What defines an ignition advance curve?

[jpnovak]

I have been thinking about ignition curves lately. What defines an ignition advance curve? I see deg vs rpm graphs for various engines and they all appear to be similar but are in fact quite different.

they are similar in the initial advance from 0 deg to about 10-15 btdc to a static advance at 12-1500 rpm. Then there is a second advance from 3K to the max advance where there is static advance to redline. IE, ramp, flat, ramp flat type of advance curve.

I realize that the curve will be dictated by the compression ratio, cam profile, induction system, etc. My question is...

How do each of these relate to defining rpm ranges for curve flats and slopes for the ramp rates? There has to be something going on so that an engineer can define a cetrifical weight and spring tension to reach an rpm range. Teach me o' great ones!

[Grady Clay]

Jamie,

Much of it boils down to cylinder pressure and flame propagation. Yes, CR, cams, piston shape, combustion chamber configuration, resonant air/fuel flow, and much more have a lot of influence.

If you look at the curves for street cars, there are a lot of other considerations. Drivability, economy, and emissions come to mind.

Best,
Grady

[Bill Verburg]

What a can of worms! fortunately if you have a dyno there is no need to understand all the theory.

First, it takes ~2msec for the mixture to complete combustion this is a relatively constant # IF mixture strength does not vary, this means all other things being held constant as revs rise the mixture needs to be lit off earlier and earlier.

Second, for each combustion chamber condition there is an ideal time, which varies considerable depending on the items Grady mentioned, here is a graph for a single state showing different timings and the results on the left


on the right is an AF ratio graph w/ different timings effect on fuel consumption.

Lastly the effects of different timing on emissions.


A manufacturer need to juggle emissions, fuel economy and driveability concerns, a tuner generally only worries about the last item.

Bosch has some great books on engine management for spark ignition engines, they are well worth reading

[jpnovak]

Great information! Love the graphs Bill.

I understand a dyno and accurate spark control will define the optimum combustion (cylinder pressure). this thread is leading to this. I plan to eventually go with an electronic fuel and spark control for my 3.2SS build which will start up shortly. I am in the process of making a static distributor to go with the trigger wheel sort of like Motronic. Just want to understand some basics before I am ready to build it all up. Yep, still gathering information.

Will the std. Bosch ignition books have this infomation?

If you assume the 2ms propagation and account for electronics and spark timing, can you assume a perfectly linear advance curve? this would just be calculating 2ms BTDC relative to crank angle as engine speed increases. This is of course, assuming that the cylinder filling, stoichiometric ratios and other factors were constant or at least adjustable enough to be relatively constant. I am hoping an integrated spark, fuel system will be.

[Mike Bonkalski]

I don't have much first hand knowledge on the subject. However, this study on Spark Advance Control by Ion Sensing and Interpretation (http://www.fs.isy.liu.se/~larer/Projects/main.html) is quite interesting. Give it a read.

[Bill Verburg]

If you assume the 2ms propagation and account for electronics and spark timing, can you assume a perfectly linear advance curve? this would just be calculating 2ms BTDC relative to crank angle as engine speed increases

No, because there are too many things changing at the same time. the only way to do it is empirically(on a dyno). The beauty of Motronic is the resulting 3d fuel and timing curves, that can be precisely tailored to each engine over a wide range of operating conditions and can be made adaprtive w/ O₂ sensors.

[Early_S_Man]

Here is a page from the 1983 1st ed. of Bosch's tech book on Motronic ...

The advent of computer-controlled dynos and ignition systems have enabled mfrs to collect much more info than was available in the 'golden years' of hotrodding in the '50s and '60s ...

Even though these specific graphs weren't taken from a 911 engine [probably from a BMW 6-cyl] ... they are representative of high-performance engines in general.

[jpnovak]

I can tell that 3D cascade is quite a bit more complicated than the mechanical setup. Its easy to see where a computer controlled integrated system is much more accurate.

This has been great information. I assume the knock sensors are the feedback for spark control. Makes sense.

Thanks for the link Mike.

[Bill Verburg]

I assume the knock sensors are the feedback for spark control

Yes, the 3.6 also used the knock sensors but the O₂ is equally important for a street engine where fuel economy is a concern.