A crash course in ignition timing

[sicivicdude]

Most people treat timing as a sort of mystical "black box" that should only be dabbled in by the most experienced of people. The truth is, timing is a relatively simple concept that is INSANELY difficult to implement. Part of the reason for the "black box" thinking is because of a lack of information or a lack of understanding about the principles involved.

The reason for the difficult implimentation is because it's much like hitting a moving target, while moving, without using using your eyes.

Timing for a spark ignition engine is absolutely crucial. It's what makes the engine do "the work" and any in-efficiencies in the system will show themselves as a lack of "potential power".

First, the theory. Most folks assume that the piston gets to the top of the cylinder, and the spark plug goes boom and the fireball pushes the piston down. That's...... naive but the simple jist of the operation. In reality, the spark plug neither "goes boom" or waits for the piston to be at the top.

When the engine is idling (or more importantly starting) you do want the spark to happen right at TDC. That way the engine isn't trying to "turn backwards" (remember, advanced means BEFORE top dead center when the piston is still rising) and is easier to start. Kickstart engines this is ESPECIALLY true because kickbacks hurt.... a lot!

Once the engine is running, there is fuel being sucked in compressed and ignited. This is all happening VERY fast (even at idle!) and while we think of a gasoline explosion as happening instantaneously (we've all seen cousin fred light up the backyard like a roman candle by pouring gasoline on a bonfire haven't we?!) a gasoline explosion DOES NOT happen instantaneously. It takes a few milliseconds for the flame front to stretch from the spark plug gap to the edge of the piston. Maximum power involves burning as much of the fuel and oxygen as you can on every stroke. Part of what makes a rechambered head so efficient is that it centers the "charge" around the spark plug meaning it takes less time for the flame front to reach about 90% of the total charge.

In order to make that explosion (and thermal and chemical expansion) happen before the piston is falling again, you actually need the explosion to start BEFORE the piston reaches the top of the cylinder. This is called timing advance. In other words, how early in milliseconds or degrees of crankshaft rotation before the piston reaches top dead center (TDC) will the spark event have to happen in order for the flame to extert pressure on the piston crown when the piston is AT TDC.

Making the explosion start too soon will result in predetonation (the explosion reaches the piston surface before the piston is at TDC and "jarrs" the piston downward). Making the explosion start too late and the flame front will "chase" the piston as it falls.

VERY rare occasions will the timing actually be retarded (the spark doesn't occur until after the piston has passed TDC) so when someone talks about "retarding" the timing they're talking about from the factory set point (or the current setpoint of advance).

The faster the engine turns, the sooner (in degrees of crankshaft rotation) the spark needs to happen in order for the flame front to meet the piston at the perfect time. This is REALLY where timing advance comes in.... because the top end power where most people try to tune their engines.

So now we have the theory... we spark before the piston is actually in position so that the flame and piston meet at the exact moment we need.


Now comes the tricky part; implimentation.

Let's take something simple, like a lawn mower engine. What we have is a magneto ignition (power generation and timing control built into one). Because MOST "lawnmower" engines (generators, pumps, weedeaters etc etc) only really are designed to operate at one specific operation or at least only really need to be efficient at one speed (governor speed or design speed)
They only have two modes of operation, one is "start" and the other is "run". The timing is setup so that it is advanced a certain amount for the "run" setting and the flame will hit the piston at the perfect amount during normal operation (3,600 rpm in the case of a B&S engine). The starting and off high rpm operation are not optimized (they're too advanced to be perfect) but that's ok. This allows a well tuned engine at the normal operation RPM with fewer parts and MUCH less complicated ignition system. It's a trade-off.

Contrast that with modern ignition systems controls. The ignition is completely fluid (the "timing" trigger is the crankshaft position sensor and/or camshaft position sensor) and controlled by the ECU based on some VERY complicated algorithms to advance and retard the timing to optimize the flame for every operational mode. Automobile engines have to be emissions compliant AND fuel efficient (those are actually hand in hand) but have the disadvantage of having to work over a BROAD range of conditions including high rpm high load (accelerating down the entrance ramp), high rpm light load (cruising on the interestate), low rpm high load (taking off from a stop light), and light rpm light load (idle). In order to optimize, the manufacturers use the crankshaft position, rpm, throttle position, engine temperature, air temperature, and even learned settings of how the driver drives (brand new vehicles believe it or not!) These are complicated algorithms for a very complicated system.

The compromise is a fairly basic system like we have on the blaster. It's a preprogrammed amount of "delay" from the maximum designed spark advance (built into the positioning of the trigger coil versus the flywheel) calculated by engine RPM. Think of it as a 2 column chart.

One column is current engine rpm (independent of load, temperature, driver) and the other column is timing "delay" from maximum advance. The "graph" is inverted to engine rpm so that there is no delay at high rpm's and delay basically equals the total available advance at idle (so the spark happens exactly at TDC)

Turning the stator plate (the "4 degree mod") increases the timing advance across the board (4 degrees more from idle to maximum RPM).

 

[CARNZ]

Good writeup, I feel like I'm back in subaru 404 class again (intro to electronic fuel and ignition management class)

 

[Triple_B]

Wow James. Great write up. Now I just want to be clear on this...

Your saying that the stock timing of a blaster is set to "ignite" the fuel/air mixture at TDC, but there is a delay (in milliseconds) that the actual "explosion" occurs, which at the time of the explosion, the piston has already started traveling back down. Correct?

So when timing is advanced, you set the "ignition" slightly earlier when the piston is NOT at TDC, so actual "explosion" (which is delayed after the spark plug "ignites" the mixture) occurs AT TDC? Correct?

It was a lot to read haha, just want to make sure I'm getting this correctly.

 

[blasterdemon7]

Good white up but this should be common knowledge for anyone that tinkers with any kind of motors.

 

[sicivicdude]

Wow James. Great write up. Now I just want to be clear on this...

Your saying that the stock timing of a blaster is set to "ignite" the fuel/air mixture at TDC, but there is a delay (in milliseconds) that the actual "explosion" occurs, which at the time of the explosion, the piston has already started traveling back down. Correct?

So when timing is advanced, you set the "ignition" slightly earlier when the piston is NOT at TDC, so actual "explosion" (which is delayed after the spark plug "ignites" the mixture) occurs AT TDC? Correct?

It was a lot to read haha, just want to make sure I'm getting this correctly.

What I'm saying is, that the time the explosion takes to reach the piston is calculated into the delay for a given RPM. The way the blaster's ignition system works is that the maximum advance is mechanically set (for high rpm) and then the actual spark event is delayed (from maximum advance) to match the time needed for the flame to reach the piston. Flame speed is constant, the fireball grows at given a given rate. Piston speed is variable and at higher RPM's the flame needs to be ignited earlier to "catch" the fast moving piston (thus the high setpoint for advance at high rpm).

At lower rpm's the fire has plenty of time to catch the piston so there is little or no advance needed so the delay time is equal to the advance time (resulting in the ignition event happening at TDC). As the rpm's rise, less delay is needed so the ignition event happens sooner (so that the fireball can catch the piston).

Good white up but this should be common knowledge for anyone that tinkers with any kind of motors.

I would agree the theory should be common knowledge.

But, it *seems* that's not the case.... there's a LOT of confusion specifically about the "implimentation".

 

[Muggzy]

... The compromise is a fairly basic system like we have on the blaster. It's a preprogrammed amount of "delay" from the maximum designed spark advance (built into the positioning of the trigger coil versus the flywheel) calculated by engine RPM. Think of it as a 2 column chart.

One column is current engine rpm (independent of load, temperature, driver) and the other column is timing "delay" from maximum advance. The "graph" is inverted to engine rpm so that there is no delay at high rpm's and delay basically equals the total available advance at idle (so the spark happens exactly at TDC)

Turning the stator plate (the "4 degree mod") increases the timing advance across the board (4 degrees more from idle to maximum RPM).

Great write-up. So basically, if I'm reading you correctly it sounds like the timing advance is most necessary at higher RPM (top-end), and less so at low RPM? I think where you almost lost me, is where you started talking about "delay". Correct me if I'm wrong, the "delay" is from the maximum advance - you want less advance (more delay) at idle, and the most advance at maximum RPM to allow the fuel/air mix time to reach max burn before the piston heads back down? I:I

One big question I've always had, "is the advance constant over the range of RPM, or does it get bigger from low to high RPM?" and my next question is if the advance is not constant, what controls the "delay" which would be higher at low RPM and lower at high RPM (delay inverse to RPM)?

The common belief (and apparently correct) is that advancing the timing improves lower RPMs (bottom-end) power. I'm guessing your point is that the factory is tuning the advance for best timing at high RPM. and what people don't seem to get is that by advancing the timing across the board, they are improving the timing at low RPM while trading off high RPM optimization (and risking pre-ignition knock - even damage )?

 

[SU420]

hello
dfffffffffffffffffffff

 

[sicivicdude]

Great write-up. So basically, if I'm reading you correctly it sounds like the timing advance is most necessary at higher RPM (top-end), and less so at low RPM? I think where you almost lost me, is where you started talking about "delay". Correct me if I'm wrong, the "delay" is from the maximum advance - you want less advance (more delay) at idle, and the most advance at maximum RPM to allow the fuel/air mix time to reach max burn before the piston heads back down? I:I

One big question I've always had, "is the advance constant over the range of RPM, or does it get bigger from low to high RPM?" and my next question is if the advance is not constant, what controls the "delay" which would be higher at low RPM and lower at high RPM (delay inverse to RPM)?

The common belief (and apparently correct) is that advancing the timing improves lower RPMs (bottom-end) power. I'm guessing your point is that the factory is tuning the advance for best timing at high RPM. and what people don't seem to get is that by advancing the timing across the board, they are improving the timing at low RPM while trading off high RPM optimization (and risking pre-ignition knock - even damage )?

Timing advance (degrees before TDC) is necessary nearly all the way through all rpm ranges (except at start and, on some engines, at low idle)

You got the idea about the "delay". The CDI box on the blaster cannot actually advance the timing..... it can only delay the timing. So they setup the flywheel and pickup at maximum advance and then use a PIC controller (micro processor) to delay the spark advance for times when the engine doesn't need as much timing advance.

The things to remember are this: When you're talking about timing, speak in terms of timing advance BTDC (not delay from maximum advance which I have yet to discover) to be on the same page as every other ignition controller system.

ONLY talk about "delay" in a sense of talking about building or tuning a CDI box to run on an A/C system (like the blaster) and even then only think of timing as "delay" when you're talking about programming the chip to run the system.

 

[Muggzy]

Timing advance (degrees before TDC) is necessary nearly all the way through all rpm ranges (except at start and, on some engines, at low idle)

You got the idea about the "delay". The CDI box on the blaster cannot actually advance the timing..... it can only delay the timing. So they setup the flywheel and pickup at maximum advance and then use a PIC controller (micro processor) to delay the spark advance for times when the engine doesn't need as much timing advance.

The things to remember are this: When you're talking about timing, speak in terms of timing advance BTDC (not delay from maximum advance which I have yet to discover) to be on the same page as every other ignition controller system.

ONLY talk about "delay" in a sense of talking about building or tuning a CDI box to run on an A/C system (like the blaster) and even then only think of timing as "delay" when you're talking about programming the chip to run the system.

That's pretty cool. I had no idea that Yamaha was using a PIC uCtlr in their CDI box. Until I read this thread I never thought of "Delay" with respect to timing (you said it first, not me ), I'm just trying to learn a little more than the little I knew. But when referenced to the hardware advance built into the flywheel/stator timing, it makes sense. Please be tolerant of my terminology, my background is in electrical engineering so I'm just using what sounds right in a layman's head.

I've never done the 4* timing advance 'cause I don't like the idea of pushing that particular design parameter to a point where I might be risking my engine. I try to stick to the things that I have a fairly reasonable understanding of what I'm messing with which is why I'm really grateful for the education I:I. I noticed you didn't comment on my last thought though...
"The common belief (and apparently correct) is that advancing the timing improves lower RPMs (bottom-end) power. I'm guessing your point is that the factory is tuning the advance for best timing at high RPM. and what people don't seem to get is that by advancing the timing across the board, they are improving the timing at low RPM while trading off high RPM optimization (and risking pre-ignition knock - even damage )? "

Any words of wisdom on that? I seems that with a microcontroller in that CDI they could probably optimize the advance everywhere along the RPM curve.

 

[Awk08]

optimize the advance everywhere along the RPM curve.

/\/\ supposedly what the hotshot cdi's do, advance/retard it according to rpm, instead of +4 everywhere

 

[sicivicdude]

That's pretty cool. I had no idea that Yamaha was using a PIC uCtlr in their CDI box. Until I read this thread I never thought of "Delay" with respect to timing (you said it first, not me ), I'm just trying to learn a little more than the little I knew. But when referenced to the hardware advance built into the flywheel/stator timing, it makes sense. Please be tolerant of my terminology, my background is in electrical engineering so I'm just using what sounds right in a layman's head.

I've never done the 4* timing advance 'cause I don't like the idea of pushing that particular design parameter to a point where I might be risking my engine. I try to stick to the things that I have a fairly reasonable understanding of what I'm messing with which is why I'm really grateful for the education I:I. I noticed you didn't comment on my last thought though...
"The common belief (and apparently correct) is that advancing the timing improves lower RPMs (bottom-end) power. I'm guessing your point is that the factory is tuning the advance for best timing at high RPM. and what people don't seem to get is that by advancing the timing across the board, they are improving the timing at low RPM while trading off high RPM optimization (and risking pre-ignition knock - even damage )? "

Any words of wisdom on that? I seems that with a microcontroller in that CDI they could probably optimize the advance everywhere along the RPM curve.

Actually, moving the stator plate moves the timing more advanced all the way through the rpm range including the overall maximum advance (moving the sensor changes how far ahead of TDC the sensor "sees" it's trigger point, so moving the stator plate may actually be better HOWEVER, I'm not certain that the stock timing is optimized anywhere across the rpm range so simply moving the stock timing forward may not actually be optimized.

/\/\ supposedly what the hotshot cdi's do, advance/retard it according to rpm, instead of +4 everywhere

The problem is, hotshot makes a CDI that's "optimized". FMF says their pipes are optimized, vito's says their bbk is optimized. The problem is, optimized for what?

Hotshot won't give out any of their timing specs so there's no way to tell whether the CDI advances the timing in the correct place. Add to that, the fact that the trigger position (maximum advance) is not changed and the CDI box really isn't as "attractive" to me as moving the stator plate.

 

[Blaner]

ok good stuff, now i get it all but doesnt ANY sort of tuning or alteration to the blasters power delivery in the way of exhuasts etc call for customized ignition curves?

surely Yamaha designed the CDI to be optimized to the stock blaster, therefore any modifications make it inefficient?

So almost every person needs a fully customized CDI! much like a programmable ECU like I run in my pickup...

 

[Blaner]

Here is a screenshot of my ignition setup for my Toyota 4AGE 16 valve that i have in my Nissan LDV.

You will notice i have it set at 13.5* at 750 and 1000 RPM, this is to keep it idling stable especially with my higher lift and different duration cams. Then, at light throttle opening percentage, (0% & 25%) you will see that i have it set very highly advanced up to 44* this is to allow the motor to rev up easily with minimal throttle opening hence saving fuel at low speeds around town, and it works very well!

Then at bigger throttle openings, i have reduced timing to a more conventional setup from 25% - 100% this then puts the motor into power mode where timing is lined up with fuel delivery to produe a solid power curve.

This setup has taken me a long time to get to where it is today with LOTS of changing to get the delivery right, this is what works best for this motor.

As you can see, it is quite varied and there is just no way a stock blaster CDI or stock toyota ECU for that matter could produce this sort of customozed map that gives the motor what it needs to run at its best!

Here is a link to the ECU I use...Standard ECU

 

[super noober]

great write up james, true asset to the forum

 

[Awk08]

The problem is........
Hotshot won't give out any of their timing specs so there's no way to tell whether the CDI advances the timing in the correct place. Add to that, the fact that the trigger position (maximum advance) is not changed and the CDI box really isn't as "attractive" to me as moving the stator plate.

as in budget build un-attractive ??
as most of this research was done for a budget minded drag build ???

or as stated above "most people tune for top end power" < that i disagree with....
far more mid tune/trail riders on here than "drag oriented top end powaa seekers"

not saying this isn't all good info SI, because it is !
is it the best set-up for eveyone on here reading this ???
just askin, don't take this wrong, as has already been mentioned
(OMG...someone is questioning civic's research !!!!)


and of course hot shot isnt gonna give out any info. R&D isnt free.
lack of that info doesn't mean the HS CDI is bunk.
another Hot Shot CDI is on my next purchase list for some
"ass dyno" testing,
possibly a timing light set-up with the flywheel cover off to confirm advances ????

/\ got any thoughts/ideas on how to do that ?
degree wheel ?
timing marks ?

 

[rideyamaha09]

Very good info, I agree with blaner that any aftermarket parts would alter your ideal timing curve, making the stock cdi less efficient, and possibly the aftermarket cdi's as well.

I also like awk's idea of setting up a timing light and checking the ignition curves of the cdi that way. All that should need done to do that would be making timing marks on the flywheel, and observing how they change throughout the curve. I'd be very interested in seeing a comparison of the stock cdi vs the hotshot cdi to see if and how the curves would differ.

 

[Blaner]

well until such time as someone designs a aftermarket electronic plug-your-laptop-in ignition ECU that the user can alter timing at specific RPM for a two stroke, us plebs will keep reving on the trusty Yami CDI not having a clue as to what its doing

 

[Larry's Shee]

Is this the "box" that was referenced before? eBay - New & used electronics, cars, apparel, collectibles, sporting goods & more at low prices

In a perfect world you would have something like this. Dyna FS Programmable Banshee Ignition CDI | eBay

 

[sicivicdude]

ok good stuff, now i get it all but doesnt ANY sort of tuning or alteration to the blasters power delivery in the way of exhuasts etc call for customized ignition curves?

surely Yamaha designed the CDI to be optimized to the stock blaster, therefore any modifications make it inefficient?

So almost every person needs a fully customized CDI! much like a programmable ECU like I run in my pickup...

I'm sure yamaha designed the ignition curves in the stock CDI to be "on the safe side" for a stock blaster, not optimized for the aftermarket parts (in particular a rechambered head as that REALLY changes how fast the explosion reaches the piston crown. The good news is, the conservative settings on the stock CDI respond well to aftermarket parts too (at least from a "safety" standpoint)

as most of this research was done for a budget minded drag build ???

All of this research was done in a quest for the dragger build but I believe a lot of this information would apply for anyone who's gone above and beyond the stock yamaha design of the engine.

not saying this isn't all good info SI, because it is !
is it the best set-up for eveyone on here reading this ???
just askin, don't take this wrong, as has already been mentioned
(OMG...someone is questioning civic's research !!!!)

I don't mind research questioning, it's only through research and counter research that we discuss and discover anything!

and of course hot shot isnt gonna give out any info. R&D isnt free.
lack of that info doesn't mean the HS CDI is bunk.
another Hot Shot CDI is on my next purchase list for some
"ass dyno" testing,
possibly a timing light set-up with the flywheel cover off to confirm advances ????

/\ got any thoughts/ideas on how to do that ?
degree wheel ?
timing marks ?

The problem I have with the hot shot boxes is that they're guessing your state of tune about as much as yamaha did. Without the ability to change what the settings are, you're only hoping that it's setup well enough to work with your engine. If not, you've wasted money on the CDI box and not improved your power output. If you can't say anything else, you can say I'm a bit of a spendthrift. I plan on getting the most (POWA!) out of the least (MONEYZ!)

When the weather clears, I plan on getting my inductive timing light out and putting a degree wheel on my yellow blaster (stock CDI, stock timing plate 0 deg advance) and "mapping" the stock ignition system. I want to figure out what's in that "black box" and now that I know HOW it works, I can see how me changing something would affect it (before my engine finds out the hard way and goes BOOM! LOL).

The yellow blaster will probably stay at stock timing stock CDI box for a long long time (it's already at the limit of predetonation, 163 psi on pump gas ) but I'll use it as a "springboard" for my research into ignition control for the dragger.

well until such time as someone designs a aftermarket electronic plug-your-laptop-in ignition ECU that the user can alter timing at specific RPM for a two stroke, us plebs will keep reving on the trusty Yami CDI not having a clue as to what its doing

They actually already have a "plug and play" CDI box out there, the MSD 4217. Unfortunately, it's quite expensive and I'm not prepared to spend $325 on a CDI box (remember, POWA vs MONEYZ)

There is another option though.... the Ecotrons system is a small standalone EFI unit with a programmable CDI controller built in. The system is designed to operate small engines (lawnmowers to 800cc single cylinder motorcycles/ATV's)

I contacted the company about mixing and matching parts on their system and they quoted me $699 for a fully setup EFI system ready to deliver the fuel and air to the dragger. Little steep for a $1k budget build so I've decided to wait.

Is this the "box" that was referenced before? eBay - New & used electronics, cars, apparel, collectibles, sporting goods & more at low prices

In a perfect world you would have something like this. Dyna FS Programmable Banshee Ignition CDI | eBay

The hotshot CDI you posted first is "the box".

 

[rideyamaha09]

MSD makes a universal programmable cdi for single cylinder engines that would work with the blaster, but its around $300. There are quite a few other programmable cdi's available listed for bikes like yz 125's which should work since they could be reprogrammed. They all tend to be around 300 bucks or more though.