I took my car to the track yesterday and aside from doing terribly (can't drive worth a crap), I set the SES light at the end of a run. I scanned it this morning and found that it had set code 43, which my book tells me is for the knock sensor. I checked the wire to the knock sensor...it's fine other than being somewhat close to a couple of plug wires. I fired up FreeScan again, cleared the code, and watched the engine data. The knock count is still incrementing, but it set the code again for knock sensor failure. I also noticed that it seems to be running in "safe mode"...the fans stay on all the time and it doesn't run as smooth as it normally should. I'm guessing this is due to the code and it's running off of default tables or something.

So...how is the knock count increasing if the knock sensor isn't working? Is there something else on the same circuit I should be checking that could be messing with the knock sensor? The only other thing I found that was messed up was the plug for the wheel sensor on the passenger side. I tucked it away a long time ago since I'm not using it (no more ABS), but it must have fallen out and landed on the header. The insulation was melted away and it was grounded to the header (I'm assuming). It's toast, but I made sure that the wires weren't touching anything and it's still setting code 43. Help!

 

Check for voltage to the connector, and check for resistance of the knock sensor itself. I can't recall what the resistance should be, I think its on Shoebox's web site.

 

I checked out that guide...guess I'll have to borrow a multimeter from my neighbor and give it a shot. Back to the original question though...if the knock sensor is setting a code that tells me it's not working, then why is it still counting in FreeScan? I mean, if it's dead, it can't function...right?

If you can't tell, I really don't want to change it out. Besides, it seems kind of strange to me that it only has around 10K miles on it (replaced it during my last rebuild) and all of the sudden it just up and died on me for no good reason. I don't see a lot of posts about the sensor itself failing, so it leads me to believe there's something else going on.

 

I was just getting to post a similar post when I came across yours. I have a 95 LT-1 that just tripped code 43. Some of the research I did shows it could be the coil pack or maybe even the opti-spark. The official manual sucks at explaining what to look for. I was hoping that somebody on the message board could provide a quick answer/solution to the problem.

Mike

 

I had that code/symstoms. I checked the wire resistance, pulled the entire harness and checked the wire by eye, and checked the sensor according to the guide. Also replaced the module in the pcm with an lt4 version. Was still throwing that and running real rich and fans on.

Replaced the sensor (cheap) and it went away. I say do all the checks, but if you can't find it replace it. Mine was flaky, still worked but was on its way out.

Coolant is the only bad thing about changing it. And I needed channel locks to get it out.

 

To fix it you replaced the knock sensor, correct? Isn't there one on each cylinder bank? What do you mena about coolant?

I don't have my book in front of me so I can't remember how it is replaced. I assume it didn't take very long.


Mike

 

There is only one knock sensor on the F-Body LT1's, on the passenger side, screwed into what would normally be the block coolant drain opening.... hence the issue with "coolant". Also hard to get to the sensor, because its sort of behind the motor mount.

There appears to be something, possibly related to burst knock, that increments the "knock count" field. I've seen data logs with large increases in the knock count, and no corresponding knock retard, and others where there is significant knock retard, and no change on the knock count field.

 

Yep, it acts as a coolant drain plug. Some people say you can swap them real quick and just loose a little fluid. Didn't happen for me. I had a bucket waiting and the water comes gushing out.

 

Just checked with a buddy...I have some "interesting" wiring going on under the hood and the harness containing the knock sensor wire is running down through the header, right along side of one of my plug wires. I'm going to try running the harness back through the stock location tomorrow and see if that fixes it...he thinks the RF from the plug wire is screwing with the signal. That would explain why it's still counting, but it's probably getting some weirdo readings (which also explains why after I clear it, sometimes the SES light sets immediately, sometimes it takes it 15-20 seconds).

 

I replaced the knock sensor this afternoon and refilled the car with coolant. Unfortunately, I am still setting the code 43 ( ) after replacing the knock sensor. When under the car, I checked spark plug and other wiring to see if anything had burned up on the headers. I could not find anything. With the code 43, I had seen some sites that indicate it could be a bad coil.

I was hoping somebody might help me troubleshoot this problem.

 

I finally fixed the problem yesterday after screwing with it on and off all week...I'm not sure which of the three fixes did it, but it works now.

1. I rechecked the wheel sensor wiring I mentioned in my original post...I didn't notice that it was burned in another section and the wires were touching each other. I clipped that crap off, wrapped it, and tucked it away.

2. Since I couldn't rerun the harness going to the starter/knock sensor/O2 sensor any further away from the #4 plug wire, I split the harness open and wrapped the knock sensor wire with tin foil. If the plug wires are putting out any RF, I figured the tin foil would shield the knock sensor signal.

3. Since I had the starter out and had easy access to the knock sensor, I went ahead and swapped it out with a new one. I'm betting this was the original problem and I'm probably a dope for not replacing it in the first place.

Oh well...at least it runs now. Hopefully I don't have to mess with it again.

Vlademar, did you reset the PCM after you replaced the sensor? Either pull the PCM fuse or disconnect the battery for half an hour or so. I don't think that code will go away on its own.

 

DTC 43 is for the knock sensor/circuit - plain and simple. Has nothing to do with the coil. First, make sure the code is cleared, as noted above. 30 seconds with the "PCM BAT" fuse is all it takes.

If you put in a new sensor and the code is still there, you need to check the wiring. Key on/engine off, pull the singel wire connector off the knock sensor. Check between the connector pin and engine ground - should read 5V. With the connector still off, check from the single pin on the sensor to an engine ground. Should read approx 4,500ohm +/-1,000ohms. If that checks out, hook up the wire to the sensor and probe voltage between the pins and the ground. Should read 2.5V +/-1V.

 

Thanks for the hints and helpful tips. I will hopefully get a chance to get back under the car tonight. I am wondering if I didn't allow enough time for the battery being disconnected to clear the code. I will get a code reader tonight and clear the code. If it comes back on, I will begin to troubleshoot with the tips you gave me. Thanks alot for the help.

Mike Vlademar

 

After checking the sensor, I determined that there was a wiring issue. Looking closer, the wire harness was against the header and had burnt through. Luckily, the KS wire is the only one damaged. Repaired it and everything is working fine again.

Thanks again for the tips and help.

Mike