MadDog

Of course there is: Things are changing MUCH faster at high RPM than at lower RPM. You yourself say that the ECU needs to switch to open-loop when things are changing too rapidly for it to keep up. Everything is changing faster at higher RPM - especially the rate of change of RPM (the rpms come faster as rpm increases). Besides, those shift points were measured , not just guessed at.

r0tor

iTrader: (1)

the car is in closed loop under 6,000 rpms (for the most part, it depends on what gear your in) and it can be proved on numerous threads in this forum.

If you take a look in the Greddy Turbo thread it looks like Greddy is using a box to control the engine temp sensor which will make the car think its cold and throw it into open-loop control much earlier.

StealthTL

iTrader: (1)

Unichip seems to have a much better reputation abroard, they are very highly regarded in Europe, the UK, and as we see, in Australia.

My sources tell me they are the tuner for the upcoming STS rear mounted RX-8 turbo kit.......I hope this means their turbo project won't be dropped (and fade away like so many others) when tuning time comes.

S

takahashi

I manage to miss this thread for a week.

APS is the head company of the unichip and they have done mine as the 2nd car. The head tuner did my job over a day and he is the one going around US to teach people tuning unichip.

I am happy in a sense... but ECU tuning only have a certain limit.

I am very confused on what tuj have said. Sounds like he wants his AFR being rich rather than lean that amazes me????

I have another dyno showing the AFR is around 13.6 at most time @WOT. At worse it is about 12.7. So it is what APS have said about their target AFR.

AS I have observed many cars via the sCANalyser. The wideO2band is about 0.75 at the lowest point when they open their throttle and drive hard. My car is lightly leaner that all during city driving. It is a very conservative tune. But the car is certainly smoother than before.

I am going to refresh the ECU with a new flash. I wonder what that does

tuj

No, what I meant was that closed-loop operation is significantly leaner than open-loop operation, at least in a normal car, to try to decrease emissions and increase fuel efficiency. I'm stating that the stock AFR's during open-loop are very conservative; I certainly don't want my AFR's any richer than when I'm driving hard! :-)

Ok, I can buy that. (Could you point me to one of those threads tho?) But I honestly don't understand why that would be. Most ECU's treat big deltas in TPS or other values as indicators to switch to open-loop in anticipation of dramatic changes in the engine. This WOT throttle mode normally happens ANY time the car goes WOT, regardless of RPM's. Why would the 8 not do this? It doesn't make sense. The only thing I can think is that the 8 may go to conservative-dynamic closed-loop, using the wideband as a rough indicator of AFR's. But even still, the ECU will have to go richer than stochiometric and well outside of the bounds of typical, constant-throttle-closed-loop AFR modulation. My point is that the ECU is going more conservative when inputs change dramatically, regardless of RPM. Do you know for certain that this is not the case?

guitarjunkie28

maybe the zero overlap has a bit to do with why it stays in closed loop so long?

takahashi

You have to see open loop and to see it real life with sCANalyser or other CAN scanner.

Otherwise you don't know what you are talking about.

Open loop is able to turn on even idling. Over a certain throttle percentage and in certain rev range, there is no pattern to it. And you are certainly in open loop in WOT despite you are in 3000 rpm.

I disagree the car is leaner at open loop. As seen in my sCANalsyer, the wideO2band is matching at 1 (Lambda) when cruising in open loop. And during WOT and open loop, my wideO2band goes to 0.8 (which is lean). Unless Hymee and sco fu&ked up the meter and mislead me (which I don't think they have), my observation tells me that open loop is leaner then closed loop. sorry :o

tuj

takahashi: i'm sure you just made a mistake, but could you clarify this?

takahashi

....... ^^"

Maybe I am confusing myself....

I have to look at it again....

r0tor

iTrader: (1)

Look in the Canzoomer section - there are a ton of warning there to only tune the computer in open-loop rpm ranges because the ecu will compensate and untweak your tuning if you hit a closed loop area. There is actually a tag you can log that tells you if your in open or closed loop as well as just taking a glance at the a/f ratio (a perfect 14.6 closed loop and 12-12.5 open loop). I've spent a ton of time tuning my CZ unit over every throttle position and rpm range and it just doesn't go to open-loop until you get above 5500-6000 rpms no matter what you do.

takahashi

........... ^^"

Oh yes it is air: fuel ratio. So the 14.6 is leaner than 12.5 that what stock is running. Mine is around 13.6 and not dropping below 12.4 at 2 dynos.

Sorry Tuj.. I am sorting myself out now. You are right! ^^"

tuj

Gah! I hate to think what this might mean. If the engine seriously does stay in closed-loop even during WOT conditions under 6k, then they must be slowing the acceleration of the engine intentionally to keep things at a pace where the closed loop can maintain stochiometric. Makes sense too that it would do it via the throttlebody. This has me wondering if:

-Mazda has intentionally slowed the acceleration of the engine under 6k even at WOT to maintain closed-loop for a longer period of time?

-Or the O2 sensor can actually react fast enough, so the car can keep up with changing conditions and run stochiometric easily accelerating as fast as possible up until the 6k point. But if this is the case, why does the car run so rich then? Does the O2 sensor run out of resolution around the 6k point?

This has me curious, does anyone know what the resolution of an o2 sensor is? Basically I'm curious as to how many ignition cycles its readings are behind the timing of the engine.