Terrys87

iTrader: (1)

I do not know the 3.0 but Toyota does share a lot of the electronics operations from the 22re to the 3.0. If you can find the exact year of your truck in the bone yard, the injector harness is easy to swap on a 22re and cant see why it would not be easy on the 3.0.

The Injector Harness goes into the firewall just beside the wiper motor and it is easily unplugged from the computer and fished thru the firewall. You will need to remove your plenum and undo several other connections.

Just throwing another idea out there for you if all else fails. I am not much help on the 3.0 other then some basic issues.

scope103

That's very close to what I get.

Not quite. The knock sensor generates an AC signal, so the it COULD be capacitively coupled at the ECU. In that case, you would measure "infinite" resistance looking into the ECU, because your ohmmeter is using DC. But that isn't the case here; both of us get a handful of kilo-ohms looking into the ECU.

moparbay

Thank you, if your getting close to the same resistance that I am then I supposed my ecu has a fairly good chance of being good. That stinks for me as my next step is to buy a OEM knock sensor and tear the top/front of my motor apart to replace the sensor.


Thank you for help.

RJR

I'm pretty sure the knock sensor is just a piezo transducer. Piezo's have no DC continuity, so will measure infinite resistance with an ohm meter. They produce an AC signal, and are best modeled as an AC voltage source of about 500mV to 1V in series with a 10nF capacitor.

If you have an oscilloscope, you can watch the KS signal when you tap the block with a hammer. I don't know of any other way to really test whether or not the sensor is producing a signal.

AKHeathen

i would be interested in what the ecu reads on a meter with ohms and diode check, also what the good toyota knock sensor reads on ohm check. i do know that a reading of infinate is a bad sensor. think of it as a microphone. generally, a microphone is basically a tweeter, but instead of feeding it ac voltage, you read what it produces from the vibration. the knock sensor is similar, but different. if you have ever heard of the cheap peizo tweeters that work on "buzzing" when the signal is applied at high frequency, then you know how it wok=rks simiarly. the vibration in the peizo makes an ac signal, which is basically clarified by sending half the power through it and reading that as good zero power and the signal is the ac that comes from that bias voltage that is seen as "good zero" by the ecu. fyi, any sheilded wire over 5 watts should do great... a good source would be the 2-way (cb radio) coax. at 15 watts, cb cable is rather thin and flexable, thus easily used without problems you would find with rj6 tv cable stiffness. i have to say, your ocal dealership is garbage. try a reputable truck shop. in big rigs, it's common day practice to repair harnesses with factory or standardized connectors, as whole harnesses are only practical with warrenty repair and total loss, like engine fire. a good truck shop should have huge bins and all the specialized crimp tools to repair most factory harness plugs. i used to do it all the time. the worst being ford connectors, but i made stuff for that..... anyways, i'm rambling, so.....


in short, if you are reading nothing in resistance from the sensor, it is bad.

AKHeathen

there are a number of potential circuits reative to many different potential icus monitoring the ks. henceforth, you could get a number of different readings that could be good or bad, depending on icu design, or specific icu function componenet design. i can tell you several different way this early system can "monitor" and measure the simple circuit, but none will tell you indefinately what is used and what the reading will be without a grwat deal of reverse-engineering an ecu, which i would rahter spend the same amount of time rebuilding a couple amps.... as a rule oif thumb, though, isolating the circuit is the best way to test it..... this often means unsoldering ic's which can be a real long pain in the ass, when troubleshooting electronics. luckily for you, it's basicaly a microphone circuit. you just need to check 3 things: 1: it can pass current for the bias (basically keeps it electrically "centered" by passing half the voltage to ground.. 2: it can generate ac voltage (that is what an audio signal is) and 3: it can "feel" the vibration. and last point: that the external device is not connected to interfere with your readings.... in this case: the ecu.
you could, but it would take a long time... you would see a cascading resistance, actually, for a few min before reaching high, or near "infinate".... and that's assuming the simplist of circuits with bias.
true and not.... it is a peizo, but also has a bias shunt. a scope will SHOW you an ac voltage, and also a knock spike on the right setting, but a good meter will show you the right info and possibly a knock if you set it right, at the right time. you just need it to measure resistance for the bias, and also measure ac volts with it powered in bias to see that it is picking up a signal with the engine running.

scope103

I measured mine yesterday: "greater than" 60megohms. So for the purposes of 90s era electronics, infinite. If there is a shunt of any sort (very unlikely), it's greater than 60 megohms, and couldn't work as a "bias" shunt.

AKHeathen

shunt is in the "sensor" for bias, just to clarify, not the ecu? it can be like 500k in some...... and it's less "era" and more "specific" engine...

scope103

I'm not sure what you're getting at here. I can confidently say that there is NO shunt resistance in the knock sensor in Toyota 2nd/3rd gen pickups. I can't speak to every "knock sensor" ever made by any manufacturer.

There IS about 2.5k resistance looking into the ECU, but there isn't any (reasonably easy) way to tell if that is a bias network or just the input resistor. And it doesn't make any difference to me.

moparbay

I have ordered a new OEM knock sensor and will replace it when it shows up to try and fix the code 52 that i am getting.

I was just thinking about my truck. My truck is a '94 but there is plenty of evidence that the motor had been replaced with a pre '92 engine along with the older engine wire harness as evident by the older style connector that plugs into the KS (KS has the round sensor connecter instead of the square connecter) pigtail. The engine also had the spring tensioner instead of the hydraulic type. When I was looking at ecms, my truck definitely has a '94 ecm.

Does anyone know if it is possible that running the pre '92 (early) wiring and KS with a '94 (later) ecm could possibly cause the code 52??

E-rockSLC

Anyone have an answer to this? I believe I'm in the same situation, 94 4runner with an apparently earlier 3vze engine and the old knock sensor pigtail. Trying to come up with a way to fix this code 52 and pulling off the throttle body and inspecting the connector I discovered the round connector instead of the newer version.

AZSPEED

I struggled with this problem for a long time on my 4 runner. Replaced everything with new parts and still had the problem. I took my ECU to a friend who is an electrical guru....the guy that diagnoses electrical problems for rockets etc..... He tested the Capacitors that are on each circuit and found the one on the Knock sensor circuit was 30% out of spec, he replaced the capacitor and all the others also. He said anything that is 20 plus years old will eventually start failing. also replaced the electrolytics I believe is what he called them. To this day I have never had the problem again. We talked for awhile how most of the muscle cars and hot rods up to the 70's had very little electrical components. Today guys trying to keep vehicles running from the 80"s and 90"s are going to have to rebuild lots of electrical components. He had all the parts he needed on his work bench and said the parts he used didn't add up to 2 dollars.........

AZSPEED

Also I would check the wire going into the ECU with it running you should read some millivolts while it's running which told me that the Knock sensor was working, the signal was just not amplified enough for the ECU to process it. My hilljack terms not actual factory diagnosis procedures. I do remember it was a black wire on my ECU. But different ECU for auto and standard and years. I had a friend with a couple old ECU and they all had different plugs.

moparbay

Ok, swapped out KS for an OEM one. took about 4 hours total, 1.25 hrs tear down and 2.75 going back together. I DID NOT have to remove timing belt to get the intake off so there is definitely time to be saved by not doing it. Use a pair of strong spring clamps to keep the belt from jumping a tooth on the cam gears when you remove the idler pulley.

I have about 200 miles on it and so far no more Code 52. Knock Sensor was bad, not the wiring or ECU.

Cookms

mholme

Cookms

mholme

That's a new one to me. Maybe try to re-seat the pigtail harness and at the ECU harness end. Beyond that, I got nothing. Code 52 can only mean the ECU has not heard from the knock sensor in a specified range, so it's something between the knock sensor and the wiring all the way back to the signal pin at the ECU or the associated ground.

RobertDriver

If your meter is a blue point, any Snapon dealer can test it for accuracy and they should carry new leads on their truck. If the accuracy is off, they should be able to send it in for calibrating at no charge.

jo23727

Hey man, were you ever able to figure this out? If so could you share your solution?

I just got back from a mechanics shop who had one of these growing up, very knowledgeable. As soon as I started describing the engine cold vs engine warm situation w code 52 he started laughing and shaking his head in agreement. He said the 3vze blocks actually have a knock in them, small but enough to trip the sensor, when the engine is cold. It’s not until it warms up when it settles in that it won’t trip that code. He said you can grab brass fittings/ couplings from hardware store and insert that into the block, then the sensor into that and having that brass buffer is enough to stop the code