high ecm output voltage

[b1500]

I've got a code 43 set indicating a problem with my knock sensor.

I haven't checked resistance yet, still looking for values, but I would expect to see 5ish VDC at the connector to the KS but instead have about 9 VDC.
Same 9 volts at ecm pin with connector disconnected. Battery is low but charging. All other sensors appear to be operating normally.

94 Chevy Blazer, 7427 ecm, BJLF.

Question is what is the proper DC voltage to knock sensor, and if it's not 9 what could be the problem/solution.
I look forward to your replies, thanks.

[b1500]

To add, KS resistance value is in range at 3.92 kohms, and I can't seem to find anywhere I can view this sensor directly from tunerpro.

[steveo]

you can't view the sensor directly, because the ECM's programming has no direct access to the knock sensor.

the ecm is signalled by a module or filter that deals with the knock sensor itself, all the ecm gets is a 'count' or whatever from that module or filter.

this is because the processors used in ecms don't deal with audio-level DSP, and the knock sensor is a microphone/transducer-like device. it has to be done with a dedicated circuit.

[b1500]

I don't recall seeing any module/filter in the schematics. Would I be correct to assume that it is integrated into the ECM or sensor?

From what I've read it transmits an AC signal over the DC power, if that waveform is interpreted by a non integrated module/filter should I be expecting to see a discrete input on another terminal? I wouldn't think the ECM would be fast enough for that.

[sturgillbd]

The knock filter is on the memcal module. (The "Blue Chip" under the access cover on ECM) Have you measured the reference voltage on say your MAP or TPS sensors? It should be 5 volts. I have never actually measured the voltage on the knock sensor connector. 3.9 K of resistance across the knock sensor is correct though.

[terpngator]

Well, I am FAR from knowledgeable about most of the aspects of the ECM/PCM. Steve is the one, to get your info from, concerning your tune. Done a LOT of reading, of late, on these things and from what I can tell to check the knock voltage there should be 5 volts on the wire with the key on, wire disconnected from knock sensor. There should be about 2.5 volts, on each side, with it connected. A code is set if the voltage is outside the range of 1.5-3.5V.

Yeah, the knock sensor module is located inside your PCM. See photo (http://shbox.com/1/lt4km.jpg).


I don't recall seeing any module/filter in the schematics. Would I be correct to assume that it is integrated into the ECM or sensor?

From what I've read it transmits an AC signal over the DC power, if that waveform is interpreted by a non integrated module/filter should I be expecting to see a discrete input on another terminal? I wouldn't think the ECM would be fast enough for that.

[b1500]

deleted

[b1500]

Well, it seems my only problem was me. I had the memcal set in the g1 adapter, but was offset one pin.

I would have expected to see many more problems than a code 43 and 9V at the sensor with the memcal inserted wrong, but I guess not.

Thanks for the help and responses.

[steveo]

woah no kidding; im suprised it didn't just fry it! thats an oddity for sure

[b1500]

It didn't fry the knock module, it's counting fine now, PCM is looking good. I don't know how I would check the limp circuits.

Hell, the truck started and ran like that, that kinda stuff makes a fella feel real lucky.