thtanner

iTrader: (1)

Currently getting a code 43 on WOT when it fails the Forced Knock test (makes your engine knock, tries to hear it, if it doesn't, error 43.) This is the type that has the ESC mounted on the drivers side firewall.

Currently getting super low voltage (.02v) @ the Knock Sensor connector. Should be 5v DC disconnected, 2.5v DC connected. (Edit: the voltage should NOT be 5v/2.5v as I state here - read down the thread further).

- Knock Sensor tests good (100k resistance)
- ESC wire to Knock Sensor good, but very low voltage (.02v) with ignition on
- ESC ground wire test good
- ESC 12v wire and voltage tests good
- ESC to ECM B7 wire tests good

I'm conflicted if the issue is the ECM or ESC module. Research and the manual are not clear if the 5v is sourced from the ECM or ESC module. Either part is $80-100 so I'd hate to throw money at the problem

Any input is appreciated!

Edit: Currently leaning towards ESC at this time, but any first hand experience would be nice.

RBob

iTrader: (1)

With the engine running and held at about 1000 RPM, does tapping on the block next to the sensor cause knock to be reported? If so then there isn't an issue with the ESC system. The forced knock test can fail due to incorrect distributor timing or too high of an octane fuel.

This is only true with the '90 - '92 TPI systems, not your '89.

RBob.

thtanner

iTrader: (1)

Timing is accurate as per the balancer markings anyways, fuel is 91 octane.

The tapping test produces no knocks counts @ the ECM.

Runs great aside from when the ECM performs the forced knock test. When it attempts to do the forced knock test, it is successful in making the car ping (golf ***** in a tin can come to mind,) but datalogging shows a big fat zero on detected knocks.


Good to know. Out of the 100 threads I read yesterday it wasn't quite clear. Assuming then it's the AC type, where it generates a small amount of AC current upon knock? Was hoping to test the KS individually before throwing another one in there and getting a coolant bath.

Edit: Going to get the test lamp back out and simply follow the book procedure to a T. I'll post the resolve either way.

RBob

iTrader: (1)

Both the early and later knock sensors produce an AC voltage. It is basically a microphone that sends the current engine sounds to the filter.

One thing to double check is which connector/wire is going to the knock sensor. Some have posted here that the fan switch connector/wire can be attached to the knock sensor. Which of course doesn't work.

Looks like the test in box 7 will do the trick.

RBob.

Vader

Do you have an oscilloscope handy? If so, you can monitor the KS input to the ESC while inducing signal by the above method.

thtanner

iTrader: (1)

Sadly I do not, but one is on my "to buy list" for my electronics hobbyist kit. Great idea, though.

Heading out to run through the flow chart and will advise further, at least to maybe help someone out who runs across this thread in the future.

thtanner

iTrader: (1)

Got an old-school test light (none of this new fangled DVM business I've been using,) and ran through the tests again.

#7 -> Yes

Faulty Knock Sensor (or) KS Wiring. Verified wiring to-from the KS has no shorts to ground/etc, so Knock Sensor it is.

Oreilly carries BWD brand knock sensors. I'll grab one of those, bench test for 100k ohms, and plop it in. If it gives me problems I'll order an AC Delco OE, as I'd prefer one of those anyways.

Galaxie500XL

iTrader: (1)

The BWD sensors seem ok...just remember 350's use a different sensor than 305's...different ESC modules, too...and careful not to over tighten the sensor.

Also, make sure that the knock sensor wire hasn't been accidentally swapped with the temperature sensor in the passenger side head..they have identical type connectors. IIRC, knock sensor is a white wire, that goes to terminal E on the ESC module.

thtanner

iTrader: (1)

Yes sir, got the 350 sensor. The ESC / knock sensor has worked in the past (this is a recent issue,) so going to assume it's the right ESC. I've read 15-18 ft/lb on the sensor installation.

Knock sensor wire is definitely correct as verified in test #7 above (as I tested from the connector itself.) I verified continuity separately. It also has worked in the past, and I haven't rewired the engine since. Good to know regardless as I'm sure lots of people searching may have done that on accident.

thtanner

iTrader: (1)

Installed replacement knock sensor @ 15ft/lb. After the wonderful coolant bath, it is now testing properly when tapping on the block.

I'll give it a drive-ability test after I get all this coolant off of me.

Galaxie500XL

iTrader: (1)

Knock sensors can be odd things. I once had one that would show failure only if going up a particular hill.

thtanner

iTrader: (1)

Test drive & data-logging confirms the new knock sensor fixed the issue. Hooray!

I also noticed the engine builder used thread-locker on the knock sensor threads, but everything I read says you shouldn't. Either way I did not use thread locker, and no leaks so far.

Vader

PST thread sealant should be safe once it cures.

thtanner

iTrader: (1)

I believe that came pre-applied to the threads of the replacement I got.



They used the standard plumbing variety of white sealant tape on the one I pulled out of the block.