mtrdrms

So I've got a lingering code 52 and I've been reading all day but I haven't found a procedure for testing the wiring for continuity. 93 3vze has new knock sensor and pigtail but but rich condition and code 52 remain. Help!

mtrdrms

Maybe someone can tell me if I'm doing this right? I tested the sensor pigtail:


And the engine harness:


And both tests yielded this:

scope103

That's almost what you want to start with. The two pins are signal and shield (ground), and never the twain should meet. So you want "infinite" resistance between them. You used your ohmmeter on the 200 ohm scale, so the "1 ." meant the resistance was something greater than 200 ohms. It could be infinite (actually, not possible in the real world), or it could be 201 ohms. And 201 ohms shield-to-signal could mean a damaged wire. So do the same test with the scale set to 20k (or so). (You could do it with the 200M scale, but just your fingers have less than 200M resistance, so you're likely to get some very high reading just from the moisture in the air!) If the resistance is greater than 20k ohms you're good.

Then check each to ground. On the pigtail side, each should be VERY high again. On the ecm side, one (the shield, probably brown) should be very low (less than 10ohms), and the other (signal, probably black) doesn't really matter. (You're measuring the bias resistor inside the ECM, and no one knows for sure what that should be.)

Now the part you're really interested in; get back to the ECM and measure the resistance from the harness KNK pin http://web.archive.org/web/201003261...e/102engin.pdf (while disconnected from the ECM) back to the under-the-hood ecm side connector (again, probably black, but assure yourself the wire color is the same at each end). This should be one continuous wire, so very low ohms is all you'll accept.

mtrdrms

This is exactly what I wanted to know. Thank you! Will post back results in the morning.

mtrdrms

Well I have a concern. I was not able to get a resistance reading from the ecu plug to the knk plug on the harness. These should be the same color wire, correct? Well the terminal in the diagram has a yellow wire and the wires from the knk plug are black and brown.

scope103

I always have trouble with the orientation of plugs/sockets and the diagrams.

First, what year to you have? While there isn't a lot of variation over the years, the FSM page I've provided is from '93.

Spread out the bundle and get the wire colors for each of them. To get you started, this is what I show:

1 G-B VC
2 L-Y VS
3 L-R THA
4 G-L THW
5 B KNK1
6 B OX1
7
8 Y-G VF
9 BR-B E2
10
11 Y VTA
12 Y-L IDL
13 G-W THG
14
15 V-W TE1
16 V-G TE2

mtrdrms

It's a 93 and I'm an idiot. I was reading the plug from the wrong direction. Good now.
I wasn't having any luck with the dmm so I borrowed a powerprobe. Awesome tool. I was able to get continuity in the harness between the ecm and the knk plug (black wire)as well as confirm the ground in the harness (brown). I now have to test the sensor and pigtail. They are new however so am I looking at a bad ecm? Thanks again for all the help.

scope103

For testing taillights, power window motors, relays, a Power Probe is hard to beat. You don't care about ohms and milliamps, you just want to know if it works!

But the knock sensor is more like RF circuitry. The sensor produces a tiny signal in an electrically noisy environment. The wiring has to carefully carry that signal to the ecm without messing it up. That's why I'd use the DMM; if the wire from the ECM to the pigtail connector has 70 ohms, the Power Probe says "meh, good enough!" But it's not. If the resistance from the shield to the signal wire is down to 400 ohms, I would think that meant a problem.

Member AZSpeed claims that he fixed his code 52 issue by replacing all the Capacitors in the ECM. https://www.yotatech.com/forums/f116...l#post52293355 Seems to me like a long-shot, but at least one person says that's the way to go when everything else has been checked.

Finally, don't overlook the obvious. Is everything plugged together correctly? Is a wire soaking in leaking coolant?

mtrdrms

Everything is clean and new. I bought this truck with a Code 52. The pigtail was missing when I got it. Not sure why. Is there a way to test voltage going into the ecm with the engine running to show that the sensor is working? No matter what I do with the dmm, I can seem to get a resistance reading anywhere.

scope103

Uh, maybe.

Here's a TSB for a later-year model (different engine, too) with a Knock Sensor testing procedure. http://www.parksoffroad.com/tacomamo...e/cip0325p.pdf Bad news: it requires an oscilloscope. You might be able to read the voltage with an RMS voltmeter (which, again, I doubt you have).

When you say you "can't get a resistance reading", what does that mean? You had a photo showing an over-range indication on the 200 ohm scale. If you touch the two probes together, do you get zero? (or something less than 1 ohm, if you're using the 200 ohm scale)

mtrdrms

Retested and now I got a reading. The dmm was inoperable due to operator error. When I touch the probes together I was getting about .5 ohm. When I tested the wire from the sensor plug to the ecm, I got .7ohm.


I am still unable to get a reading from the sensor to ground on either wire.

scope103

Aaugh! Please don't say "I am unable to get a reading." I'll bet you're getting "overrange," which is not only a reading, it's probably the one you're looking for. (You're looking for "infinite," but in the real world you can't measure infinity. So as I discussed above, you're looking for "overrange" on a sufficiently-high range. For automotive, the 20k range is just fine. Not good enough if you're testing MOSFETS, but you don't care about that.)

If the plug is disconnected at the ECM, you should get "infinite" to ground on both sensor wires, and the sensor itself. The wires don't connect to ground, and the sensor (a piezo element) is very, very high resistance. If the ECM plug is connected, I don't know what resistance you get looking into the ECM (and you don't care).

By the way: good job putting the leads together to measure 0.5ohms. That's about right for ordinary test leads. And a good sanity check (for stuff like "operator error"). Your 0.7ohms on the wire is good; the meter is probably only accurate to about +/- 0.5 ohms, so you're reading a good wire.

mtrdrms

Cool. Sounds like I'm on the right track. So in your opinion, how likely is this problem going to be ecm related? Additionally, Would you trust a used ecm to replace mine or should I send it out to be rebuilt?
Thank you again for all of your assistance.
Andy

mtrdrms

Well I took out the ecm (started another thread) and the case was nasty corroded but the inside circuitry looked fine. No burns, corrosion, breaks, etc.. Ugh, why can't this be easy...