Send questions or comments to [email protected]


Depending which end of the armature holds the condenser in a given magneto, the central bolt through the points plate may bolt directly to it. If this is the case, you will have to manufacture an adapter in order to replace the condenser with a modern one. The original condenser itself could serve this purpose if its top were machined away, but it is filled with a now-banned carcinogenic PCB so I definitely do not recommend doing this.

At the top of the next photograph are Bosch (left) and Lucas (right) condensers with the threaded "nut" on the underside to which the bolt through the center of the points plate attaches.

[Linked Image]

At the bottom of this photograph are the end caps of two armatures. Note that the cavity of the one at the left (Bosch) is much deeper than the one at the right, making it easier to deal with when adding a modern condenser. Also note that the one at the right has a "square" opening for the condenser nut, so the Lucas condenser above it does not go with it.

My approach to making an adapter "nut" for the replacement condenser starts with a 1/4"-20 brass screw as shown at the left of the next photograph. This screw came from a local hardware store and there is nothing special about its diameter or pitch (other than the diameter of the head, as discussed below), or even that it is brass. Although brass is a nice material to machine and tap, a stainless M6 screw would function as well.

[Linked Image]

I use a lathe to reduce the thickness of the head, shorten the screw, then tap if for whatever screw is used to hold the points plate in the magneto I'm working on. Depending on the magneto this could be BA, metric, or some now-obsolete American thread so a pitch gage is needed to determine the required tap. The next photograph shows a jig I made to hold the screw in the lathe while doing all the modifications to it (although everything could be done without such a jig).

[Linked Image]

The points plate mounting bolt needs to be insulated from the magneto housing, and the next photograph shows a nylon insulator in the shape of a top hat that accomplishes this. Delrin, acetyl, phenolic or some other insulating materials would work as well. The OD of this insulator fits the ID of the hole in the end cap of the armature, and the thickness and OD of the "brim" of the cap fits in the larger diameter recess. I tap the insulator to match the brass screw (i.e 1/4"-20), but I haven't listed other dimensions here because they depend on the particular magneto. Also, as noted above, the necessary insulator might not even be cylindrical, and anyway anyone who has the tools to make this piece will have calipers to make the measurements.

As shown in the next photograph the brass piece threads into this insulator. One of the condenser leads is soldered to it to make the necessary electrical connection to the points via the central mounting bolt.

[Linked Image]

Although it isn't clear from the photographs, the OD of the head of the brass screw is slightly larger than the OD of the main section of the insulator (i.e. slightly larger than the ID of the hole in the end cap of the magneto). To minimize the height of the final condenser assembly the top of the brass screw can be reduced in one dimension in order to fit between two condensers, as seen on the right of the above image as well as on the left of the next one.

Although fabricating this "nut" might sound time consuming, it actually doesn't take much time at all. Also, since it doesn't take much longer to make a half-dozen of the individual pieces than it does to make just one I have a stockpile of the components ready for assembly whenever needed.

[Linked Image]

The above photograph shows the final condenser assembly with two 0.083 uF Panasonic capacitors (left) next to an original Lucas condenser (right). In this particular assembly the capacitors are being held a little above the insulator by the springiness of the lead, but they would be in direct contact when epoxied into place in the end cap (not that direct contact is important, other than keeping the overall height low enough that interference with the armature isn't an issue). The assembly would look essentially the same if made with two of the Vishay capacitors I recommended in Appendix V.

The next photograph shows what the assembly looks like from the bottom. Although when epoxied in place in the armature the leads shouldn't be able to touch any part of the housing, I like to insulate them anyway. The black substance I use for this is "corona dope."

[Linked Image]

Although I used a 1/4"-20 screw, and mentioned an M6 would be fine as well, it can be seen from this photograph that there is plenty of room for an even larger one if desired for some reason.

As for strength, because of the design my only concern would be if the "brim" of the insulator could break because of pressure from the points plate mounting bolt. Although nylon should be strong enough to avoid this, to be sure it isn't an issue I use brass screws whose heads have an OD slightly larger than the OD of the major section of the insulator (i.e. slightly larger than the ID of the hole in the end cap). Because of this, although at very worst the "brim" might compress slightly with time, slightly reducing the clamping force of the central bolt, it cannot pull through.

If brass screws with large enough heads are not available for a given magneto it would be easy enough to build up the diameter of smaller ones with silver solder, or just to make ones entirely from scratch. Finally, it wouldn't hurt to put a few drops of glue on the brass screw before tightening it into the insulator just to be absolutely sure there never will be an issue of the brass/nylon assembly trying to separate even if the mounting bolt is screwed in and out a number of times over the years.

It can from the above photograph that the new assembly has basically the same dimensions as the old Lucas condenser it replaces. As mentioned several places in this thread, the similar size is no coincidence. For a condenser to survive high pulsed currents without burning out requires thick capacitor plates, to survive high voltages requires thick dielectric spacers between each of those plates, and to have the proper capacitance despite the thick dielectric spacers requires many layers of large area. All of these factors add to the volume, which is why these capacitors that I have tested to survive the high pulsed currents of a magneto for many years intrinsically are large.

The first photograph in this post shows that the cavity in the Bosch housing is much deeper than in the other one. In such a case two, thin 0.083 uF capacitors don't have to be connected in parallel to create the necessary capacitance, but instead a single, thicker 0.16 uF or 0.18 uF can be used. However, no matter what, the larger capacitor must be from the same "family" as the 0.083 uF ones that are rated for high pulsed currents. The blue bars in the next photograph show that one of these higher capacitance units is quite a bit thicker than the 0.083 uF ones which is why it won't fit in the cavities of many magnetos.

[Linked Image]


Finally, as a reminder, removing the armature from a magneto for even a few seconds "permanently" reduces its magnetization making the motorcycle more difficult (or impossible) to start because a higher kickover speed is required to generate sufficient voltage for a spark. The only way to restore full performance after removing the armature for any reason is to remagnetize it after it has been reassembled. To do this requires an appropriate electromagnet ("magnet charger"), and one that was designed for pre-WWII magnetos does not have sufficient strength to remagnetize post-WWII alnico magnetos.