Understanding ignition dwell and the Model T timer

[Been Here Before]

I am not certain how to present this question and not feel dumb.

Dwell is defined as - Dwell angle is the amount of time, measured as degrees of rotation, that contact breakers close in a distributor. Unless dwell angle is accurate, ignition timing won't be accurate.

The discussion on the benefits of setting the dwell of an induction coil with the ECCT, is all four coils will share the same dwell to allow for better performance.

Using a traditional timer, can dwell and performance be effected.

Is the timer roller, not acting as a contact breaker?

[Norman Kling]

Using Model T coils, the dwell would be the time it takes for the current in the primary to build up enough to cause the coils to vibrate. They should be adjusted to approximately 1.4 amps Which can be determined with an analog ammeter. I made a box in which I place a coil. with a spark gap on one side connected to the high voltage lead from the coil. The primary is connected to 6 volt battery and the ground (connection which would go to the timer) is grounded to the chassis. I lay this box on the running board and I can connect the primary to either the battery connection on the terminal block or to the mag connection. So I can check the coil on both battery and magneto. I adjust to 1.4 amp to get the coil to spark. On battery the current will begin to flow when the rotor makes first contact with the timer. So if this point is off, it might have uneven dwell when running on battery unless you replace the timer. When running on magneto, the current will rise with the speed of the engine, so it is kind of an automatic spark advance on magneto. Probably why the cars run faster and smoother on mag than on battery.
The most efficient test would be on a hand cranked coil tester, and if you can find someone who has one, use it. Or a tester which can be had from some of the parts stores. One such tester is called Strobospark. It is said to be good. I have not used one.
My method works quite well and my T's run smoothly.
Norm

[speedytinc]

I am not certain how to present this question and not feel dumb.

Dwell is defined as - Dwell angle is the amount of time, measured as degrees of rotation, that contact breakers close in a distributor. Unless dwell angle is accurate, ignition timing won't be accurate.

The discussion on the benefits of setting the dwell of an induction coil with the ECCT, is all four coils will share the same dwell to allow for better performance.

Using a traditional timer, can dwell and performance be effected.

Is the timer roller, not acting as a contact breaker?

YES. Having coils tuned that each fire @ exactly the same time is one large part.
Whatever timer you run must fire exactly every 90 degrees as well.
One minor effector is the centering of the timing cover. Mathematically insignificant on its own. Possibly critical if it effects roller bounce.

Anderson type timers can be fine tuned by contact tab bending.

New day timers can have timing error due to their contact locations during the molding process. This can be cured by machining accurate stop & start points on the contact pads. (this was my correcting method)

Roller timers should be pretty accurate,(assumed) but I have not done testing. Their issue is needing a smooth running path with no bounce.

I assume the electronic timers are designed for perfect firing. E & I timers

[Craig Leach]

For the most part points & timers are simular but very different. A timer completes the circuit so the coil will create a spark during the time
the circuit closed. Single point ignition the points close the circuit to complete the primary coil winding to create a magnetic field then open
the circuit to collapse the magnetic field into the secondary winding creating the spark. So dwell is the time the primary coil has to biuld the
magnetic field. When the points open is the time that the spark occurs. In a timer the time the contact is made is dependent on the design of
the timer and not adjustable but as I understand is effected by speed & vibration. Points are effected also by speed but that usualy doesn't
accure below 5000 rpm. Or thats the way I was taught.
Craig.

[Ron Patterson]

Craig
You might find thisaricle of help with understanding the Model T Ford Ignition System and Spark Timing.
Ron Patterson

[Jerry VanOoteghem]

Craig
You might find this article of help with understanding the Model T Ford Ignition System and Spark Timing.
Ron Patterson

Ron,

Thanks for once again posting this informative article! It supports a comment which I earlier made that was readily refuted. Since I don't wish to make inaccurate statements, I am thankful for this resource that reassures my thinking, (and hopefully that of others).

[Larry Azevedo]

This thread has a lot of interesting observations, questions and comments. At the risk of motivating a lot of interesting feedback I'm going to chime in with observations of my own on coil performance. Obviously, one must have the cushion spring set to about 0.005 in and the point gap to 0.030 to start. Next the points must be clean and meet in good contact with minimal (or none) light observable when looking through the closed points. Capacitor must be good as well and all contacts clean. I have used the ECCT extensively, and as noted below, the computer interface is important to have.
'
When using the ECCT with the computer interface, 100% in a given histogram is not necessarily the most stable Time to Fire (some in this thread call this dwell but I tend to use that term to apply to the rotational time that the timer is in electrical contact with ground. Nevertheless, I will use the term TTF (Time to Fire) in the following. A jump of 1 degree in the histogram on the ECCT at 1000 rpm represents a time of 0.166 milliseconds since the crank is turning at 6 degrees per millisecond at that rpm. Therefore, the time between 2 histograms is 1/6 or 0.166 ms. That means that differences inside that range MAY NOT be counted by the ECCT software as an assignment to an adjacent histogram. One can actually get, for example, 60% at 0 degrees and 40% at 1 degree and that coil could be more stable than the 100% histogram.

The easy way to really determine the stability or consistency of the coil TTF is to shoot a number of single shots on the ECCT, recording each time to fire value and then calculate averages and standard deviations. Really stable coils can have stabilities on the order of 0.01 ms, well within the width of a single histogram. When doing this you will see evidence of arcing as some values abnormally long. Arcing is a problem because it changes the TTF and therefore the time the spark plug will fire. Arcing on the order of 0.2 ms can be common, that extra time relates to an error in ignition timing of 3 degrees at Montana 500 speeds (2500 rpm, 61 mpg and 15 degrees/millisecond for the crank). Arcing can be caused by a number of factors including too weak cushion spring, poor point contact and pitted points. The idea to reduce or eliminate arcing is to get the coil points to separate QUICKLY. I also have a benchtop system where I can directly observe the current buildup on a coil and the ultimate ignition. With scope the arcing is easy to see.

Another important issue is that after making a coil adjustment there is a 'settling time' that has different time scales. Immediately after an adjustment it takes the coil a few minutes for the wood mounts and the metal to relax a bit which changes the TTF. To get good stability results (for competing in the Montana 500, for example) one must monitor coils for several weeks to months and then treat them very carefully! The coils I used last year in the 500 were monitored for a couple of months and NO adjustments were made during that time. Obviously, one needs a number of coils to find 4 really stables ones.

Another important fact is that even if you tune 4 coils to be identical on the ECCT, that does not ensure optimal performance at high speed. Note that the ECCT operates by driving the coil at 12.4 volts, this usually results in a time to fire on the order of 2 milliseconds. Coil time to fire depends on the driving voltage and TTF decreases with increasing voltage. In a running engine on magneto at high speed typical TTF's are on the order of 0.8 to 1.2 milliseconds. If you are running your car ignition with a 12V battery then your tuned coils from the ECCT will perform well. However, if you run on magneto with its varying voltage then the time to fire at high speed can be different for the 'matched' coils. Therefore, to obtain optimal performance all coils should have very similiar behavior with respect to TTF at various voltages.

I need to remind everyone that the information above most likely will make minimal difference for your touring experience! It is important, though, if you are trying to squeeze out a fraction of a horsepower at high speed.

I will be giving a presentation at the Winter Workshop at McPherson College on January 20, 2024 discussing the above and also timer performance using my benchtop measurement system which fully simulates the Model T ignition system running on either battery or magneto. I will also bring my full test system to demonstrate. Additionally, I am planning to bring the system to Chickasha in March and can test your coils and/or timers on the system for free! Hope to see you!
Larry Azevedo