Here is a thought that I had. Perhaps someone more knowledgeable than I can chime in.

I have a JCW MCS coming in a next few month. I have been thinking about upgrading the ECU with MTH. I know there are a lot of threads going around asking for a comparison on the same car, same day, etc. It seems that it is not easily done since the ECU needs time to "learn".

So, as in the case of the MTH, when the software is being downloaded from the ECU, is the "learned behavior" part of the file?

 

This is part of how the engine control compensates for driving style wear etc. Whether MTH upload wipes the old values, or puts in new ones ins't really important. It's has not adapted to your particular car yet. I have no clue about the magnatude of the effect. there is something called a DME reset (DME=ECU) that wipes all the adaption values, and some more (like limit positions on the TB), but you need special stuff to do this, and the on-board computer reset isn't the same thing.

But after all this, I don't know if the adaptive values are part of the download. The MTH team says you need to drive your car for a while to get full benefit, so even if they are downloaded, it doesn't seem that the upload comes "pre-adapted".

Matt

 

What I was thinking is this:

1. Save the original file that you dowload prior to MTH.
2. Upload the MTH and drive it long enough for the learning process to take place.
3. Download and save the now "learned" MTH file.

At this point you should have two files. One original and one MTH. Both should have your driving habits as part of the file (assuming that data gets saved). You should now be able to quickly swap the files while on the dyno to get a comparison that is somewhat fair. So, hence the question. Does the "learned behavior" somehow gets written into the ECU software and is it possible to save the file?

 

No. The file that you retrieve from the ECU does not contain any learned behavior. It only contains data that was flashed into the ECU either by the factory, the dealer, or by MTH or the like.

 

Where is it stored than? And if it is stored, than, should there not be a way to retrieve it?

 

Adaptation values are stored in a separate area and are not readily extractable or readable. The significance and importance of "learned" behavior has been highly overstated by certain tuners and members of this board. If you reflash the ECU with a program that makes more power, it will make more power on the dyno ... instantly. If you reflash back to stock, it will make less power ... instantly.

 

I hear you. But you have to admit. If it was possible to replicate the "learned" portion of the software, it would put a lot of conjecture and speculation to bed.

 

But it's not possible. The way to put conjecture and speculation to bed is to do this:

https://www.northamericanmotoring.co...0&postcount=44

and then this:

https://www.northamericanmotoring.co...1&postcount=45

 

Yep, I've seen that before. Oh well, it was just an idea. On to the next project...

 

Adaptive memory can only be reset by the dealer at this time.

 

Not true - from my own experience on the dyno. I had been running the first tuner version on my MCS for about 900 miles. Because I was planning a comparison of the different version of MTH, I took it back to stock ECU, reset through speedo, ran 250 miles, reset the cache again and then dynoed at what I thought should be stock mapping. The "stock" results came out with 168.1 hp and 145.1 torque - at the wheels. That would indicate a drivetrain "gain" over stock rated bhp numbers. So the tuner version was still there and affecting the output. No mods other than a Dinan intake.

At this point I didn't know exactly how long it took to overwrite the cache - so I assumed I had a really strong MCS. Upon completion of the dyno of what I thought was stock, I loaded the latest "standard version," reset the ECU and got in one more run. Results were exactly the same. After one more week and approximately 275 miles I went back to the same dyno. Five runs all showed a slightly lower output - around 164 HP and 142 torque. Standard version should make lower numbers that Tuner but it was not as low as I expected. Conversations with Volker resulted in the following aha's. The performance cache is not clearable by the speedo reset, but it can be reset by the dealer using the BMW/MINI computer program - not sure I would want to go ask an SA to do that for me. The performance cache can be overwritten by driving, as we all know, but 250 miles isn't enough. The rule of thumb that Franz and Volker use at MTH is three tanks of fuel.

I'm not arguing that downloading a new map has instant results. Anyone I've loaded the MTH for in the past year, and both of my MINIs, feel stronger instantly after going from stock ECU to "standard" MTH or to Tuner MTH. It just takes a few miles (or a dealer reset) to develop its full potential. And, an MTH downgrade - tuner to standard or stock, doesn't clear as quickly, or even make a noticeable difference in the drivability of the car.

FWIW - I will be dynoing my wife's (stock) CVT MCC on the latest tuner version, and my (stock) MCS w/ Dinan intake on the latest tuner version this weekend. Then after a 15% pulley install, and three tanks of gas on each mapping I'll dyno my MCS on "standard" and later on "Tuner" version to see what happens at each step.

 

Your results differ from mine. In a controlled dyno environment where IAT and ECT are stabilized prior to each run, I haven't seen any change in power from clearing adaptation (using the routines that the factory scan tools use). This is on multiple cars.

 

I don't think that's a reasonable conclusion to draw from your results. When you reflash to stock, the tuner version is GONE.

 

Then I have a one-in-a-million MCS that is putting down 168 hp to the wheels on S-lites and runflats no less. Since I am re-dynoing tomorrow on a well broken-in Tuner version, I guess we'll see what I get. Unfortunately, I'm not willing to go back to stock one more time and drive it 800 miles to find out (I've been spoiled - and want to get past this and move forward to the pulley upgrade).

All I'm saying is that even though the mapping software had changed, the adaptive portion of the ECU "apparently" didn't finish overwriting the way the car was "accustomed" to running in a span of 250 or so miles. Tuner-> Stock = 168 HP; "stock" to standard - slightly less. Doesn't seem right. If I see around 168 at the wheels tomorrow then IMHO it had to have been remnants of the adaptive cache that got me there.

 

I agree, it does seem strange that your stock MCS put out 168 whp. What do other stock MCS dyno on the same dyno? Does the shop have good control over temperature management for repeatability? There have been other unusually high stock dynos as well, for instance the one from Sport Compact Car:

http://www.sportcompactcarweb.com/pr...old/index.html

If you had said that you dyno'ed completely stock at 145 whp, then installed the tuner file, then reinstalled the stock file and dyno'ed again at 168 whp, well then something fishy is going on.

 

I would be really surprised if my MCS really put that power to the wheels in stock configuration and I know that the Dinan intake didn't get it there by itself. This is a Dynojet 248C running the latest software and they have been compensating for conditions and trying to manage the I/C temp with a fan assembly directly on the top of the I/C. Four runs in a row (each time) were all within a few points of each other. Guess we'll see tomorrow....

 

Andy - I ran the dyno again yesterday. This was after allowing the ECU to "adapt" over about 800 miles since I reloaded the Tuner version on my MCS. The car is stock other than an intake. Numbers were right around 168 HP - same as my previous runs. So, I can only surmise that one of two things has happened. Either my baseline "stock" dyno several weeks back was still being affected by the tuner version that had been on the car previously, or the tuner file I installed and tested yesterday (which was the same one I had played with before) didn't affect the car that much. I know it is purely subjective, but the butt dyno said the car felt much different after installing the tuner version, than it did for the first 2000 miles before I ever touched the ECU.

MINIUSA no longer states whether the 168 HP rating on the '05/'06 MCS is BHP or WHP, so I'm not sure what I saw was still stock numbers, or a bump from what one should see at the wheels. In any case, I can't go back and try to get another set of runs in under the same conditions. After the runs on Saturday, I decided to go forward and installed my 15% pulley. In a couple of weeks I'll try it again to see where it ends up this time with the pulley and a map that accounts for the pulley mod.

 

It is bhp, not whp. HTH

 

Again, what sort of numbers did other MINI's put out on the same dyno? I think you are choosing to ignore the most likely possibility.

 

I'd have to check with the shop. I've been doing this independently so not there to see what others have done and with what mods (i.e.: dyno days). I'd be a lot happier if I didn't feel like I screwed up my original stock/baseline numbers.

BTW, I also found out just hard difficult it is to dyno a CVT Cooper this weekend. Even in sport drive/manual mode the power output blips just before red line as it prepares to shift itself. My wife's totally stock CVT MCC numbers were all over the place depending on how well we were able to hold it just shy of redline before the ECU took over and shifted for us. Numbers seemed to be between 98 and 103 HP running on Tuner MTH. I'd really like to know what a stock 5-speed Cooper can do totally stock and again on the tuner version of MTH.

 

Unless you loaded the tuner file instead of the stock file by mistake....