PDA

View Full Version : Knock values and related voltages w/ PFC


boardernr
03-22-2007, 12:45 PM
Well, my hydra should be here in about a week, and i know that it only shows knock values in raw voltage instead of a number value like on the Power FC, so i'm just wondering if anyone knows what the voltage might be at certain knock values on the PFC, mainly around the 20-25 knock count. Is the scale linear?

Boosted2.0
03-22-2007, 01:53 PM
Pretty much. Assumign you have a single wire sensor you will have a 0 - 5 volt signal. You will see a normal linear progression of voltage as engine noise increases at higher RPMs - look for big 1.5 or 2 volt jumps above that current average - thats knock.

boardernr
03-22-2007, 02:15 PM
Thank you boosted :) so say a 1.5-2 volt spike would be roughly equal to what knock value with the pfc?

Boosted2.0
03-22-2007, 03:27 PM
Thank you boosted :) so say a 1.5-2 volt spike would be roughly equal to what knock value with the pfc?


depends on the starting point, but probably 50 or 60 - knock is relative - knock at lower RPMs will tend to have lower peak voltage. Thats doesn't make it less dangerous.

The Hydra map that Denny has has a trendline for acceptable knock threshold that seems pretty good - we had some knock above the line during tuning and it was not audible knock. Then again thats a 2 wire sensor - your car has a one wire sensor so its going to be a whole different ballgame.

boardernr
03-22-2007, 04:06 PM
so when you were tuning you had the ecu setup so that it would not pull any timing if it sensed knock, and then whe you were done tuning you turned it back on, right? did you get a chance to notice how good that function worked once you were done tuning? like does it catch the knock very quickly? i'm used to using the J&S, i'm guessing it's a somewhat similar program, just built in, right?

Boosted2.0
03-23-2007, 12:12 AM
so when you were tuning you had the ecu setup so that it would not pull any timing if it sensed knock, and then whe you were done tuning you turned it back on, right? did you get a chance to notice how good that function worked once you were done tuning? like does it catch the knock very quickly? i'm used to using the J&S, i'm guessing it's a somewhat similar program, just built in, right?

It didn't have any knock after tuning when I was in it so I couldn't say how well it works. I doubt its as nice as the J&S though. I didn't delve too deeply into that software regardless - we were REALLY pressed for time (he left at like 1 or 2 AM and was driving back to MS that night)

Jesse IL
03-23-2007, 07:34 AM
What you should always do while tuning is look at the knock sensor voltage charted over time. The number used by the PFC is slightly misleading. When you look at the voltage over time, you can easily see where spikes are occurring.

As far as how I like to tune, I start with a conservative spark number, look at some logs to see where the knock voltages are falling and then slowly bring in the spark while watching knock to see how its affected.

I'd suggest WEB 3.0's numbers as a starting point: 15 degrees at 10 psi with timing retarded 1 degree per pound of boost after that. Its always better to start with timing low and bring it in rather than relying on the ECU to retard it.

Boosted2.0
03-23-2007, 09:23 AM
What you should always do while tuning is look at the knock sensor voltage charted over time. The number used by the PFC is slightly misleading. When you look at the voltage over time, you can easily see where spikes are occurring.

As far as how I like to tune, I start with a conservative spark number, look at some logs to see where the knock voltages are falling and then slowly bring in the spark while watching knock to see how its affected.

I'd suggest WEB 3.0's numbers as a starting point: 15 degrees at 10 psi with timing retarded 1 degree per pound of boost after that. Its always better to start with timing low and bring it in rather than relying on the ECU to retard it.


yep yep - I normally start even lower than that on the timing JIC there is a major AF problem.

Jesse IL
03-27-2007, 02:52 PM
^^^

Also a good suggestion. I should have added, I like to make sure its nice and rich when I start. I do some driving at cruise and part throttle just to get a sense for where the AFR is falling (since running lean at cruise isn't going to harm anything or cause knock). I'll then just ramp up the fuel in a pretty linear fashion and then try and work backwards from there.

I would try to limit the number of runs at very rich/retarded timing, as you can heat the hell out of the manifold doing that. I'd stay start with the timing pulled, make one third gear pull just to make sure you're nice and fat, then go back to a more realistic IGN value.