massn8v

I'm driving a 1995 toyota 4runner 4wd with the 3.0 and automatic transmission. Code 52 has been driving me nuts. I pulled the intake and installed a new sensor with pigtail, and I was still throwing the code. Ran new wires all the way back to the ECM...still had the code. Relocated the sensor to the lift hook on the engine, and it worked for a while. Now it's back.

Here's what makes it different from every other post I've been able to find. My CEL only comes on when I'm going through neighborhoods, coasting for a while, and then hit the gas again. Other than that it stays off. When it comes on, I turn the truck off and the CEL goes off. Next time I'm coasting and then hit the gas, it comes back on.

I've tried the sensor on both engine hooks, tried a Toyota sensor, tried an Advance Auto sensor, throwaway sensor from my shop, ran coax cable to the ecm....any other ideas???

Co_94_PU

iTrader: (-1)

If you doubt it's accuracy. Check your shield grounds, and maybe try some different routing. Once you've eliminated wiring and sensor issues, the only thing left is the computer.

Also make sure the harness plugs are tight and clean, I mean the metals bits I don't expect you left the plug half hanging out.

If you can get your hands on a portable O-scope you can eliminate the ECU as a source for the problem.

Does it do it consistently, in a certain place/area. eg, Road texture canceling out the piezo effect?

massn8v

I appreciate the help. I think I've finally figured it out. This has only been my truck for about a month, and it has ran like an absolute DOG since I got it. I thought it was the 52 code that was killing the power. In the time I've been working on it, I've probably checked the timing about five times. Every time it was 10 BTDC with the diagnostic terminal jumped.

Today I finally ran out of guesses, so I really thought about how the knock sensor works. My understanding of it is the pinging and knocking is what makes the sensor create the voltage that it sends back to the computer. I also tried to think about what might cause the computer to see a fault if everything was working correct in that circuit. I thought maybe my motor wasn't producing enough pinging to create that voltage.

So, on a hunch, I went out and advanced the timing as much as the distributor would allow without pulling it. I can definitely hear some pinging now, but the CEL light hasn't come on for about twenty miles. I used to be able to turn it on on demand by doing what I described in my opening post. Now I can't turn it on no matter how I drive it. I feathered the throttle, floored it, ran it hot, ran it when it was cold...pretty much every scenario I could think of. Like I said it's only been about twenty miles, but so far it hasn't turned the CEL on or thrown the code in the computer. Also, it has tons more power.

Right now my timing isn't even in the ballpark of correct. I literally can't even see the timing marks on the pulley when I hook up the timing light. Depending on how busy it is at work on Monday, I'm going to pull the distributor and timing belt cover to check everything out. My only guess is that someone before me put the timing belt on incorrectly or stabbed the distributor wrong. I've been chasing every problem I could think of and the timing was so far off, I was going in circles

And, like I said, I'm working on a hunch here. I'll let you guys know what I find when I tear into it.

j2the-e

Any updates here? Dealing with a code 52 myself. I replaced a long block(and knock sensor and pigtail). I know the distributor is in correctly, but when I set the timing it almost maxed out retarded. Truck is a dog. Then it set the code 52. Haven't tried advancing the timing yet.

massn8v

If you're looking for advice on this, I would say redo that wire all the way to the computer and leave the sensor in the block where it's supposed to go. Then do it again until you're 100% positive that you have good solders and your grounds are good. Also, use a TOYOTA sensor. In my particular case, I'm pretty sure the timing belt isn't lined up correctly and that was giving me half my problem. If you think your truck might be doing the same, advance the timing a little. When you say it almost maxed out retarded, what did you mean? The way I find my timing is by jumping the diagnostic terminal and then hooking up my timing light. Stock timing should be perfect at 10 degrees before top dead center. When I did that to my truck, it ran like a dog and just seemed like the timing was off. There's really no way I can describe why I thought that, it's just an experience thing. So I ran the timing till I thought it sounded good, checked it, and I was way past stock timing. That got rid of the code 52 for a while. But it still came back on at random times every few days.

So I tore into it again, and I think I chased it down to bad soldering when I ran a new wire to the computer for the knock sensor. The only shielded wire I could come up with was the solid core coax cable. Since then I found some strand core shielded wire at a guitar shop. The guitar cable they use to plug into amps is perfect for this. I put the new wire in and I haven't seen the light at all.

On a side note, when I went to have my truck inspected, it failed because of HC count on the low side. When I set my timing back to stock, it passed with flying colors. This further reinforces my thinking that my timing belt is off. And, again, I just haven't had the time to get the timing cover off.

I'm still not sure if the timing issue affected the code 52 or not. That is purely a guess, and it might have just been coincidence that it helped. Book theory says the timing shouldn't affect it. Book theory also says that there are only four options to check when you get code 52: Knock sensor, pigtail, wiring to computer, and computer itself. If you have replaced the sensor and pigtail, I would bet it's the wire to the computer. When I fixed my wire...actually fixed it beyond doubt...the code went away and hasn't been back for 300 miles.

j2the-e

By timing is maxed, I mean the distributor is maxed. Timing is set to 10 degrees btdc with the check connector jumped. I'll double check the distributor. And I'll check for a ks signal at the ecu. If those check out I'll get some shielded wire. The harness is old and did get moved a bit when doing the engine. Pulling that intake back off is last option!
Thanks!

BamaYota1

I had a code 52 problem a while back. I used GTO wire between the knock sensor and computer. It is the same wire used on neon transformers which produce 15,000 volts at 60 milliamperes. It has a thick silicone sleeve with a very pure copper wire. If you are near a city, any sign repair shop would give you 10 feet or so to rerun the wire. If not, I could send you some for the price of postage.

massn8v

You bet, man. This ran me ragged for a while, so anything I can do to help, I'll do. I've got one more tip that I just thought of. When you run the shielded wire, run the shielding part of the splice longer than the core. Do the same on the side that goes to the computer. This will make it so you can splice/ground the shielding and barely expose the core wire. This sensor sends out millivolts, so any interference can theoretically mess with how the computer reads it. Try to have as little of that core wire exposed as possible. Maybe even wrap any that is exposed with some tin foil or something if you still have problems.

Let me know how it goes or if you have any questions.

j2the-e

Thank you both. I'll update when I see the truck next. It belongs to a friend and he's fixing the front end right now so I might not see it again for a week or so.

j2the-e

All wiring checked out so I went ahead and pulled the intake. The knock sensor was from Napa. Junk! I put a factory sensor in and all is good to go. What a waste of time! Use a factory sensor and be done with it!