ZR6 knock sensor issue

[mklaaijsen]

Hi all,

Been reading up on the issue we experience, but many issues regarding the knock sensor are due to winterizing activities not replacing the knock sensor or proper connecitons.

In our case the issue is a bit different.

We own a 2006 210 SANTE with ZR6, the knock sensors (2 sensors) are located underneath the intake manifold.

We use the boat in the Netherlands, running on 95 octane unleaded fuel.

The issue we experience is when pulling a wakeboarder revving up even under 3000 rpm the tach display starts beeping and showing "knock sensor see dealer".
When we place the RINDA CodeMate on the engine it throws a code 44 (Knock Sensor 1 or 2 Inactive).
Resetting the ECM helps for about 10 mins or so, and again the ECM throws code 44.

Testing the resistance from ECM Terminals J1-1 and J1-17 return 99 kOhm, well whithin range what the MEFI-4 diagnostics manual tells us.

Last week we replaced both the knock sensors, noticed that the old ones were torques to over 50 Nm, the new ones we torques to specs (20 Nm).
No teflon tape, thread bond or anything used, cleaned out the tread and seat of the sensors in the engine block before installing.

Today we took her out for a session of wakeboarding, unfortunately after about 15 minutes the errors started showing up again.

My question is if someone else around here has experienced the same issues?

The last resort in the MEFI-4 diagnostics manual sais replacing the ECM....

Any information appreciated.

Thanks.

Marc.

[AirTool]

Did you track the water temp during the 15 minute warm up?

It seams like the problem is coincident to the transition from the warm up mode to operating mode. I don't know specifically about MEFI-4, but Knock Sensor input might not be used during warm up.

I'm 99% sure there are no cat convertors....correct?

[mklaaijsen]

AirTool,

You are correct on the cat converters, it does not have catalytic converters.

I agree on your comment that the 15 mins is very suspicious, after the warm up mode finishes, that would make sense and more easy to work to a solution.

But after two wakeboard sessions, we got annoyed by "snoozing" the alarm by pressing the up- and down button on the tach that we decided to clear codes while on the water.
After resetting the ECM codes with the CodeMate, we started the engine, and within 10 minutes or so (dont know exact as I was 3rd rider in the water) the error popped up again.

Not sure the ECM, after resetting, starts up in warmup mode, could make sense if it does in some way after reset.

Given your information above should mean that if we hook up a DiaCom to the ECM we would see one or two "lost" Knock sensors, and only triggering the error codes after warming up mode where it might ignore the knock sensor signals.

Thanks,

Marc.

[AirTool]

I need to find the number of modes in this system. I may have at home.

Funny one document I have on my computer states "When the ECM does not receive the “window” voltage for a
calibratible length of time, a DTC will be stored." Which means severe knock might actually turn the mil on. I would have never guessed that like I said in the other post.


BUT, later in the document "DTC 44 will set only if the ECM does not see any activity on
the KS signal circuits."

So I could be right again.

I think you have a problem with the wiring, ecm connector or ecm.

One test could be to pull out the two wire temp sender and put in a plug. Leave it connected but secured so it doesn't warm up and see if the mode doesn't change because of the low temp. But mode trivia doesn't really matter,....again I think your problem is wiring, ecm connector or ecm.

[mklaaijsen]

You are right, I believe DTC 43 will be stored when continuous knock is detected.

One thing I should add is that we did order and replaced a different part number, as we discovered when we pulled the old ones and compared part numbers.
Old ones were 12589867 (ACdelco 213-3521) the new ones are 10456603 (ACdelco 213-362) but both have the same Standard Motor Product Ref. KS116, so we decided to go with the new ones.

This is where your story really supports what we experience now, since the old sensors had an internal resistance of 99 kOhm, and the new ones none, so now I think from the start on the ECM is receiving readings beyond the voltage window because of the difference in resistance?

This reading can only be monitored with a software tool as MEFIScan? or DiaCom, we know a guy with DiaCom, but he is on holiday now, we need to take a close look at the operating voltage window of these sensors.

The test you propose is a good one, haven't thought of that yet, to fool the ECM that the engine is not warming up, and keep it in warm up mode to see if DTC 44 is logged.

I measured the resistance from the ECM connector terminal to a good ground source and measured the 99 kOhm, which means to my opinion that the wireharnass is good.

But to make sure i will construct a new wire harnass and replace the old (correct internal resistance) knock sensors with correct torque and see if it still happens.
If the DTC 44 gets back again we should try disconnecting the temperature sensor, if im not mistaken it is the two wire temp. sensor in the cylinderhead, we might have te old one, left over from the rebuild.

Thanks.

[AirTool]

You are right, I believe DTC 43 (AT: should be 44) will be stored when continuous knock is detected.

Re-read that comment. That would make me wrong. I'm saying I don't believe knocking throws the code...only a problem with the sensor. But I don't know and can't confirm.

I don't think live knock sensor data can be seen on the scan tool. That's from my OBD1 experience.

You can check those GM and Delco part numbers on the Delco site or maybe rockauto.com to see if the first number was superceded. That is common.

I can't comment on that resistance measurement but that seems like a lot of resistance.

temp sendor for ecm is two tire. (it is one wire for gauges and lights)

[mklaaijsen]

Oh.. now im confused..

To my understanding DTC 44 is logged when no activity is detected (which is the code we have) and DTC 43 is logged when continuous knock is detected.
I dont really understand the post above, to your understanding it might be possible that continuous knock logs DTC 44 as well?

Anyways, I had to little time to get out on the lake tonight, but went to the boat to check wire harnass as you suggested, peeled back most of the tape protecting it, all looks fine.
Also tried pushing back the connectors J1-1 and J1-17 to see if they might slide back in the housing when pushed on the ECM, both are fixed in place inside the J1 connector.

Regarding the resistance value I got that from PCM manual L510005P (Page 5-117) is a determination table to narrow down the issue, thats where i read the resistanc should be between 93 and 107 kOhm.
While the new ones dont have any, will follow your lead if it might be a superceded version, but the difference is a lot indeed.

One thing i noticed tonight is weird too, and also supports your theory, I started the engine with the knock sensors disconnected... no fault codes for the 5 mins I ran it....

We hope to get out on the lake tomorrow and test your suggestion with the temp sensor outside the engine block.

Thanks

[AirTool]

I'm looking at an older manual and maybe the codes changed. Can you post a link to the PCM doc you are looking at? so I can download. I checked their site and that one isn't there. my book says 44 is knock sensor problem and there is no 43. Earlier, I googled 43 and found ignition/starter codes so I dropped it.

Are you going to check the delco numbers or do you want me to try?

I'll also note that I don't think you can damage your engine running it this way. When the DTC comes on, the ECM is supposed to use values that are safe (e.g. retard timing). I just wouldn't load the heck out of it an lug the engine.

[mklaaijsen]

AT, thanks for the update.

I got the DTC 43 code from the CodeMate manual (http://www.pirate4x4.com/tech/billav...r%20Manual.pdf), it is the same manual shipped with our CodeMate we ordered.

When we receive the DTC 44 when wakeboarding, we "snooze" the warning by and continue wakeboarding, only we try not to revv it over 3k rpm, it still runs strong, bitt less power probably due to the fail-safe ECM tune that is activated. The "Check Engine" warning light is not lit by the way.

On the ACDelco site the 213-3521 and 213-362 both mention to be interchangeable, only the 213-3521 have an application list.

On rockauto.com the result is the same, both 213- numbers are usable for the 12589867 and 10456603.

Since on the AC-Delco site only the 213-3521 has application list attached to it, it looks to me the 3521 is indeed an superceeding partnumber for the 362.

[AirTool]

Questions:

1. When do you get 43 versus 44

2. Are you getting any message on the dash display

3. How are you "snoozing" the codes

I'll download the manual at home tonight - the office I'm sitting in blocks that site.

[mklaaijsen]

AirTool,

1. We only getting the DTC44 (hope it means we dont get continuous knock)

2. Yes, in the Tach LCD display it sais "Knock Sensor See Dealer"

3. We "snooze" the alarm by pressing the up and down buttons on the tachometer simontaneously, it dismisses the alarm for 1 minute.


We took the boat out on the lake tonight, hooking up the old temp sensor outside the engine block, only forgot to properly ground the sensor, resuting in immediate error regarding Coolant Sensor.
Next time will bring a stretch of wire to properly ground the sensor to fool the ECM it is in warming up mode as you suggested.

Thanks for all your efford so far AT.!

Thanks

[mklaaijsen]

Update:

One lead I followed today is checking wheater all ground connections are okay, specially those coming from the ECM.

The ECM has 4 ground connections:
J1-13 (ECM Ground)
J1-28 (ECM Ground)
J1-29 (ECM Ground)

J2-3 (Sensor Ground)

All wires are color coded black

I tested all pin-outs to a proper ground source, all J1 connections are okay (0 ohm) only the J2-3 is not connected to ground.

Does anyone know if this J2-3 connection should be connected to a ground source, or is this the return wire from the sensors' ground terminals?

Im planning to add a wire to this connection and connect it to a proper ground source, can this be harmfull?

Thanks

[mklaaijsen]

Today the mechanic has put the DiaCom software on the engine.

We read the total engine hours: 175 hour and engine knock time 0,1 hour.

We started a wakeboard session behind the boat with him keeping close eye on the parameters.

All time the engine knock parameter was reading OFF it never turned to ON during the run
Suddenly after 9 minutes in the recording the engine / ECM throws 2 fault codes:
* Knock sensor 1 inactive
* Knock sensor 2 inactive

All the rest of the parameters were correct, reference voltages, MAP sensor.

Only RPM and speed governor was triggering ON and OFF during the run, this might be the PerfectPass which was on during the run?

Thanks.

[AirTool]

What water temp was displayed with the faults came on? Water temp is a parameter to change control modes.


I don't know about the grounding question. I need to study that document and can't do that until next week.

[mklaaijsen]

AT, The following parameters were there when the knock sensor DTC was thrown:

Temp is 162,5 degrees fahrenheit came from 170,6 degrees (thermostat open)
3091 rpm
14,3 volt
MAP 3,88 volt and going down with rpm.