Distributor vs Coil-on Plug

[sonikaccord]

TLDR: COP gives almost 10 times more dwell time than a distributor based ignition on a 4 cylinder engine. Should a stronger spark at high rpms give more horsepower with all else being equal? Why not upgrade to a COP system if it was cheap and easy to install?

I need someone to check my math really quickly. I have an idea, but first I need some baseline numbers. This is going to be for the distributor ignition system of a 4 banger. I want to compare it with individual coil based iginition. So here we go:

For an engine rpm of 8000, the camshafts will be spinning at 4000 rpm and the dizzy will be firing 4 times per cam revolution. That's 16000 ignitions per minute, 268 ignitions per second, or an ignition event every 3.7 ms. Each cylinder will be firing 67 times every second or one firing every 15 ms. For our distributor based systems, that equates to the coil getting less than 3.7 ms of dwell at 8000 rpm. From scanndanner's YouTube page, it looks like our stock coil saturates at about 5ms? or was it 8ms? Meh, I'll look it up later.


Now let's take it one step further. The stock distributor rotor radius is 30mm, while the arc length of the metal "tab" is 20mm. The contact points inside the rotor cap are 5mm wide effectively making the metal "tab" arc length 25mm. This equates to 47.7 degrees of ignition contact time. Our engine is still spinning at 8000 rpm, the camshaft is covering 24,000 degrees per second. A 47.7 degree long tab will pass a single point 503 times per second. (That's 2012 total points per second if you are still counting, 4 points in the dizzy housing). To hit a point 503 times per second, our metal tab will be contacting that point for a total of 2 ms. From the other paragraph above, it takes 3.7 ms for our rotor to travel 90 degrees. Of that 3.7ms travel time, it will be contacting a point for 2 ms. That leaves us with 1.7ms to charge our coil for the next firing event. (I checked my work here, 90 degrees travel between points with a 47.7 degree tab gives us 42.3 degrees of empty space before the next point. Traveling at 24000 degrees per second, that leaves us with 1.76 ms before reaching the next point plus the 2 ms contact time equals a total of 3.7 ms, it really should be 3.8...my numbers may be off by one or two thousand ms)

For individual coils, assuming a spark duration of 2ms for simplicity and comparison, each coil gets a whopping 13ms to recharge for the next spark event. The next ignition event will have maximum spark amplitude and duration.

[F22Chris]

Well said. The only concern I see is for an F22a, you would need a stand alone management system, toothed crank, figure out crank sensor mounting, etc. unless there's another way to get a trigger?

I know this is about COP, but what about a wasted spark setup? Then one would be able to run stuff like MSD instead of a full on aftermarket ecu, right?

[sonikaccord]

I'm going to make another post on the stock triggering, as well as some MSD "claims." I may need to change the title depending on how far this goes.

But for now, wasted spark is essentially a halved COP system for us, so we would get 6.5 ms of dwell time per coil, 2 coils instead of 4 and no distributor required. I can include this setup in the trigger post as well. Everything else is still the same as far as the math is concerned. Just divide by two from the COP setup.

[deevergote]

This is all far beyond me, but I'm curious to see what comes of it!

Since our distributor systems are effective even with turbo and high-revving NA applications, would there be any genuine benefit to a COP setup? Would the improved performance actually help, or would it be beyond the needs of all but the most highly modified engines?
My understanding of this stuff is limited (hopefully I'll learn a thing or two), but it almost seems to me like an omelette maker that can bench 1000lbs. Sure, it's impressive, but he would have the same ability to crack eggs with or without that level of strength.

[sonikaccord]

Yeah, I think this may get a bit more involved than what I imagined.

That is the question that I want to answer objectively. I have read numerous times that the stock ignition system is fine for high-output turbo and high-rpm operations. But how do we know? It may very well be ok, but what if it can be optimized? What if there is a bit more horsepower left on the table? Can a COP setup help the casual driver gain more low end power and improve gas mileage?

From what we see so far, there are a few benefits to going to COP:
No distributor needed
Longer potential dwell time
A smaller coil can be used to minimize the dwell time necessary to achieve maximum output from the coil.
More spark timing control.
Durability/reliability?

The negatives are:
more expensive?
Difficult to implement/retrofit with stock ECU
Hopefully we can eliminate both of these limitations.

That's a good analogy! But would he be able to crack ostrich eggs better than quail eggs? Will he run out of strength or become less efficient as the egg cracking difficulty increases? How much better will he perform vs a standard omelette maker in the same scenarios?

[deevergote]

Good flip of my analogy! I suppose it's definitely worth investigating. I mean, many modern cars use COP, so there must be a reason for it. If there's a way to get more control, more power, or even just greater fuel efficiency, it'll be a worthwhile thing to do.

I know it's been discussed before, but your first post is the most in-depth exploration of the topic I've seen yet on here.
(sorry, typed this after your first reply, and forgot to send it... I got distracted by learning how the coil pack on my Miata works, as that's semi-relevant!)

[GhostAccord]

Great write up you have going here. It's nice to see people still doing investigative work on these older systems.

I've run a few COP setups using the stock Honda cam and crank VR sensors and a MagaSquirtIII EMS with great success.

[sonikaccord]

The DIYAutoTune people are a smart group of guys! They actually aren't too far from where I stay. I'd love to have lunch with a few of them.

[F22Chris]

Well isn't another purpose of this to get more dwell for a stronger spark? You have the coils, but they'd be putting out the same energy right?

[sonikaccord]

I wouldn't say more dwell, but rather optimal dwell. If we do a COP setup, we can run smaller coils that need less dwell to achieve the same output energy across the entire RPM range. That really needs to be my next calculation, the stock output energy, voltage, secondary current, etc.

I also thought about going to the junk yard and grabbing 4 of the v6 accord distributor system coils, 4 ignitors and have an unnecessarily overpowered ignition setup...just because I like excess...and I think you all like to see excess too.

Single coil part numbers:
TC-08a Honda internal coil
TC-19a Honda external coil
TC-24a Honda V6 external coil

[sonikaccord]

I split the threads.
For all the previous information, refer to this one:
http://www.cb7tuner.com/vbb/showthread.php?p=3271469

Staying on topic with this thread:
I feel the next comparison and calculations should be spark energy for both setups. Is more, better? Do we really need a longer spark duration at high rpms?

I found some guys from down under that give some coil testing tips. I think I can do it without shocking myself too badly.

http://www.dtec.net.au/Ignition%20Co...alibration.htm
http://www.dtec.net.au/Ignition%20Co...%20Testing.htm

Why is dwell important:

Inductive ignition systems store their energy in a magnetic field and the correct dwell will allow this field to reach its maximum strength within the design limits of the coil.
The energy stored is measured in Joules and is ½ the inductance x the current squared.

[sonikaccord]

Measured inductance and resistance of primary and secondary. LCR set to 1kHz for inductance and 0.1kHz for resistance. Resistance double checked with Fluke DMM. They are almost exact.
Calculator used:
http://www.megamanual.com/ms2/configure.htm#dwell

Internal coil
Primary
Resistance: 0.77 Ohms
Inductance: 5.18 mH

Secondary
Resistance: 17 KOhms
Inductance: 32.5 H

Dwell(calculated):
Amps Time(ms)
1 0.45
2 0.92
3 1.44
4 2
5 2.6
6 3.27
7 4.01
8 4.84
9 5.8
10 6.9*
11 8.23
12 9.89
13 12.09
14 15.38
15 22.09

V6 External Coil
Primary
Resistance: 0.434 Ohms
Inductance: 3.26 mH

Secondary
Resistance: 19.75 kOhms
Inductance: 80 H

Dwell(calculated):
Amps Time(ms)
1 0.28
2 0.56
3 0.86
4 1.17
5 1.5
6 1.84
7 2.19
8 2.56
9 2.95
10 3.36
11 3.8
12 4.26
13 4.75
14 5.28
15 5.85
16 6.46
17 7.12*
18 7.85
19 8.66
20 9.56
25 17.15

External Coil
Primary
Resistance: 0.94 Ohms
Inductance: 7.7 mH

Secondary
Resistance: 20.27 kOhms
Inductance: 94 H

Dwell(calculated):
Amps Time(ms)
1 0.67
2 1.4
3 2.19
4 3.08
5 4.07
6 5.2
7 6.51*
8 8.07
9 10
10 12.53

Asterisk represents where I would cut the dwell time as a maximum value for that particular coil. Hopefully I can get some primary current and voltage waveforms soon to compare the calculations to. Maybe even secondary(high tension) waveforms!

I manually attempted to switch the igniter, and ended up with some nasty waveforms due to the bounce of touching wires. I have a function generator at work, but I don't want to lug that thing around, and I want to be somewhat discrete. I'm thinking about using my old phone as a square wave function generator with the headphone out set to 67Hz = ~15ms. Should give me a clean picture. The duty cycle may have to be adjusted to fine tune the dwell because of the bench power supply. I'll take a pic of my setup too! I'm not working with anything fancy.

[sonikaccord]

I have posted some of MSD's claims as well as my arguments below. I'm assuming MSD is talking about distributor based ignitions, but I'll try to argue both sides of COP and Dist.

Inductive ignition systems are used due to their simplicity and inexpensive production. For stock applications these ignitions are adequate, but when it comes to gaining performance, factory inductive ignitions fall short.

The questions to be asked here are Why and how do inductive ignitions fall short? We'll try to answer these below.

The primary weak link of a stock ignition is because the coil serves double duty. The coil must act as a step-up transformer to create a higher voltage spark, plus it needs to store this power until the ignition is triggered. As engine rpm increases there isn’t enough time to completely step-up the voltage before the ignition is triggered resulting in a weaker spark. This low voltage spark may not be enough to light the fuel mixture in the cylinder which will result in a misfire and loss of power.

MSD is saying that coils are the weakest link because they have to step up the voltage AND store that energy until it is ready to fire. Well, by nature ignition coils with a Primary and Secondary winding ALWAYS do that by default. It's not a weak link. It's a natural physical process that can't be changed, similar to taking a dump. MSD's CD system relies on only the step-up part and not the energy storage part of the ignition coil. We will talk about that later. As explained above, the lack of "step-up" timing is true for Dist but not for a properly designed COP.

A CD ignition, like an MSD 6 Series, is capable of producing full power sparks throughout the entire rpm range. It draws its energy directly from the battery where a custom wound transformer steps it up to over 460 volts. This voltage is then stored in a capacitor until the ignition is triggered. At this point, all of the voltage is dumped into the coil where it is transformed into even more voltage, anywhere from 30,000 - 45,000 volts depending on the coil, which is sent to the distributor and finally to the plugs.

A COP system is also capable of delivering full power sparks throughout the entire rpm range. A Dist system may not be able to, again as explained above. Instead of having the ignition coil's primary winding store the energy in a magnetic field, MSD takes a more direct approach and stores the energy in an electric field. This is good because a capacitor can be dumped and recharged very quickly. This is bad because most 600 volt rated capacitors have a small capacity(affects spark duration) and it takes a DC-DC converter circuit(adds complexity) to step up 12 volts to 400+ volts and then that 400+ volts is sent through another ignition coil essentially.

Conventional CD ignitions supply one spark of intense energy but for a short duration (time). An MSD uses multiple sparking technology to take advantage of the quick rise time and power of a CD Ignition by producing a series of sparks. More sparks equal more heat in the combustion chamber resulting in complete combustion of the fuel mixture which produces more power. At lower rpm, there are many benefits to multiple sparks including a smooth idle, improved throttle response plus the spark series prevents fouling plugs or fuel loading up in the cylinder when air/fuel distribution is poor.

The multiple spark series of an MSD Ignition Control lasts for 20° of crankshaft rotation. At lower rpm, 1,000 rpm for example, there is plenty of time to fire the plug a number of times to ensure ignition of the fuel mixture. As rpm increases, the piston travels up on the compression stroke faster resulting in a shorter amount of time available to fire the plug so the number of sparks that occur decreases. By about 3,000-3,300 rpm, there is only enough time to fire the plug once. From about this rpm range on, an MSD ignition control delivers one intense, full power spark.

MSD uses more sparks because of the short pulses that are inherent in the CD ignition system. They have to or they would probably suffer with poor low rpm performance. With a strong coil, a longer spark duration is achievable with both COP and Dist. COP, and technically dist, can also have a multi-spark system. I think BMW and Mercedes have been doing it for a while, like since early 2000s

As a bonus, I found this on the corvetteforum (through google search, I don't own a Corvette...have thought about though )

A spark igniting the compressed fuel/air mixture in the cylinder is a problem in probability. The plasma (spark) must come into contact with gasoline/air mixture in the correct proportion and only as a gas in order to cause ignition. The longer the duration of spark, the higher the probability that the cylinder charge will become ignited. If you look up the spark duration for capacitor discharge ignition systems you will find them even shorter and this is why some CD systems have multiple spark so as to reduce misfires. You have found yet another reason why an aftermarket coil may not be as desireable as the OEM coil !!!!!!!

[sonikaccord]

I can't use my phone as a function generator like I first imagined. On to plan B. Building a custom circuit to create an adjustable dwell time. Based on the calculations I need a circuit that can trigger up to 10ms pulses. Enter the 555 timer IC. If you don't know about this simple IC, just know that it is the standard when it comes to anything involving time or counting time.

My only issue to work around is that I can't have a normally high output. The ic is normally low and goes high when triggered, which means if I connected directly to the ignitor, it will turn it off when triggered. I need the opposite. I may have to add a component or two extra... Shouldn't be that big of a deal. That's where I am so far.

[sonikaccord]

Pretty much said F it, and go astable mode. I'll use my scope to single capture the coil primary voltage. If I can find a good resistor, I'll try to get current too. I'm not messing with the secondary measurements just yet. I'll need some solid 800v zeners for that side of the circuit.

I have my 555 set up to run at 50% duty cycle at an adjustable frequency. An ignition event shouldn't last more than about 3 ms, but I have a potentiometer to adjust "dwell time" just in case. It's technically adjusting frequency, but for this testing set up it won't matter much.