Re: Ignition Analyzer?

[Tom]

Here’s a little more information. Your mileage…, my two cents…, and all the other caveats apply:

The early Merc-O-Tronic and Stevens testers (like the one in the Johnson service manuals from the ’50’s) had a vibrator tube to generate a series of voltage pulses. These pulses were fed to the primary side of a coil. Wires were attached to the ground and the secondary (or both sides of a dual output coil) and then to a spark gap. With the pulses applied to the primary, sparks should be generated by the secondary. These testers had the added functionality of being able to adjust the amount of current applied to the coil, and a meter to read the amperage. Each coil has a current rating at which it should work, so you dialed to the lowest current setting at which the coil fired. If the value was at or below the specified value, great. If it was a lot more, the indication was that the coil could be on its way out. The coil specs were listed in the manual that came with the unit, and specified manufacturer and model number of the motor and/or the coil.

These testers also had a 110 V circuit for testing condensers. Some condenser testers will tell you the capacitance of a condenser, some will tell you leakage. These can be useful for some applications, but not necessarily for magneto diagnostics. The most useful test is to apply a high voltage across the condenser and measure the leakage. These testers had a neon tube that flashed when the condenser discharged. If it flashed fast, condenser was bad. If it stayed on, condenser was bad. If it flashed slowly, the condenser was probably OK. The difference between what the Merc-O-Tronic and Stevens testers did, and what many testers do today, is the voltage applied to the condenser. They really need to be challenged at higher voltages than the ones generated in a $15 tester.

Later models had electronic circuits to generate the pulses, since the vibrator tubes had points inside, which tend to corrode. I’ve brought back a handful of the vibrator tubes by opening them up, cleaning the points, adding some desiccant, and sealing them back up.

The Merc-O-Tronic had a neat feature, a separate probe for measuring voltage leaking from a coil. This was a grounded probe, and you passed the tip near various locations where secondary voltage might be leaking. It could find problems with insulation of the coil, the wires, or any part that could be a problem, like the insulators in the early Eisemann mags and the Johnson alternate twins that fire from one coil. I guess you could do the same thing with your finger if you held the Stevens unit in your other hand, although this would not be recommended practice.

Another way to generate the pulses is to put the primary of the coil to be tested in series with the primary of a Model T coil and a 6 volt battery. The primary voltage of the Model T coils is switched on and off as its points open and close. Since the coil to be tested is in series with the Model T primary, that one is getting switched on and off too. The problem with this method is that you can’t vary the current below the operating current of the Model T coil. But just seeing if the coil under test can generate spark is usually enough. You need to set up a gap for the Model T coil secondary, too, so you don’t burn that out.

If you have a Model T coil set up to test coils, and you want just a decent condenser tester, look into an old Heathkit C-2 or C-3 condenser tester (or the equivalent Knight, PACO, or EICO capacitance bridge testers. They measure capacitance and test leakage at high voltages. My re-capped C-2 is just about bullet proof, and the magic eye tube is pretty cool. Don’t buy the ones that say "In-Circuit" testers, those are a different animal.

T