Ls3/480 2002 z28

[AutoWiz]

We have done a couple of these ls3 480 swaps. The last was on a 2007 corvette. This one is on an old Camaro. One thing that I like to do when doing these swaps is I like to keep the knock sensors correct to the controller. I know there are wiring harness adapters to allow for using the new sensors with the older controller, but I believe it is more correct to remove the newer style knock sensors from the sides of the new motor..



And drill and tap the block to recieve the gen 3 knock sensors that were removed from the ls1 valley..



The LS3 has different connectors on the injectors and map sensor, we moved the knock sensors from the valley to the sides of the block, and we are deleting all the secondary air stuff so there are some changes that need to be made to the harness..



Here the motor is fully dressed with new ARH longtubes and ready to go back into the Camaro..




And of course..



Better than any watermark. And here is where we stopped for the holiday weekend..



I should have this completely wrapped up by lunchtime or a little after on Tuesday and hopefully by the time I leave work on Wednesday I will have that .hpt file to post up here. I will save a log of its final pull on the dyno.

[Pekka_Perkeles]

We have done a couple of these ls3 480 swaps. This one is on an old Camaro. One thing that I like to do when doing these swaps is I like to keep the knock sensors correct to the controller. I know there are wiring harness adapters to allow for using the new sensors with the older controller, but I believe it is more correct to remove the newer style knock sensors from the sides of the new motor..

And drill and tap the block to recieve the gen 3 knock sensors that were removed from the ls1 valley..

Very good indeed!

Since the knock sensor are now in a different location, have you validated their response to knock (with Gen III PCM) and if so, what was the end-result?

Just recently spend some time to do the exact thing and it was somewhat surprising. :-)

[AutoWiz]

We make every attempt to leave all knock sensor and sensor settings alone. And re-use the original knock sensors. It is a testament to effort that we can say this and we are also a magnuson dealer and installer. But it is true. And the kr sensors haven't led me astray yet. I never slack off my tables. The new location is just as close to the crank as the valley its just side by side rather than fore and aft. This method that we have settled on is far better we believe than using the adapter harness and newer sensors. At least there is a known good sensitivity calibration, ya know? Anyways, if you have a way or procedure to tune a knock sensor that is not just removing it's sensitivity then please speak up.

[Ben C]

There are different thoughts on the matter however if you look into knock sensor tuning (resonant frequencies of detonation) it is the bore diameter that has an impact.
Something to think about

[AutoWiz]

I wish there was a clear cut and dry path to exact knock sensor calibration. That I knew about. I mean that sensor is so damn important. Especially when you are trying to tune for power and pull after pull on the dyno you go a touch leaner you add 2 more degrees. You really REALLY want good data from that sensor as you try to bring your tune to the edge. And everybody everywhere just seems to want to disable or slack off it's ability to do its job.

[dermotw]

For what its worth: I kept the original sensors (resonants) when I went to LS376 block etc. in my 99 TA, but I re-positioned them closer to the block front (at side of course) as there were also holes there, and in my case that was easier. I have not really noticed any difference in their performance (surprisingly) but then again, I do not have the technology to listen to the block ringing, lol. I could tell with a delay 'scope (oscilloscope) since thats my real job, but I don't have the time really...

I have edited my KR level table, but NOT generally to reduce its sensitivity, in fact I wound it up in quite a few places, (compared to a '99 TA table remember) the only place I changed sens significantly downwards is idle, where I personally don't really care if it detects anything or not!

[Jason B]

I wish there was a clear cut and dry path to exact knock sensor calibration. That I knew about. I mean that sensor is so damn important. Especially when you are trying to tune for power and pull after pull on the dyno you go a touch leaner you add 2 more degrees. You really REALLY want good data from that sensor as you try to bring your tune to the edge. And everybody everywhere just seems to want to disable or slack off it's ability to do its job.

What about the old carb and distributor way......run a lower octane and tune til you get slight knock. Then raise the octane and make sure theres no KR.
FWIW I have a gen IV block with the knock sensors set up this way also.

[AutoWiz]

Well here's the problem with that. 2 and 3 degrees of kr is something that we see on the datastream. It is something that we feel and hear only because of the timing pulled when it happens. The fact is by the time you are actually hearing pinging with your ear-crometer there is actually a great deal of detonation going on. These sensors are meant to be precise beyond what we are capable of with our limited senses. It needs to be calibrated with or against another known good calibration. For a guess or a close to or we think it is better, I would much rather just keep the factory settings just as they are.

There is another method of detecting detonation that came out on a few vehicles. But I think Ion Sensing/ Ionic Ignition as a standalone device like a wideband o2 sensor would make calibrating a knock sensor a simple thing. Ion sensing uses the spark plug on the exhaust stroke to measure the levels of ionization in the combustion chamber. This can provide much more specific knock detection on a per cylinder per combustion event basis.

[Pekka_Perkeles]

Anyways, if you have a way or procedure to tune a knock sensor that is not just removing it's sensitivity then please speak up.

Sure,

Just don't calibrate them at WOT. Here's my post from another forum:

Well, you do not need to do any WOT testing to find out if the sensors are working as intended.

Now, make high and low octane tables identical for now. Then add several degrees of advance somewhere between 1200...2000 rpm and maybe around 0.30....0.60 grams of Cylinder Airmass. Then purchase a decent knock headphones/headsets or make those by yourself.

Go to a test drive and use suitable gear (maybe fourth if you have manual) and try to target those cells where you raised ignition advance. Again, no WOT needed.

With modest pedal (and too much ignition advance in your tune) you should both hear knock and see it in the logs.

If you hear it, but don't see it, make your knock sensor settings more aggressive.

Happy hunting.

https://www.corvetteforum.com/forums...post1593953139

[AutoWiz]

When I am talking about knock sensor calibrations I am thinking more about the resolution that is a percentage,... that is a tenth of one degree of kr. That is the resolution of these sensors from the manufacturer and I still have yet to hear a single good way to tune this resolution. Forcing detonation on your car is just beating your engine to death for what? Does that some how correlate to a percentage of or degrees of kr? If not then what is the point of forcing your car to knock? Besides just hating your engine?

Forcing the engine to knock and then seeing a change on the kr sensor is more of a functionality test than it is a calibration procedure. And a dangerous one at that. functionality can be verified with lightly tapping the side of the block with a hammer. And there needs to be no explosions at bad rod angles.

[AutoWiz]

The z28 is all back together. Here is the engine bay..







I got the tune good enough to drive the car off the rack and onto the dyno roller. And wrapped up the day getting the car strapped down..

[Ben C]

Inducing knock as suggested isn't going to damage an engine. It is commonly suggested as the only real way to tune aftermarket knock detection systems. And useful when using knock listening devices so you are aware of the exact sound of knock.

[AutoWiz]

I am sorry that it is commonly suggested but it is very wrong. Any level of detonation at any rpm under any load hurts your engine. This is evident from looking at a spark plug after you force your detonation. And it has nothing to do with tuning because you don't know exactly how much knock you are inducing to compare to the noise level you are hearing. So you don't get the accuracy. Just some level of noise upon some level of knock. But certainly far from anything that we would actually call a calibration. The factory knock sensors as close as possible to their original location with the factory settings untouched is the most accurate we will ever get our knock sensors to perform.

[Jason B]

I bet I'm not older than anyone in this thread, but.....Back in the days of carbs and distributors everyone advanced til they heard detonation, then retarded. I never blew an engine up or even knew of anyone that did.
You're not going to get a PCM so accurate in knock detection that it beats the old way. You're only going to think you did.

[AutoWiz]

Leaving the knock sensor calibrations alone is definitely more accurate than the "old way"

[Pekka_Perkeles]

I am sorry that it is commonly suggested but it is very wrong. Any level of detonation at any rpm under any load hurts your engine.

Take some time and log a perfectly stock LS engine. Run high octane for a while. Maybe a full tank.

Then switch to a low octane gas.

Take a log.

See, IT WILL KNOCK.

But it won't burst into flames. Your engine will be okay.

Just like millions and millions stock LS engines do everyday.

This is evident from looking at a spark plug after you force your detonation.

So then never switch to a low octane gas. Because I can guarantee, you then force the engine to knock with low octane fuel.

But that's why knock sensors are there, right? Otherwise they would be useless.

And it has nothing to do with tuning because you don't know exactly how much knock you are inducing to compare to the noise level you are hearing. So you don't get the accuracy. Just some level of noise upon some level of knock. But certainly far from anything that we would actually call a calibration. The factory knock sensors as close as possible to their original location with the factory settings untouched is the most accurate we will ever get our knock sensors to perform.

Again, you have two ways to detect knock. By listening and by logging. Then just tune accordingly.

Anyway, just ask Bob Morreale.

[mbray01]

I have done this several times, with relocating the single wire knock sensors to the side of the block, however in my experience, running known good timing, and knowing where knock sensor sensitivity is, and how its actually being read by the pcm, on a dyno. Every time without fail, i have had to increase the knock sensor sensitivity, when located to the outside of the block. I assume its due to the sensor being within the tube on the top which in itself likely creates more resonance. Lets remeber, the knock sensors are sending a signal to the pcm at all times, knock is determined not by actual knock, but how high that signal reaches. Knock sensor don't read actual knock as some people like to believe, but read the natural harmonic frequencies within the engine, knock increases those harmonic frequencies and that is what its reading. If you have ever monitored the live data from a knock sensor you will know exactly what I'm talking about. But again, when moving to the outside of the block, I have always had to increase sensitivity to get it in line with how it was from the factory.

[Area47]

I am sorry that it is commonly suggested but it is very wrong. Any level of detonation at any rpm under any load hurts your engine. This is evident from looking at a spark plug after you force your detonation. And it has nothing to do with tuning because you don't know exactly how much knock you are inducing to compare to the noise level you are hearing. So you don't get the accuracy. Just some level of noise upon some level of knock. But certainly far from anything that we would actually call a calibration. The factory knock sensors as close as possible to their original location with the factory settings untouched is the most accurate we will ever get our knock sensors to perform.

location and dynamic have been changed of the sensor from the original orientation. You will need to calibrate the sensors accordingly. Leaving them stock will only induce more issues. A factory calibration does show knock retard for many different reasons. Some of those you are not even remotely grasping. The motor itself will take a lot more abuse from many forms of "KR" before it comes crashing down. 20, 50, 75k miles of it is not unheard of and still warrantied by the manufacturer. Silence is not the answer in a optimized engine calibration.

[mbray01]

location and dynamic have been changed of the sensor from the original orientation. You will need to calibrate the sensors accordingly. Leaving them stock will only induce more issues. A factory calibration does show knock retard for many different reasons. Some of those you are not even remotely grasping. The motor itself will take a lot more abuse from many forms of "KR" before it comes crashing down. 20, 50, 75k miles of it is not unheard of and still warrantied by the manufacturer. Silence is not the answer in a optimized engine calibration.

hit it right on the head

[AutoWiz]

Man, Everytime I re read this thread I am just so thankful and grateful that I went the way of a real education back at the start of and all throughout my career. The opinions and ideas expressed here in my thread I made to teach you guys the educated way of doing things has been met with such ignorance. Comparisons to older engines that were made with lower compression and heavier metals with thicker material everywhere is like comparing apples to oranges, no? They are completely different animals. But none of that matters. What is truly amazing here is the belief that a human ear is a good tool to calibrate a knock sensor. This befuddles me. It is not laughable, It is dangerous. A human ear is not sensitive enough to pick up less than a degree of knock. By the time you hear change rattling you are a few degrees or more into knock. So just going by the above suggested methods we are utterly dumbing down what these piezoelectric sensors are capable of. And anything else suggested is a guess at best but far from a real calibration. This industry does not expect even a master tech to have this understanding or to be knowledgeable in this area as L1: Advanced Level Engine Performance Specialist is not one of the 8 a: series tests required for Master Tech status. Which is why it is so important you guys listen to what I have to say on this. Because I am a re-re-re-re certified L1 tech. I was that even before I was a master. All you have to have is engine repair and engine performance and you can test for L1. But you need a good working knowledge of the engines control systems to pass this test. As always I am so proud of what I have achieved in life so I can even show you my test scores..



Now the reason I wanted to bump this thread was because the car cam back for a 102mm throttle body. I wanted to add to this thread these pictures..







And then is gonna come the dyno video, results, and files. And I would attach links to this thread everytime this car comes up in conversation everywhere on the internet I go. At this point surely you all can see how that works. I direct traffic to my threads. Yes it is to show off but ultimately it draws business and hits to the threads I make. But let's be honest here. I am not going to link anybody to a thread where everything I say is questioned or shot down and with such ferocious ignorance. "No, You have to make that ultra sensitive piezoelectric sensor work stupid and lazy and slow like your human ear. And find a way to tie your ear into your mpvi for logging." I mean that seems to be the implied logic here. the stock calibration with the sensor moved a few inches is a lot closer than anything you are going to come up with in your mind. I assure you. And there is no other real way to tune this sensor accurately or anywhere near the accuracy it has in its stock calibration.