AFR Meter Issue

[cwestra]

I just finished setting up my Innovate LM-2 AFR meter on my 340 6-barrel car.  I welded in bungs on both pipes at a location near the manifold flange at the 12:00 position, as specified.  I calibrated the sensor at around 20.8 in air and then proceeded to check my initial A/F ratio.  On the right pipe I am getting around 13.5 at idle which seems reasonable, but on the left pipe I get over 22.  This does not even seem possible, as it is leaner than the calibration measurement in air.  Any ideas from the experts on what could be causing this or if anyone has seen similar readings on their system?  Pipes are sealed good with no air leaks. 

[edl94]

Try swapping the sensors and see if the lean side swaps with it. Might be a bad o2 sensor or wiring issue on the lean side.

[cwestra]

Try swapping the sensors and see if the lean side swaps with it. Might be a bad o2 sensor or wiring issue on the lean side.

Actually, I'm using the same sensor for each side.  Just capping the side I am not monitoring.  I can get the mixture to show richer by rapidly pushing the gas pedal several times, so I know the sensor is picking up something.

[Chryco Psycho]

possibly the pipe is not sealed & drawing in air

[cudabob496]

ya, possible leaking weld.

or leak in exhaust system up stream of weld?

no welding needed anymore. I used a new clamp on bung.

Obviously not real if all your spark plugs are same color.

[cwestra]

ya, possible leaking weld.

or leak in exhaust system up stream of weld?

no welding needed anymore. I used a new clamp on bung.

Obviously not real if all your spark plugs are same color.

I did double check that and there are no leaks at bung weld or flange gasket.  Even if it were caused by a leak, why would it register a ratio greater than the calibrated air measurement?

[cudabob496]

I did double check that and there are no leaks at bung weld or flange gasket.  Even if it were caused by a leak, why would it register a ratio greater than the calibrated air measurement?

a ratio of 22 would mean excessive air

[Strawdawg]

I did double check that and there are no leaks at bung weld or flange gasket.  Even if it were caused by a leak, why would it register a ratio greater than the calibrated air measurement?

The calibrated air measurement is a % of oxygen in the air and is not an air to fuel ratio because there is no fuel in the open air...or there is not supposed to be any

[dodj]

I believe you are supposed to get 22.4 when you calibrate in free air out of the pipe. I would power up the Innovate with no sensor attached (to kill the last calibration), then re-attach the sensor and re-calibrate.

[Strawdawg]

  I recall that mine read 20.9% which is the percentage of O2 normally found in air

It's easy to confuse this % with an A/F number which is what is displayed once the sensor has warmed up and is in the exhaust stream

[cwestra]

  I recall that mine read 20.9% which is the percentage of O2 normally found in air

It's easy to confuse this % with an A/F number which is what is displayed once the sensor has warmed up and is in the exhaust stream

I understand.  It is easy to confuse them, and that's exactly what I was doing.  In my scenario, the sensor is indicating (most likely falsely) that I have an extremely lean condition coming from the left bank of cylinders.  Not leaner than air but at a mixture ratio of around 22 lbs. air to 1 lb. fuel.  My guess is that the engine is not actually that lean but that somehow the sensor is not picking up on the correct ratio.  I'm not sure a car can run when it's that lean (what is the maximum A/F ratio that a car can run on, by the way?).  I'm not sure there is a problem with the calibration since the right bank shows a slightly rich condition at around 13.5.  The problem is occurring only on the left bank of cylinders for some reason.

[Strawdawg]

Somewhere around 18-1 is about as lean as one can run.  In the low 17's they usually begin to surge at cruise.

Initially, you said it was running leaner than the initial calibration of 21.9 but as I pointed out 21.9 is the percentage of oxygen in the air and is not a ratio.  Therefore I wondered if you knew this-no matter

Now, is this a single channel set up or is it a two channel set up with a different set of inputs for each bank and you are moving the single sensor you received to a different set of inputs?  For some reason, I was thinking that the LM2 allowed you to run two sensors at once so both banks could be monitored at once if so desired.

If it is a single input system, does it work when you put the sensor back on the original position?

[cwestra]

You are correct.  The LM-2 is a dual channel system allowing the user to monitor both pipes at the same time, if desired.  In my situation, however, I am just moving the sensor from one pipe to the other, capping the side that does not have the sensor.  I am keeping the input connection the exact same each time.  I'm aware that on this system if you plug into the second input without having a cable plugged into the first input the system will assume a narrow band condition instead of wide band.  Since I am not using this port it should not be affecting my results, as I see it.

[Strawdawg]

So we cannot blame it on a defective second input as you are just  moving the sensor and not the input.

If the car was that lean on one bank, it would not run worth a darn because it would effectively be running on four cylinders.

I take it that you put the sensor back on the side that was working, that it once again works.

It would take a massive air leak to show lean full scale on the unit.

If you bring the engine up to about 3000 rpm, does it still read wacky?  This pretty much diminishes the negative pressure pulse on the tail end of each firing cycle...or at least makes it small compared to the total flow from all four cylinders

[cwestra]

So we cannot blame it on a defective second input as you are just  moving the sensor and not the input.

If the car was that lean on one bank, it would not run worth a darn because it would effectively be running on four cylinders.

I take it that you put the sensor back on the side that was working, that it once again works.

It would take a massive air leak to show lean full scale on the unit.

If you bring the engine up to about 3000 rpm, does it still read wacky?  This pretty much diminishes the negative pressure pulse on the tail end of each firing cycle...or at least makes it small compared to the total flow from all four cylinders

I first installed the sensor on the left pipe and got the high reading.  Then I put it on the right side and got the 13.5 reading.  Then I put it back on the left pipe and confirmed that I was still getting the high (22) reading.  I have not put it on the right again but can do so.  When I rev the engine up on the left pipe the A/F does try and come down but is still high (19 - 20).  I can get it to come down to near stoichiometric by pumping the pedal while maintaining a mid range RPM.  Thanks for all your assistance on this, by the way.