts can vary so drastically. If so, then physical mods are only a factor and the duty cycle should be adjusted first. It's frustrating not knowing how it all works...

Do you check misc times to see how a tune works? I check two ways; one is I start at counting at a given speed, like 60, then ffwd one second and note the speed. The other is I count the time it takes to get from x to x speed, like 60-70. I set my Scanner speed to read two decimals to better read, and only use the front wheel speed, which I also calibrate with gps. I know it's not perfect but it's the best option I have.
My 1-2 shifts are also quicker on avg, and 2-3 avg pretty close to it, then they get longer in higher gears. My rpms have never flared fyi. It does, however, drastically vary the shift point. Sometimes shifting ~5200, or as high as 6400. I'll never understand why. The speed at which if downshifts varies too. It used to be as long as I was going 83 it would downshift to 4th, now I need to be ~88. Otherwise it goes to 3rd, which is a pointless waste of time. One run I checked recently I floored it at 87 and it took 3 seconds to get to 88. It shifted to 3rd, then instantly saw that was too low a gear and shifted to 4th, then finally started accelerating.

Still trying to get my launches perfected but it reminds me of the "Three-Body Problem", too many interactive factors and thus impossible to predict. It seems that my best (on avg) launches happen when I set the brake to hold it, then simply stab the gas full throttle and hold it down for the run. It works better on avg than holding the brake manually, for whatever reason. If I power brake it doesn't seem any better, but on avg worse because the odds of a bad run go up. Part throttle launches delay boost so I lose again.
Aside from my butt dyno, I gauge it using the speed after the first second. It can be well below 10mph, even 5mph, or as high as my record of 16.35. Often it takes a half second to get to 1 mph, which doesn't happen when I set the brake and stab it. I assume yours does similar? I've also been using rather sticky tires that last ~5k miles, or I'd never do as well.
Starting off in 2nd, which some swear is better, is always a fail for me. Did you have any luck with that? Plus I have to manipulate it to get it to start in 2nd. I have to put in in M, then shift from 1 to 2, 2 to 1, then back to 2. If not it'll shift to 1st as soon as I gas it. Then when I do gas it I can't hit the downshift button or again it'll downshift. Ugg, too much work and doesn't help anyway.