TpCpLaYa

iTrader: (10)

Ok so I did some work about a month ago removing injectors, new OMP lines, ect ect. After all was said and done I got a code 05 (knock sensor). I assumed that the wires were broken ect ect. My sensor did not look all that good either. So I ordered a low mileage sensor from a reputable seller on the forum.

Today I put the new sensor in and repaired the crispy/broken connector on the wire that runs to the ECU. Still had the code after clearing. So I then ohm'd the wire from the ECU pin up to the connector for the Knock sensor (not the connector on the knock sensor) and I read open. So I figured that I had another break somewhere. So now I have run a completely new shielded wire from the ECU to the new sensor and I am still throwing the Knock sensor code after clearing.

I am sort of stuck now as to what I should do.
https://www.rx7club.com/showthread.p...ck+sensor+wire
this guy seems to have gone through exactly what I have, except of course he never posted his solution.


I am home with the car for a few days and would love to have this resolved otherwise I go back to driving with the shitty knock sensor code ( car is a DD 6 hours away at school). This code sucks and brings on fuel cut at like 8 psi in 3rd+ gears....so I can barely drive the car.


ANY help would be greatly appreciated.


ONE WEIRD bit of info and hopefully someone can clear up the reasoning behind the results.
Test: Ohm the single wire lead of the Knock sensor to ground with sensor connected to block/ground.

Old sensor: .07 ohms or something like that ( so there is continuity)
Replacement sensor: OPEN??

So I wanna say that there has to be some kind of resistance that the ECU uses to flag the code but why does the "good condition" sensor come up as open when connecting one meter lead to the battery ground post and the other to the knock sensor wire? This is with the sensor screwed into the block.


I'm desperate

TpCpLaYa

iTrader: (10)

basically if someone can confirm this for me?


Hold the threaded end of a knock sensor to engine ground/block/UIM then attach an ohm meter lead to the knock sensor wire and the other lead to battery ground post. What resistance do you read, if any?

On one sensor I got .07 or .7 ohms and the other sensor I got OL or open.

TpCpLaYa

iTrader: (10)

omg someone please help....

TpCpLaYa

iTrader: (10)

so nobody has quick access to a knock sensor or anything that can help me out with the test I did?

How the F is it possible to throw a code when the wire completely checks out and the sensor is ok.

David0ff

iTrader: (6)

try swaping ecu`s with someone near you

i will try to get you a reading this week when i have some time to pop the hood

brent clement

I believe that the issue is not with the knock sensor itself but another wire that causes the ecu to believe it's the knock sensor. In other words another wire to the ecu is causing the knock value to be affected. Just a theory though, I'm not an electronics guru. I've had the same issue as you know, and swapping ecu's did nothing but it's not to say it wouldn't work in your case. I'm running a pfc in this particular car so I just disabled it for the time being. With the ohm meter hooked up as long as your base value is the same as the fsm when you gently tap the sensor you'll see a change in the reading when the sensor is good. Wish I could be more help, I'll definitely let you know when I find a solution.

Brent

TpCpLaYa

iTrader: (10)

thanks man that's the first bit of help I've got.

I'm pretty good with auto electronics (i'm ASE certified) but this has got me scratching my head. I'm leaning towards it being the sensor AGAIN.

The wire runs from the the sensor to the ECU. The only other related circuit is the shielded wire that gets grounded. It really has no direct affect on the knock sensor wire, but it used to block out interference.

spoolage

iTrader: (2)

This is a very long shot but did you by any chance switch the power steering connector up with the knock sensor connector? I think I remember them being similar when I dropped my engine a few months ago... again, LONG SHOT! haha

spoolage

iTrader: (2)

Here, just found a random link pertaining to what I'm talking about

https://www.rx7club.com/showthread.p...ck+sensor+code

IRPerformance

iTrader: (41)

See if you can borrow as power fc and plug it in temporarily and watch if the knock sensor is reading.

TpCpLaYa

iTrader: (10)

ok well I'm positive I'm working with the knock sensor connector/circuit. My problem exists even running a new wire directly to the ECU so that is out of the question.

RE, I'm only home this weekend and it was such short notice so no chance of swapping a power FC. I'll be driving back to school tomorrow morning.

I just want to get a resistance SPEC for the OEM knock sensor since Mazda brilliantly decided not to publish one. I haven't even gotten far enough to care if the knock sensor is working or not because I can't even key ON without throwing the Knock sensor code so it's definately failing the continuity/reistance check that the ECU performs with KEY ON.


like i mentioned before my old sensor reads .7 ohms and the new one reads open. I believe it should read somewhere in the 300k ohm range. Anyone who has a knock sensor laying around could easily double check this for me (which is what I'm asking). One ohmeter lead goes to the single wire and the other lead goes to the metal body of the sensor. There should be some continuity.....but how much is the question and the engine code wont stop setting until the ECU sees the resistance it's programmed to see

zaltec

Not to hijack this thread but i'm having 05 code too, my knock sensor is shorted so i must change it, not too easy to find in Europe.
What can happen to my engine with knock sensor out? It's going to limp mode? It's retarding timing? It's more prone to detonation? What should i do in the meantime i receive the new sensor?

Thank you.

Speed of light

iTrader: (8)

Knock sensor should have a very high resistance. My recollection is that it is in the 500k to 600k range, this is because the type of transducer used has a very high impedance. The ECU sees the sensor by the bias voltage that results when the sensor loads it. It should be a nominal 5v.

If your sensor has low or 0 ohms resistance, then it is busted. Best way to check this is to hook everything up and check the voltage from sesor lead to ground for a nominal 5v, +/- 1.5v.. If you don't have this, check the ecu output with no sensor: it should be aprox 12v; if so, look to the sensor or wiring for the problem. If no output look to the wiring or ECU.

Hope this helps.