I'm getting information from lots of people saying zero is the only acceptable answer and then some from much more reputable people saying some KR is ok and its a nuanced answer. I'm getting a huge amount of KR at idle but I think its my Y-pipe hitting my SFC (they are very close). But I'm more concerned with the 100% load column where is have a few degrees in the 3k RPM range. I have my fuel AFR dialed in at 12.5:1 the whole way through WOT.

I will attach the WUD display for a 20 minute drive to work with a few WOT pulls once traffic cleared. One screen is SR Peak and the other is Total.

 

I'd probably shoot for zero myself.

But one thing you could do to verify it really is knock would be to reduce the timing at least by the amount of retard in those areas and then make another pull. If the ECM is still pulling out a lot of timing, it may be false knock being generated by something else.

If the ECM is pulling that much timing, you may as well try to get to the bottom of it, otherwise it's essentially defeating your desired spark map.

Your screen name implies a supercharger? What's your setup? Compression ratio? Octane of fuel you're running? Cam?

 

This car had a P600B Procharger on it long ago. I'm running the L98 with all forged bottom end, 355ci, L98 heads milled 0.010" for 10.05:1 compression and just a ported plenum and a comp cams 210/220, .500/.510, 114LSA cam and SLP 1 3/4" headers and open exhaust after a custom Y-pipe. Ford SVO 24lb injectors running the AFPR at 45psi. I run 93 octane. So pretty mild setup really.

Building a 421ci LS3 stroker with an F1A Procharger and Frankenstein Stage 2 893 casting heads, Stage 2 Tick Performace Blower cam and a Holley Terminator X so I'm just dicking around until I get the parts for that.

 

10.05:1 on 93 octane could be pushing it a bit... particularly with that cam, which is going to build quite a bit of dynamic compression.

You're about where I was at 10.1:1 several years ago with a cam about that size, and I was also getting quite a bit of knock retard. So my suspicion is that's real knock you're getting (particularly at the high MAP region).

At the time I was running a .028 thick head gasket, so I put in a .039 which got me down to ~9.8 and it was good there. If you're running a .039, you could go to the .051 and do the same thing. Maybe spark plug with colder heat range could help.

 

Grumpy would have said 0 KR. A tickle of the knock sensor ie < 1* is ok. Could be noise depending on valvetrain like my DOHC LT5. I would go along w the recommendation of determining whether its false knock or not by either pulling out timing or using 104+ Octane booster.

 

i usually unplug knock sensors lol seen stock stuff just get way over sensitive and false

Its hard to tell if its real or not sometimes. Def arent gonna have detonation at idle so i would say any at idle or low rpm is probably false.

how much timing you run? Tpi def has high cyl pressure in the 2800-3200 range at high loads so you can take several degrees out there. A dyno will show any detonation in the power curve. You take timing out and add one or so back in till power stops climbing and then back off 1. Tune all other motors without a knock sensor just fine.

10:1 on todays 93 gas aint bad, but it may want less timing than stock and a step colder plug than stock is def a good idea, maybe even 2 steps. Have to read plug to see. Ngk 4 heat is typically stock level. I like 5-6 heat range for more modded stuff

 

Don't tune for knock, tune for peak torque (and there won't be knock). Time on an eddy current dyno is time well spent for dialing in part throttle timing curve.

 

Well I'm trying to get it close before I pay for dyno time. No one in the entire DFW will tune an OBD1 GM vehicle so I need to have a few tunes on hand to switch between when I get my 3 pulls on the dyno for $90. LOL

 

Your type of knock sensor gets excited by vibration at its resonant frequency so can get excited by other vibrations in a non-stock drivetrain not related to combustion. More likely to get false knock with those kind of sensors and there's not much you can do about it except desensitize it in software (if allowed). That's probably where the knock retard is coming from at idle.

I think you have plans to use Holley EFI in the future and that's the time to change to a 2-wire non-resonant knock sensor that is based on bore size. It'll be more adjustable for a non-stock engine so that you can keep the safety net of knock retard but not interfere with normal operation.

 

Ah, yeah, not being able to flash an ECM quickly really sucks when you're paying by the hour.

 

Not sure if you are being sarcastic or not, but it takes no more than 4 - 5 seconds to flash in a new tune on an EBL Flash system.

RBob.

 

Dyno time being as much as it is, kinda wanna just do wot runs

street driving will do those part throttles. You can see if the KR is consistent in those low map areas and then yank a bunch and see if it goes away. If it doesnt then i would suspect its false. But without knowing how much timing you have its hard to say. Anywhere from 34-38 deg always worked well
for me in most areas in cruise. Some like abit more like alum L98 heads, some like lower end of that like Afr 195 and maybe vortec heads

 

It's false, it's picking up chatter... 36 Knock Counts at 800-RPM is happening at engine startup and shut down. If you click on Counts you might see it above 6000-RPM in the same 40-kPa column.

- Rob

 

I generally do not even worry about occasional knock retard caused by random burst knock. To me you are leaving power on the board if the timing is not advanced enough to occasionally trigger knock retard events. My engines often end up slightly octane or knock limited. I will only address it with retarded timing if I can consistently make it knock in the exact same manner.

You should see how much knock retard happens on a 100% stock tune and 100% stock engine, especially the newer ones. I hear many of these small 4 banger crossover SUVs spark knocking pulling out onto the road in front of my house, yet the vast majority of them seem to be living 250K or more.

 

You can gain torque and be detonating at the same time. If you use non filtered dyno results, you see detonation as jagged lines similar to saw teeth in the torque curve.

 

Knock kind of messes with the cylinder pressure vs. crank angle. If you set fueling first and then sneak up on timing and stop at peak torque then I'm not sure you can get knock unless it's for other reasons like lean fueling, carbon, or fuel soaked oil that didn't burn last cycle, etc..? You're an expert compared to me, I've got my ears open.

My engine absolutely will not knock if fueling is right. Knock is waaaaay far away from my optimal timing (5 degrees or more). But I have a much better combustion chamber than SBC so that helps a lot I suppose. The only time I've ever seen knock retard is when my fuel pump was failing and engine went lean.

 

As I mentioned previously some engines/combustion chamber designs are knock limited on the fuels available today. That means the timing on them has to be set retarded compared to where it makes peak output. Some modern engine designs do not even use 20° of timing at WOT. My buddies LS3 for instance makes peak output at 21°. That being said some engines with old combustion chambers and domed pistons ran best with leaded fuel and 38-40° of timing, you can absolutely bet a DZ 302 would rattle your teeth out on 93.

The other time you will often see knock is if the engine is loaded more than 100% aka lugging. By that I mean cruising along at low rpm and go up a hill in top gear where the engine is giving full torque output and is actually getting pulled lower in rpm because its output is less than the required torque in that gear. Think of a truck with a relatively small engine pulling a trailer uphill or RV going uphill before the trans downshifts.

 

OP forgot to mention his exhaust hits the sub frame connectors, but ONLY at idle.......

 

1. Never tune by knock sensors or depend on knock sensors for anything
2. Tune the engine based on VE and RPM values using common sense and established chamber-fuel safe timing values per application
3. Compare current fuel (93 octane or whatever) with high octane fuel (110 or 116 octane fuel) to establish 'realist' knock values

For #2 acceptable LS 02+ Natural aspiration 9:1 compression modern chamber timing WOT peak value is 24 to 26* btdc when using 93 octane
All in by 2800-3200rpm
Remove timing for idle-2800rpm regions based on load, application, vehicle weight, gears used, etc... each gear ratio has its own ideal timing value for low rpm, with numerically higher gears demanding fewer timing btdc values (More time under low RPM load = more heat = Less timing wanted)

 

Yup. Because if I pull timing I never get any knock at high load. The engine is torqued enough to keep the Y-pipe away from the SFC. If it was causing knock signal no matter how much timing I pulled, it would be false knock. Deductive reasoning isn't hard.