Water Injection and Lambda Sensor

[pyaap]

Hi Everyone,

I've got a turbocharged Volvo which has been running WI for nearly a year now. The car was running fine until recently, it was found to be running very lean (AFR 13.5 at full load open loop; I gather it should be in the low 11s). All fuel components eg. MAF, fuel pressure reg, fuel pump etc. were changed but no change at all to AFR. Injector duty cycle is around 100%. When running closed loop, AFR is 14.7.

Yesterday, it was found that both lambda sensors were giving inaccurate readings. At WOT, the sensor before the CAT was registering a very rich mixture ( 2V over what it should read) and the sensor after the CAT was registering a very lean mixture. Both were well out of normal accepted values and it has been suggested that both sensors may have been contaminated.

I am running a system 1s with 0.6mm jet and would like to know whether its possible that some of the atomised water (from windscreen reservoir, hence mixed with screenwash) may have gone straight out the exhaust valve and onto the lambda sensor, thereby causing inaccurate readings when the pump suddenly kicks in and lowers the intake charge temp.

Has anyone come across a situation like this before? I am really stumped because nearly everything to do with the fuel supply has been changed but not made any difference. The car is detonating severely (even with the WI system working) and I'm having to be very careful when driving it. Any suggestions or advice will be greatly appreciated.

[AutoX47]

Are you running an 850? What are you comparing your lambda sensor readings to, do you have a seperate 02 sensor from the factory ones? The front o2 sensor may still be correct if you have a lower line ecu where the creator maxes out the boost maps & tries to guess the fuel or have a blown intercooler hose then the ECU dumps all the fuel it can

when you said 2 volts are you talking about the lambda reading, the highest i've ever seen anyone have on the 850 is 1.250 according to the stock engine management

[pyaap]

Its an S40 2litre high pressure turbo. I've not seen any readings myself, the info is merely what I've been told has been recorded using VCT2000. The car is mechanically sound, I do think the problem is purely electronic and the wild lambda sensor readings seem to be a plausible cause. I'm just trying to ascertain whether anything could have gotten through to the sensor. The readings were taken using the stock sensors and I also think a reference point was taken using a wide band lambda sensor.

[Richard L]

Just in case pyaap doesn't post the conclusion to this thread, I have the a possible answer from him regarding the lambda problem:

It looks like we may have identified the source of the lean running problems I've been experiencing as I discussed with you last week. I had a RACELOGIC traction control system wired into the car and it involved using a multi-plug connector to connect the fuel injector wires from the ECU to the Racelogic module. It seems like this connector was introducing a resistance into the injector wires which was causing them to hold back on injector duration. This explains why the engine was being starved of fuel eventhough the injectors were running at 100% duty cycle. This also affected the lambda sensor voltages as the pre-cat sensor was registering voltages of between 1.7V and -0.5V!! Volvo specs stipulate a cycling between open and closed loop of between 0V and 0.9V. Removing the multiplug connector and soldering the injector wires directly together seems to have cured the problem. The car is now running a lot richer and the lambda readings are exactly what Volvo quote. I cannot be absolutely sure what the AFR is until I get the car on a rolling road but it is a lot better than it has been.

In any case, thanks for your advice and I will keep you posted on what I find after I get the rolling road results.

[pyaap]

I can confirm that the source of the lean running problems my car experienced was due to the RACELOGIC traction control system. I stripped the system out of the car and the car immediately was running a 12.2 AFR when it was 14.1 previously.