P0325 - Which knock sensor is the culprit?

[Mike534x]

Hey guys! Added another GMT360 to the family, a 2008 TrailBlazer LT. Its been well kept up and maintained. There is one issue I'm having, the CEL came on for a knock sensor. The code is P0325 Knock Sensor 1 Circuit according to Torque.

I've got paperwork from the previous owner showing one of the knock sensors was replaced by the dealership a few months ago. It doesn't say which location it was, and I can't seem to pinpoint which sensor it is. I know there is one on the Left side of the engine block by the A/C compressor, and the other somewhere behind the block by the starter. Anyone dealt with replacing this one before?

 

[TJBaker57]

somewhere behind the block by the starter.

Not "behind the block".

Knock sensor #2 is ON the block near the starter.

 

[mrrsm]

Perhaps a bit easier to visualize it without being "wrapped up" inside of an SUV... :>)

 

[Mike534x]

I appreciate the visual representation! This makes it a lot easier after looking at the service manuals (thanks @Mooseman !)

So am I correct in assuming Knock Sensor 1 is located behind the AC Compressor, while Sensor 2 is by the starter?

 

[mrrsm]

Yes... with the *popular prejudice* of having them arrayed in such a manner as to split the "Cylinder Sensing" into Three per Sensor. IIANM... The one depicted with a Red Arrow pointing in THIS second image was probably damaged when Electrical Connector portion of it Snapped Off) during the Engine's Removal at the Salvage Yard... but NTL (Nonetheless) as you can see, its placement is adjacent to and behind where the A/C Compressor's position is located.

 

[Mike534x]

Posting an update, so this thread doesn't "disappear" without a definitive answer.

The CEL turned off when I parked at work for my shift, came out and the light was off. Drove a few hours the last few days, and its remained off. Took her into the dealer, and knock sensor isn't showing as a "pending" code. There is a pending code/history for the Steering Wheel Angel Sensor, which I had already confirmed when the service stabilitrak light was coming on, and the warnings would disappear upon restarting the truck. I'm not in a hurry to replace the sensor, so that's low on the priority list. Back on track, the sensor was confirmed replaced. My only assumption, is that the dealer didn't clear the code and the PO didn't drive it around long enough for the computer to reset.

I'll be keeping tabs on this thread, in case the code comes up again in the future and if current/new/lurking members/guests run into a similar issue, they'll have up-to-date info.

To be on the safe side, since dealers are hard to trust nowadays. I purchased a new knock sensor to have on hand, should I need to replace it.

 

[mrrsm]

Very Cool. And for anyone arriving late to this matter... You probably Do NOT require a Fancy - Shmancy Tech 2 after replacing the Knock-Sensors in response to the KS Codes as any Scan Tool capable of Code Re-Sets will do the Trick. This is because the "What?" of what is actually being *sensed* here by those Knock-Sensors ...is so very Asymmetric in Nature.

Brian from Briansmobile1 YouTube Channel produced THIS excellent Video explaining the "How It Works"...about Knock-Sensors while providing excellent "Info-Graphics" showing the "What Does the PCM have to DO after *sensing* Engine Knocks?" aspects in response to these conditions.

If you've ever wondered "Why Should I be Using ACDelco Top Engine Cleaner?" in order to De-Carbonize your High Mileage Engine Combustion Chambers, Valve Seats, Compression Rings and Piston Tops... THIS is THE Video for You because it answers the Question, "Why is the Knock-Sensor Code Setting in the FIRST Place?"

 

[Mike534x]

So after about 200 miles of driving, the CEL is back on with the same code. It seems like my task this weekend is to replace the knock sensor, the one the dealer replaced was the one near the starter, so that leaves the front one by the compressor. Will report back with updates if this solves it or not.

 

[mrrsm]

...and on the outside chance that by some fluke...they have their Harness Connections ...Reversed...Changing out BOTH Knock Sensors might not be such a bad idea since just "As Carts Follow Horses..." it will be the OTHER One that fails next.

 

[Mike534x]

...and on the outside chance that by some fluke...they have their Harness Coonections ...Reversed...Changing out BOTH might not be such a bad idea since just "As Carts Follow Horses..." it will be the OTHER One that fails next.

That's an excellent idea! Honestly, I don't understand why both weren't replaced at the same time. For the labor and cost they charged the previous owner ($300), both should've been replaced as the most logical thing to do...

Since I need to replace the thermostat anyhow, that knock sensor at the very front should be relatively easy to reach down and get at vs going underneath the truck I'd imagine?

 

[Mike534x]

Managed to replace one of the knock sensors. After climbing up underneath, I took a good look at the front sensor. It does "look" like it was indeed replaced, the sensor and bolt were clean and looked like it came straight from the package. The rear one by the starter had some dirt/road grime on it, so I opted to replace that to see if the code went away. There was some minor corrosion around the face of the sensor that sits against the engine block. As far as I can tell, I don't believe the rear sensor is the factory original. The part number looks like it belongs to an NTK sensor (has a JAPAN stamp on it too). The front one by the A/C Compressor matches the AC Delco Part Number on the one I bought off Amazon. I know the torque specs were something like 29 in pounds, but the bolt came loose without even a hair of a turn from the wrench. So I began to wonder if the sensor wasn't making full contact, and the code was set off due to a false reading?

After geting that squared away, I cleared the code using Torque. I'll be driving her around a bit this week to see if it comes back on or not. On the off chance the front one is a problem, I'll either get the dealer to replace it since its under warranty, or throw the second one I have coming in a few days.

Will continue to keep this thread updated.

 

[Mike534x]

Welp 20 miles driven and the CEL is back on once again. Same code again.

Looks like I'll be going back under this weekend, to change out the "new" one that's in front. I'm hoping this is a case of just having a faulty sensor out of the box. I've got the second one on hand, its made by Delphi. The AC Delco was out of stock when I placed the order.

If this code comes back on again, where else should I start looking to track down the culprit?

While under there, I'll undo and retighten the knock sensor bolts. I don't know how I came up with 29 inch pounds in my last post lol. After rechecking the service manual again, its 18 ft lbs. I oversnugged the rear one by 3-4ft, so I'm half certain thats causing a partial false reading given how sensitive these things are. Fingers crossed the front one is either a faulty sensor/under (or over) torqued. I've seen faulty new O2 sensors out of the box, but how common would it be with a knock sensor?

 

[mrrsm]

There is one other possibility here...you might be having a Legit Knock somewhere in the Rotating Assembly. Does the Vehicle occasionally *Hesitate - Shake -N- Shudder*; whether at Idle or while *At Speed* ...?

 

[Know1else]

Very Cool. And for anyone arriving late to this matter... You probably Do NOT require a Fancy - Shmancy Tech 2 after replacing the Knock-Sensors in response to the KS Codes as any Scan Tool capable of Code Re-Sets will do the Trick. This is because the "What?" of what is actually being *sensed* here by those Knock-Sensors ...is so very Asymmetric in Nature.

Brian from Briansmobile1 YouTube Channel produced THIS excellent Video explaining the "How It Works"...about Knock-Sensors while providing excellent "Info-Graphics" showing the "What Does the PCM have to DO after *sensing* Engine Knocks?" aspects in response to these conditions.

If you've ever wondered "Why Should I be Using ACDelco Top Engine Cleaner?" in order to De-Carbonize your High Mileage Engine Combustion Chambers, Valve Seats, Compression Rings and Piston Tops... THIS is THE Video for You because it answers the Question, "Why is the Knock-Sensor Code Setting in the FIRST Place?"

Ac Delco has two top engine cleaners, 10-3007(8) and 10-3014, off the top of my head, may be wrong or updated. Which is the way to go?

I’ve been fighting carbon on cyl 4 with sea foam, and gumout…both seem helpful, acdelco is special order at Oreilly, and 3 times the price of off label…

 

[Mooseman]

You should check continuity of each wire from each sensor to the PCM. Maybe one's broken?

 

[Mike534x]

There is one other possibility here...you might be having a Legit Knock somewhere in the Rotating Assembly. Does the Vehicle occasionally *Hesitate - Shake -N- Shudder*; whether at Idle or while *At Speed* ...?

I thought so too, but this engine is deadeset quiet and smooth. No hesitations, and feels very healthy throughout the power band whenever you nail the gas. Compared to the I6 in the 02 Bravada, its a night and day difference (the Bravadas engine would occassionally hesitate from a stop, or shake/shudder on shutdown. )

You should check continuity of each wire from each sensor to the PCM. Maybe one's broken?

I was thinking that as well. If a wire is broken, shouldn't it trigger a CEL right away since its not getting a proper readout from the sensor?

 

[azswiss]

If you have access to an ELM327 adapter and the Torque Pro app the following custom PID's might give some insight, especially the sensor voltage readings:
- Knock sensor active counter: 22125E
- Knock Sensor 1 voltage: 2212F6
- Knock sensor 2 voltage: 22147A

Mode/PID, formula, min/max, header, & scale data for input into Torque shown below:

 

[Mike534x]

If you have access to an ELM327 adapter and the Torque Pro app the following custom PID's might give some insight, especially the sensor voltage readings:
- Knock sensor active counter: 22125E
- Knock Sensor 1 voltage: 2212F6
- Knock sensor 2 voltage: 22147A

Mode/PID, formula, min/max, header, & scale data for input into Torque shown below:
View attachment 110400

Thanks! I do have a ELM327 adapter, I haven't messed with Torque in awhile, so I'll need to figure out how to add in the custom PIDs. With those added in, am I just checking to see if there's any kind of voltage reading being registered by them?

 

[MD_Envoy_Driver]

There is one other possibility here...you might be having a Legit Knock somewhere in the Rotating Assembly. Does the Vehicle occasionally *Hesitate - Shake -N- Shudder*; whether at Idle or while *At Speed* ...?

My 2003 does the shudder sometimes....what's up with this? Usually when warming up in the morning at a stop. I thought it might be some settings switching over to CPU feedback and taking over. No codes, ever.

 

[mrrsm]

@MD_Envoy_Driver ... in your situation, since you have a 4WD SUV... look to Changing the Fluid in your Transfer Case first ...and see if that remedies the issue.

 

[MD_Envoy_Driver]

@MD-Envoy_Driver ... in your situation, since you have a 4WD SUV... look to Changing the Fluid in your Transfer Case first ...and see if that remedies the issue.

OK. I had the fluid sitting around for both front and back replacement and should have done this a few thousand miles ago. I'll let you know. Thanks. I never drive in 4X4 mode, but I'll take any help here.

 

[mrrsm]

While you are servicing things underneath your Envoy... Give a thoughtful Look & Check to all of the "U" Joints in your Drive Train ...jik... anything is amiss. Also... Grab the Upper and Lower ODs of the Two Front Tires and while rocking them Back-N-Forth & Side-2-Side .. with Jack Stands Supporting the SUV... Listen and Feel for anything excessively loose or *noisy* as well.

 

[azswiss]

With those added in, am I just checking to see if there's any kind of voltage reading being registered by them?

Although I have not played with these PID's myself I would start by seeing how the two sensors compare with each other by graphing the outputs. These two PID's were id'd by @TJBaker57, so he is probably the next contact point on this.

 

[Mike534x]

Sadly I can't test the PIDs in Torque, since funny enough my bluetooth adapter stopped working. Swapped it between the Envoy and Trailblazer, and its refusing to read any data.

Did replace the front sensor today. Got all the extensions on to reach inside, and went to turn the wratchet with one hand and I couldn't get the bolt loose. Took two hands, and a little bit of muscle to break it free. This thing was cranked on there. Swapped sensors, got it connected and "snug", and started the truck up and the light was off (it was on this morning when moving it). So the code turned off the moment the new sensor was plugged in on its own. Interesting...so either the sensor was damaged from the tech cranking it down or the sensor was dead out of the box.

New bluetooth adapter's been ordered so I can get a good reading from torque on whats being reported by them, but like always I'll be posting further updates once I get a good drive in this weekend to see if it comes back.

 

[TJBaker57]

Swapped it between the Envoy and Trailblazer, and its refusing to read any data.

These ELM327 will try to use the protocol it last used. So if you switch vehicles and the two vehicles use different protocols the ELM327 might not read.

Best practice in Torque pro is to set up different vehicle profiles for each of your vehicles, then in the vehicle profiles set the correct protocol for that vehicle.

The automatic protocol scan will often fail when switching from a CANBUS (ISO 15765-4) vehicle to a Class II Serial Data (SAE J1850 VPW) vehicle.

 

[Mike534x]

These ELM327 will try to use the protocol it last used. So if you switch vehicles and the two vehicles use different protocols the ELM327 might not read.

Best practice in Torque pro is to set up different vehicle profiles for each of your vehicles, then in the vehicle profiles set the correct protocol for that vehicle.

The automatic protocol scan will often fail when switching from a CANBUS (ISO 15765-4) vehicle to a Class II Serial Data (SAE J1850 VPW) vehicle.

I appreciate the tip! I did setup different vehicle profiles earlier, but its still not correctly working it seems. I've got a new one on the way, because aside from it not reading Torque reports its not connecting to it half the time so I believe it might be time for a replacement.