Just an FYI: I used this when my brand new sensors were causing problems and it was how I figured out they were junk.

 

^ Great video above. Good way to test for multiple issues that could plague the system. Thanks for sharing.

 

I just put everything back together. I bench tested both knock sensors on a vice for both resistance and voltage increases when tapping with a hammer. I installed both knock sensors hand tight and then torqued to 15ft/lbs, then tested both knock sensors for resistance and voltage again. I then connected the plugs, RTVed the edges, and tested the ends of the harness for resistance and voltage again. Taped down the wire and ran it in a way that it definitely isn't pinched, and I even have access to the connector itself after the manifold is installed. If it doesn't work now, then the Fbody spirits hate me. I installed an LS6 valley cover while I was doing all this, so I have that improvement at least. I did snap the brittle little plastic evap hose that goes into the driver side of the intake manifold, so I need to replace that before I test drive it tomorrow.

 

^ Seems like you have covered all the bases quite well. I expect it to work nicely this time, fingers crossed!

Is that EVAP elbow still available? I thought I remembered it being discontinued, at least by GM. Maybe someone else is making it now.

 

I can’t find it online, so my only hope is finding it off a part out car otherwise I’ll need to replace it with a normal rubber fuel hose. I think it’s 1/4” size?

 

I did my KS a while back.....used delphi......throws a code every now n then.....using GM KS next time ...and GM is NOT CHEAP

 

Part number for this is 12573362, just in case that helps in your search. Looks like the last time I ordered one was 2010, not sure when they were discontinued though. The tube itself appears to measure 5/16" outer diameter, but inner diameter is probably close to 1/4".

I imagine Delphi was the original GM supplier of these...I would have thought that would be the best option available. AC Delco isn't always the same as "original GM" these days, sometimes they outsource, but Delphi was most likely the original assembly line supplier.

 

My saga continues. No detonation at WOT which is the goal I had among all this of deep cleaning my intake manifold and intake port area. I swapped a new LS6 valley cover while I was doing all this during my most recent intake manifold removal. Still getting a knock sensor code, it’s bank 2 now, which I don’t know how is possible since I tested resistance and voltage with a multimeter even after they were torqued and connected to the harness. Now additionally i notice I am idling a little high, around 1,000rpm, and when I come to a stop while driving and push the clutch in the rpm’s float at around 1400 for a few seconds and then drop to 1000. Getting code P0507 for high idle. I did break the evap elbow on my last install and am using 1/4” fuel tube on that connection to my TB now, could that be the cause or is something else likely causing the high idle and the Rpm float? Searches suggest it’s a vacuum leak somewhere, but all hoses look fine

edit: fairly certain it’s somewhere in the PCV:
-if I isolate the PCV system by removing the clean air line from my TB and cap the TB barb and remove the hose going from my catch can into my intake manifold and cap the intake barb, idles normally and doesn’t hang
-if I attach the clean air line from my TB to my passenger side valve cover barb and remove my catch can by having the valley cover connected directly to the intake manifold barb with a fresh line of hose, still idles high and hangs
-if I attach the clean air line from my TB to my passenger side valve cover barb and have the hose from the valley cover or the catch can outlet going out to the atmosphere and cap my intake manifold port, idles normally and doesn't hang

I'm fairly certain the LS6 valley cover I got should have a built in PCV since it came off a 2004 C5, however if that is incorrect and there is in fact no PCV itself, could that be the issue? If so I can test by connecting my old PCV in the line going from the valley cover outlet into the intake manifold and see if that makes any difference.

edit 2: saw another post searching that says running without a PCV valve itself creates an equivalent of a large vacuum leak, so this supports the theory that my Ls6 valve cover isn’t the model with the built in PCV. I’m going to try installing my old PCV valve into the loop later and hope that fixes it. Still have no idea why I’m getting knock sensor codes though.

Video here:

 

I think your idle issue is definitely a vacuum leak, but I have no experience with the LS6 valley cover as I've never installed one. I would suggest that there could also be an intake gasket leak since the intake has been on and off so many times, but it seems like you have found the PCV system to be suspect with your tests above. Hopefully someone with experience using the LS6 valley cover can offer more advice in that regard.

As for the KS code now moving to Bank 2, I really don't know what to say. It's hard to imagine that you have yet another bad sensor after performing all those tests, but the fact that the code has moved from Bank 1 to Bank 2 would suggest that the problem is not exactly the same as before. I wonder if the sensor can still be faulty even if those bench tests showed it to be good?

 

Just a quick thought...do you still have the original sensors that you removed? I seem to recall that you only replaced them as a "might-as-well" while the intake was off (not because you were getting any codes). So if all else fails, what about just going back to the originals? They might have countless years of life left in them. I'm just not too sure of the durability and quality of these new sensors.

 

Idle issue fixed by adding a PCV valve into the loop, I’m guessing the version of LS6 plate I bought is the earlier one that didn’t have the valve built in so I needed to add it back. I was able to make it look fairly OEM and clean, I basically put some 3/8 hose on the outlet of the PCV valve and then used a 3/8 brass coupler on the inlet to join it with 3/8 hose to get female ends on both side as required, and used clamps for good measure on every fitting.

 

Sort of, they don't look atrocious but they don't look great either, and the plastic connector was so brittle that it crunbled when I removed the original harness. If I do take this mf'ing manifold off again I think I may try AC Delco knock sensors based on other threads where people said it was the only thing their car would accept. Surprising since Delphi is not junk, in fact prior to 2017 they were a division of GM. Photo of my old sensors:

 

Agreed. Delphi was the OE assembly line supplier for many parts on these cars (from the fuel pump to many of the electronics), so I'm surprised it wouldn't be equal to or better than what is sold over the GM parts counter today. But, the fact remains that you're still seeing a code, so if the current AC Delco ones are the answer then I guess it is what it is. Just sucks that you have to keep pulling this thing apart.

 

Seems like I’m not alone, a half dozen people with the same issue in this thread and no conclusive solution.

https://ls1tech.com/forums/pcm-diagn...voltage-5.html

The best theory comes from post #81 where someone bench tested AC Delco vs Delphi and said AC Delco put out .520mv at 2000rpm while Delphi throws .430 and the computer wants .5 or above, so if that is correct I’d need to either replace the sensors yet again with AC Delco or see about trying to get tuning done to change the range of mine to only throw a code below .4 instead of .5

 

I wonder if there is a way to make the old sensors workable even with the broken plastic connector? In reading that thread, it seems like some folks have had trouble even with the AC Delco brand, so I'm not sure if that would be a certain fix. But the old ones were working fine, so if they are able to be made usable then it might be the best bet.

Seems like having knock sensors is more trouble than it's worth at this point. My old SBC and SBO have survived for 50+ years without them, I think I'd rather just delete them from the tune then change them countless times trying to find a set that will work. Swapping them wouldn't be a big deal, if not for having to pull the intake every time. Maybe that knock sensor relocation kit is the best answer.

 

I wonder if there's a way to get Gen IV knock sensors to work with a Gen III computer. I haven't seen anyone having any issues with those, and they're probably still available new from GM.

 

I looked for original GM knock sensors and wasn’t able to find any readily. I did find a thread where someone kept getting codes with AC Delco sensors and ended up fixing the issue with Duralast knock sensors, so I am giving that a try. For what it’s worth, the Duralast sensors say made in the USA. Will take it for a drive tomorrow and see how it goes. Both Duralast sensors show AC voltage nicely when I connect them to a multimeter and they seemed to torque a lot more normally. Basically hand tightened them and got clicks on the torque wrench with minimal
extra turning of the wrench.

thread here:
https://ls1tech.com/forums/general-m...d-quality.html

 

I see that I posted in that thread last year. Sorry I didn't remember about it, I should have linked it to you much earlier. I hope this is the answer, but it might just be that different batches are better or worse from different suppliers at different times, because it seems like there are some sketchy reviews for just about every brand at this point. But if you find that this is still the "fix" one year after that thread above was posted, then perhaps that brand has remained consistent at least for now.

 

I don't want to jynx it, but just went for a 30 or so mile drive and no codes so far. Hopefully the car likes these Duralast sensors and that's the end of it.

 

Took it for more driving today, and so far so good. Only code I have is the 4030 for catalyst efficiency which is a common issue on this cars and unrelated. Hopefully the knock sensor bank codes stay away, if so then the fix was going to Duralast sensors.