Lean Misfires at 2700-2800 rpm

[TurboBob]

On my car I run with disconnected O2 sensors now and let the MAFT Pro set the AFR for me. I run 15.5:1 for lean cruise, 14.0:1 idle, etc....... But previously, I used to tune for bottomed trims for best driveability.

Bob

[keithmac]

Cheers, that`s worth knowing. I`m going to try for 90 trims and see what the drivability`s like. Had my instant o2 trim in the 40`s today while messing about, amazing how much fuel the stock ecu can pull when it wants to!.

[TurboBob]

the trims are not true %, they are about 4x the actual fueling adjustment.

The ECU modifies fuel delivery using the following equation:

base fuel * (slow trim + fast trim +200)/400 = final fuel

Bob

[badass3000]

So you just richened it up at 2800rpm?
I am using a second ECU from Avpro, they told me they couldnt figure it out so they sent me another unit, and I still get the stumbles. IIRC when I was getting my car dyno tuned, the tuner told me that he could not tune the stumble out no matter how much fuel he added.
Guess I will have to try it myself.

[DoctorDex]

Originally Posted by TurboBob

With the OBD-1 cars, you should be able to tune it to the rich side so that the trims are bottomed, and in that area of the map the O2 trim is running about 60 - 80 or so. That way, when it goes into open loop for the 16:1 area, it will be a little richer than it wanted when it sets the O2 trim at 100.

I'm not sure if I fully understand what you mean. The problem I have isn't during open loop, its during closed loop under very low to light load, ie. at highway cruise, in 5th gear.

I've tried richening it up for the given rpm points (up to 30% more fuel) and it has helped to some degree (the car doesn't lurch as bad), but it still lurches. The only real fix I've been able to come up with is to cruise outside of that RPM range.

[keithmac]

You do have an extreemly strange fault, almost between the points where the ecu wants to correct to stoich, or go into lean cruise mode (16.8:1) maybe trying to correct to stoich with fuel trims while reading the lean spot off the fuel map?.

Shame you can`t just hit the 2800 rpm range specifically, but I didn`t add anywhere near 30% fuel to correct mine..

[DoctorDex]

The misfiring occurs anywhere between 2600 and 3200, just more often around the 2700-2800 range.

Does anybody know what are all of the parameters the ECU uses to determine load? Is it just airflow vs. rpm? Or are there others parameters, such as TPS, speed, coolant temp etc.?

Just thinking out loud here:
If this problem was ECU related, you'd assume that by changing the crystal in the ECU, ie. to a faster crystal (8200 rpm) then rpm range you'd misfire in would change, but from what I recall it had no effect. This kind've points away from ECU.

Could a bad TPS with poor contacts cause these problems? I checked the adjustment of my TPS last summer and it checked out, however, I never checked to see that the changes of resistance are continuous.

[badass3000]

Im starting to think this is some sensor acting stupid also. I mean my car didnt always do this then just out of the blue started to do it. I thought it was gone when I tried a friends ECU but mabey it wasent. Think I will have to try that again. mabey start swaping all the sensors, TPS, CAS, O2, ect.
I dont use a MAS so I dont beleive this can be related to that, pretty sure I had SD working well at one point.

[keithmac]

Cooant temp is used for warmup enrichment, but when the temp gets above it`s threshold the car will run normal trims etc.

I beleive the ECU used airflow + air temp to get an air mass reading, then references against rpm and throttle position to calculate the actual engine load. What does your TPS signal look like while you`re logging the miss?.

[TurboBob]

actual engine load is airflow (mass) / RPM. Fuel delivery is calculated from this since it represents the amount of air in one cylinder (properly scaled).

However, there are a bunch of airflow vs TPS style thresholds that control the fuel pump relay, decel enleanment (related to this thread), decel fuel cutoff, Power Enrichment (WOT) mode, as well as others I'm sure.

The ECU cannot run 16:1 in closed loop, it is impossible with the stock O2 sensor to detect any AFR other than 14.7:1. So, to run those AFR's the ECU must open the loop. If the trims are bottomed, and the O2 trim is ~80. When the ECU opens the loop and commands 16:1, the actual AFR will be 15.5:1.

Make sense?

Watch the O2 trim when the ECU acts up and see if you are switching in and out of closed loop, or some other action. Also look at the TPS reading.

Bob