14k on o2 sensor and giving a

[robertg]

1221 over and over. it has been a year almost to the day and it only has 14k miles on it. what should i do. thanks

[Martin in Bellevue]

Check for vacuum leaks. If the intake has a big enough leak, the o2 sensor could be unable to make up the difference, keeping the burn within parameter.

[winfred]

O2's can die early, the one in my 535 crapped after a year, i broke it apart and found the ground wire was loose, so it's not set in stone that a new o2 is a good o2, just unlikely it's bad

[Elekta]

O2's can die early, the one in my 535 crapped after a year, i broke it apart and found the ground wire was loose, so it's not set in stone that a new o2 is a good o2, just unlikely it's bad

Should one replace 02 sensors as a preventative maintenance, or is this item an if it aint broke....

I did TX DOT cats last year, but my exhaust is beginning to smell different, and my mileage hasn't gotten above 20mpg in a couple of tanks. Am I going to set off a bunch of repair $$$ by just r&r the 02 sensors? As the summer gas scam kicks in early, Im thinking I might want to get back to 24 mpg hwy again.

Can just doing the 02's get me there? Car runs pretty damned well as it is, and I'm [not] looking for cans of worms to open. [edit]

[Bill R.]

100k if that could be considered routine
Bmw specifies that you change them as routine maintenance every 50k miles.. but I think that might be a little more frequent than needed..

Should one replace 02 sensors as a preventative maintenance, or is this item an if it aint broke....

I did TX DOT cats last year, but my exhaust is beginning to smell different, and my mileage hasn't gotten above 20mpg in a couple of tanks. Am I going to set off a bunch of repair $$$ by just r&r the 02 sensors? As the summer gas scam kicks in early, Im thinking I might want to get back to 24 mpg hwy again.

Can just doing the 02's get me there? Car runs pretty damned well as it is, and I'm [not] looking for cans of worms to open. [edit]

[winfred]

if you have access to a graphing multi meter you can test the sensors, set it to a 1 volt dc scale, connect the + the the black on the sensor and the - the gray (or the body of sensor if no gray) clamp the sensor in a vice and heat it with a propain torch till it's glowing, while watching the graph it should be above 700mv with in 30 seconds and should not drop below 750mv with full flame on it when up to temp, after you've been heating it for about a minute and a half or so remove the flame and watch the graph the voltage should drop to near 0 in less then a 1/4 second (10ms is usually spec) and it will be very sensitive to what you do with the flame, constant flame should be a nice even voltage .8-.9 or so. you can do this with a non graphing meter but it's hard to see the reaction time of the sensor and that's what you are looking for, a nice sharp drop off in voltage, most of the time the sensor will still give a reading but will be slow to drop off 1/4-1/2 second this is where a lot of the full throttle surging comes from, the computer is seeing delayed information and not adjusting correctly.

[robertg]

t