Engine light always on
Hi i am getting code 13, i know this is the oxygen sensor but this is what is happing.About two weeks ago i changed the inlet manifold gasket and thermostat, before i did this, i would get the engine light coming on and off when i stopped at traffic lights, when i drove off it would turn off, as i was changing the thermostat i noticed it was stuck open,so put a new one in and now as soon as the car warms up the light comes on and stays on till i turn the car off . Now for some reason i think because the car didnt get up to proper temp with the bad thermostat it wouldn't make code 13 come on unless i sat at the lights and the car would warm up.I have put in a new oxygen sensor, a new temp sensor and reset the computer and have the same problem.i took the plugs out and its running rich. i know for some reason the computer thinks the car is running cold, i just cant fine it, if anyone has had the same thing happen can you point me in the right direction Thank.
which temp sensor did you replace? there are two side-by-side. The one on the left, or the one on the right(standing at the front bumper looking at the engine)?
The one on the right the one with two pins
well that's the sensor that the ECU uses.. how bizarre. have you checked the wiring for that connector between the engine and the ecu to make sure there's no short/increase in resistance as the engine heats up ? As in, check it before you start, then get the car hot, check again, compare values?
i thought for shore the temp sensor would fix it,its doing my head in ,how do you test resistance ?
Spend $15 (not $10 - the meter must have an internal impedance of > 10M ohms) on a multimeter and read up on electrical circuitsOriginally Posted by peter p
.
A code 13 is set because the ECM is "seeing" a voltage on the O2 sensor wire that doesn't oscillate from a low - ~100 mV - to high - ~800mV - but stays around 450mV.
To test the O2 sensor output, find the ECM under the kick panel in LHS of the passenger footwell. Disconnect the bigger of the two connectors (connector 'C-D') to it and remove the anti-back out comb from the back to allow access to the wiring terminals.
The terminal assignations are moulded into the plastic connector block.
With the connector unplugged, use a dressmaker's pin to access terminal D7, carefully to avoid bending the terminal contacts. D7 will have a purple wire to it.
Unplug the connector to the O2 sensor in the engine bay. Connect the multimeter, set on "ohms", lowest resistance setting, between the dressmaker's pin and the other end of the same purple wire at the O2 sensor. You can use lengths of electrical wire to cover the distance but ensure the connections are good - test by checking the resistance of just the test wiring before checking anything on the car. The ohms reading should be "0.00". If it is not, the wiring is broken between the O2 sensor and ECM.
Perform a similar test between D6 and the earth at the battery.
Plug everything back together but leave out the "comb". Use two dress makers pins to access pins/terminals D7 and D6 at the ECM from the back of the connector. Set the multimeter to measure 0-2 Volts and connect across the dressmaker's pins. You can use short lengths of wire wrapped around the pins and the meter test probes to make this easier. Test the test wire resistance first, as before.
Start the car. From cold, the voltage should read 450mV. This is a voltage generated by the ECM before the O2 sensor starts to work. As the sensor warms up and begins to generate voltage the reading should start to jump from above to below 450mV and back.
You will have to raise the engine rpm a little (just above idle) to get the oscillation as the ECM runs slightly rich at idle (slightly higher than 450mV) rather than switching rich-lean continuously.
No oscillation = dead sensor or dead ECM.
Coolant or some silicon type sealants will kill a new O2 sensor.
Last edited by Cheap6; 22-04-2010 at 09:40 AM.
Thanks for the info got a multimeter today, will give it a go tomorrow
Did the first test today seemed fine, didn't really know how to do the second, will the car start with the comb out ?, also i disconnected the 02 sensor today and the engine light came on at the same time,so looks like its running without a signal from the 02 and if i start the car and let it idle up to temp the engine light wont come on till i drive off
The comb is only there to stop the terminals in the connector from backing off the pins in the ECM while/when the connector is plugged in. Usually that's not an issue.
The second test is just measuring the voltage from the O2 sensor at the ECM end of the wires. It's conceptually no different from measuring the voltage across a car battery, only the terminals are smaller and closer together (if it helps to think of it in that way). Also, in this case the voltage from the "battery" (O2 sensor) changes with air:fuel ratio (actually the difference between the oxygen partial pressure in the exhaust and that in the atmosphere).
Did you check the earth wire (to/from D6) also?
The O2 sensor codes won't set until the car is driven and a particular set of circumstances is met; basically the ECM is sure that the O2 sensor, if it is working correctly, should be generating a voltage. The sensor has to be hot enough to generate the voltage too.
Except when initially coming to rest from moving, the O2 sensor is ignored at idle (= no codes set too), which is why I suggested bumping the engine speed up a little when checking the sensor voltage. Driving the car will achieve the same end.
A code 13 indicates a constant mid level O2 sensor voltage. A consistent high or low voltage - corresponding to rich or lean - or an erroneous O2 sensor signal (eg. a high resistance somewhere) into the ECM would set a code 45 or 44.
If there's only a code 13, the O2 sensor circuit must be broken somewhere or the O2 sensor is not generating a voltage at all (which would be unusual). If the circuit outside/up to the ECM is OK i.e. there is an O2 sensor voltage across D7 and D6 that alters with air:fuel ratio, no break in the wiring to and from the ECM, the break must be inside or at the ECM.
When/if you get a voltage reading across D7 and D6, you might also try altering the air:fuel mixture and seeing if the O2 sensor voltage tracks the changes. eg. with the engine running, pull an injector plug off - voltage should drop (lean), or squirt some carb cleaner into the intake duct or pull the manifold pressure hose off the fuel pressure regulator (plug it with a finger to avoid a minor manifold vacuum leak) - voltage should spike higher (rich).
It might make it easier to tape the dressmaker's pins or short lengths of wire to the meter probes.
Ok finally worked out what you are saying,did the test from cold start and got a reading of .650mv let it warm up went for a drive,no change still .650mv.pulled an injector plug off still 650mv,did the d6 to earth at battery, didn't get 0.00 but i had to connect some wire together so i dont know if this is a real reading, d7 was 0.00. i can get hold of a computer, but it from a vn manual will it work ? Thanks for all your help
As long as the D6 to battery earth is close to 0 ohms, that sounds OK. The circuit goes through a few connections to get there. You might also try the same check from the exhaust manifold to the battery terminal and D6.
Did you put any thing, like sealant, on the O2 sensor threads when you changed it maybe? The sensor earths to the exhaust manifold through the threads, then back to the ECM (D6) via the engine.
As another check of the O2 sensor, you might try measuring the voltage (between the sensor and an earth point) directly at the sensor pigtail, sensor disconnected, and also pulling the injector plug/adding carb clean to see its response. The sensor must be hot though.
If you want to try the ECM, if you can, its better to swap the suspect part into a good running car and see if the fault follows as that minimises the possibility of damaging the good part.
Do you still get the 650mV across D6 and D7 with the O2 sensor unplugged (at the sensor)? That would suggest the ECM is OK but the sensor voltage isn't reaching the ECM.
It's worth looking at the ECM pins and the ECM connector in the car for bent pins, damaged or corroded terminal connectors too.
All ECMs from VL Walkingshaw (not going to see too many of those
The MemCal is easily swapped; it's the blue (and brown) plastic module inside the ECM. Undo the two screws on the ECM case sides furthest from the terminal connector and you'll be able to work it out from there.
Well after all the testing,I went to the wreckers today and got a second hand 02 sensor and the problem is fixed, the one part that was new was the cause all along Thanks to Cheap6 for all the help
Posting Permissions
LinkBack URL
About LinkBacks
Reply With Quote