AFR consistency?
 

Well, i know this forum is kind of dead, but i thought i'd try posting this here anyways...

I have an 89 base, stock S4 engine, bunch of mods (see sig). but, the important part of this is a stock ECU, an SAFC-II, and a Zeitronix ZT-2 wideband.

anyways, i have been trying to tune my car on the street. i thought it would be simple, but i am running into some wierd AFR inconsistencies.

Basically, i went out and tuned the car for about 13.0 AFR across the range at WOT. the correction on the HI map is pretty leaned out, getting to about -20 at redline. now, i did a run or two and had it pretty good, but i have noticed that AFR will change slightly from run to run, and on different days the AFR will be different also.

Has anyone else noticed this? My car seems to be running fine, with no error codes from the ECU. everything seems to be within spec. maybe the stock ECU is just not very consistent?

It kind of sucks because i have seen as much as 1.0 AFR discrepancy between runs. so, going rich to 12.0 i loose power, but going lean to 14.0 at WOT can get dangerous...

I can't help with the varying afr, but I got my 86 n/a restickered and put a SAFC out of my other car into it and in the upper ranges, say five grand and up I have it leaned out in the minus 20-30 range just to keep it in the mid 12'S.

I think it helps to pich a uphill section of road and going full throttle. Puts more of a load on the engine vs picking a flat or downhill section of road.

I'm not much of a *turner*. I set the series four for high at 100 percent and low at 99 percent and more or less have both high and low reading the same at a given rpm. Except at the 1000 rpm I have it at zero and use variable resistor to lean/richen the idle afr. Series four cars have the variable resistor. Like I said, not a tuner here.

yeah, i am definetly realizing how much these cars need to be leaned out in NA form, even though i have full bolt-on exhaust, intake, etc. mods.

i'm not a tuner either, so i have just been going from what i have read in terms of AFR.
some people have said 13.5 for max power, but i find that a bit lean. so, i figure 13.0 is a good number to go for. it's really hard to say though if there is a difference between 12.0 to 13.0, i guess i would need a dyno to really tune properly.
there are two main points i have found that have been hard to get smooth: the
- secondary injector kick-in (about 3600-4000)
- VDI valve opening (about 5500)

at both these points, it is a little lean before, and then after it is rich. i think i will have to put the NE points closer at those RPMs, rather than just having one NE point at the problem spot.

that's a good point about picking an uphill. i also find that 3rd gear i am a little leaner than 2nd, although maybe that is just coincedence. i have been tuning in 2nd gear though, because that is what i am always in at autocross.

i have a Series 5, so i have the low throttle point at 25%, and the high throttle at 50%. i figure that the NAs reach full load pretty fast, so there is not too much difference between 50 to 100% throttle in terms fuel mixture. but, i might try setting it higher.

so really, i am wondering why it is not very consistent. HAILERS, were you able to log consistently the same AFRs run after run, day to day?

More and more I think the biggest problem with RX7s is the ECM instead of something else.

If I ever play with a N/A FC Im definitely going with a megasquirt. Pretty soon UMS will be out, which has ion sensing, so no knocking! :D

yeah, i actually have a V3 Megasquirt sitting in my room, but for now i am just trying to see what sort of tuning i can get with the stock ECU + SAFCII.
they haven't really gotten widespread implementation of even MS-II yet on rotary, so i don't think UMS will be all that "soon".