Help with Knock Sensor Code

[clemsonteg]

I have a 1970 C10 with a 5.3 from a 2003 Silverado installed. I had a friend program the computer to remove the general items like the VATS and rear O2 sensors to make it operate in my setting.

He noticed during some cycling that I was receiving some false knock and he suggested I swap the knock sensors because he had seen a corroded rear sensor cause knock in the past. I have now changed the knock sensors 3 times, all with new AC Delco units (2 sets from Amazon to start and recently a brand new set directly from my local GM dealer). I also replaced the short harness with a new GM unit the first time I replaced the sensors. I have checked all of the wiring from the sensors to the PCM, ensured that the knock sensors have the correct resistance and that they respond to tapping on the block, but I still keep receiving the P0327 and P0332 trouble codes for 'low voltage'.

Is there anything in the tune that could cause the PCM to think the voltage is too low? Possibly something I can point my friend to and have him check? I have a wireless OBD scanner and app on my phone, but with no PID's for knock I don't have a way to log the sensor output.

Is there anything else I can do besides having him remove the sensors from the tune to prevent the codes from occuring again?

[5FDP]

Nothing in the tune should cause any of that.

I'd be looking at the wiring past that little short harness, go all the way back to the PCM. Even run new wires all the way back just to make sure they aren't the problem.

It is not a good idea to turn the knock sensors off in the tune, unless you're extremely gifted and can hear the mainly non audible spark knock I wouldn't suggest it.

[clemsonteg]

I'm not all that keen on turning them off either. I've ohmed the wires from the main harness connection to the PCM and checked for shorts to ground and shorts to 12v and didn't find anything. I'll have to see if I can get my hands on some new pins to test out running new wires.

[dermotw]

You can adjust the sensor sensitivity in the tune, according to the manual, those codes are for "too much or too little activity" from the sensors...

[clemsonteg]

You can adjust the sensor sensitivity in the tune, according to the manual, those codes are for "too much or too little activity" from the sensors...

I was assuming that the low voltage meant too little activity, would that be right? I am also assuming that desensitizing the sensor wouldn't cause the PCM to trigger the trouble code as much as it simply wouldn't retard the timing until a higher voltage, right?

[dermotw]

Hi,

From the GM manual, it says that the codes are for front (327) and rear (332) sensors and the code is set when, quote; "the PCM determines that this frequency is less than or greater than the expected amount for less than 3 seconds".

So it's neither low nor high. Just "out of range". And it's a pulse train (frequency) of course, not a level.

Lots of things could make knock output change: headers, cam, engine mounts even, anything that increases mechanical noise (or indeed reduces it, but thats less likely). Plus things rubbing, rattling etc. Even over or under-tightening the sensors themselves.

The odd thing (in your case) is that apparently the PCM learns the knock sensor frequencies, and only sets the code(s) if they drift away from the values it learned - and I don't see why it would be doing that, unless you have some odd PCM setting or wrong setup somehow (and the knock wiring is dead simple).

The actual knock retard is caused by the LEVEL of these frequencies the sensors output. Probably, desensitising them won't fix this. It might fix "false knock" though. It's common to have to desensitise them in a tuned engine I think, I had to.

kr D.

[clemsonteg]

Hi,

From the GM manual, it says that the codes are for front (327) and rear (332) sensors and the code is set when, quote; "the PCM determines that this frequency is less than or greater than the expected amount for less than 3 seconds".

So it's neither low nor high. Just "out of range". And it's a pulse train (frequency) of course, not a level.

Lots of things could make knock output change: headers, cam, engine mounts even, anything that increases mechanical noise (or indeed reduces it, but thats less likely). Plus things rubbing, rattling etc. Even over or under-tightening the sensors themselves.

The odd thing (in your case) is that apparently the PCM learns the knock sensor frequencies, and only sets the code(s) if they drift away from the values it learned - and I don't see why it would be doing that, unless you have some odd PCM setting or wrong setup somehow (and the knock wiring is dead simple).

The actual knock retard is caused by the LEVEL of these frequencies the sensors output. Probably, desensitising them won't fix this. It might fix "false knock" though. It's common to have to desensitise them in a tuned engine I think, I had to.

kr D.

Thanks for the additional explanation. I even went as far as making a new harness from the pcm to the knock sensor sub harness last night, but it didn't fix the issue. I know for a fact that the wiring from the PCM to the connection on the knock sensor is good, it is all new wiring with no kinks or nicks. I torqued the sensors to 15 lb-ft and I am seeing all of the right stuff - 100k ohms from sensor to ground with the engine off and changing mv levels when tapping near the sensor.

I also think that it is odd that the PCM 'learns' what the level should be and somehow the sensors are drifting from that range. I believe all of my rattles and rubs are fixed as well because I went searching for those originally with the false knock.

We are going to take a look at the tune tomorrow. Since I know some items have changed, I may have him put it back to the bone factory tune, make sure I got rid of the false knock, and go from there.

The only other item I found that could cause the error that I haven't tried is replacing the PCM. I really want to avoid that if I can.

[dermotw]

Could be that e.g. one of the sensor type values got set wrong in the tune, they wont work properly at all then, and I don't know what kind of errors you'd get.. in the tune, there are some entries for things like sensor count, type, detection mode, they vary by year though. I'm assuming you have resonant sensors, the kind that go under the intake of LS1's....

[2xLS1]

Setting the 0327 and 0332 DTCs to no error reported does not turn off the knock sensors functioning. Many times they work fine detecting knock while failing their diagnostics.

[clemsonteg]

Setting the 0327 and 0332 DTCs to no error reported does not turn off the knock sensors functioning. Many times they work fine detecting knock while failing their diagnostics.

Would that keep it from pulling timing the way it does when the trouble code sets?

[2xLS1]

Would that keep it from pulling timing the way it does when the trouble code sets?

yes

[BuckNasty92]

Sorry to dig up an old thread guys but this is the closest I've gotten to the answer I've been seeking. My truck (2001 Silverado 5.3) has me chasing the usual KS codes. I've done everything from replacing sensors, cleaning the threads in the valley cover, torqued to 15 lb ft, all new wiring from sensor to PCM, along with checking and cleaning all other fuel and spark related sensors. So my question is if I disabled P0332 and P0327 in the tune (uncheck box in engine diagnostics) would my knock sensors still be active and pull the necessary timing commanded from the tune in the event it begins knocking?

[dermotw]

Those 2 codes are totally disabled on my car (no SES, no error reported) and the sensors carry on working just the same.

[LRI]

Sorry to dig up an old thread guys but this is the closest I've gotten to the answer I've been seeking. My truck (2001 Silverado 5.3) has me chasing the usual KS codes. I've done everything from replacing sensors, cleaning the threads in the valley cover, torqued to 15 lb ft, all new wiring from sensor to PCM, along with checking and cleaning all other fuel and spark related sensors. So my question is if I disabled P0332 and P0327 in the tune (uncheck box in engine diagnostics) would my knock sensors still be active and pull the necessary timing commanded from the tune in the event it begins knocking?

I'm thankful that you did! I've been dealing with this same issue and took the same corrective actions you did. Just scrubbed the 332/327 from dianostics. Hope to see the car respond accordingly. My issue has been the darn thing pulls timing and dumps fuel while cruising down the FW. Randomly at that. Yet the scans show things as they should (within reason, still working on a few bugs.)

Anyways, thank you.

C.

[Jermoslin]

im dealing with these two codes also, have about 1500 miles on my iron 5.7 build with stock 799 heads, TSP 218/218 .550 on 114, stock manifold and stock intake with e fans. what about adjusting the knock sensor global gains to increase sensitivity since motor use to idle around 35 kpa and now around 55 kpa. the number is 4.000 in 35 range and 1.000 in the 55 kpa. what about setting the 50-55 range with 4.000? will this allow the low voltage to set?