ElScorpio

Hello All
I have posted a similar thread asking the adive for "should I go ahead with the software upgrade."
We had alot of feed back and we are grateful for the advise.I sent the car in for a "free flow exhaust system" removed the CAT.(will reinstall when car goes into service to save warranty) they installed the software aswell. Before they removed the CAt and installed the soft were the car was put on a dyno.It read 167kw at the engine (223HP) after the installation of the Software and exhaust the reading was 190KW (254HP) at the wheels. It seemed to have worked but at what cost.
The software is on trail and can be restored to factory setting if required.We need the following questions answered please from any tech guys.

They have tuned the car and adjusted how rich the car burns. We have adjusted the car so that it burns only 5% rich.This was sugested by tuj because if the car runs to lean then there will be a "ping effect" that will damage the seals.
1.Is 5% to lean do we have to increase it to 7%
2.The second question is we live at sea level and the tuners have informed me that there are sensors in the engine that can detect pressure decrease at higher altitude (inland) and that the ECU will adjust across the board.If we have tuned at 5% rich burning will it be to little and cause insufficient fuel supply to the engine & causing "pinging".
These are the questions....any advise...
Thanks from South Africa.
A.

Audioslave8

223hp at the wheels is pretty rare.... most are not even breaking 200whp. Nice job though.

khtm

Wait, did you say 223hp at the *engine* (crank) BEFORE and 254hp at the *wheels* AFTER?

That's HUGE...if that's the case...

r0tor

iTrader: (1)

most dyno's outside of the US correct wheel hp fo flywheel hp... this is what you are seeing with his results. But anyway, nice gains.

Also, yes the car has a barometric pressure sensor that the PCM looks at as well as a temperature sensor that allows the readings from the MAF sensor to be accurate at different temps/pressures.

ElScorpio

Thank u...i am stoked about the gains but will this cause damage to my engine, eg pinging that will damage the seals should I return to the orginal setting....we must remember the power gain to get from removing the CAT and adding a free flow exhaust system....

therm8

That's what he said. And I think he's mistaken. Just not possible to get...what 40-50hp...out of an ecu tune, on a naturally aspirated Rx-8, at least not starting from 223hp. 254 at the flywheel, i'd buy, but not at the wheels.

Edit: just saw "removed the cat," but that's turbo gains at the wheels, still betting on flywheel numbers

Richard Paul

iTrader: (5)

If someone told me that they would give 3% gain with new software I MIGHT believe them. I might even try the product if they offered a money back guarentee. If you tell me you have software that will give me 15% I'm ROFLMAO at you.

Will it do the same for a Zo6 Corvette? That would be nice, 500 HP + 15%=575HP.
Man could you charge those guys a bunch. charge $5000 and get it.

tuj

RP, have you seen the AFR's on the stock Renesis? Pig rich. He's claiming a 31 hp gain, which if the car's rwhp is really around say 185, then that's a 16.7% gain. You might notice that canzoomer claims a 40+ hp gain from their stage 2 and stage 3 units, which is a 21% gain. These numbers are not unreasonable. Would you laugh at someone who claimed a 15% hp gain from a different carburetor? Understand that the factory ECU tuning is not optimized for power, but rather is a compromise. Some systems on the 8 were heavily optimized for power production, like the intake and exhaust (hence why we see such small gains from aftermarket systems), but the ECU tune was not one of them.

I believe Mazda tuned the ECU to run rich during WOT to reduce EGT to extend cat life, especially since the EPA mandates how long OEM catalytic converters must last. This is what was primarly responsible for the 'missing horsepower' issue that hit the 8. Keep in mind that ElScorpio lives in a place where a cat is not mandated, therefore cat life is not a concern for him.

Also, I did not suggest specifically the "5% rich" number, but I did suggest a conservative tune since no one likes blowing engines. I explained that the leaner one runs, the higher the potential for knock (ping), especially at high loads.

From what I'm hearing, it sounds like this is the first credible ECU reflash by a tuner. I am looking forward to seeing dyno plots w/ AFRs from ElScorpio.

Richard Paul

iTrader: (5)

I think that is wishfull thinking. I too dislike not having the control over the damn box. If I had the time and money I'd be all over putting a Motec in full control just to see what is going on.
It may be possable to find out what the power is for real on a Mazda Star race engine. We would have to allow for the lack of street exhaust and other street going accesorys. It would be a great clue as to the potential of peak tuning. I may know some people in that series but I think they don't get to know the real fact. Yet these things have a way of trickiling out.

You'll also find they correct to Std Corected power. But that can be adjusted back.
Also next week I'll tell you for sure what a completly stock 8 makes to the rear wheels. I'll know the corection factor and the control conditions. I sure wish that I had a shop with a Superflow dyno. They have a better correction program.
I spoke to Superflow last week and they directed me to a shop in Camirillo which is not all that far. But is not as handy as the one quarter mile down the street from here.


That sure came out confusing. If you didn't notice some of my answers are included in the original quote area. You'll have to go back and read the whole thing by line. Sorry I don't know how to work these things.

Fanman

All of what you say is true, except no one has ever seen those kind of #'s coming out of a simple ECU upgrade. What Canzoomer claims, & what has been reality have been 2 different things. The claim of 40+ hp has never been seen. Most of the tunes have yielded around 16-22 whp WITH OTHER mods (midpipe, intake, pullies, etc), & tuned for that specific car. That is about 10%..at most. Good luck to him, it's just never been shown here on this board. I would tend to agree with Richard in that adjusting the AFR's & timing will get you about 3%-5%, with mods about 10%. Remember this isn't a massive large displacement V8, this is an out of the box highly tuned engine. Similar to the S2000, you are not going to find massive horsepower with minor mods (like exhaust, headers, intake, ECU re-program).

Razz1

So then a Unichip claim of 10 HP isn't that bad.

tuj

Ok now here is something I really really want to point out: piggyback units are insufficient for a ECU such as the 8's. Why is this? Well its because the 8's ECU offers significantly more complicated functionality than the ECU of even 3 years ago. The ECU is controlling:

primary fuel injection
secondary fuel injection
leading timing
trailing timing
sec. port acutation
aux. port acutation
intake length actuators
throttlebody position
abs
tcs
dsc
egr actuation

We also know that the ECU must control closed-loop operation as well as open-loop, and uses different maps for acceleration and deceleration. There is also a throttle-pedal position to throttlebody position map for both acceleration and deceleration. Everyone knows how hard this car is to dyno. Its because the ECU is very sensitive to the operation range of the parameters its receiving.

So far, all porting attempts I've read about on this board are on-hold because of ECU issues. When MAF or EGT or O2 values exceed normal characterstics, the ECU moves to a cautionary map. And the long term fuel trim of the Greddy units eventually starts to lean out much of the map as it detects the car is running too rich in the closed-loop portion of the rpms. This leaves Greddy owners having to reset their ECU's to keep the lower rpms from leaning out too much. Greddy tricks the ECU into going into open-loop mode earlier by making the engine think its cold, but this may have some other side effects that we are not aware of and still doesn't entirely fix the problem.

The whole point of a piggyback is that if a ECU's function map between input and output is known, a new input can be devised to create an arbitary output. Any parts of the output that need modification can also be modified if necessary. BUT: this assumes that the translation function of the ECU is known, which it is not. Things like long-term fuel trim, and possibly other adaptive characteristics this ECU has mean that it is not a simple mapping as it may have been with older ECU's.

Now, I don't want to discredit canzoomer since I really don't know if 40hp is possible or not firsthand, but it is limited in that it can only make the modificaitons to the input and output of the stock system, and it cannot know how the stock system is adapting to conditions, at least given what little we know about the ECU thus far. Therefore, the true power potential lies in utilizing the existing software to fully tune the car in every way, rather than just simple open-loop timing and fuel. I'm saying that piggybacks can never hope to fully optimize the engine; only a full stand-alone or modification to the stock unit can.

I have no idea if these tuner have done things like changing the throttlebody positions, port actuation timings, etc to optimize the engine. But, I fully think that if this were done, a conservative tune (ignoring cat life) for 93 ocatane could achieve said 15% goals. Remember, if the ECU is really running closed-loop up to around 5k rpms, then the AFRs targeted by the ECU are almost certainly leaner than ideal, due to the fact that during acceleration, richer AFR's are targeted for peak power production. And it runs way too rich at higher RPMs, not to mention the throttlebody positioning may not be optimized for optimal air velocity (followed by a corresponding fi duty cycle) to achieve maximum volume and still maintain a slightly rich acceleration mix at all rpms during acceleration.

But no, I haven't had the priviledge of examining and modifiying the stock ECU software myself, so I have no idea how much potential hp is in the software. However, before you say that you've never heard of a 15% gain on an engine, consider how many of those engines were the first of their kind to meet modern emissions standards, and produce more hp than the previous generation of fuel-injected NA rotary? I don't want to seem like I'm justifying Mazda loosing hp on this engine due to their programming (both by choice and by error) but I'm just saying the idea is not so far-fetched. Also, remember that he doesn't have a cat, and he has a free-flow exhaust, which may work in synergy with the ECU tuning.

Sigma

The thing is he's saying he got a LOT more than 31hp.

He said the baseline was a corrected 223 at the engine (probably ~175whp) and, with the ECU tune (and "free-flow" exhaust), he was at 254 at the wheels -- that's a gain of over 75hp. That's more like a 40% gain. Or almost double what many are getting with a GReddy turbo.

Even if we presume there was some error in the translation and that's not what really happened, I still have to wonder why some tuners that have apparently been able to do what rotary experts have not in over a year, have to come on here to ask such basic questions about the ECU in this car. If they were able to what no one here could, one would think that they shouldn't need those questions answered.

tuj

Help from an insider at Mazda. I don't really know if these guys did or not, but my guess is that they had to have, in order to be able to compile a reflash that is accepted by the ECU. My guess is that they got access to the source and modified all of the necessary fueling and timing maps to tune the engine to a higher spec. In fact, I wouldn't be surprised if such a flash already existed at Mazda (ie. a high-tuned set of maps designed before cat-life became a concern).

Remember, a 13b for Formula Mazda makes ~175 rwhp, 240 e-shaft, and that's at 6300 rpms with carburetors. And those engines can run 3 seaons easy without a rebuild. Yes, there is a powerloss from the side-exhaust on the Renesis. But the engine is still dyno'ing 180+ and there's no way its tuned as aggressively as a race engine. Plus it benefits from FI and throttle by wire.

Now someone may point out that Racing beat has only achieved mild gains in their own ECU tuning. This may be true, but I'll bet they intend on keeping the cat also. Until more people tune their cars with canzoomers or full ECU's without regard to the cat, I guess we won't really know. But I'm betting that in race tune, the Renesis could put out well over 230 rwhp in the 8.

rexi

Elscorpio,
Who is the company that has reflashed the ecu?
This program could potentially be applied to rx8's from all over the world as long as similar fuels with similar octane ratings was used.
Did you mean to say that ENGINE hp went from 228 to 254 at the ENGINE,as this would be feasable?


This I feel is very ,very promising.

Regards
Rexi

mcpheeg

In my expericence of tinkering with the ecu and cats in/ out.
Standard RX8 straight out of the showroom will dyno at approx 215HP at the flywheel.
RX8 with cat out and stock ecu with free flow exhaust system will dyno at 220HPat the flywheel.
RX8 with cat out and unichip piggyback with free flow exhaust will dyno at 227HP at the flywheel.
RX8 with high flow cat (more restrictive than cat out), canzoomer unit modified as per his own instructions for full rpm and his own stage two map, free flow back box and exhaust header will dyno approx 240HP.
How do I know this, I have done it.
Please see this thread for the proof.
So that is approx 9% from the canzoomer unit. Based on stock rx8 without cat. 240 against 220HP.

So bearing in mind that not all dynos are 100% acurate this could be the gains he is seeing. The only reason I run with a cat is the RX8 without a cat is very smelly. (also illegal in this country)
Dang forgot to mention the 240 run was with a lightened flywheel.
Canzoomer unit does work wether it acheives 40HP+ is another issue. For all my mods I got 25HP. Although this doesn't seem a huge gain, it is a gain from an engine that mazda has gotten alot out of already.
Also some of the turbo kits are only getting 50-75HP gains. Turboed I would expect alot more, but hopefully someone will get a supercharger project finished so I can buy it, install it and enjoy it.
Heres to the developers that give me all the fun of spending loads of cash and tinkering with my car.
Keep up the good work.

derwankel

Even if someone has managed to re-flash the ECU, I'd have to believe with the encryption that is in place that they have to either have authorized access to the encryption keys and therefore legal limitations placed on them and what they can do by MAZDA, or they have defeated the keys. Personally, I would not want someone elses code who managed to crack the keys managing my engine, much less my power steering and traction control.

I have often speculated to myself that in today's world of multi-billion dollar aftermarket opportunities, that a manufacturer like MAZDA that had plans to roll out their own forced induction, would likely make it very difficult for competitors to get any upper hand. If you think about it, their implementation is quite ingenious ... that is to tie all of the various systems (power steering, radio, engine, transmission (auto), etc.) into the ECU. Therefore, even the best third-party ECUs like motec, have no chance of managing all of these systems.

I would not be the least bit surprised to learn that MAZDAs FI (speculation) is managed by the ECU as well. And I would not be surprised to find out MAZDA is a test bed for FORD and future practices in the art of encryption.

And think about this ... all MAZDA has to do is change the encryption keys on the ECU the next time you are in for an oil change and bang ... back to square one.

The world of high performance automobiles is rapidly changing. If I were in MAZDAs place, and I saw an enormous market for aftermarket performance, I'd make sure my products had every advantage available.

Unlike Microsoft, I doubt that we will see the MAZDAs of the world sued by the likes Acosta or RP for anti-competitive practices. Particularly when their ECU does much more than manage fuel, timing and ignition.


Just my speculation ... and of course, I could be wrong ...

dmp

btw folks - dyno's don't have to be accurate - just consistant.

ElScorpio

Gentleman
This tpic has seemed to got everybody wondering what if...Can it be...how have they done it...Well we have removed the CAT yes and have a free flow....that did put the HP up...the software upgrading did work..there is no lag on the accelerating...its instant and constant....but we stand to lose 2% of the power gain as we need to install a silencer or muffler..to reduce the noice from the free flow...it is echoing into the car...dam loud...but the Check Engine Light did come on today...we think it might be due to the sensor for the exhaust...as the air flow out is not constant with out the CAT...but there was the same problem with the software in Germany due not the software but the exhaust system sensors...we have the exhaust senors installed on the line but the air is running out very quick...you should here the backfire from the system...powerful...

We will do more test this week and will everyone in the loop regarding the outcome...we will try and get the name of the software to u guys...

Cheers...

Red Devil

Thanks for the info and please keep us updated, ElScorpio.

dmp

The only 'exhaust' sensor measures the air/fuel ratio. That sensor doesn't measure exhaust velocity, thus, it would have no idea if the exhaust is flowing faster or slower or not at all.

army_rx8

hmm interesting thread...i think i shall hold my opinion in reserve untli i hear more of what was really done...etc.etc. still very exciting...hehe but every thing on here is exciting to me

Nemesis8

iTrader: (3)

Are you going to post how you connected to the PCM? And what software you used to write your own maps?

After watching the ProMazda at Buttonwillow 2MB video, I agree that the Renesis has yet to show us what it can do on a street car. The ProMazda Renesis cars are running Motecs, and have the ability to change the AFR from within the cars cockpit. The only other modification was the addition on the lightened Mazdaspeed flywheel. So, leaning these cars out to the sweet spot will create RWHP for sure. How much is not known yet.

So, if this truely is the FIRST RX8 other than the Racing Beat test bench 8 to get at the actual PCM maps, I am impressed with your efforts. Please keep us informed on your progress and how you accomplished this feat.

Cheers

tuj

I do not think this is the case. The ECU flash is most likely encrypted or signed with a Mazda key, which corresponds to the other part of the key-pair that is embedded in the ECU. If you don't know both sides of the key-pair, you really can't do anything. But, if the algorithm for the encryption is known, and the key in the ECU could be changed, then it would be easy to generate a new key-pair, and put half of this into the ECU, so then it would accept your new signed code. Basically what I"m saying is that one side of the key is probably embedded in the ECU and cannot be easily changed, meaning it would be hard for Mazda to lock us out IF a tuner already had the Mazda key-pair.

derwankel

You presume it is difficult to change the encryption keys ... which it is not when you are the OEM. Then there is the practice of randomly rotating the keys. There are any number of ways to prevent easy access.

And sure, if you know the cryptography AND you could reset the primary key, of course you can generate your own key pair ... game over. But I imagine that would be very hard to come by, and as I said, it would likely come with serious restrictions from MAZDA. If it were obtained through clandestine means, well, then you have to ask yourself if you trust the hacker to flash your ECU. Which is why I am suspect of any "legitimate" claims to remap the ECU. Emphasis on legitimate.

If Racing Beat is "playing" with the ECU maps, they likely obtained access through MAZDA given their relationship. I don't see any crypto junkies at RB. Consequently, it is unlikely they would produce any maps that MAZDA would not have control over, not to mention the risk Racing Beat would take in creating a map that caused, for example, pre-mature cat failure. If it were that easy, and there was no cat concerns (i.e. our friend here) then, sure, MAZDA could easily lean out the AFRs in open loop. It's not that simple, I'm afraid.

I can certainly see RB doing research and testing maps for "off-road use only" ... but then you've basically given up your warranty anyway. You're not going to see those guys advocating pulling the cat for leaner AFRs on the street. ... I could be wrong though.