Burst knock retard... what is it?

[Backside180]

Can someone please explain what burst knock is?

Thanks

[dc96819]

A big transitions when you hit the gas

[Backside180]

A big transitions when you hit the gas

Is it real knock, or is it a preventative measure to avoid drivetrain abuse?

My tune has 2 tables (enable delta cylinder air and vs. ECT) and both are set to 0 in stock form. Does this mean I will never get any burst knock retard, or are there more tables that effect this?

[dc96819]

I think its for fuel economy
My truck has one table min 0 max 8.The bigger the # gets less sensitive.
I added .03 across the whole rpm range and it went away.
Do you have BK in your scans.I tought only trucks had that problem

[WS6FirebirdTA00]

It is timing that is pulled in ANTICIPATION of knock. When you see large transients, it will pull timing to prevent knock from happening. This is not knock, it is just timing pulled. This is nothing for fuel economy, it is to protect your motor.

I have my burst knock tables zeroed out.

[Backside180]

It is timing that is pulled in ANTICIPATION of knock. When you see large transients, it will pull timing to prevent knock from happening. This is not knock, it is just timing pulled. This is nothing for fuel economy, it is to protect your motor.

I have my burst knock tables zeroed out.

OK... I'm clear up to the point where one person says zero the table out and the other person says the higher # is less sensitive...

I assume one person has the timing number zeroed out and the other raised the gm/cyl threshold?

Also, when logging "knock retard", is that a cumulative value that would include burst knock or is "knock retard" only in response from the knock sensors?

[EC_Tune]

Knock Retard is real knock on LS1's. Total Knock Retard (LS2+) is anything that is knock related pulling timing (eg. burst KR, or real KR).

Like Mike (WS6FirebirdTA00) said, it pulls timing to prevent knock during throttle transients. You can either zero the Burst Knock retard values or increase the G/cyl to reduce it's effect.

The best way to understand this is to log all the spark advance and spark retard values on the scanner table view and see what is doing what.

[01 MidMet WS6]

So is burst knock something that should be turned off when tuning or all together to eliminate false knock from registering when logging? Or is it something that we should continue to use? If one were to desensitze it using the enable cyl. delta air table how much of a change should be made?

[TiredGXP]

I zeroed out my burst knock retard table completely, logs indicate that I'm not getting knock on large throttle transitions, so for me this helped maintain more consistent timing and didn't cause any knock issues. Your results may vary

[BlackSS]

The best way to understand this is to log all the spark advance and spark retard values on the scanner table view and see what is doing what.

Can you elaborate here? I'm losing spark somewhere and I'm not sure where.
I'm logging KR, & IAT advance. What PID's should I add other than burst knock?

If it is burst knock, what tables need to be modified, and what do I set them to?

Thanks,

Craig

[EC_Tune]

There's a ton of them. Eliminate things from your table view that aren't spark related (Except airflow values and Engine RPM) and save it as a new configuration (.cfg file). Then go log some data and see where the missing timing is coming from. Could even be Knock Learn (HO table has more spark than LO table and the PCM is interpolating between them.) changing your spark advance.

[John_F]

So either raising the gm/cyl threshold or zeroing the tables will have no ill affects?

[jimmyblue]

Burst knock retard is a predictive retard that is meant to
prevent a knock-burst following from one "problem area" -
an abrupt air surge, which may not be well covered by
fuel or by spark advance pullback.

So you see a delta cylinder air value as the trigger. This
is pretty nearly the same as a MAP step, intake / cam
particulars aside.

Trucks suffer from an excessively eager burst knock
setup, it can kill your first gear acceleration.

To look at the threshold that is used, you need to use
Excel and create a pointwise DCA column where DCA

=(CA2-CA1)/8

The 8 being to line up the 12.5mS time-tick used by the
PCM, with the 10/sec frame rate of the data. If you are
logging more than 24 PIDs then it's /16.

Now scan through that column for a good throttle tip-in
and see what the maximum DCA is running. If you see
ping at some steps and not others, see if there is a clear
"ping threshold" and put your delta cylinder air values for
burst knock, just below it.

If you are tuned nicely you do not need this feature. But
it can compensate for lags in the system or for you wanting
to not make the VE table over-fat or play with transient
fuel, to cover up tip-in ping.

If you don't ping on any tip-in event, and you are not seeing
burst knock retard in the scans, then you can leave it be.
It doesn't seem to be a big problem for cars.

[GMC_DUDE]

Thanks for posting up this thread. I was experiencing the opposite problem. After installing the 6.0L in my truck I kept getting 1.7 - 3 degrees of KR anytime I would accelerate while going uphill at 60 - 70 MPH. The KR would be just a spike with a rapid taper-off. After reading this thread I checked my cal and found the previous owner had zeroed out the burst knock values for a warmed up engine. I copied over the values from my 4.8L cal, and voila'! No more KR spikes going up the hill.

[John_F]

Thats just what I did for now. I never had any knock at throttle transients so for now I zeroed out the last line for a warmed up car. I'll keep an eye on it for awhile and see how it goes.
As for jimmyblues reply, I always read them and try to understand him but man.... does he speak the same language as me

[GMC_DUDE]

I hope Backside180 doesn't mind if I divert somewhat but this is still on the topic of burst knock. After giving it some thought and reading Jimmyblue's post, I'm wondering if my "fix" isn't just a band-aid, and the real problem isn't a messed up VE table or something? Thinking out loud for a bit: I am running closed-loop w/ MAF (i.e. stock configuration) so both of these control systems should hold the AFR to stoich during part throttle and the MAF should cover PE mode at least in steady state. This is verified by my wideband meter. The fly in the ointment as I understand it is in the transient events, like in my case doing 1500-1700 rpm in 4th gear and then giving it enough gas to promote about 2psi boost, but not enough to cause kickdown and the attendant higher rpm's. The MAF, being lazy, does not respond at once to the increased air flow so for a split second there exists a lean condition which causes brief detonation and a KR spike. There are (at least) two ways to eliminate the detonation. One way is to use burst knock which pre-pulls the timing in anticipation of detonation. This eliminates the ping and KR but sacrifices power and you still have a potentially dangerous lean condition under boost. Would a better approach be to increase the values in the VE table? That way a sudden increase in airflow would instantly create more fuel, eliminating the lean condition? How would I know if I got the VE table values too large? Or is there another parameter(s) that I should be tweaking? I have two books on EFI tuning on order and plan to learn a lot from them so sorry if this is all kid stuff. The How-To's don't really work for me because they are cook-book oriented and I like to understand what exactly is going on. Thanks for putting up with me.

Edit: Why is this more of a problem with trucks than with cars?

[EC_Tune]

The MAF, being lazy, does not respond at once to the increased air flow so for a split second there exists a lean condition which causes brief detonation and a KR spike.

Chris had an excellent explanation of MAF/MAP relationship HERE. Everyone should read it first.

What happens when you open the throttle quickly at low engine speeds?

First, air flows in to the intake manifold, the MAF measures this "intake fill" but it hasn't necessarily entered the cylinders yet. So the MAF change *should* make the next cylinder firing 'rich' because it is measuring an increase in airflow but it's just filling the intake manifold not the cylinder necessarily.

So now you have a big 'huff' of air to fill the intake. The MAP Kpa finally increases so the PCM knows there is more pressure in the intake thus more airflow. Finally the air moves into the individual intake valves and fills the cylinders. Injector pulse width is calculated based on the air entering the cylinder from the MAF/MAP/transient fueling relationship.