I am about to tackle the feedback thing now. I stillk have a few baseline cells to get right, but not too many. The parameters that I can control are the MAP readings that trigger the O2 feedback off, and then back on again when it comes back down. Ther another parameter is the RPM range to make it active.

Any ideas to get me started? The big question is should it be active at idle? Since I cruise at 1200 RPM often, it seems it should kick in lower than 1200, so that is not a problem. But should it kick in at idle. It seems like it should to me, but I am no expert on this. The other idea is at what MAP reading the O2 correction should cut off. It seems like maybe around 85 KPA, and then cut back on at something just under that figure. There are are a number of fine tunings to be done, but they are things to be just worked out with tuning. One of them is what ECT for the O2 feedback to become active. Another is how many seconds after startup for it to become active.

Any thoughts on this subject are appreciated.

Sam

_________________

Keep in mind the fact that I dont really know.....


Idle is its own thing....no o2 at idle.

WOT is its own thing....no o2 at wot

MAP is how you differentiate between idle, cruise(O2 sensor) and WOT...
you must have a read on that by now.

The O2 sensor should start imputting as soon as it is hot enough...

did I mention I dont actually know?.....

_________________
Yeah....Im the one who destroyed this rare, vintage automobile.....

Why don't you want O2 at idle. Is it because you want it to idle richer? This engine idles nicely at 14.8. Thanks for your honesty here. YOur signature pretty much says that also.

It seems like there are cells that are low vacuum that would want to be richer than stochiometric that are not quite wide open throttle either. Maybe it needs to go open loop somewhere sooner than WOT?

Sam

_________________

Is the O2 sensor heated?

The reason for open loop at idle is that with an unheated sensor it may not stay hot enough and you can't trust the readings......

I'd say goto open-loop at about 90% of WOT (or the MAP equivalent)

_________________
Ed
64 Valiant 225 / 904 / 42:1 manual steering / 9" drum brakes

No o2 at WOT is a narrowband thing. With wideband start excessively rich (but not as rich so that it misfires) and lean as desired.

No o2 feedback at idle is for stability. At that low rpm with no load it doesn't take much adding/subtracting fuel to cause it to waver. If you can get a steady idle with feedback during all driving conditions go for it.

Ed brings up a good point about heat, but assuming wideband or even heated narrowband that shouldn't be a big issue.

Thanks guys. I have a heated narrow band providing ECU feedback. The wideband serves a gauge only. There are trim tables that allow me to miniize the feedback percent at low air flow rates. I suspect that Pierre's comments explain why. There are also self adjusting features that change these trims, with time spent driving, based upon the percent of correction going on. You have some control over this activity, but I don;t understand it yet.

When I first got the car running the idle was very unstable, and I considered correcting this with O2 feedback, but did not. As it turned out, this instability was due to three things: 1. TB blade adjustment; 2. timing cells at idle range being different from one another; and 3, fuel cells adjacent to each other in the idle range being different from one another.

As the RPM would rise slightly, the ECu would give the different neighbor cell's VE value more weight in the fueling formula, and the RPM would go up even more if the neighbor VE value was higher. The idle control would then grab it, and bring it down. When the pointer got in the lower KPA cell, then the RPM would crash until the idle control program recongnized the lower value, and would bring it back up, at which point the cycle would start all over again and continue in a rolling fashion. The ECU adjusts the idle RPM with timing adjustment as well as idle solenoid. The percent of adjustment of timing is also selectable.

With the turbo, the MAP readings are not always directly proportional to throttle position. If the rpm gets high enough with the throttle open just half way, it will go into boost, and run right up into high KPA values. I guess that is why they chose to relate the O2 feedback to MAP reading and not throttle percent.

For the sake of those reading to gain cnfidence to install their own EFI, here are two prinicpals to remember when setting initial tuning:

A block of VE and timing cells adjacent to one another in the idle range must all be adjusted together as a group. For me this was 750 and 1000 RPM and the highest 3 vacuum (or lowest KPA) cells. And, the throttle blade adjustment must be adjusted so the idle solenoid can operate on both sides of this TB adjustment. If the TB is so closed that the idle solenoid has to be fully open to maintain the target idle speed , then the idle solenoid has nowhere to go if the idle drops. The same principal applies to the how closed the idle solenoid is.

The Accel splash screen displays what percent the idle solenoid is open all the time. Each maker will have his own recommendation for the percent the idle control solenoid should be at idle. I seem to recall that Accel wants the idle solenoid to be 10 percent open at idle. So it appears that the solenoid functions primarily to raise RPM if it falls. You must then adjust VE values, and timing vlaues in these blocks, all together, and adjust the TB blade to compensate, watching the percent of idle solenoid opening. It is back and forth until you get it right, but this is a base line that must be set before anything else is done. In the Mega Manual, it says get the car idling well before you do anything else. Well, you do have to get it started. Then adjust the start up calibrations until it will start reliably.

Also, be sure to save your new calibrations as a discrete file after a tuning session. This will allow you to go back there later if you change something incorrectly later.

Sam

_________________

Sam,

I do not have direct experience with your EFI system, however I have been developing my own system for a few years and might be able to help.

I desire to get the open loop performance the best I can before going closed loop.

I have found that closed loop O2 at idle can be enabled after the following:
O2 sensor is warmed up ( this will happen quickly with your heated sensor).
Initial starting fuel enrichments are complete.
Engine is approaching 60C or greater.

If you try too quick the engine may be too lean and stall. If so increase the time before going closed loop.

I have found that O2 feedback should be inhibited when coasting (MAP values are lower than normal idle). Otherwise you will be too rich.
Do not close loop at high load, boost, or WOT, you will want 12:1 A-F.

On the TB, typically the throttle plate stop is adjusted to fully close the throttle plate without touching or binding in the bore. An air by-pass screw is adjusted with a fully closed idle air solenoid to give an idle slightly slower than desired when fully warm. The idle air solenoid can then adjust the speed. If your TB differs from this please ignore.

On idle quality: The base fuel is important. It is the fixed component of fuel that does not vary with MAP. For instability increase it a bit at a time, however no more than needed.

You are on the right track of keeping the fuel table smooth and VE cell values fairly close adjacent to idle. Just off idle most engines take a dip in the fuel requirement when lightly loaded. Try leaning a bit, then fuel increases as RPM's increase (follows engine torque curve).

Hope this helps.

Thanks for the feedback. That would be 140 F. I will start there with the O2 feedback. I will also keep the O2 feedback out of the lowest map cells, where drifting or coasting would be happening. Then, of course I can tune from there.

Sam

_________________

Most engines, especially ones with batch fire systems, like to idle a bit richer than 14.7:1. 13.5:1 is fairly common, but you'd need to try different settings and see where it runs smoothly. This is why you don't want narrow band O2 feedback at idle.

_________________
"Mad Scientist" Matt Cramer
'66 Dart - turbocharged 225
My blog - Mad Scientist Matt's Lair

I was not aware that O2 sensor is out of picture at idle! WHY?

What about the cars out there like this?

Cheers, Wizard