Weak 7MGTE

[ForcedTorque]

The electronics can be swapped between the housings.

How could I possibly have overlooked this?????????? Just put your stock housing back on, using the electronics in the Lexus. Do NOT try to clean them during the swap! Let me know if you need a stock housing.

And, while we are on the subject of me overlooking things.......What part of Mobile are you in?

 

[IRONMAN]

Theodore, I stay off of DIP a lot though. Mine is a red targa top. I'll check with the stock housing thanks for the info and the offer. I used the stock wire for about an inch or so from the ecu plug.

 

[Dirgle]

I used the stock wire for about an inch or so from the ecu plug.

Did you fully seperate the thin shield braiding from the small inner core stranded wire on those coming out of the ecu? This is something a lot of people miss the first time they do this rewire.

 

[IRONMAN]

Really? I may need to check that. Thanks. I have found and replaced my AFM and housing is stock again.Still get code 52, but I will definitely check that wire for that shielding.

 

[7mLove]

As to your code 52...i stole this from steve_mk3 thank you steve hahaha. And i quote:

"The other night I searched for information on KS. What i found is that in order to properly test the sensors themselfs you would need a oscilloscope.
The KS put out a wave signal that you cant test with a conventional VOM.

So i performed test that are similar to what you may have goon thru-
1- end to end test of the rewire itself making sure I had contiuity between the KS end and the ECU plug
2- Tested each rewire to make sure there was no continuity between the sheilding and the core of the wire.
3- I tried tapping on the knock sensors with a vom connected to the wire and I could not get any voltage- As expected but I thought I would try.
4- I tested the shield ground to the battery ground, I found that I did not have a ground.
5- Found a good ground connector under the dash to the far right side, It was a 12 mm nut.
6- connected just one side of the rewire to the ground

* success!! no more code 52.

So what I learned from this is the signal is extreamly sensitive. Properly grounding one end of rewire should work better than grounding both ends. Grounding both ends may actually induce noise.

I would thaink that you can connect both inputs to the ecu to one good knock sensor for testing purposes only. I have not verified this, it was going to be my next step after rechecking the wireing.

Keep in mind that it can only be 3 things that cause a 52
- Bad sensor
- Bad wire
- bad ecu
Most probable cause it the wireing, double check and verify the ground is good."

Hope this helps

 

[ForcedTorque]

Well was hoping it wouldn't be the knock sensors. It is a GE block so there is only one sensor and two plugs. Is there a way I can bypass it with a resistor to "trick" it into thinking it's ok.

Wondering why this information was never added to this thread? Does anybody know how to remedy this situation?

 

[Dan_Gyoba]

Buy another knock sensor. The bosses for the sensors are on the block. There are 3 of them. The GE got a sensor in the middle, the GTE got the front and rear. My '89 GE block actually had the bosses for the sensors all tapped, so I could have installed sensors in the front and rear. Not all of them are tapped, however, so you might have to do it.

I know of people who've done the NA-T thing using a GE block and found a way to only use the 1 knock sensor. I wouldn't recommend it, it was really more trouble than it's worth. I'd far rather run a tap into the block to thread in another knock sensor.

 

[IRONMAN]

I have two brand new ones. Both throw code. The wiring was checked and I rewired again for the hell of it. The shielding is grounded on one side and it is 100 percent shielded. The core isn't grounding and the shielding has 0 ohms resistance to the battery. The engine does not knock or tap loudly and anything the might tap was removed or secured. I still have a code 52. Car runs fine the code just causes it to runs rich and be a little sluggish. (both knock signal wires go to same sensor.)

 

[suprra_girl]

Did you check continuity of core to shield wires? Seems you checked ground from shielding and that's it.

 

[jetjock]

...both knock signal wires go to same sensor....

 

[Dan_Gyoba]

Buy another knock sensor.

Let me clarify that... Install another knock sensor. Buy a tap, tap the threads on the block for the front and rear knock sensor locations. Move the GE knock sensor from the middle to the front, install another one in the rear location. Wire appropriately.

 

[IRONMAN]

If the knock sensors work on oscillation and not resistance would it not be true that as long as there is not a knock (so no oscillation) there should not be a signal at all? Which would mean I shouldn't get the code. I'm not saying that's how it is. I'm just trying to make logical sense of the situation and fix the problem without pulling my engine to tap a thread.

 

[Dirgle]

A common misconception is that a code 52 means that the engine is detecting knock. This is not the case as detected knock is taken care of by the TCCS behind the scenes. A code 52 means that there is a fault in the system for detecting knock. This prevents the TCCS from actively controlling knock.

To set a code 52 the system checks the circuit to the knock sensors for an open or short. If you leave a knock sensor unplugged that would be an open. Remember, each knock sensor only has one wire going to it, the return ground path goes through the engine block. The shielding on the knock sensor wire is solely for electromagnetic shielding. It is not part of the knock sensor circuit.

 

[jetjock]

Unlike some other knock sensors there's no path to ground internal to these. No resistor, etc. Why they read open when rung out. Without going into too much detail the system verifies sensor and wiring integrity by listening for normal ignition events at particular times.

 

[3p141592654]

If the knock sensors work on oscillation and not resistance would it not be true that as long as there is not a knock (so no oscillation) there should not be a signal at all? Which would mean I shouldn't get the code.

Not correct. The system is smarter than that. It listens for normal combustion sounds during the ignition event. If it does not hear them, is sets code 52.

 

[jetjock]

Are you talking to me Jon? If so isn't that what I said?

 

[3p141592654]

Sorry JJ. I must have had a stale view of the thread because I was actually responding to IRONMAN. I added a quote to make it clearer.

 

[jetjock]

OK thanks, and thanks for clearing up my poor choice of wording about sampling combustion versus ignition. You are of course correct.