The "Franken-Tester"

[browning]

I have made several additions to the test rig that I have been playing around with this winter and have more in mind. I have added an AC volt/ammeter to the inverter supply to make it possible to monitor the power consumption with changes in flywheels weights and magnet types. One toggle switch allows me to change from operating the coil box to operating one or two light bulbs or turning them both off and comparing voltage with load or no load with a more accurate digital voltmeter. The variable speed VFD along with the digital tachometer makes checking with various speeds possible to within 10 rpm. I have added the spark plugs gaped to .073 to simulate the operation under compression mostly to find out how slowly the motor needs to operate to provide adequate magneto voltage for starting. This week I intend to add an indicator/pointer to the magneto frame and make two index plates attached to the flywheel at 180 degree opposition which will make it possible to use an induction timing light to verify timing accuracy of coils and timers at various speeds. I am curious to see how ECCT timed coils compare with Strobospark or HCCT ones in that regard. I think I will be able to check the accuracy of Anderson type timers and make adjustments much more easily than with installation on a car. I have found out that too much magneto voltage turns the coils into fireworks displays at just around 1000 rpm and I am certain that operation like that would destroy the points in short order so I am looking for the "sweet spot" to make low speed starting as good as possible while preserving the points at higher speeds.

[TXGOAT2]

CORRECTION: Adding a choke to the mag output wire would limit voltage at higher RPM while leaving it about the same at lower RPM. They were used on late model non-starter Ts to regulate HEADLIGHT voltage, but NOT the ignition supply voltage)

** A CHOKE IS NOT NEEDED FOR THE IGNITION CIRCUIT**** See below.....

[browning]

I have studied that option some but, quite frankly, lack enough understanding to spec a choke to do the job. Any suggestions are very welcome.

[TXGOAT2]

Actually, a choke is not NOT needed for the ignition coil circuit. The coils' points will open when the applied voltage reaches a certain point, and how much higher the available voltage goes once the points are open is moot. The headlights and other bulbs are a different matter, since they have no points to open and no significant inductive reactance. The rising magneto voltage as engine speed increases will force more and more current through the headlight bulbs until their life is shortened or they burn out. Adding an inductor in series with the headlight circuit can prevent too much voltage being applied to the lights at higher engine speed, since its impedance will increase as the frequency of the magneto's AC current increases with engine speed.

[Mike Silbert]

I am not sure I understand your setup and issue correctly in spite of reading it 4 times.

Your setup has a VFD running an electric motor turning the crankshaft and you are testing magneto and flywheel performance and setup??
You measure output with gauges , coils, and bulbs is this correct?

If I got that all correct then I have some comments.
Magneto bulbs have been a problem since day one due to the huge variance in voltage over the RPM range.
Over the years all kinds of things have been tried so I will be watching to see what you do and how well it works.

My concern is you say @ 1000rpm the coils (and I assume the spark coils) turn into a fireworks display.
1000 rpms is 25mph on a stock rear ratio. 1800 or 2000rpm is the rated max and you are no where near that.
I don't think anyone, including me, drives their T never to exceed 25 miles per hour.
They were designed to be hooked straight to the magneto and drive @ 30mph, up to 35mph all day long.
I will say I don't watch the coils while tooling down the road so I can not see what is really happening.
But the coils live untouched for years and thousands of miles rarely doing anything but working fine.
There seems to be something different here.
Maybe someone can make and post a video of a stock Model T coil box tooling down the road to see what really happens.
I am curious to know what really happens myself.
It probably would be best to video a car with a firewall coil box and a cowl to limit sunlight for maximum visibility.
I can not do this, my cars are put away for winter and I don't do videos.
There are several T owners that post lots of videos maybe one of them can do it?
Maybe we can get a video of your fireworks display?

I don't know your setup but in a stock car the coil fires 2 or 3 times then close to 2 cycles of the crankshaft before firing again.
The timer limits the coil duty cycle of each coil to somewhere sorta about 10% on, 90% off.
I am sure someone can calculate the exact number but it changes depending on which timer is used.
It's well documented that 100% duty cycle will overheat a coil.
So how are you controlling the coils? Timer? Always On? Other?

I love head to head comparison tests with data, I would like to see what your tests tell us.
Mike

[TXGOAT2]

Coil capacitor problems will cause excess arcing at the coil points. As far as I know, the coils don't much care what the supply voltage is, be it 6 volts DC, 12 volts DC, or 4 to 30 volts AC, so long as they are adjusted correctly, the plug gaps within reason, and the capacitors are good.

[browning]

Mike, Your assumptions are almost entirely correct. In the interest of clarification I will restate that the genesis of this project was to create a mechanism which would allow me to test various versions of rare earth magnet replacements, primarily as an experiment. There are many options for rare earth magnet types (high temperature neodymium and samarium cobalt) and many sizes and shapes, each with their own properties with respect to pull and polarity. At the same time there can be any clearance built into the system. The Ford spec is in the range of .025 to .040 but with stronger magnets that clearance can be increased to make adjustment less critical. So the problem is matching the proper magnet with the proper clearance and it seems that trial and error is the best path to ascertain that combination. It occurred to me that I couldn't build a new engine/transmission assembly and install it to test all of the possibilities so I found the junk engine block, crankshaft, and camshaft and readjusted the bearings for proper fit. I turned the block upside down, mounted it on the frame, and drilled the main bearing caps for oil cups. Now I can change flywheels and modify clearances at will and in just a few minutes. Measuring voltage is done either with a digital multi meter or with the AC voltmeter pictured. When the ignition coils are activated by the installed timer they fire the four spark plugs the same as if installed in a car but the digital meter doesn't like the EMF generated by the coils and reads erratically so I can shut off the power to the coils and divert it to the light bulbs, either one or both. All I am attempting to do with the light bulbs is create a load on the magneto that would simulate the coils, but without the interference which makes the voltage measurement go crazy. I have built three flywheel/magnet assemblies with various magnets and can create an output that mimics the stock flywheel very closely with about a .200 clearance and by reducing the clearance can increase the voltage substantially. One of my flywheels (1/2" by 1/2" samarium cobalt magnets) is the one which causes the coils to do the fireworks display so I have disqualified that as a workable option. I think the next iteration will be the one which I will have enough confidence in to build into a working engine. I am shooting for a combination that provides higher low speed (cranking speed) voltage without exceeding a safe voltage for the coils and without robbing the engine of more power than necessary. It has been recommended to me by the gentlemen on the forum to use a choke coil to reduce the higher voltage and I am looking into the best way to test that. Today's experiment is to create a pointer and two index plates on the flywheel so that I can confirm timing from coil to coil. I am not trying to influence anyone or trying to sell anything, just passing the time and staying out of my wife's hair. I appreciate your interest and wish you well.