Tpx

Alright. Some of the frequent visitors to DIY PROM probably know me by now as the new beginner of the tuning bunch. I have yet another question to ask of the knowledgeable.

Short background, I've been tuning my '88 this past week pretty intensely, have burned 8 chips and made huge progress over the previous tune I started working with, mostly with AE and the main spark advance table. However, I've come up against a wall so to speak. Knock and knock retard. I understand what it does and what causes it, but my issue is telling false knock from real knock; how can you tell false knock from real knock? Is it even possible?

I have an exhaust system that when I floor it, tweaks just a little bit and smacks up against the subframe connector. Every time I give it too much gas, or WOT, I get insta-knock. I understand this most likely is the FALSE knock I see in the dataloggs from aggressive acceleration, but I wanted to ask you guys if I should ignore it or what.. I feel hesitant to ignore knock on my engine even if I know I have a reason to. To help, I have attached my main spark table, and PE spark add is zero'd out. Base timing is set to 0 degrees. (Yes I know it's a flat curve, I'm working on it)

Second question, when does knock usually occur? I am not sure if this is a silly question, but in my limited understanding knock is predetonation, and that happens when there is either a lean mixture or too much timing. I know the too much timing is false, since at WOT I have only about 20-26 degrees of timing. Not sure about lean mixture, I only rely on my NB, which says it is fine. Anyway I am asking because I would like to know that I could "ignore" knock from, say, 0-3000 RPM and only start retarding knock after that.. but that may be a dumb idea to get around my exhaust system.


Ok, I'm not sure how many people run the $6E mask, but for those that do, I have a .adx for Tunerpro v5 with a nice dash set up. I know that when I was getting into customizing my tunerpro dash (gauges) I had a tough time with the percentages to get the stuff them where I wanted them, so here's a free one for you guys. Just rename the file to 6E.adx or whatever, I had to change the name to 6E.adx.txt to bypass the upload restrictions. I'm sure this could be translated over to other masks as well.

Cflick

Well, no !

Knock is the sound made by detonation. There's no "pre" in this at all. There is preignition, where the charge is lit prior to the spark occurring, but that's completely different, and usually far more devastating as there is no noise associated at all.
Detonation occurs usually between 10 and 25 degrees AFTER top center !
What happens, is the charge gets lit, and starts burning. Pressure and heat build, hopefully reaching maximum just about 15 degrees ATDC.
If pressure builds too high, often caused by to much spark advance OR too low an octane rating, the end gasses, the as yet unburnt fuel, is deteriorating rapidly in its ability to not detonate as heat builds. *IF* the flame front gets there and burns it, no problem. If it deteriorates too far, before the flame arrives, it literally explodes like a bomb ! It "detonates." Octane causes this deterioration to occur slower. That's ALL octane does ! It does NOT cause a slower burn as some think. It still deteriorates regardless, which is why knock frequently occurs at wide throttle and low RPM. There octane simply can't hold out that long. Later timing causes the burn, and the heat deterioration of the fuel to happen later, so that hopefully you get within the time the octane *can* last and prevent detonation before the flame arrives.
The end gas trapped between the piston wall and the cylinder wall above the top ring rarely burns at all, as there is too much metal absorbing heat, so the flame can't get there. This is where detonation occurs and breaks that top ring. If you're lucky, detonation blows off chunks of piston, and you find bits of aluminum on the spark plug. If you're not so lucky, you get broken rings which you can't see without a tear down.
If you're really, REALLY lucky, the detonation occurs when there isn't much unburnt fuel left, so the bomb is less powerful, and only blows the carbon off of the piston, not the aluminum itself. Ideally, the flame burns all of the fuel before the octane deteriorates completely, and there is no knock at all.
SO, if spark happens too early, gas burns too soon, and pressure rises before the piston reaches top. The rising piston combined with burning gas causes the pressure to go well above normal. Instead of maximum pressure around 1200 PSI, you get 2000 PSI or more. It's this abnormally high pressure that cause even more heat, and detonation to occur where you hear it ringing the block. You want the timing as late as possible to avoid this, but you want it early enough so that 90% or so of the burn has happened, and maximum pressure reached about 13-17 degrees ATDC. That's the trade-off. Too early, and you have damage. Too late, and you have lower power and poorer mileage.
Preignition is completely different. That's where something ignites the gas, which lites REALLY easy at low pressure, like 120 degrees BTDC, and it burns normally, but way too soon. Piston rises, pressure rises ( something like 4000 PSI ) heat rises, and you get these nice view ports in the top of the piston that allow you to inspect the rods and crank shaft from the top of the engine.
So, can you ignore knock ?
I wouldn't.
If you have false knock, you're far better off to correct that problem. That's merely rearranging a few things.
Remember, late timing and you buy fouled spark plugs, poor performance, and poor mileage.
Too much advance, and you buy internal engine parts, like pistons, rods, bearings....

Tpx

Read it all, trying to process it, thanks for clearing up a misunderstanding between pre-ignition and detonation I didn't know I had.. I'm gonna think on this before replying further.

zen2

iTrader: (1)

very well wrote cflick, good read thanks

Tpx

Ok, after reading that my next question is, can a single "knock" as read by the ECU, where it increments the knock counter up by 1, be a harmful occurrence to your engine? What about harmful at or below 20 degrees timing? My car idles at 18-20 degrees timing, should I even bother paying knock attention at such low spark advance at idle?

If I understand what you're saying correctly, a knock is detonation, NOT pre-ignition which is the very very bad one. Detonation can be harmful, but is it only harmful in large quantities, such as 5 to 10 knocks all at once? Do knocks as seen by the knock sensor mean nothing in 1's but something serious in multiple counts at once?

Craig Moates

Nice...

Cflick

Can it ? Yes.
Is it likely ? Probably not.
Do remember that a count of 1 means one millisecond that knock has been occurring. It does NOT mean one and only one knock.
It's a function of the ESC filter module.

The timing at which detonation occurs is not really meaningful, EXCEPT that excessive advance can and does promote knock, AND that retarding timing is pretty much the only weapon the ECM has to stop knock when it is detected.
Detonation is detonation at whatever RPM and timing it may happen.
The explosion is a true detonation, so the RPM is completely irrelevant. By that I mean that the time it takes for the detonation, from beginning to end of the explosion, the piston moves maybe .001 up the bore, whether at 100 RPM or 10,000 RPM.
Having said that.....
At idle, the amount of air/fuel in the cylinder is pretty low, so there isn't much fuel there TO detonate, so the explosion will be less powerful over-all.

Correct !

No !!
Detonation is LESS harmful than pre-ignition, but it's like comparing a hand grenade or a mortar to a nuclear bomb. Both are harmful. One is absolutely immediately devastating.
( pre-ignition can destroy an engine in about three revolutions ) The other is more like termites. There *IS* damage, but it's slight per-knock, and it is cumulative.

Incipient knock, the kind that is weak enough to only blow the carbon off of the piston can actually be helpful by cleaning the chambers, but you really don't want to be running on that ragged an edge. ( OK, I don't )

That counter is counting the number of milliseconds that the knock sensor has been hearing the block ring, and assumed to be that knock is happening.
It is NOT counting individual knocks, so you don't know if it's every cylinder continuously, or only one, or several but not continuous or consecutive, or even that it's actually detonation that caused the count ! It could be a slapping timing chain, or a loose lifter.

Think of it this way....
The knock demon is a little guy with a hammer and a torch hiding inside your combustion chamber. The weak knock that humans don't hear but the sensor does, he's heating up your internal parts with a brazing torch, and rapping on it with a 4 ounce tack hammer. How much of that will your parts stand ?
That LOUD unmistakable knock that you hear from the driver seat is that demon using a cutting torch and a 4 pound sledge.
Of course, the sensor might be hearing your exhaust pipe hit the cross member, but it has no way to tell the difference.
Can you ignore any of them ?
You decide.

Cflick

P.S. Hey, Craig ! How's it going ?

Tpx

Cflick, I highly value your time and the advice you posted here. I will take it to practice. If I may ask, does your engine always produce between 2-10 knocks at start-up? I believe that mine does this either due to the starter causing false knock or the exhaust being moved at the unsteady start-up routine. After that, I see between none and 10 knocks on a typical thirty minute drive (could be exhaust). Anyway, I will fix the issue with the exhaust pipe and treat knock more seriously, as I do value my engine and want it to last me.

Cflick

At key-on, not at start up. And again at key off.
Key on, wait 5 seconds, key off, wait 5 seconds, key on.....
Then check the logs.

Pull coil wire, so it can't fire. Crank it a bit, and check the log.
If it shows knock, it's false.

For a real treat, the knock sensor itself is basically a microphone.
Clip a headphone on the sensor wire, and listen to it.
CAUTION: You will hear all sorts of stuff you never expected !!

Previous engine ( also a built wildcat ) showed similar to yours.
Turned out it was internal mechanical slapping here and there.
Worse, plug cuts showed evidence of detonation, but I found none during the tear down.
One broken ring which could have been detonation, but it did blow out the side of a cylinder, bent rod, hammered a piston to oblivion, bent valves, so that one ring is not conclusive.
This time, it's blue print clearances and that's all gone.
I didn't know, and had some timing as low as 12 degrees where it's now comfortable at 42.
( high RPM and low throttle, high vacuum. Internal combustion pressures show that it likes it this way )
Chasing false knock can certainly lead you down a primrose path !
Timing advance +/- about 4 degrees is around 1/2 percent in performance. That's an 8 degree window, so you can afford to stay away from the ragged edge until you're sure.

ddahlgren

A little carbon can be your friend and the emphasis on a little dusting not substanial deposits. An engine, not to be confused with a 'motor' as is commonly used and incorrectly, is a machine that works on heat and expansion. Anything that aids in transfering heat into the piston cylinder head or cooling system is reducing the power available because of cooling. It is what piston top and combustion chamber coatings are all about, keeping the heat in the cylinder rather than reject it.
Dave

Craig Moates

Doing good, hanging in there. Enjoying your postings here!

Cflick

Quite true, so long as it's just a little dusting. It's the deposits where you really don't want them. They tend to heat up and can cause pre-ignition. It's blowing off those deposits to which I was meaning to refer.

My first experiments with that were around 1974 or so.
A rather substantial improvement in power output, ( something like 20% ) but we didn't have the tech at the time to easily coat with anything that could regularly and routinely stand the heat of normal combustion. ( around 4000F )
Because so much more heat was retained in the mix, higher octane was required.
( I think we were running 104 VP at that point ) The deterioration of the fuel is accelerated by the extra heat retained. You still need enough octane to resist detonation until the flame front can arrive and burn the fuel. Backing off timing in this situation doesn't really help. It reduces detonation yes, but at the expense of all of the gains, and heats up the exhaust valves and ports CONSIDERABLY. Better fuel is required.

If memory serves, I mentioned something to someone in the pits about our sudden and substantial increase one day, who said something to someone else, and the next thing we knew Dyno Don ( with considerably more money than we had ) was running my ideas.

Next I tried mirror finishes inside the combustion chambers, but that didn't really do anything at all. It must be a heat insulating coating of some sort.
BUT, that has little to do with knock, causes or cures....

Cflick

Well, winter is approaching and my feet were getting cold, so I decided to stick 'em in my mouth for a while and warm 'em up a bit.

Cheers !!!

JP86SS

iTrader: (1)

Good to hear from you, its been a while.
Thanks for sharing, I know you've done some pretty involved testing over the years. You have a good handle on what is really going on in there.
Jp

MikeT 88IROC350

Very cool Dashboard! Yes I am running TP ver5 and the $6E, so hope you dont mind if I steal your dashboard. I tried myself to make one, and all i got was 4 analog guages, and 1 digital in the middle, which likes to flash on/off very fast, so it is not very useful.
As far as the download, i did it and saved the ADX as something else. But when I tried to open it in TP5, said it was created with a newer version of TP!?
Is there really a newer version of TP RT5? I am at build 5.00.6356.00.
Also, very good reading on the spark knock. I am trying myself to figure out retard, only during WOT. My post is down the boards a bit. I need to get out and test my changes.