Well well..I knew it wasn't that lean!

[user-23911]

You can't go wrong with plug cuts.......that's how you find your dud electronics.
Electronics are for lazy people and don't tell you which cylinder is leanest or which is richest.

[MadBill]

...
So that being said..If I'm at .073 on the Secondary Metering plate main hole and showing a 11.3 reading going back to .070 might get me there?
My choices are drill size .070 or all the way back to stock 1850 which is .067 and of course smaller if needed but wouldn't think so.
I have drilled and tapped one metering plate for a brass 8-32 set screw which I could drill and use for a jet. If not buy two new 134-9 metering plates.
Not going to reuse the AED jet plate since I don't like the .030 space between the jet face and the notched area on the float...looks to tight for good fuel flow.

It's useful to disconnect the secondaries and tune on just the primaries (to a reduced RPM equivalent to their max flow with all four open), then reconnect and finalize the tune by adjusting only the secondaries. I've seen as much as a 15 hp gain with the same overall AFR by leaning out over-rich secondaries and richening the secondaries.

[Caprimaniac]

Funny these old threads pop up.

This time OP's findings are exactly what I saw while logging with an older Innovate unit on my carbd car this fall. Very high AF¨s. 17,5 cruising. 12 at flooring throttle and 14 at accelerating to nrxt gear. +1-3 poinys from expected.Unfortunatrly do not have another sensor/gauge to try.

[rebelrouser]

I will probably get in trouble for this, but how about a couple WOT blasts, clean cut and pull plugs and see what they show? I believe a spark plug can be more honest then any politician and even some electronics.

The wideband, plug color, exhaust temp and MPH time slips all should agree if you are right on the money, any tool in the tool box is a help.. I agree nothing beats old fashion plug reading. One thing an O2 sensor helps a lot with is driveability part throttle, and just cruising sometimes.

[Firedome8]

My PLM does free air calibration. I recommend a chassis dyno session to sort out the full throttle mixture jet for best HP and then look at the mixture strength. Plug readings are great especially if your heat range is appropriate. My plm has served me well for 17 years now. I season the sensors in the exhaust of a stock car before I use them on the dyno. I would seek out a loading dyno with an experienced tuner it cant hurt...any feedback is appreciated.

[David Redszus]

One thing an O2 sensor helps a lot with is driveability part throttle, and just cruising sometimes.

Adding to your list. The O2 trace should be smooth and not jittery or irregular, across a stable rpm band.
Many O2 readouts have been smoothed or slowed, thereby obscuring the mixture variances.

To see the real O2 readings, the sensor should be placed in a single exhaust pipe (not in a collector or
combined section) and sampled at a high sample rate; 500 to 1 kHz.

Samples taken from a collector show the average values which is quite misleading.

A man with one hand in a bucket of boiling water and the other in a bucket of ice water is,
on the average, quite comfortable.

[Belgian1979]

500 Hz to 1 kHz is incredibly fast. Don't think any sensor can do that. Don't think it is really necessary to control mixture as you would be looking at a trend anyway.
Agree on the 1 sensor per cyl though. Amazing how much things can vary when you measure like that. And even after you set them equal at idle, doesn't mean that in some other part of the table it remains like that. You can see how one cyl affects another, how one bank affects the other, how the exhaust affects the cyl,....
Major PIA to get them installed and working. And even then, you'll have some that you cannot place in the best position which makes them prone to damage due to moisture at startup. Have them now starting with a delay and that seems to have done the trick.

[David Redszus]

500 Hz to 1 kHz is incredibly fast. Don't think any sensor can do that. Don't think it is really necessary to control mixture as you would be looking at a trend anyway.

No, it is not that fast at all. The Bosch lambda sensors easily do that.

The problem is not sensor response, but sampling speed. It is a logger problem, not a sensor problem.
A quick check would be to run the sensor output signal into a recording scope.

Mixture ratio is only one issue. Cycle to cycle variance is a major factor that affects performance.
Even mixture variance within one cycle can be seen with adequate sampling speed.

Ideally, we would like to see readings in one degree increments. That would require sampling
at 100kHz, which is beyond most all data loggers.

Once you see how irregular and sloppy the burn actually is, you won't be happy.

[Xnke]

On top of that, O2 sensor placement in the pipe is fairly critical as well-and only gets more critical as the engine RPM and airflow get higher.

Too big a pipe diameter, and reversion effects start to show in the wideband traces, too short a pipe diameter, you start to get fresh-air suckback at low RPM and the sensor goes nuts, etc.

They are excellent tools-but just like an oxy torch will remove any nut or bolt, it's not the only tool for the job.