KNOCK sensor tested EA82T 3 plug ecu
Posted: Fri Sep 24, 2010 1:56 pm
I thought I heard pinging, knocking from under the bonnet of my L Series sedan GLTA. EA82T running dedicated LPG at standard boost. Theory for spark advance for gas is that it likes a lot of advance down low revs and no more than 28 maximum total advance. When you add boost as in turbo, things alter again, not been able to find too much info other than a gas guru telling me his Mazda running NA compression with turbo LPG simply staked any internal advance and set things at static fixed 16* BTDC. Have been hoping the knock sensor system has been taking care of the retard side of things, started to suspect it was not. So......
Started fiddling with a dizzy looking at how to reduce maximum advance, stuffing around more likely
It was so quick and easy to follow FSM to test the three plug ECU's separate knock sensor control system - not even remotely works with ECU. I think 4 plug ECU L's are integrated in the ECU.
To test the system I found the test connect lead, a black wire with white plastic insulator over a female spade terminal, just below the coil in the dizzy knock system wiring loom.
Shoved POS+ test lead of my cheapy DMM in here, the NEG- to ground.
With engine idling I was to get 1.7 Volts. I did
Stopped engine to hook up a timing light to #1 lead
Then to knock engine block near sensor at rear of RHS block with the back end of a piston remover ( I used a breaker bar)
If the system was working correctly I would see 1.7 volts increase to about 4.5V on the DMM ( I did) and see the timing decrease by about 10 degrees and slowly back up again to where it was set (factory 20* BTDC, I had it at 16 for safety)
I saw this and was happy -system OK
Today I hooked up an extension lead to test lead to run incabin DMM taped to dash, earthed its pin inside plug in cig lighter. Got 1.69V all the while driving through load and rev range.
I am assuming that if the sytem was to detect a knock the voltage should rise to 4.5V and the timing would decrease as I saw before underbonnet. As I saw no voltage increase I am assuming there was no knocking ( heard none today) so knock system not activating retard function and all is OK
Must have been other noises from under bonnet I assumed again, to be pinging, knocking, must be mistaken. So, next time I register the baby I will try wind timng back to 20, see if get same good results and may consider mounting a volt reading panel to dash for continual obs
Anyone else got experience with knock sensor testing of this model or others?
please add comment for archive sake
Started fiddling with a dizzy looking at how to reduce maximum advance, stuffing around more likely
It was so quick and easy to follow FSM to test the three plug ECU's separate knock sensor control system - not even remotely works with ECU. I think 4 plug ECU L's are integrated in the ECU.
To test the system I found the test connect lead, a black wire with white plastic insulator over a female spade terminal, just below the coil in the dizzy knock system wiring loom.
Shoved POS+ test lead of my cheapy DMM in here, the NEG- to ground.
With engine idling I was to get 1.7 Volts. I did
Stopped engine to hook up a timing light to #1 lead
Then to knock engine block near sensor at rear of RHS block with the back end of a piston remover ( I used a breaker bar)
If the system was working correctly I would see 1.7 volts increase to about 4.5V on the DMM ( I did) and see the timing decrease by about 10 degrees and slowly back up again to where it was set (factory 20* BTDC, I had it at 16 for safety)
I saw this and was happy -system OK
Today I hooked up an extension lead to test lead to run incabin DMM taped to dash, earthed its pin inside plug in cig lighter. Got 1.69V all the while driving through load and rev range.
I am assuming that if the sytem was to detect a knock the voltage should rise to 4.5V and the timing would decrease as I saw before underbonnet. As I saw no voltage increase I am assuming there was no knocking ( heard none today) so knock system not activating retard function and all is OK
Must have been other noises from under bonnet I assumed again, to be pinging, knocking, must be mistaken. So, next time I register the baby I will try wind timng back to 20, see if get same good results and may consider mounting a volt reading panel to dash for continual obs
Anyone else got experience with knock sensor testing of this model or others?
please add comment for archive sake