Ukraine Train

I’ve read through some of Grumpy’s posts on timing but I feel like I still don’t have a good understanding of spark advance. On a carbureted engine, vacuum advance adds timing at idle and decel, but otherwise the mechanical advance only adds spark based on engine speed and maintains it at a fixed point above 3x00RPM, all without regard to engine load.

Looking at some timing maps in EFI, though, it seems that advance peaks out at about 3500RPM and is then retarded at higher RPM’s. Why is this? I would think that as engine speed increases you have to time spark earlier and earlier to ensure peak combustion pressure at the right moment ATDC. I would think that higher engine load would require more spark also because it takes longer for a bigger a/f charge to burn.

I guess I’m just looking for some general insight on when you need to add more timing. Any takers?

305sbc

iTrader: (3)

When looking at EFI maps, are you taking into account the advance added by the PE vs RPM table (power enrichment spark-adder) ?

Ronny

iTrader: (1)

That would be true if load did not build up exponentially, wind and mechanical resistance.

Putting that much heat in cylinder leads to detonation. There may be other reasons as well. Emishions?

My earlier experience 40 years ago had mechanical advance with no vac modification. So when first viewing the SA maps it was a puzzler.

I dont have my laptop here at office to look at my current SA in .bin but it is a LT-1 iron head for SA. Old school may have 32d SA at WOT. I may have that in my .bin right now or 30(PE zeroed out) BUT i have 40d SA at light cruise rpm/map. So modern cars have more timing at cruise.

xch3no2

Savvy tuners understand there is power to be gained with the proper high rpm retard.

I'm sure that it is favorable to emissions as well, or they wouldn't do it.

Ukraine Train

No, just the main spark advance table. Here's the one from S_AUJP. What's with the hump at 100kPa? Regular AUJP doesn't have that.

dimented24x7

iTrader: (2)

In regards to why most timing tables stop at 3600 RPM, its important to note that the total time available for combustion is proportional to (1/RPM). There is a much greater difference between the time available at 500 and 600 RPM than there is at 3400 and 3500 RPM. The variation becomes much less as the RPMs increase. At a certain point, the required increase in advance is too small to notice, so most tables typically go flat above 3000-4000 RPM. Its not always so, however, if the cylinder pressure varies. An example of this would be an engine that only has good VE above 4000 or 5000 RPM. In that case, the timing would vary a bit at higher RPMs to achieve the best torque.

Timing, however, is much more strongly dependant on cylinder pressure than it is on the actual time needed. The pressure greatly effects the propigation rate. The combustion speed increases rapidly as the pressure increases when the pistion goes to TDC. If it was strictly a matter of time, than engines would require nearly no advance at all at low RPMs, but we know this is not so. Less is needed, but its still going to be at least a couple degrees at high load, and maybe 15-25 degrees at low load.

dimented24x7

iTrader: (2)

To get an example of how much the cylinder pressure effects the propigation rate, the timing can be given in terms of the time the actual combustion takes. For example, if the engine needs 20 degrees of advance at 800 RPM for peak torque, the time it takes for the mixture to fully burn is (20º/360º) x (1/13.3 revs/sec) = 4.2 msec.

The first table Ive attached is my regular spark advance in degrees. The second one gives it in terms of milliseconds required for the mixture to burn. Notice how much more time it takes at low RPMs?

dimented24x7

iTrader: (2)

That looks like a spark advance table done for mean best torque. This is the optimal timing for best engine performance.

For an LS1, the PCM has both the standard timing table, and the one for mean best torque to use as a reference for retarding the timing (spark knock, transmission protection logic, etc.)

The normal table is the first, and the one for mean best timing is the second. The second looks like the one you have posted. The first one is much more smooth.

GM no longer does the spark like they did in the older ECMs as it can cause instability. The engine is prone to surging and lugging, especially with a manual, if the spark advance changes too quickly. Changing it smoothly causes a slight loss in performance, but increases drivability. Ditto for emissions as high SA will cause increased NOx.

dimented24x7

iTrader: (2)

At the risk of going into gory detail, its better to think of timing in terms of the density and propigation rate.

Under normal conditions, the flame front progresses based on interaction between the fuel molecules. Basically when a neighboring molecule decomposes and recombines with the oxygen, it releases heat and the combustion products carry that heat away in the form of kinetic energy. When they bump into a neighboring molecule that has yet to combust, the transfer of kinetic energy leads to it decomposing as well, and this progresses across the cylinder.

The rate that this happens is proportional to how closely the molecules are packed together. If they are far apart (low MAP, lean mixture), then it takes longer for each interaction to take place. When they are squished tightly together (high MAP, richer mixture), the distance is much less, so the process proceeds much more quickly. This is also true as the piston compresses the mixture. It squeezes everything closer together, so interactions take less time.

Under detonation, though, the fuel molecules no longer need a nudge from their burning neighbor to combust. The temperature and pressure across large portions of the cylinder reach the threshold at which the fuel spontaniously decomposes all on its own. This means that a flame is no longer needed for combustion, and large portions of the mixture burn all at once rather than in a controlled fashion.

305sbc

iTrader: (3)

I'm not seeing a hump at 100kpa.


edit: you must mean the 2* drop between 4400 & 4800.
ref- the PE spark adder table.
You add the two tables together for total SA.

I just looked at a regular AUJP bin and the PE spark adder adds 2* at 3200 and 4* at 4800 rpm.

JP86SS

iTrader: (1)

Keep in mind when working with the $8D table is that the 100 Kpa row is used until the engine warms up and transfers control to the closed throttle SA table.
It is no just for WOT.

dimented24x7

iTrader: (2)

Are you sure about that? In the code, it looks like the main SA is used for warm-up, and then the closed throttle SA is used when the engine is fully warmed. It doesnt specifically use that row of the table, but just the corresponding cells for the current MAP/RPM in the main table. The main reason being that theres no cooresponding coolant SA correction table for both the open and closed throttle spark advance, so the main table is used during warm up.

Edit: The hack does imply that like you say. But, it also has the wrong addresses there for the cool temp threshold, and states the 100 kPa row is used on the first pass. The cold SA threshold also points to the start of the SA table, and not the thresholds above like it should. Id take the hack with a grain of salt. If it really is like that in the actual binary, then it should be patched. I certainly wouldnt use it like that. Its an obvious logic flaw if its so.

JP86SS

iTrader: (1)

Looking at the description in the hac, I'm not seeing the incorrect address, logic, etc you mention.
Is this the same code breakdown you looked at?
https://www.thirdgen.org/forums/diy-...k-advance.html
Post 16 has the code breakdown.

305sbc

iTrader: (3)

Yeah, but that's still only true for idle (closed throttle) correct ?

I'm not sure the OP was talking about idle SA. It seemed to me he was wondering why the WOT SA started dropping again at the higher RPM.
I thought he may have been discounting the PE spark adder that gets factored in on top of the main spark table at WOT.

dimented24x7

iTrader: (2)

Thanks, thats better than whats in the hac posted on moates.net. I figured that the hac was in error instead since it didnt look right.

The use of the 100 kPa cells until warm-up is achieved is a really poor way of doing things. It may sort of make sense in emissions terms to retard the SA and keep the combustion going in the manifolds so the AIR can do its job, but for our intents, its crap as far as Im concerned. It couples your warm-up SA and your WOT SA together, resulting in that funky looking SA table at WOT.

A better and simple approach to handling warm-up would be to simply direct the ECM to use the open throttle table during warm-up so the cool temp SA compensation can be used. This can be done by simply changing this:

bra LB946

to this:

bra LB93B

which would send the ECM to use the main table in full, and give the proper advance during warm-up, rather than coupling it like that. Thats what I would do, anyway. Theres stuff like that in all the GM ECMs and PCMs. Either by mistake, or just by poor coding. I usually just go and fix it so it doesnt cause any headaches.

ninetyone

iTrader: (2)

ok, i have a question for you guys. I was looking in tunerpro and comparing my bin to a JET bin. I see that the changes made in the Spark adv. main table was only in the 90-100kpa's. What are the kpa's supposed to represent? Is this 90-100 supposed to be related to PE or WOT? Ok,what i am trying to do is this: I noticed that i can run a lower octane gas and i do have better part throttle response,but when i get up past 3500rpm it will ping(of course i run 93 to prevent this) but, what changes to the spark table would i make to prevent this pinging when using the lower octane gas? Should i pull out like 2 degrees? and where would i be pulling this 2 degrees from? the 90-100 kpa areas in the table? Let me know please.

xch3no2

Kpa represents manifold vacuum.

Remove 3-4* at a time (from knock point) till knock is eliminated.

Determine that there no lean spots contributing to the problem.

Ronny

iTrader: (1)

That would be the corresponding axis point of RPM/MAP.
90-100 kpa is WOT and Yes you would be in PE(usually).
Might want to try a datalog with 93 octanne to see if it lessens. Often KC's are mechanical especially if a modded engine/exhaust or drivetrain slop.

xch3no2

Thanks Ron,

The source of mechanical knock (ie. other than detonation) can be from anything loose on the car as well as noisy engine components, (roller-rockers or headders).

305sbc

iTrader: (3)

KPA is the % atmospheric pressure in your intake manifold. Higher number = more pressure.

First, take a look at your PE spark adder in order to determine what your total advance would be when in PE mode.

Next, the point on the main spark table where you start reducing timing depends on where on the throttle you start to have knock. If you're at part throttle, maybe 1/2 throttle opening, then you probably want to start pulling spark away around the 50 - 60 KPA area of your table, 3500 RPM and up.

Many people find it easier to zero out the PE spark adder and just do all your spark work in the main table.