A challenge for vrg3 (now that he's almost done his masters)

boostjunkie · Post by **boostjunkie** » Tue Apr 20, 2004 2:26 pm

I was talkin on the celica boards (I really should get to work around here) and asked a question about reflashes to the ecu. The response I got was kinda puzzling:

quote:
--------------------------------------------------------------------------------
Originally posted by boostjunkie26
I'm just curious. Has anyone tried cracking the ecu for reflashing puposes? I did a search, and all of the aftermarket companies seem to be dealing with the power fc, for the most part. It is a good unit, but seems like reflashes could be just as beneficial to tuning.
--------------------------------------------------------------------------------

Our ECU is too smart for that.......it will just overwrite the reflash.

I'm no EE genius (which is why I posed this question to vrg3), but it would seem to me that if the ecu is reflashed that it wouldn't be able to access the original programming and would just use the new settings. Could someone explain this to me?

vrg3 · Post by **vrg3** » Tue Apr 20, 2004 3:50 pm

You're absolutely right that if you replace all the code in the ECU, then it can't access the code that you replaced. That's common sense.

However, it is possible that the ECU has some protection schemes in it to make aftermarket reprogramming harder. It may have some code in ROM built into the processor, for example. That code could have some way of recognizing whether the code in the external memory (be it ROM, flash, or whatever) is valid Toyota-written code or not. It could even have some error correction information stored somewhere that allows it to "recover" from "corrupted" flash memory.

If Toyota (for whatever reason) made it a goal to keep hackers out of their ECU, it could be very, very hard to do.

Or maybe he meant Toyota keeps the maps separate from the code, and the ECU can recognize when the maps have been tampered with. So if you try to change just the maps the ECU will go into limp mode or something.

boostjunkie · Post by **boostjunkie** » Tue Apr 20, 2004 6:12 pm

Interesting little tidbit of ecu technology. Looks like it might be Unichip for me . . . or the awesome (yet expensive) apex-i power fc.

evolutionmovement · Post by **evolutionmovement** » Tue Apr 20, 2004 6:17 pm

What did Lotus do for engine management on the Elise? They got an extra 10 horse, but I believe they helped out the torque curve a little also.

Steve

vrg3 · Post by **vrg3** » Tue Apr 20, 2004 6:26 pm

They use the same ECU but with modified programming. I remember reading that Toyota asked Lotus for a copy of their modifications after seeing the results.

I think the Elise also has a different intake and exhaust setup, which helped them get the extra power.

evolutionmovement · Post by **evolutionmovement** » Tue Apr 20, 2004 8:35 pm

I think they may have also done some head work, but I was pretty sure they reprogrammed it, my point being that it can be done unless Toyota supplied them with some special code or something tat would allow the ECU to recognize the new map and use it.

Steve

vrg3 · Post by **vrg3** » Tue Apr 20, 2004 8:54 pm

I would expect that Toyota at least gave them some tools to make the job easier. But, yeah, of course there is some way to recalibrate boostjunkie's new car's ECU.

boostjunkie · Post by **boostjunkie** » Tue Apr 20, 2004 9:25 pm

evolutionmovement wrote:I think they may have also done some head work, but I was pretty sure they reprogrammed it, my point being that it can be done unless Toyota supplied them with some special code or something tat would allow the ECU to recognize the new map and use it.

Steve

No headwork.

The intake and exhaust were changed, as were some additional timing advance in the midrange.

Is it any wonder that people with full exhausts and intakes are getting almost 20whp out of their celicas!

There's one guy running a 13.7 with intake, full exhaust, and power fc . . . with a FULL interior!!

evolutionmovement · Post by **evolutionmovement** » Tue Apr 20, 2004 11:04 pm

Damn! Is that with high rpm clutch drop? That's pretty impressive. They must be lighter than I thought.

I just assumed head work just because that's pretty much SOP for Lotus.

Steve

boostjunkie · Post by **boostjunkie** » Wed Apr 21, 2004 4:49 am

I dunno how he launched but I'm hazarding a guess it was with slicks. Believe me, 2500lbs is ALOT lighter than 3000-3100. I can't imagine what that Elise is like!! 1900lbs!!

NemesisEJ22t · Post by **NemesisEJ22t** » Wed Apr 21, 2004 5:23 am

The last i remember, people were still trying to crack the Toyota ECU for the 2ZZ-GE. Toyota may have helped out Lotus when they did the motor for the Elise, but they will not help out aftermarket companies. That is why there really aren't any ECU reflashers or chips for the 2ZZ ECU. Also, Toyota ECUs cannot be reflashed, i believe that (at least for the older ECUs) a new chip must be installed and then flashed with the desired information. This is what most of the aftermarket companies do for Toyota cars when they modify the ECU. To more directly answer your question, i think toyota uses some sort of special code in the new ECUs to (i use this term very loosely) "encrypt" the data for processing and sending it back out to the electronics. This is the major block on reflashing the ECU, i don't think that it rejects the new fuel maps, etc. rather it just won't work with them. Don't quote me on that though.

evolutionmovement · Post by **evolutionmovement** » Wed Apr 21, 2004 6:04 am

Makes sense.

Yeah there's no substitute for less weight. My 2100 lb '83 sedan was so tossable it was amazing. No power but a blast to drive.

Steve

boostjunkie · Post by **boostjunkie** » Wed Apr 21, 2004 1:05 pm

evolutionmovement wrote:Makes sense.

Yeah there's no substitute for less weight. My 2100 lb '83 sedan was so tossable it was amazing. No power but a blast to drive.

Steve

Prime candidate for a swap!

Thanks for the info, NemesisEJ22t. I'm still really new to Toyota tuning and mods. I'm even skeptical about puttin on an intake for the sheer fact that it makes the engine run lean. Until I can find an appropriate ecu (or come up with the money for either the apex or unichip) I will stick to modding the stock intake.

NemesisEJ22t · Post by **NemesisEJ22t** » Wed Apr 21, 2004 1:45 pm

Cool man, just remember that you may get a CEL with the intake. I believe that a few people have had this happen, its no big deal though, doesn't affect performance.

omahasubaru · Post by **omahasubaru** » Wed Apr 21, 2004 2:18 pm

If only Cobb tuning would have the desire to take some time and work on the Legacy Turbo ECU.

I use their AccessPort on my WRX and it is amazing. I had just their reflash to begin with, then upgraded.

Something like that would be very benifical for you legacy turbo guys as they can tune/offer maps for different turbo & injector set-ups like they do for the WRX. My guess is that they don't feel it would be in high demand... I think they are wrong.

With more and more people finding, rebuilding EJ22T's to swap into other subarus, it would be something good for them to offer.

I also think the simpiler OBD I ECU would require less work to re-program as well. But what do I know.

boostjunkie · Post by **boostjunkie** » Wed Apr 21, 2004 2:31 pm

NemesisEJ22t wrote:Cool man, just remember that you may get a CEL with the intake. I believe that a few people have had this happen, its no big deal though, doesn't affect performance.

Another reason I'm modifying the stock intake. The reason why the ecu throws a code is because of the variance in piping diameters at the MAF. The stock MAF housing is a 2.5" diameter. The aftermarkets are 2.75"-3". This throws off the calibration of the MAF sensor (just like wrx intakes) causing the engine to run lean. The MAF "thinks" that a certain amount of air is coming into the engine, when in reality, because of the larger housing, more air is coming in. The ecu freaks out because it's readings at the o2 sensor don't jive with the MAFs, so it throws a code either for the o2 sensor or the MAF.

omahasubaru, Kelly is in the process of putting together an ecu for the turbo legacys, IIRC.

vrg3 · Post by **vrg3** » Wed Apr 21, 2004 3:26 pm

omahasubaru - Actually, Cobb did offer me some help with my project of reverse engineering the Legacy Turbo ECU. I just haven't put enough together to bring to them yet.

NemesisEJ22t · Post by **NemesisEJ22t** » Thu Apr 22, 2004 2:16 am

Thats weird. I figured Toyota would have kept using MAP based speed/density sensors like they did with the older versions. At you figured out how to fix the problem. I know here you can't pass inspections with a CEL on.

mTk · Post by **mTk** » Thu Apr 22, 2004 2:18 am

vrg3: the mwsoc irc channel (location can be foudn in a sticky on nasioc) has a few cobb employees every once in awhile, i'm there 24/7 stop by.

MK

vrg3 · Post by **vrg3** » Thu Apr 22, 2004 2:19 am

Cool, I will sometime.

boostjunkie · Post by **boostjunkie** » Thu Apr 22, 2004 2:50 pm

NemesisEJ22t wrote:Thats weird. I figured Toyota would have kept using MAP based speed/density sensors like they did with the older versions. At you figured out how to fix the problem. I know here you can't pass inspections with a CEL on.

Dunno why, but I'm hazarding a guess it has to do with the VVTL-i. This system adjusts timing for both exhaust and intake based on speed, throttle position, and incline. I'm assuming they'd need a more accurate means (I THINK MAF sensors tend to be a little more precise than MAPs, although flakey) to measure airflow.

vrg3 · Post by **vrg3** » Thu Apr 22, 2004 3:53 pm

From what I understand, VVTL-i allows continuous change of valve timing.

If that's true, it'd be a pain to have to predict the engine's volumetric efficiency at whatever position the VVTL-i gizmo is at all the time. If the engine doesn't already have it, it would require some kind of encoder for feedback so the ECU could know the precise offset added to the camshaft's position. And a lot of math.

In such a situation, mass flow is just easier than speed density.

Do other cars with continuously variable valve timing use MAP sensors for fueling?

boostjunkie · Post by **boostjunkie** » Thu Apr 22, 2004 4:13 pm

vrg3 wrote:From what I understand, VVTL-i allows continuous change of valve timing.

In such a situation, mass flow is just easier than speed density.

Do other cars with continuously variable valve timing use MAP sensors for fueling?

You are correct about the VVTL-i system. As far as I know, all cars with continuously variable valve timing use a MAF: Toyota, BMW, Range Rover, and Porsche.

DOA · Post by **DOA** » Sat Apr 24, 2004 7:23 pm

Quick (and utterly useless point) but the elise weighs a damn sight less that 1900lbs, more like 600kg (1200ish lbs). Another useless point is that Toyota offers an off the shelf supercharger kit for the engine lotus used! Gimme, Gimme, Gimme!

legacycentral bbs

A forum for 89-94 BC-BF(BJ) Legacy Owners and Fans

A challenge for vrg3 (now that he's almost done his masters)

A challenge for vrg3 (now that he's almost done his masters)