Short Term Fueling Compensation

Looking to query the brain trust as to why an ECU operates a particular way.

I run a big block Ford with ITB's, O2 sensors on each side. The ITB's balance to where the airflow is matched at idle and also when I open the throttles. At 2k rpm constant, my airflow measured with a STE Synchometer matches exactly on all 8 cylinders.

At the same time, my target lambda is 1.00 and the sensors fluctuate between .98-1.02 on both sides.

The engine runs fine and I am happy with its performance.

But now the question.

When I look at the live data or any log, I see that the Short-term fuel trim on cylinders 1-4 is constantly around 120% and cylinders 5-8 show it around 80%. And that remains pretty constant, no matter what the rpm is.

Can someone help me understand why this is? Assuming the injectors are all matched as close as typical production allows and there are no major leaks in the headers or intake, I would expect them to be around 100% and both sides relatively equal.

What am I missing

Thanks,

Paul

I would say you are using bank-based fuel trims, and you have the O2 Sensors reversed (Bank1 sensor in Bank2). The system is just dialing itself to the maximum allowed short term fuel trim. See if you can swap the sensors in the software configuration, if not, then perhaps you can swap the pins at the ECU connector. If that's not possible, then maybe you actually have to move the physical sensors.

That was my immediate thought as well. When the ECU attempts to trim 2 banks and the sensors are assigned backwards, that's usually how the trims end up.

If that's not it please let us know though as it's not the only possibility, just the one I've seen most often when that's the symptom.

Thanks for the input guys. I went out this AM and checked the wiring, sensors and ECU. It is changing fueling by 'banks' as you said however it is wired and configured correctly. Side 1 is the passenger side, Side 2 the drivers side all the way thru in the hard wires and ECU settings.

I've sent an email to the manufacturer but don't expect an answer very quickly as all the knowledgeable guys are on vacation over the next couple weeks.

Any other ideas I can consider during the wait for their response

Paul

Just for grins, try swapping (what cylinders are in what banks, or which lambda is in which bank) in the ECU settings -- the result will either be better, worse or the same!

Could be the old programmer "off by one" error - computers tend to start counting at zero, so instead of Bank 1 / Bank 2, they would be Bank 0, Bank 1. Now you can see how it someone could code bank 1 and "not bank 1", and get them reversed.

Again I'm in agreement with David. I'd at least try it.

Alternatively, is there any chance your ECU is associating incorrect cylinders with a bank? For example some ECUs can make this a bit confusing by using firing order number, cylinder number, or output number in various places.

OK, confirmed the hardware is correct. Pin A connected to Wire A. Settings in the ECU are correct - But short-term fueling is still opposed.

I did some rudimentary testing today switching sides and didn't see anything but realized I got sloppy in my procedures. Doing a more proper test in the morning.

I did find that turning off Long Term, the engine ran fine, Lambda on the right bank dropped about 20 points, left bank stayed right where it was.

I ran some more definitive tests and confirmed the engine/ECU is wired properly, setting configuration (Left,Right,etc) is correct.

1) With the LTF turned on, the right side Short Term removes 18% and the left side adds out about 22% to maintain an average lambda on both sides of 1.0

2) when I turn off the LTF and Short Term is disabled, the Lambda values for the right side go to about L.86-.92 and for the left side, L.1.13-1.20

During this test, the air flow on all ITB's using a STE flowmeter are matched exactly - the width of the pointer is centered between two dial marks the same on every cylinder.

Just to verify it was correct, I then swapped the side to side settings of the ECU so it thought O2 left sensor was monitoring cylinders 1,2,3,4 rather than 5,6,7,8 like it really does. The actions of the short term did reverse as you would expect. The engine continued to run however you could tell it wasn't 'exactly' right.

So, I've learned the engine is configured properly and the ECU is trying to do its job. I have the ITB's balanced properly yet my short term trims are opposed, 22% one way and -18% the other way.

Is this just an anomaly of the injectors/ITB system or is there something else I should be looking at? My ECU limits the short-term corrections to a max of 25% which I have it set to. The right side continually bumps or stops at that limit and the left side does occasionally touch it.

I would think that a properly adjusted engine would see these fluctuate +/- 10% as it adjusted to heat/humidity/tolerances/etc.

What am I missing? Below is a screenshot of the log showing RPM, Lambda and Short Term

This must be a mechanical issue -- The clue is the with no closed loop, the banks don't seem to be balanced mixture-wise even though you've verified the airflow seems to be identical. Here are some ideas:

- Could there be another air leak (bad / missing gasket, non-flat making surfaces of the manifold / head) that doesn't show up on the flow meters.

- Does the fuel rail somehow supply more pressure to one bank than another. I wouldn't expect that to show up at idle, but maybe the differential pressure is more sensitive then.

- Oh, one last wiring thing -- are the injector firing on the correct cylinder?

Since the engine is running well, you could consider compensating for this... I looked at the PE3Monitor software, I see that you can enable a per-cylinder fuel compensation (Engine->Engine Setup->Enables->Fuel Compensations->Cylinder). If you did that can gave the left side cylinders 20% more fuel, and right side cylinders 18% less fuel. You may just need this at idle, or perhaps the values change at higher engine speeds.