Using the O2 Sensor to Diagnose Lean or Rich Cylinders

I've been putting my scope on everything lately, and ran into some odd O2 sensor waveforms on two of my cars. Basically, they both have below average fuel efficiency, and I noticed they both have very erratic O2 signals. Sometimes flatline lean, sometimes pegged rich with very fast oscillations and sometimes pegged slow with very fast oscillations.

This inspired me to do a little experiment on my third car, which has good fuel control. While recording the traces for the O2 sensor and the #1 injector, I disconnected the injector and inserted a 47 Ohm resistor into the injector connector so I could still see when the pulses where occurring relative to the O2 waveform. The results where interesting. Here is a link to the video:



Basically, I saw two interesting results. One was a small trough in the O2 waveform which always occurred exactly at the time the injector was firing, and the other was a very large trough occurring approximately 45 - 50 milliseconds after the injector fired.

So, I was thinking this might somehow be useful for diagnosing say, sticky injectors, or maybe misfires on cars without misfire counters.

I'd love to hear other peoples ideas.

Here are some captures at different time bases:

Nice captures! You might be interested to know that OEMs were thinking the same thing as you. I know of it as Fuel Trim Cylinder Balance, and GM started introducing it on some models as early as 2008. I believe it's now federally mandated on all cars past 2010? Don't quote me on that. :lol:

They use the upstream O2 and complex modeling of the exhaust system to detect lean cylinders on the same bank. After it detects enough lean events on the same cylinder, it flags a code. P1171 and P1174 are the codes for GM that I know off the top of my head.

I got this response from a guy on iatn:

members.iatn.net/forums/read/msg.aspx?f=...12&m=93671&fv=0&ar=0

There are some vehicles that use the front Heated Oxygen
Sensor to monitor individual cylinders for a fuel trim
imbalance.

Here is an OBDII fault code for it: www.obd2-code.com/2017/08/p219b-air-fuel...inder-imbalance.html

"...the cylinder imbalance detection monitor matches up the
upstream O2 sensor signal with the corresponding
cylinder...Each cylinder is monitored and adjustments are
made as needed for each cylinder to correct an air/fuel
ratio imbalance. If an individual cylinder reaches a
calibrated adaption threshold 15%), the diagnostic monitor
will be considered failing..."

Here is an SAE paper about it: www.sae.org/publications/technical-papers/content/2015-01-0869/

Something to keep in mind if you do try to use your scope
captures is that the placement of the O2 sensor in the
exhaust and the length of the exhaust runners (the amount of
time a specific cylinders exhaust charge takes to go from
the exhaust port to the O2 sensor). The design of the
exhaust system can affect things (unequal length runners),
as well as VVT could affect the exhaust pulses. I suppose a
collection of known good waveforms could prove useful.