Knock Retard

[TransAmDan]

I have been doing a bit of data logging recent. Not really tuned it much since the stroker kit apart from getting the fuelling correct.

The red line on the 3rd part of the graph is the knock retard scale. Retarding the timing from up to 10 degrees. Why does it happen? well there are 2 microphones in the valley of the LS1 engine. These are like microphones, detecting detonation of running bad fuel, and then retarding the timing to protect the engine.

In the following graph we have a couple of short full throttle burst, we do get 1.2 degrees of timing taken away in 2nd gear, but not in 3rd.

The fuel if anything is a little rich as 12.38AFR, not far off of the target of 13.07. Going too rich is okay, yeah you will loose a little power, but its safer than going lean. Lean can cause detonation aswell as timing too far advanced.

My spark timing is a couple of degrees less than stock. Has going to a 383ci stroker kit mean more timing needs to be removed? is the knock caused by external forces like the exhaust banging(I dont think it is, but it has happened to others)

My spark timing table for high octane fuel

After doing some research the sensors can go bad. They are 20 years old now. Perhaps time to get new. Rock auto do them for about £35 each (AC delco) and you can but the wiring loom too. I may go for that, nice easy solution which eliminates the doubt.

I can turn them off in the ECU, but I don't want to run without protection.

 

[Will]

I must admit, the TA usually smells quite rich when I've been following you, but if you have set it up on the rich side then I guess that's why. Not sure how the stroking will change things, but if the knock sensor mics are faulty that could throw things off. There must be some way to test them but I would think that would involve testing them acoustically as well as electrically. Depending on what type of microphone they are, they may show the correct impedance and resistance, but if their diaphragms are clogged or have gone hard due to age, they will be a bit deaf lol.

 

[TransAmDan]

These are the little sensors.

They sit in the valley under the intake manifold.

They thread into these holes.

The wiring loom for it.

 

[TransAmDan]

I must admit, the TA usually smells quite rich when I've been following you, but if you have set it up on the rich side then I guess that's why. Not sure how the stroking will change things, but if the knock sensor mics are faulty that could throw things off. There must be some way to test them but I would think that would involve testing them acoustically as well as electrically. Depending on what type of microphone they are, they may show the correct impedance and resistance, but if their diaphragms are clogged or have gone hard due to age, they will be a bit deaf lol.

The rich bit should only be on full throttle. Other times the O2's are switching around 14.6AFR so shouldn't smell. Perhaps the lack of cats not helping. My fuel injectors are around 10 years old. I do have a spare set in the loft that have been flow tested. I'll swap them over at some point. Perhaps one cylinder is richer than others.

 

[TransAmDan]

The LS1 KS's are a single wire broadband sensor that the PCM sends a certain amount of voltage to, as the sensor detects knock, it changes the voltage reading and the PCM then knows how much knock is present by the amount of feedback voltage change.

 

[Will]

Yeah, I remember seeing the knock sensors when we stripped it. Interesting how they work then. I guess they must be tuned to a certain audible frequency (the pre ignition noise). I am just wondering if that frequency might be different for different capacity engines, therefore, might yours need different sensors because it has been stroked. Hence the wide band sensors presumably?

 

[TransAmDan]

It must listen to a particular point before its told the relevant coil to make a spark, if it heard a bang, then it went of too early. That is some clever timing, how many times does a cylinder fire at say 3000rpm? I guess its 1 in every 4 strokes? and then 8 pistons, that's quite a narrow window to listen and even smaller are higher RPM's.

These are some of the settings. The values in the boxes represents. Knock Sensor Level vs. RPM vs. Cyl: Used to increase, decrease Knock Sensor sensitivity per cylinder. Knock sensor level that must be exceeded to consider the engine knocking. The knock sensor is compared against this table's levels based on current cylinder firing. If found to be below the table result, no knock retard can occur. It is the base knock sensor table when the TPS and RPM tip in tables are not in use. Knock energy table level is from 0 to 64.

 

[AmericanThunder]

Valve train noise can provide false knock to the sensors as well. If your valve train geometry is quite aggressive you could be triggering.
Headers definitely increase false knock too.
The 90’s LT1 knock sensors were more sensitive the man the same era LT4 knock sensors for for those reasons. The LT4 has more aggressive valve geometry though I don’t recall if it came with shorty headers

 

[TransAmDan]

Yeah valve train is agressive. Cam is a BTR Stage IV, 235/242 .621"/.592" 111+3. The dual valve springs are good for .650 lift.

Been out for a drive to get fuel and a quick 10mile round trips while data logging. nice weather so I took the t-tops off. For some reason I have one frozen finger, lol.

I'll check on the datalogs later. After filling up the timing relearn happens. Will be interesting to see if the spark retard is in the same spots and same amplitude.

There are people in the states that have turned off the knock sensors and say they know what to listen out for. I wouldn't want to take that chance.

 

[AmericanThunder]

So, the LS6 was basically an LS1 with better flowing heads and more aggressive valve geometry. Are the knock sensors the same part number between the two?

 

[TransAmDan]

I have different heads, Ported by Patiriot performance. They are a different casting number from LS1. I think they are LS6 heads with larger valves etc..
Stock cam is under 200 degrees for intake and exhaust, so I have many more degrees of open time over stock. With the valve moving at different times and if the knock sensors can feel this then some adjustment is needed. Its hard to find info on this on ls1tech.com and other sites. Its been around for 20 years, someone must have encountered this before.

 

[AmericanThunder]

All the heads interchange except LS3?
I’m also wondering about knock sensors from other LS variants perhaps being less sensitive.

 

[AmericanThunder]

Looks like the KS changed dramatically in 2005 for the LS2 but all variants of LS1 & LS6 share the same part number.
Aftermarket?

 

[AmericanThunder]

Hawks motor sports sell an adaptive wiring harness to enable use of the LS2 KS. Now what advantage does that offer?

 

[AmericanThunder]

Just a thought... what octane fuel are you running?

 

[AmericanThunder]

Also found this: https://forums.holley.com/showthread.php?2534-Connecting-amp-Tuning-Knock-Sensors-Holley-EFI

 

[TransAmDan]

Just a thought... what octane fuel are you running?

Nipped out at lunch and fulled up with V-Power, I data logged, not checked results. My thoughts are I'm not expecting any change as already retarded timing, but it could prove something.
Before this full up, it was standard unleaded from Shell.

 

[TransAmDan]

Okay, logs from a couple of days ago, WOT in 2nd gear, gave KR at 4500rpm spark timing at 18.4degrees. Fuel mixture good at that point.

Also seen in 3rd gear 88mph. Fuel mixture also good at that point.

Filled up with V-Power, I have a 26min data log since then.

Wasn't a lot of good places to to full throttle, but manage to do a bit in 2nd gear 100% throttle, no spark retard, felt good. Partial throttle before that was knocking a bit.
Another partial throttle below.

Overall I think there was less spark retard in that run. It doesn't occur at full throttle.

I think V-Power has helped (unless its in my mind and I want that to happen) WOT timing is fine. Perhaps need to adjust partial throttle timing and keep and eye on fuel mixture, which appears pretty darn good right now.

The engine used to be 10.3 CR, it then went to 10.8CR with the heads, with the stroker, i really don't know what compression ratio I am, they were shooting for 10.8. With the ratio being close to 11:1, then it would make sense that super unleaded would help.

It doesn't matter running regular unleaded as the ECU checks the octane and drop to the low octane spark timing table. I'll do some tweaks, program it in and drive with some more data logging.

 

[renchero]

ranchero runs better on v power. can make it knock on normal unleaded but can't make it knock on v power. don't know if you are running timing chain or gears but summit racing say don;t use some of there gear drives with a knock sensor

 

[TransAmDan]

ranchero runs better on v power. can make it knock on normal unleaded but can't make it knock on v power. don't know if you are running timing chain or gears but summit racing say don;t use some of there gear drives with a knock sensor

Timing chain on mine. That is a good thought though.