A Novice's Magneto Charge

My 1915 runabout was like a lot of other Model T’s.  It ran better on magneto than on battery, and there was enough juice to light the headlights, but I could never start it on MAG as Ford intended. It started easily on the first pull of the crank when I cheated and used a battery, but I could pull all day on MAG and the only result would be lots of exercise. This was not a big problem for me because battery starting was so easy, but when some other needed work prompted me to pull the engine it was a good opportunity to do a magnet charge while I had the engine out. So here’s a story for novices, by a novice.

I had never done this before, so the first thing was research. I read about several approaches to skinning the magneto-charge cat. All of them take multiple batteries in series to produce 24 or 36 volts. Some folks use a DC welder. The easiest method is the basic in-car charge. It involves only hooking up jumper cables in the right way.  The main tool required is a compass. This was covered by Dan Conder in the November/December 2010 Vintage Ford. It’s also in the excellent MTFCA book, The Electrical System, which also gives instructions for the other methods.

The simple in-car charge is enough for lots of folks, especially when their magnets are so weak before charging that it makes a big improvement in the car’s running. But I knew from my reading that other methods produce a stronger charge, which is what I wanted. The basic in-car charge uses the coil ring as the charging electromagnet, and as it’s used in the vehicle it doesn’t come in direct physical contact with the magnets. It’s close, but doesn’t touch. I’ve seen different estimates of the strength of the charge. Some say the simple in-car charge gives the magnets 60% of  the charge produced by full contact. I’ve also seen a claim of 80%.

The next method is a variation on the first. It’s also done in the vehicle, but with the hogshead removed so a charging coil can be put in direct physical contact with each individual magnet. Of course, this gives a stronger charge. You can test the magnets by placing a piece of steel on them and using a fish scale to pull it off. When it takes two pounds of pull, you have your charge.

If the engine is out and the transmission is off, you can lay a field coil on the magnets to charge them. If the transmission isn’t off, you can unbolt the ring from the block and move it back against the magnets. If everything’s apart, and the magnets are off the flywheel, you can charge them individually with a charging coil.

As my transmission worked just fine and appeared to be in good shape, I didn’t see any reason for complete disassembly. I decided on charging the individual magnets with a charger made for the purpose. I based my charger on the one Jim Golden posted on the MTFCA forum. The important part of the device is the electromagnet. I used a core of one inch hot rolled rod six inches long, and wrapped a hundred feet of 12 gauge wire around it with electrical tape to keep the wire in place.

I mounted the coil on a wood base with aluminum hardware held in place by brass wood screws. A pair of toilet float rods with brass nuts hold the coil on the base. Pieces of aluminum flashing folded in three layers keep the rods from digging into the coil. I don’t know if all that non-ferrous stuff is necessary or helpful, but that’s the way I did it.


The control box is very simple. In fact, the one I made could have been even simpler. I included an ammeter as an indicator to show me that the switch was working properly. That proved a waste of effort. When the current is applied you can see the charger jump against the magnet. Even though the core is touching, it moves to make full contact, so the ammeter is superfluous.

The working part of the control is a starter switch I bought at a swap meet for $2.50 with a lever across it to make it easy to push down. The rest of the charging set-up is just wires to the coil and the batteries, and cables connecting the batteries together.

The additional tools are a test weight (I didn’t have a fish scale) and a compass. The traditional recommended weight is an iron piston, said to weigh two pounds, but I didn’t have a loose one. So I took a perch with nut and a spindle to the post office and had them weighed. The perch was a little over a pound, but the spindle was 2 pounds 2.1 ounces. That would do. For a compass I just zapped a nail with the charger a few times and hung it with a bit of brass wire and a thread.

Some may find this next part unnecessary, but I’ve had bouts of senioritis all my life. Not wanting to lose track of what I was doing in an absent-minded moment, I used the compass to identify all magnetic poles, and marked them with chalk. As we learned in grade school, opposite poles attract. The pointy end of the nail aimed north, so that was its south pole. All the poles that attracted the point got an N, and the ones that attracted the blunt end got an S.

                  I also identified the poles of the charger and marked them with a felt pen.

Next came testing. I happen to have an old magnet that’s been here on the farm since I was a little kid. Until I got into Model T’s, I never realized what kind of magnet it was. It had only enough magnetism left to pick up a few small nails. After several zaps with the charger it would hold the perch, and with a few more it held the spindle. I was ready to move on to the real thing.

The first day I used just a single twelve volt battery. I had read that the magnetizing is done not by a steady current, but by repeated fluctuations between full current and none. So I applied the current with a dozen quick taps of the lever. In this first session I set the pole of the charger against each pair of magnet poles, south pole of the charger against a pair of north poles, then the north pole against a pair of south poles, and so on, all the way around the flywheel.


When I turned the engine horizontal and applied test weights, I found that all the magnets would hold the perch (one pound), but only a few pairs of  magnet poles would support the spindle. It was time to up my game.

Overnight I charged some batteries, and on the second day I used four of them: two twelve volts and two sixes. I suppose other combinations to make 36 volts would do as well, but these were what I had handy

This time, instead of setting the charger core against each pair of poles, I charged one pole at a time. My aim was to get each pole of each magnet to hold the spindle without any help from a nearby pole. Yes, I realize there’s a steel plate connecting two adjoining poles, but I pretended there wasn’t and dealt with each pole individually.


After once around the flywheel, three individual poles held the spindle. After another time around, six held it, then nine, then seventeen, and finally I got all thirty-two individual poles to hold the weight. There was some variation. Some were strong enough to hold it immediately with no adjustment. Others weren’t quite as strong, and the spindle would drop off until I touched it to the pole at just the right angle. But eventually all poles supported the spindle and I declared victory.