Ignition Timing Mechanical Advance

[Captain Obvious]

So the recent chatter about timing and distributers has me thinking.

The stock distributors include a mechanical (centripetal) advance function and the spec varied some over the years, but as an example here's the spec for 77:

     0 degrees at 600 distributor RPM (1200 RPM at the crank)
     8.5 degrees at 1250 distributor RPM (2500 crank)

There should be no mechanical advance below 1200 RPM, then start ramping up to a max of 8.5 degrees by the time you reach 2500 RPM. Once above 2500 RPM the mechanical advance is topped out and holds at 8.5 degrees all the way up to redline.

Specs like this are given in degrees because that's what can be measured and that's what you can see with a timing light. But if you convert the timing specs from the angular domain (degrees) over to the time domain (seconds), it looks like this:

     At 1200 RPM, the engine is spinning at 20 rotations per second.
     At 20 rotations per second, each rotation takes 50 milliseconds.
     If each rotation takes 50 milliseconds, each degree of rotation (out of 360) takes 139 microseconds.
     So at 1200 RPM, the 10 degrees of base advance provide about 1.4 ms of advance.

Doing similar math, if the engine is spinning at 2500 RPM, then each degree of rotation takes 67 microseconds. The 18.5 degrees of advance (10 base + 8.5 centripetal) provides about 1.23 ms.

If you go through all the math and plot the results, you come up with something like the following. Notice that the advance value where the centripetal advance is active is relatively flat, but then the advance falls off above that once the mechanical advance has topped out:

Studying the numbers, I come up with a relatively constant average of about 1.2- 1.4 milliseconds of advance in the centripetal advance region.

So my question is.... Why didn't they continue to advance the centripetal timing above 2500 RPM? Is that just a limitation of the mechanical functioning of the distributor? They just couldn't get the springs, weights, slots, etc to swing a wider range all the way up to redline? Isn't necessary for some reason? Or wished they could have, but just couldn't come up with a cheap enough way to do it?

[cgsheen1]

CO, you just made my head hurt...

[Zed Head]

Maybe the benefit of more advance tails off itself.  No need for more.  Or...black hole.

You could compare to what the pros run using electronic engine control.  Seems like there must be some dynamometer studies out there somewhere.

[Zed Head]

Got $33?

https://saemobilus.sae.org/content/940546/

https://www.sae.org/publications/technical-papers/content/940546/

https://www.sae.org/publications/technical-papers/content/940547/

[jonbill]

The purpose of firing the spark before TDC is to get max cylinder pressure at the optimum point - I believe I read that the maths works out at about 22 deg ATDC is the optimum point. If combustion was instant, then 22 ATDC would be the spark timing, but combustion takes time and so the ignition is started at the appropriate time before hand.

Since the speed of combustion is reasonably constant, as engine speed goes up, you have to start ignition earlier. I.e as the engine goes faster, you need more degrees of rotation for the time it take for combustion to reach its peak at 22 ATDC.
So that's why the advance curve is set to give about 15-20 advance at idle and 30-35 at 3000 rpm.

The question then is, why not keep advancing the ignition timing after that? The speed of combustion is still pretty constant and the engine is getting faster....
And I think the answer is something to do with by the time you're igniting at much more than 35 BTDC, the amount of cylinder pressure you've developed while the piston is still rising becomes significant and a problem (detonation) , so you have to compromise and accept that peak pressure must come increasingly after 22 ATDC as the engine speeds go up beyond that.

Edited March 10, 2020 by jonbill

[Zed Head]

It's an interesting topic.

http://www.iaeme.com/MasterAdmin/UploadFolder/IJMET_06_09_005/IJMET_06_09_005.pdf

[AK260]

I think JB said it perfectly.

 

The only other thing I’be read in addition to what has been said already is that above 3kish rpm the VE increases which helps with better air / fuel mixing and not as much time required for a full burn. I could be wrong so happy to be corrected. I have to dig out the articles again.

 

[AK260]

This is a great read ...

 

http://www.s262612653.websitehome.co.uk/DVAndrews/timing.htm

 

[Zed Head]

But who is Dave Andrews?

http://www.s262612653.websitehome.co.uk/DVAndrews/

[jonbill]

The Internet has come on quite a lot since 2002.

[Av8ferg]

Well, that was educational....

[Captain Obvious]

6 hours ago, jonbill said:

 The question then is, why not keep advancing the ignition timing after that?
The speed of combustion is still pretty constant and the engine is getting faster....

Yes. That's exactly the question. And thanks for the input into potential answers.

I get the need for the advance simply because the engine is spinning faster. That's the easy "first order" part. The more difficult part is "Why is it all in at 2500? Would it be better if it went further?" I guess I'm looking for the second order kind of effect, right?

5 hours ago, AK260 said:

The only other thing I’be read in addition to what has been said already is that above 3kish rpm the VE increases which helps with better air / fuel mixing and not as much time required for a full burn. I could be wrong so happy to be corrected. I have to dig out the articles again.

http://www.s262612653.websitehome.co.uk/DVAndrews/

And thanks for that input as well. That website is a good read, but unfortunately, not much info to answer the question about "why did they stop". About the only part that pertains to that question is the table of "Establishing maximum advance requirement" - "Notwithstanding the compression ratio and other factors, the characteristic that determines the maximum advance setting is the shape of the combustion chamber and the position of the spark plug."

So it seems that the VE could be a factor. That's one of those "second order effects" I was talking about.