Now, there's a lot going on here and I'm going to try to clarify all of it. A handful of people have apparently been watching primes video and not understanding what they're seeing.
This post is to clarify exactly what we do, why we do it and what it means to both the regular customer and the "I'm chasing the S4 world record" customer.
Originally Posted by
Bianco
So what are the benefits of tcu tuning on a 2010 or 2011 car?
You really think anyone whose audi dsg shifted like this would not care or notice ?
https://youtu.be/5hjvBYvtcS4
This video is Prime Time running 11.45. First off he doesn't seem to even use launch control.
Yes, that's the video of the record being set. What you're missing is what's actually going on and what the issue is.
First of all, prime has L/C disabled. The Gen 1 boxes don't launch well enough for world record level runs. They have a smaller oil pump and can't lock up the clutches as fast as the later units. So, we offer Gen 1 flashes with or without L/C. Prime has loads of experience launching the car and his "smash and go" technique is the best available for launching a Gen 1 car. It's as good as L/C on a Gen 2. All of these are assuming good tires and good track prep. We've done Gen 2 cars with street tires or slicks and the results are basically the same.
Best 60's:
Gen 1 L/C: 1.69 60'
Gen 1.5 L/C, Gen 2 L/C, Gen 1 Primetime driver mod: 1.5x 60'
Also note that Ron's launch technique is rather difficult to duplicate. He's a master at it. +1 for driver mod
The following applies to all generations of DSG. We do it basically the same on all of them.
Originally Posted by
Bianco
And then his shifts hang up. It doesn't sound like it's only costing him 'thousandths' of seconds, no offense. If he's coasting during those hangups, he is losing valuable time under acceleration. That's lost mph. And as a result he's losing valuable elapsed time. It's nowhere near thousandths.
TLDR: Those shifts are as intended. The problem he's having is much more subtle.
No offense taken. Others have called it "granny shifting" and said that he's coasting for half a second on that run. Neither of those are correct. What you're hearing and seeing in that run is called amax. It doesn't shift crisply, because it's not supposed to. Amax uses both clutches simultaneously and slips between the gears.
A normal shift de-rates the engine by telling the ECU to pull timing. It does this to prevent free-revving and to lower the torque at the clutch to make shifting easier. It has the engine de-rate while it's opening the first clutch. The engine free-wheels for 50ms or so while the second one is being closed and then the TCU tells the engine to bring back the power. That de-rate timing retard is what causes the DSG "farts" you can hear on videos.
Amax doesn't work like that. When in this mode, the TCU keeps the beans on the entire time. There is no de-rate at all. It controls the engine RPM by having both clutches engaged in the linear (slip) region together while it transitions between the two. The upshot of this is that there is no de-rate, the engine is making full power the entire time and is transmitting it to the ground even during the shift. The downside is that it's extremely delicate. If the transition time is too fast, the clutches won't slip enough, will be working against each other and will rip themselves out of the transmission. On the other side, if the transition time is too slow, the engine will free rev, hit the limiter and de-rate via the rev limiter. It comes back as soon as the clutch closes a bit more and pulls it off the redline, but it still spent some time in de-rate.
That is the problem that prime is talking about when he says he's having "shifting issues". If you watch the video VERY closely, you can see it peg the limiter twice during that run. We've been dialing the transition length back a couple ms at a time with him. Unfortunately, we've not been to the track with him and he doesn't have internet at the track. So, we have to flash it at his house the night before and only get one try per track day. That's made it a bit of a slow process finding exactly the right value.
As to what this is or isn't worth:
During one of our previous track days, we tested Amax vs. non Amax shifting at the drag strip. Amax was consistently good for a tenth. Even if it does drill the limiter occasionally, it still shows significant gains. That's why it's on in our production files (only after a launch or a smash/go on Gen 1). We tested, collected the data and enabled it as a result. We could disable it and give crispy shifts (like the car does during normal acceleration with our software) instead. But, the data says this is faster.
The hitting of the limiter results in a de-rate for about 25 ms out of the 50ms transition period (we can see it happen in the logs). That occurs during the absolute center of the shift, when the clutches are both the least engaged and the least amount of power is being transmitted to the wheels. It DOES have a negative affect, but it's not a very big one. We estimate, at best, that we could translate about 1/10 to 1/5 of that time into E.T. if we can get the transitions all to be perfect. That's 5ms to 10ms worth of ET pickup (1ms = 1 thousandth)
We're hoping to go visit Ron in the spring and bring our tools, his car and his driver mod to the track. That should give us the chance to dial it in perfectly and answer once and for all what it's really worth.
Originally Posted by
Bianco
Which is good news for you as it means he should be able to extend his record margin when his TCU tune is fixed.
That sounds good.. We'll find out in the spring exactly how much. Based on the data posted above, I don't see tenths.. maybe a hundredth.
Originally Posted by
Bianco
He went 11.45 right? What about rynurz who went 11.5 in a 2010 with only one pulley.
Rynurz put up a nice time. But, it's quite a ways off from Ron's. Picking up a tenth in drag racing is no small feat.
Originally Posted by
Bianco
Prime time has two pulleys and a beta tune allowing him to take advantage of considerable power advantage over any other stage 2 car. He also has just about every mod that works. So he's leaving very little on the table other than his TCU malfunctions.
Agreed. Ron's in it to win it. He's also faster with the TCU flash than he was without it. How much is left to be had is anyone's guess until we go up there.
Originally Posted by
Bianco
But to say nobody else with a 2010 or 2011 is close is both inaccurate and a bit misleading since we have only one other 2010-2011 car running two pulleys, and that is 2000s4. His car went 11.66 on a pump gas tune with some 98 octane mixed with 93.
I don't see that run posted on either of the prominent sites. Can you link it?
I don't see it as misleading. I said that nobody else in the top slots is running a Gen 1 car. That's true: Auditude isn't, Jones isn't, Eurocode isn't, we aren't. The nearest is rynurz, and he's over a tenth behind the lead.
None of us know how many pulleys are out there. I have an idea how many are on APR cars, because the people have been PM'ing me to get Prime's file (which we're not quite ready to give out). I don't know how many are on Giac cars (since their Stage 2 seems to run well with both pulleys out of the box).
Edit: I had the wrong kind of tires listed for Ron (primetime). He emailed and corrected me that he runs Contis.
Bookmarks