How do you know when o2 sensor is BAD

87T

New Member
Joined
Sep 21, 2001
I was told by a GM dealer teck. that when you see the o2 values go to 000 when you let off the gas suddenly the o2 sensor is bad soooo I replaced the sensor and within about two weeks the new sensor will do the same thing,,go to 000 when you let off the gas. Is my sensor bad ???????:)
 
Originally posted by 87T
I was told by a GM dealer teck. that when you see the o2 values go to 000 when you let off the gas suddenly the o2 sensor is bad soooo I replaced the sensor and within about two weeks the new sensor will do the same thing,,go to 000 when you let off the gas. Is my sensor bad ???????:)

That's the DFCO in action.
Decleration Fuel Cut Off.
turns the fuel off to min emissions.
there are some parameters for it to be enabled, so it doesn't always apply when you lift off the gas.
 
IMHO, a digital storage oscilloscope (DSO) or lab scope is the best method to check an O2. The DSO can measure O2 response in milliseconds. Mainly you wnat to see if the O2 has the ability to range full rich (above 850mv) when induced and full lean (less than 150mv) when induced. Also you want to know that the O2 can respond from full lean to full rich within 100 milliseconds. If you know someone that has a DSO, I can explain the procedure, but the cost of a DSO runs around 1500-2500. Kind of pricey to measure O2 response. We have one at work, so I borrow it periodically.
 
Engine running @ idle observe O2 function, should be switching between .2 - .9 quickly. You can also check your cross counts. While idling, spray an amount of carb clean thru a small vacuum port and watch the O2 go to .9 till the spray clears. If so, I would not condemn O2 sensor. Low O2 reading upon decel is normal under about 80% of the time called Decel enlean

What part of west central fl R U? I am in St.Pete.
 
The usual reason for failure around here is lead poisoning, and that seems to both slow down the response so the cross counts per frame or per second start dropping off (keep an eye on this after you put in a new sensor to get a baseline so you will know what is fast and what is slow on your car under your test conditions, either cruising at some constant speed or idling; cruising is better because you don't have to wonder if the sensor is really warmed up), and to start limiting how high the sensor will read when very rich (it may show 700 mV for example when it should be showing 800 mV so you think you are leaning out when you really aren't, especially with the heated sensors). [Now that's a run-on sentence :).] Oh, all of this is based on the narrowband stock style sensors and GM ecm. I have no idea how you would test a wideband unit without some expensive equipment, but the good news is that the widebands seem very lead tolerant.
 
Top