Knock sensor questions 2.0 (3.0?) for 1996 (1 Viewer)

[b7610]

So..... my ordeal started with a PO 115 for the Engine Coolant sensor a couple weeks back. Easy-ish fix, but in the process I smacked the knock sensor by the oil filter and basically destroyed it and the fitting, resulting in a PO325 code. Replaced both with Autozone parts and still retain the code. So dug a little deeper, red some forums, and on the advice found here bought a used OEM one from an Ebay seller with good reviews. Still no luck. Time to dive even deeper. Checked continuity from KNK1 at the ECU to the new fitting. All good. Checked continuity to ground (seen where this can be an issue) no reading so I assume not accidentally grounding. Not sure what else to check. The only other culprits I've read about is shielding. I question this as the trigger for the PO325 was breaking the sensor, and not some random issue. But maybe? On that a few questions.

• The wire in the harness is not shielded at the ECU terminal, nor at the sensor terminal, nor is the replacement shielded. Is this correct? I assume it passes though some sort of shielding "over jacket" where needed.
• If my issue is shielding, where goes it go to ground? Wondering if somehow I messed that up splicing in the new wire and harness? I don't see any shielding on the harness side wire.
• Would inadequate shielding trigger a PO325, or just send an incorrect signal to the ECU?
• My wire is red as opposed to black as stated in the 1996 wiring diagram. Common to see this?
• The PO325 only comes on while driving. Does not come on idling in my driveway. Does this mean anything?
• Seems terrible luck to have two defective/incorrect replacements but maybe?

I know there are a ton of threads on this already and I'm pretty sure I've read all of them. I'm quite stumped where to turn to next as on the surface this should have been a simple swap of known damaged parts....

Thanks - Chris

 

[ppc]

I would suggest reading the SFI section in the FSM. There is an overview of all engine sensors and logic what conditions trigger codes. Explains two-trip logic where all codes don't get set when an error occurs. It's very understandable why the code isn't occurring at idle from my understanding. With a different color wire, it might point to previous repair not using a shielded wire. I don't know if the wiring diagram shows grounding points but normally it shows a suedo shield.

 

[mudgudgeon]

I would stay suspicious of the used sensor, until you are 100% certain it is not an issue.

Follow the FSM as suggested. If nothing else stands out, pony up for a new sensor.

My memory is the knock sensors only have s single wire?
Ground via the block. Was the block clean where the sensor screws in?

 

[Malleus]

I'm thoroughly confused. Your description of the problem doesn't mention a splice but your root cause analysis (bullet points) does. Where/why was a splice installed?

Which coolant sensor are you referring to? There are two.

If it's the ECT cut switch sensor (EWD: E3 ENGINE COOLANT TEMP SENSOR), it should have two wires in its connector housing, from the engine harness, neither of which is shielded. One is R/W (red, with a white tracer) and the other is BR/B (brown, with a black tracer), according to the EWD.

IIRC, the EWD is wrong; when I refreshed the head on my '95 a couple of years ago, I also unwrapped the engine harness to check for burns in the injector circuits, and the ECT circuit to the connector housing was broken. I had to depin the connector housing and crimp new terminals onto the engine harness wires. Neither wire color matched the EWD. FWIW, the EWD wire colors are the same in every 1FZ-FE EWD.

The knock sensors, as well as the crankshaft position sensor and distributor circuit wires are shielded in the "B" and "C" connector housings, at the ECM. The wires at the source terminals are not shielded.

 

[b7610]

Mud - I tend to agree that it could still indeed be the sensor. Might swap with the rear but leery of creating the same issue a second time. Probably just buy a new OEM as suggested. Correct - single wire to the knock sensor. My question on ground was it is my understanding shielding needs to go to ground.

Malleus - the ECT was not really part of the issue. It's fixed and no code. The splice is in the single wire going to the knock sensor which needed replacing as well.

I am no expert at this, but as I look at the wiring diagram above, I interpret the shielding represented as a dashed box to protect part of the wire but not the ECU end or terminal end which are rendered outside the dashed box? In any case, there is definitely no shielding at the ECU end nor past the point where the harness is fastened to the block at the knock sensor end.