Welcome to the SR20 Community Forum - The Dash.
SR20 forum logo

Thread: Knock sensor code 34.

+ Reply To Thread
Posts: 1-10 of 35
2008-07-25 18:27:29
Knock sensor code 34.
This is the code I am getting. So far I have done everything the fsm says to check for sensor function. I have 567k ohms resistance.This shows good for the knock sensor. I checked for continuity from the knock sensor connector and pin 54 (97 ser computer) and I have 6 ohms resistance. good continuity.

I will be checking another knock sensor later to test to see if that is a problem. right now I am going to make sure I have good signal at idle. I think I am supposed to have 2.5V at idle? AC current right?

Also, I rewired the sensor a while back to extend the harness, I noticed the signal wire is wrapped with another wire. I am guessing this is to reduce noise in the circuit? It must not be completely necessary to have that wire extend all the way to the sensor, because the signal wire to the ecu is unshielded as well.

Anyone have any suggestions? The knock sensor is not very old and was torqued over the spec I think. I am going to try another mounting place for it because the stock location is a pita.

any comments appreciated.
2008-07-25 18:32:17
figured I would add that the sensor was clocked at 11 oclock, not 4-5 like the FSM says. I don't think this would cause a circuit malfunction though.
2008-07-25 21:41:41
I tested to see if there was a short to ground caused by the shielding wire, there is no continuity from pin 2 on the knock sensor connector and ground, but when the computer is plugged in there is 500-600k ohms resistance and continuity exists. This only happens when the computer is plugged in. Is this normal? I checked voltage at the pin on the ECU KOEO and I get 2.5V when the knock sensor is unplugged. When the car is running I get 1.5-2.0V at the knock sensor.

I should be seeing a higher voltage correct? I should have 2.5V at idle?

I redid all the grounds on the harness just like the FSM says, the knock sensor checks out ok, and I have continuity from the sensor to the ECU. What am I missing? I have a spare sensor, should I try that to see if it works?
2008-07-25 21:44:04
This happens to alot of VE's Im not sure why. I thought it might be the harness thats going bad.
2008-07-25 21:47:06
Originally Posted by Coheed
I should be seeing a higher voltage correct? I should have 2.5V at idle?

Yes. The knock signal actually 'swings' around 2.5V, so the 2.5V is essentially '0'.

This is a very common problem btw for B14s, I've never found the root cause.
2008-07-26 06:17:16
Here is the consensus. I did a test on the ks harness at the sensor. I unplugged it with the KOEO and tested voltage. The reference was 3.31V, not 5V like it should. Also, when the sensor is plugged in, the voltage drops to 2V even.

Now the only thing I can't figure out is why the computer is not giving a 5V reference. There is no short in the wiring at all. Could it be that another connection (that's currently unplugged) is not completing part of the circuit that the computer needs to provide the right reference? I have probably 5-6 connectors currently unplugged and zip tied to the harness.

Hmmmmmm, The 2V base could be confusing the computer into thinking that the knock sensor has excessive resistance, hence is throwing the code.
2008-07-28 15:34:54
ok I figured it out. I removed the sensor and placed it on the intake manifold to make it easier to get to (i don't recommend this as it can change the signals sent to the sensor). I redid the 2 grounds that go to the sensor and bolted them back up to the manifold with clean connections. The reference voltage is still low however, but the code has not been thrown since.

So here are the steps I took to fix the problem.

Checked sensor resistance. Got .560M ohms approx.

Checked for continuity between pins 54 on the ECU and the knock sensor connector. Good.

Checked for continuity between ks connector and ground with computer unplugged. None, good.

Checked for continuity between ks connector and ground with computer plugged in.
.560M ohms. Good.

Checked reference voltage at pin 54 on ECU with KOEO, sensor unplugged.
got only 3.31V or so, not 5V.

Checked ref voltage at ecu again with sensor plugged in and got 2.0V.

The voltage is low, but it does not throw the code anymore. Just figured I would help those out and post my findings. Good luck guys!
2008-07-28 21:36:50

Tho this is ODB1.
2008-07-28 22:05:40
I've dealt with this problem on S13 KA24DE, its quite common, this link may help out:

2008-07-28 22:13:16
+ Reply To Thread
  • [Type to search users.]
  • Quick Reply
    Thread Information
    There are currently ? users browsing this thread. (? members & ? guests)

    Back to top