Last week, I took my 2004 RX-8 to the dealership to replace the sun visor. When the service advisor entered my VIN, he found that Mazda's MSP16 service program applied to my car so they flashed my PCM.
Afterwards, my car stalled when I stopped for traffic lights. I immediately took it back and they verified that the parameters were correct. They said that it needed to "learn the way I drive." This sounded bogus to me because what kind of program would allow the idle speed to drop below 500 rpm. Nevertheless, I decided to drive the car for a few days and see what happened. The car has not stalled once since yesterday.
I don't know much about ECUs but before they became popular there used to be a screw that was used to set the idle speed. Isn't this one of the parameters?
Right after flashing, does frequent stalling at idle suggest some other possible issue such as dirty plugs or cracked wires? I have not checked these.
The car seems to be running well now but I should mention that I use regular gas so the accelleration (while fine for me) is not as good as when I used premium.
Bob Simon wrote in news: snipped-for-privacy@4ax.com:
Idle speed is set by the ECU. The stop screw is obsolete.
No, the computer has to "get used to" the particulars of the vehicle. Believe it or not, it adapts over time to the vehicle's "personality" (for lack of a better term).
Many vehicles designed to run well on premium can be run at a reduced performance level on regular. I believe the RX8 is one of those.
There's more power in "regular" gas though... Premium is needed if using regular gas causes pre-ignition due to an engine having high compression. Premium is LESS volatile than regular. But you knew that... ;-)
The big mistake many make though, is putting premium into an engine that was only ever designed for regular gas, as all you are doing is REDUCING your performance, not improving it!
If that RX-8 engine was meant for premium, he should be careful as there is possibly very hot conditions in the combustion chamber with some pre-ignition that he isn't hearing... A quick check of the plugs should determine whether this is happening. How much is a rebuilt motor going to cost you? There's always mid-grade... probably a better choice to go with than just regular grade.
OK. I presume the idle speed is set to 750 or 800 because that's where it idles now (which is just fine).
I'd still like to understand why the idle dropped to less than 500 rpm for the first few days after the flash. It seems like the ECU program should prevent that even BEFORE the computer gets used to my car's "personality". No?
Anonymous wrote in news:anonymous- B11AA6.08012710092008@[74.223.185.199.nw.nuvox.net]:
The best choice is to just go with what the owner's manual recommends.
2 cents: If you buy a car that's designed to run on premium but can't/won't pay the extra dime or so per gallon, you shouldn't have bought the car in the first place.
Bob Simon wrote in news: snipped-for-privacy@4ax.com:
The ECU is doing a lot of stuff all at once. Until it's learned all the particualrs, it's going to be changing all kinds of parameters, which affects things like idle speed.
I'm sure they could've engineered the ECU to behave better while it's learning, but the improved programming might've cost more than the bean counters would've wanted. Everything's a trade-off.
Generally, the thing to remember about idle speed, idle "dip" and ECU vs. mechanical adjustments-- Something (usually the mechanical adjustments) establishes the minimum idle speed when the ECU is running in "open loop" mode. When the ECU "hunts" for proper settings, it can really cause some odd behavior. Minimizing this may be as simple as increasing the open loop idle speed mechanical adjustments. The basic idea is to make the conditions/combinations/tuning that the ECU is trying to find and set to more easily reached. Part of the problem has to do with tolerance/accuracy of the various sensors. The ECU is stupid, and only understands what the sensors are telling it. The tolerance on the sensors is perhaps wider than it might be. This means that the ECU must learn how they respond in relationship to each other, and decide what the "proper" output is.
(Not to mention that the OBD2 standards were setup to subsidize dealer shops with Mfr specific testers.)
Chuck, Are you saying that the RX-8 has a mechanical idle screw? If so, I just need to take my car back to the dealer and ask them to adjust it to a min of 500 and my problem will be permanently fixed.
(I forgot to menti>Generally, the thing to remember about idle speed, idle "dip" and ECU vs.
Adjusting the mechanical points (e.g., idle stop screw) from factory settings is not a good thing. The computer takes care of all that stuff.
If there is in fact a mechanical stop screw and it's still set from the factory, a reputable dealer won't (and shouldn't) adjust it.
If the car's not running right, take it back to the dealer and let them diagnose the condition. Chances are that the ECU is to blame and not a physical stop screw setting.
That said, if the dealer that flashed the ECU changed the stop screw (if there is one) to begin with, I'd find another dealer.
Bob Simon wrote in news: snipped-for-privacy@4ax.com:
Joe, Thanks for the suggestion. I will take it back tomorrow.
I'd like to understand the reason for your comment. Since the car has been intermittently stalling ever since the dealer flashed my ECU, and since you said youself that Mazda engineers, "could've engineered the ECU to behave better while it's learning", it seems reasonable to give the ECU some assistance in this situation.
If the car has a mechanical idle screw, shouldn't it be set at the factory to prevent the idle from dropping below 500 rpm? This seems like a simple solution to a predictable problem. Bob
Bob Simon wrote in news: snipped-for-privacy@4ax.com:
A wise decision, especially if the car's under warranty.
Introducing another variable (i.e., changing the idle stop screw) simply compounds the problem IMO. After changing the stop screw position, you now have two issues: (1) a misbehaving ECU, and (2) a misadjusted stop screw.
The stop screw doesn't really control the idle point - the computer does that (in conjunction with the physical attributes of the engine).
Not knowing the particulars of the RX8 - in general, the stop screw should be the point at which the throttle is completely closed, i.e., your foot is off the gas pedal. Without considering computer control, if you turn the stop screw to the point where the engine revs increase, it's much like pressing down on the gas pedal a bit. At that point, you're no longer at "idle". The throttle has opened a bit, and more fuel/air is delivered.
Given this scenario, introducing a computer into the picture complicates things tremendously. With the throttle partially opened, the computer senses that engine is not idling so it starts to change things like the fuel/air mixture, timing, etc.
In summary, the stop screw is set at the factory to the correct position. If, after changing/flashing the computer, the car doesn't idle properly, it stands to reason that the problem is solely within the computer. Changing other things that weren't "broken" in the first place compounds the problem, making the solution that much harder to achive.
The whole thing boils down to what was the car doing before the ecu flash. If there was a problem, it should have been corrected before flashing the ECU.
As to an idle dip--- The throttle body and butterfly likely need cleaning. Since there is a bit of wear over time, the open loop idle may have changed due to the changed tolerances. This is something that the ECU normally would correct for as a long term trim. This process is involved in the "learn mode" others mentioned. and since a flashed ECU usually starts with factory defaults, wear (and crud), as well as sensor age and output shift can change the trim away from the defaults.
Also of concern is proper operation if the IAC (idle air control valve) If the (usually stepping motor driven on mazdas) valve is malfunctioning by sticking, it can cause idle dip problems. An issue on the Miatas chased back to crud getting into the passages between the IAC valve and the throttle body, as well as intake manifold passages related to the EGR valve and other emission control functions. Dealers like to change the IAC valve. Unfortunately, this is not pocketbook friendly.
MotorsForum website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.