Britbike forum Forums Projects started by Britbike forum members Members bike projects Restoring a Rotating Armature Magneto

Restoring a Rotating Armature Magneto

Table of Contents
(Note: I added this Table of Contents after uploading the final installment of this thread. Clicking on an underlined link will open a new window with the individual post)

Table of Contents, Background, Introduction

Initial Inspection and Initial Tests

Disassembly and Detailed Inspection

Disassembly and Detailed Inspection (Continued)

Disassembly and Detailed Inspection (Continued)

Repairing Damage Done to Armature by Previous Restorer

A Sidebar About Screw Threads

Repairing Damage Done to Armature by Previous Restorer (Continued)

Preparing the End Cap
-- Removing the Bad Condenser
-- Testing the Armature's End Cap

Tests and Repairs of the Electrical Components
-- The Slip ring
-- The Coil
-- The Condenser
-- Sidebar About Replacement Magneto Condensers

Tests and Repairs of the Electrical Components (Continued)
-- Reassembling the Armature
-- Testing the Reassembled Armature
-- Brushes

Tests and Repairs of the Electrical Components (Continued)
-- Brush Spring Pressure
-- Contact Breaker Points

Mechanical Components
-- The Bearings

Mechanical Components (Continued)
-- The Cam

Reassembly, Remagnetizing, and Extended Stress Tests
-- End Float
-- Contact Spring Pressure
-- Magnetizing
-- Sidebar about Magneto "Chargers"

Reassembly, Remagnetizing, and Extended Stress Tests (Continued)
-- Strobotac

Reassembly, Remagnetizing, and Extended Stress Tests (Continued)
-- Long-Term Tester
-- Elevated Temperature Test

Final Tests
-- Low Speed Test
-- Distributor Tester
-- Disassembly, Inspection, Remagnetization, and Reassembly
-- Timing a Magneto Using an Inductance Meter
-- Performance of the Magneto on the Road

Epilog
-- The Major Problem Areas in Restored Magnetos
-- Tools
-- Magneto Repairers
-- Armature Winders
-- How Much Should it Cost to Have Your Magneto Rebuilt?
-- Cosmetics
-- Final Comments

Appendix I: Post-WWII Magneto Condensers

Appendix II: Replacement Condensers for Post-WWII Magnetos

Appendix III: Anatomy of a Post-WWII Lucas K2F Magneto

Appendix IV: Diagnosing a Failed Aftermarket Slip Ring

Appendix V: An Alternative Replacement Condenser

Appendix VI: Adapter to Attach Points Plate to Condenser

Appendix VII: Rebuilding a Lucas KNC1 Competition Magneto

Appendix VIII: Rebuilding a Lucas Magdyno

Background
This "bike project" will be different than most on this site, in that it will be about just one component. However, the magneto is a component that is of interest to owners of a variety of classic American, British, and Continental machines.

In addition to Harley-Davidson and Indian (and others), Bosch magnetos also were used on a number of very early British bikes as well as ones made on the Continent. Also, all Lucas and BTH rotating armature magnetos are direct copies of the Bosch design, which was appropriated by the Allies at the start of WWI. For these reasons, my restoration of a c1920 Bosch ZEV might be of interest to a variety of motorcyclists. Because of this, although the "original" will be on BritBike Forum (britbike.com), I have given permission for it to be cross-posted on caimag.com and classic-harley.info. However, those are the only locations that have permission to post copies of this material.

While I won't be able to reply to individual emails, people who are reading this on other than BritBike Forum can send questions or comments to [email protected] and I will address them in future posts if they are of sufficiently broad interest. To save people time, no, I am unable to recommend someone to restore your magneto. Sorry.

Introduction
Recently a good friend purchased a 1923 Harley-Davidson Model F twin to use for an event. He shipped the bike to someone to rebuild for him but, despite it having a freshly rebuilt Bosch ZEV magneto, he asked me to check it out while the engine was being rebuilt. Unfortunately, the magneto was non-functional when it arrived at my house even though it appeared to have had 0 hours on it since it had been "professionally rebuilt."

Although I finished rebuilding it a few weeks ago, only now am I finding the time to start converting my notes to prose. I'll post this in installments as I complete each segment, and will try to write it as if it were a "real time" narrative, but it probably will end up in both present and past tense.

Because my friend had specific plans for his motorcycle, I was working to a hard deadline to get this magneto fixed, so I didn't always interrupt my work flow to photograph every step. However, so there wouldn't be gaps in the visuals, I later made additional photographs to reconstruct the missing steps. I've noted these where ever they occur, such as the following of a single-cylinder Bosch ZE1.





The ZE1 shares the same construction and many parts with the V-twin ZEV, so the magneto I rebuilt looks almost identical to it.

Although I try to work only on my own magnetos, I've disassembled about a dozen "professionally restored" magnetos over the past fifteen years. Without exception, what I found inside each of these failed magnetos in no way deserved the term "professional." I'm not going to name names, but two of those failed magnetos were rebuilt by people who many riders of classic British motorcycles would recognize (I don't know/remember who restored the others). However, it's important to add that I've never looked inside a restored one that was functioning properly, so what I've observed cannot be taken as representative of the work of all magneto restorers.

My goal in restoring any magneto, including this one, is simply to return it to the condition it had when it left the factory. No better and no worse. In some cases improvements are required because original components or materials are not available, or there is some intrinsic defect that has to be addressed. For example, I use a modern resin on the coil rather than whatever old resin was used 50+ years ago, because those old resins had several undesirable characteristics.

The steps I will describe in subsequent posts are the ones my experience tells me are necessary to return a magneto to its original performance. To do less would be to produce a magneto that is less reliable, has a shorter life, and/or produces a lower output. As I think will be clear from my subsequent posts in this thread, the magneto repair and testing equipment I routinely use to accomplish this are well beyond those employed by people who do this for a living. Whether someone could find customers willing to pay what it would cost to restore a magneto as I will describe is not an issue, because I am my only customer.

I'm guessing that this magneto restoration tale will take a dozen installments to tell, not counting responses to any questions, and it might take two months to get to the end. So, please be patient.

Send questions or comments to [email protected].

INITIAL INSPECTION AND INITIAL TESTS:

When I did my initial inspection of the Bosch ZEV I found the lock nut on the fixed point was loose and the gap was ~0.025", which is about 0.010" too large. A shim under part of the points assembly was in the wrong position resulting in the points being offset from each other by ~1/4 their diameter. I didn't photograph it that day, but the next day I reconstructed it roughly as it had looked, which is shown in the next photo.



Since the slip ring was new, and its condition indicated the magneto could not have run for more than an hour at most, it's possible the previous owner tried adjusting the points when the magneto failed and this is why they were loose. However, whether they were left this way by the restorer (which seems unlikely), or by the person who sold the motorcycle to my friend (more plausible), I don't know.

Another problem with the points that is not apparent from the above photograph is that the ID of one of the insulators was too small to fit over the OD of another, leaving a space between it and the brass housing. It's a little difficult to see, but from right to left at the bottom of the next photograph is the head of the screw that mounts the fixed point assembly to the rotating base, a brass washer, a flat insulator (with the gap mentioned above between it and the brass assembly), the assembly itself, and a cylindrical insulator that keeps the mounting screw from shorting the assembly to the rotating base. The OD of that cylindrical insulator is larger than the ID of the other insulator, resulting in the gap. Because this can be very easily fixed by slightly enlarging the ID of the flat insulator with a round file, the fact the restorer didn't notice it is the first of the direct evidence we have about the "quality" of his work.



The magnets didn't feel as strong as they should have felt, but this is hard to be sure of without an actual measurement. However, it wasn't worth the time it would have taken to measure the field strength with my magnetometer since I already knew I would be completely disassembling the magneto. Although the Alnico magnets in post-WWII magnetos retain enough of their magnetism for the magnetos to still function (although, at reduced capacity) after a restorer removes the armature and replaces it, that isn't the case for steel magnets. If it does have reduced magnetism in the present form it likely means the restorer did not understand this essential point about pre-Alnico magnets, or didn't have the facility to remagnetize them. Or both. But, taking it apart means I will have to remagnetize it no matter what.

The brass inside the points plate was not tarnished, indicating the magneto was restored sometime relatively recently (i.e. within the last decade or so). I assembled everything in the points cavity in their proper positions, set my test rig with the minimum gap (so it would spark even with weak magnets), and fired it up to see what things were like. The magneto exhibited classic bad condenser sparking at the points almost immediately (it probably had already failed before I even looked, but it took maybe 30 sec. to make sure everything was properly spinning before I looked at the points).

The next photograph shows the rig I use for extended tests. I built it around a 1/2-hp, 1725 rpm motor that operates either clockwise or counter-clockwise, and it has a board with adjustable gaps for magnetos of up to six-cylinders. Adjustable plates and brackets allow all types of platform- and flange-mounted magnetos to be easily attached.

Thanks very much. I'll be spending a lot of time on airplanes this month, but some of that time should allow me to get material in shape for some of the upcoming posts in this restoration thread.

Agreed

We are lucky to have someone on the forum prepared to share this kind of information

Thanks for the comment. I'll be away on a ride this weekend, but should have my next post ready to upload by early in the week.

I think people will find the information in the upcoming posts pretty useful. I have never seen anything as comprehensive as what I will be posting published anywhere before. Even for someone who doesn't plan to rebuild his own magneto, my hope is that having the mystery removed will help raise their level of expectations and, as a result, the quality of restoration work. As long as so many motorcyclists accept having their rebuilt magnetos fail, and even regard it as somehow inevitable, the bar will remain set pretty low for work done on these devices.

Send questions or comments to [email protected].

DISASSEMBLY AND DETAILED INSPECTION:

After my initial inspection and test to characterize the condition of this magneto when it arrived, I disassembled it, making notes of what I found to refer to when I started the actual restoration. Disassembly was quite easy. Two screws hold each HT pickup, and a bolt presses the points assembly into a taper in the end of the armature. If this were a later BTH or Lucas magneto there also would have been a safety gap screw to remove (failing to remove it accounts for the characteristic piece broken from so many slip rings).

The photograph in the first of these posts shows there is fairly large gap between the points assembly and the housing so I tried gently prying at the edges and it immediately popped off. Had it not come off so easily, and rather than prying with ever-bigger screwdrivers, I would have taken the time to fabricate an appropriate "pusher" to slip under the assembly and that used the housing to push against (a two-jaw puller would have to push against the screw threads in the armature, which would be fine as long as a the assembly released before the threads had too much force on them). BTH and Lucas thoughtfully provided threaded holes in their later points assemblies that are used to push them off the armature using the same bolt that holds them in place. After removing the points assembly, and the four screws holding the front plate, the armature was free to be removed.

Overall the components were in reasonable condition for their age, although there were some issues to think over before attacking the rebuild:

1) It has a new slip ring, which looks good. However, some aftermarket slip rings were made with inappropriate plastic that is partially conductive so this requires measurement with a megohmmeter.

Had the slip ring been chipped I have a Teflon mould I use to repair ones on BTH and Lucas magnetos using a non-conductive epoxy with 3000 psi shear strength. If I had needed a mould for this restoration, but had this one not been the correct size for the Bosch, I would have made one specifically for it. The photograph shows the mould on the left, a new Lucas slip ring on the right, and the chipped slip ring that is on a spare Bosch ZEV armature. However, for a slip ring as damaged as the one in this photograph I would fabricate the missing portion on my lathe and attach it to the existing piece rather than build it up entirely with epoxy (unless it were an "emergency" field repair).



To remove the slip ring on this armature would have first required removing the bearing race. Although I didn't need to do this, I will describe in a later post how it is done.

2) The bearings appear to be new, but I will inspect the races and the balls with a microscope for any evidence of Brinelling.

3) Although I suspected the leads were of the cotton resistance type, because they felt so limp, their resistances were less than 1 Ohm so they must be made with copper wire. However, even though the insulation looks a bit old I will have to leave it to the engine rebuilder to replace these leads once the magneto is installed, because I can't guess the correct lengths to use. This will be in the "installation notes" I will send to the engine rebuilder when I return the rebuilt magneto to him. I also attached a yellow tag (from an office supply store) with a note to one of the leads to be sure it wasn't overlooked.

4) The HT pickups are old, but look to be OK. I made a note to do a more careful inspection of them later, but all three brushes in the magneto looked like they were new.

Unfortunately, there are aftermarket brushes that are much harder than OEM, and these would grind a groove into the slip ring if this is what is in this magneto. An example of this is shown in the next photograph, which is on a BTH armature that was in a box of parts I bought a few years ago.



While the above photograph shows the damage hard brushes can cause, soft aftermarket brushes also are a problem. These slough off carbon too fast and can cause an internal short in less than a few hundred miles. Since there is no sign this magneto ever has functioned since being rebuilt, there is no way to know from the appearance of the slip ring itself whether the brushes have the proper hardness. Since this is problematic, I will replace all three with NOSs Lucas brushes if the diameters allow that to be done. If not, and although I am equipped to measure Rockwell C hardness and could convert to the Shore scale used for materials like carbon, the Rockwell test Shocks the sample with a pointed indenter so it is not used on brittle materials. Since this measurement itself could destroy the brush, if the Lucas brushes won't fit, I have my own tests I will make. Although not as precise as a proper hardness test, it still allows me to determine whether they are appropriate to use or not.

[detailed inspection to be continued]

Totally interesting article MM, I look forward to the next section.

Thanks very much. Because of my travel schedule, it looks like I will be posting sections roughly one per week. The best is yet to come.

DISASSEMBLY AND DETAILED INSPECTION (CONTINUED):

5) The magneto has a rewound coil in it, as well as a new condenser (although, as I wrote in a previous post, the latter already had failed). The electrical properties of the primary and secondary with the armature in the magneto were:

Primary___: R= 0.58 Ohms; L= 3.40 mH
Secondary: R= 4.67 kOhms; L=16.38 H

The inductances would be somewhat lower if outside the housing. For comparison, the primary of one of my two spare ZEV armatures is still good, and it has essentially identical values of R=0.57 Ohms and L=3.52 mH (next photograph).



Although I don't have a functioning Bosch ZEV secondary to measure, the resistance of the secondary is reasonable, although the inductance is about 50% higher than that of a comparable Lucas armature. My conclusion was that the person who rewound this coil neither made it with significantly too many windings of too fine wire, nor too few of too coarse.

--------- Sidebar (not in chronological sequence) ---------
I am writing the following three paragraphs after the fact. I realized after having sent the magneto back that I should have measured the transformer ratio, which would have told me if the secondary has the correct number of turns. All that is required is to run a small AC voltage from a signal generator through the primary and measure the voltage developed in the secondary. This transformer ratio is ~35 for similar Lucas and BTH magneto armatures. The actual ratio of turns is 50, but the transformer loss due to generation of eddy currents is ~30%.

The reason the number of turns in the secondary concerns me is that, not having x-ray vision, I have no way of knowing if the rewinder used insulation between its layers. Not doing so would result in a less robust coil. The R and L values for the secondary are consistent with wire that is ~20% too large in diameter, but with ~50% too many turns. Combined, that would give the correct resistance, but an inductance that was 50% too high. The resistance and the inductance of the primary are as they should be, consistent with the primary being made up of the correct number of turns (which determines L) of the correct diameter wire (which determines R).

In order for a coil with too many turns of too-large diameter wire to fit in the volume available means interlayer insulation likely would have been omitted. Omitting this insulation makes it easier to wind coils, which is all the more reason to think it might not be there. The secondary winding would have about 30 layers, and in normal operation the coil output would be limited by the spark plug to no more than 6 kV. This means there would be ~200 V between adjacent layers, which the insulation would be able to handle without problem. But, if a plug wire fell off and the voltage rose to 18 kV, adjacent layers of the secondary would have to withstand ~600 V. That still is within the limit a good quality insulation can withstand. But, I don't know the quality of the insulation, or if the coil was vacuum impregnated so the insulation can't abrade. Unfortunately, the limited time I had to get this magneto back to the engine builder made me overlook this test that I otherwise would have thought to make, so this will be a lingering uncertainty.
--------- end of sidebar ---------

Unfortunately, the restorer got carried away with his use of epoxy on the coil. The coil itself is not only potted with epoxy, it is firmly attached to the armature by excess. However, what is not possible to tell is if only the outside of the coil had been slathered with epoxy, or if it had been vacuum impregnated. I made a note of this because it is a concern, but I will have to think about how I want to deal with the uncertainty.

The reason vacuum impregnation is an issue is that the motion of any wire through a magnetic field subjects it to a sideways force. If the armature coils are tightly wound, there is no place for any of them to move, so that force wouldn't matter. However, if there is any room for movement at all, the oscillating motion due to the field changing direction twice every revolution can result in the insulation on a wire slowly abrading away, eventually shorting it to its neighbor. Such shorts do not necessarily cause the armature to fail completely, but they do degrade the output. Vacuum impregnation using an appropriate low viscosity epoxy or other potting compound fills the voids, making motion of the wires impossible.

Winding an armature coil is a tedious job, and I don't begrudge the ~$150 that people charge for doing this. However, I do begrudge the quality of some of the coils that are produced. Proper coil winding is a craft, and not all magneto coils I have seen were wound by people who mastered it. I'm reminded of something I learned from an instrument maker 40 years ago: "There's a difference between handmade, and homemade." Too many coils I've seen over the past 15 years are of homemade quality (and, again, I've only seen the insides of magnetos that have failed, so this is not a representative sampling).

Magneto armatures were not my first exposure to coil winding, so my disappointment with the quality of rewound armatures is based on first-hand knowledge of what is possible to do oneself. In any case, I felt I had no choice but to purchase the equipment needed to do this in my own workshop. Having a coil winder meant I was no longer dependent on anyone but myself for any aspect of a restoration. Anyway, I have my own German-made coil winder as well as equipment for vacuum impregnation of potting compound. However, the short time to get this magneto restored, interrupted by the trips I have to take before it has to be shipped back, means I would much rather use the existing coil if it survives my stress tests (to be described later).

I haven't used my coil winder for a few months so the area around it has accumulated some clutter. But, the photograph shows it with the left half of the pair of brackets I use to hold armature like the one in the Bosch (and Lucas and BTH). A mating piece is held by the tailstock whose dead center can be seen at the edge of the photograph. The very thin wire of the secondary is fed from the spool through a spring-loaded tensioner to help keep it from breaking (partially shown at the top, with a pulley at the end of a pink rod that is connected to the tensioning mechanism), and is automatically layered by an auto-reversing mechanism whose pitch is adjusted to match the diameter of the wire. The speed is controlled by a rheostat in the foot pedal to allow starts and stops to be made smoothly, and a counter keeps track of progress.



After winding a coil I pot it with an appropriate low viscosity compound and use a vacuum pump to impregnate the liquid into the voids between the windings. I use thick-walled plastic containers to hold the compound and the armature under vacuum, allowing me to see what is going on without the risk of implosion if a glass container cracked.

I set up the next photograph using a partially-disassembled armature in the plastic container to show the scale. I am not being coy about not naming the potting compound I use, but recently I've been experimenting with several new ones and don't want to "endorse" any until my tests of those are completed.



The excess epoxy on the present coil would be OK (except, see below), except part of the coil is slightly too large and has rubbed on the inside surface of the body. I can't tell at this point in the restoration if it is simply excess insulation that is rubbing, or if wire is involved. But, it's possible to check the output of the coil with my Merc-o-tronic tester. The output of the coil can be seen sparking across a 5 mm gap in the window in the tester, so it passed this test (however, I'll do a longer test of just the coil itself before declaring it to be good).



As an aside, I have at least a dozen magneto testers, but I almost always turn to a Merc-o-tronic 98 or 9800 first when checking a coil. Although these can make six types of tests, all but one of them -- the current required to generate a spark across a 5 mm gap -- can be better made with more modern instruments. However, if I only could have one instrument for testing magnetos, this is what I would choose.

6) As long as I can remove the end cap that contains the condenser, removing the epoxy that encases it, although a headache, and replacing it with a proper capacitor should be straightforward (note, the photograph above was taken during a later test I made after I had removed the end cap).

Send questions or comments to [email protected]

[detailed inspection to be contined]

DISASSEMBLY AND DETAILED INSPECTION (CONTINUED):

7) In the next photograph the protrusion at 7:00 in the taper locates the points block with respect to the armature to ensure the magnetic flux reversal coincides with the opening of the points. The missing chunk at 11:00 in the taper appears to be damage caused by some heavy-handed restorer. The black cylinder at 4:00 is a carbon brush that completes the secondary circuit back to earth when the points open. To do this, the carbon brush has to make good contact with the surface it rubs against.



Unfortunately, as can be seen in the next photograph, the surface the earth brush rides against is in pretty poor condition. Although the brush can make a good electrical connection to such a surface, its roughness will wear away the brush more rapidly than it should, eventually resulting in failure of the magneto. I will have to turn it on my lathe to make it smooth.



Since this surface needs to be parallel to the back surface of the plate holding it I tapped four holes in a piece of scrap Al, faced it in the lathe, then mounted this plate on it in order to face the surface for the earth brush.

After turning any rubbing surface on the lathe I use my surface roughness meter to verify it has the correct smoothness. I didn't take a photograph after turning the ZEV surface, but the following one shows the roughness meter being used on the ring used for the earthing brush on a Lucas armature from my box of spares. I had the meter set to display the roughness in micro-in. (59 micro-in. is 1.5 um). In fact, a "perfectly smooth" surface is not desirable since some abrasion of the carbon during operation is required to ensure good electrical contact. Industry practice for the roughness of commutators calls for values 1.5-2 um, so the earthing ring on this Lucas armature happens to be fine as-is.



I measured the earthing ring on a BTH armature several years ago when I restored the magneto for another friend. It had failed shortly after he paid quite a bit to have it professionally restored by someone whose name would be familiar to many riders. The roughness measured 190-210 micro-in., which is nearly 3x too high. The grooves the restorer left in it by using a too-sharp lathe tool would have abraded the earth brush too rapidly had the inappropriate condenser he installed not failed. In fact, the grooves were so pronounced that it was easily possible to determine the feed rate he used on his lathe. I still have this armature because it was easier for me to rebuild an unrestored armature I had than it would have been to spend the time to first undo what the restorer had done to it.

8) The condenser in the Bosch armature is completely potted in the end cap using some red epoxy that I'll have to remove. If you look closely you can see that the clear epoxy used to pot the coil also overlaps with the armature, firmly attaching it. This excess epoxy is sloppy workmanship and would significantly add to the effort required to prepare this armature if I had to wind a new coil on it.



9) A problematic aspect of the armature itself is that some of its laminations have suffered at the hands of the restorer (or an even earlier restorer -- there is no way to know when over the past 90 years some of the damage was done). Armatures are built up from laminated steel plates that are insulated from each other to minimize eddy current losses that otherwise would severely reduce the output from the magneto, especially at low rpm. Someone beat quite a bit on one end of the plates on one side, and there are two holes drilled 180-deg. apart in them as well. I don't have any prior experience with Bosch ZEV magnetos to know whether or not these holes were drilled at the factory for some reason, but neither of my spares have them. Also, although the hole in the above photograph appears to be round, the one on the opposite side is fairly crude.

I have two Bosch ZEV armatures of my own, both with their original, rotten coils, that I could press into service for winding new coils if needed. But, no matter which armature I end up using, there are issues to consider. I'm going to think about this for a day before proceeding. The armature I am repairing is in the first photograph, and one of my own is in the second.





In contrast to the armature I am working on, the photograph of mine shows what an unmolested one looks like (complete with 100 years of patina). However, the coil in mine is original, looking about as fresh as the wrappings on a mummy in the British Museum.

The dents in the armature will have shorted the outer edges of those laminations together. Also, the holes someone drilled through the laminations in the center will have shorted those laminations that are in contact with the holes. Although these are electrical shorts, each increases the eddy current losses rather than stopping the output entirely, so it's a judgment call at this point whether their cumulative effect on the magneto output at low rpm will be unacceptable. All things considered, my experience tells me it is worthwhile proceeding with this armature despite the damage to the laminations.

Despite the issues noted in the past several posts, detailed inspection has not revealed any problems that would keep me from restoring this magneto to full operation. Unfortunately, its poor condition means it will take longer than the couple of days I had hoped would be the case had it been in good shape, since first I will have to undo all the damage done by the person who restored it. But, now have to decide how to proceed (e.g. use the present armature if tests show it is good, or wind my own replacement coil using one of my old armatures). Send questions or comments to [email protected]

REPAIRING DAMAGE DONE TO ARMATURE BY PREVIOUS RESTORER:

The story of what the professional restorer did to this magneto just keeps getting better…

Although the end cap came off without difficulty, often they are a very tight fit. Because of this, I designed a set of separators consisting of four pieces held in place with hose clamps that lets me gently press armatures apart by turning two screws. The next photograph uses one of my spare ZEV armatures to show how it works.



The next photograph shows the end containing the condenser of the actual armature I am rebuilding, with the four screws holding this end cap in place. Or, so it appears. The photograph also shows a screwdriver set with bits in small increments of thickness and width to perfectly match the slot in any screw head, as well as a small impact driver for those bits if any extra encouragement is needed (which it wasn't in this case). Depending on circumstances I use a 1/4"-drive air impact driver (35 ft-lb. max.), but this 1/4"-drive manual one also is perfect for "emergency" repairs away from home.



Even though the heads had been staked to keep them from coming loose, the screws came out easily. Too easily. Normally, staking them would seem to be the mark of careful work, except....

...three of the screws were broken!!! The staking held the heads in place so it looked like the screws were as they should be, but they weren't. Comparing the screws to those in my spare ZEV armatures shows the three broken ones are the originals, while the one screw doing all the work is a modern replacement. But, that's not the end of the bodge.



The only screw from the set of four whose diameter and pitch I can accurately measure is the unbroken one. As can be seen, it's threaded all the way to the head, whereas the three broken ones are not, showing the one good screw isn't original. This screw's OD is close to the fairly uncommon U.S. size 6-40, but according to my pitch gauges it's close to, but not quite, 38 tpi. It can't be metric, because its diameter is ~3.6 mm, and its pitch would correspond to an implausible ~0.66 mm… Damn!, after breaking three screws the magneto repairman crammed a British 4 BA screw in the one good hole, and this is all that was holding it together!! The condition of the slip ring indicates the magneto couldn't have worked for more than a few minutes after having been rebuilt, which is a good thing because this single screw would have been the only thing keeping the armature from flying apart while it was spinning at up to ~2000 rpm.

This motorcycle lived on the British Isles in recent years, which is why it didn't take me too long to check for BA threads. This is truly an international magneto: designed in Germany, manufactured in America, and bodged in England.

Luckily, I have a mill, so getting the three screws out shouldn't be too much of an issue, although it would take time. First, I had to fabricate a fixture with 2.000" ID at one end of it to hold the armature precisely vertical in the mill. I turned this on my lathe, as well as a base with a 5/16" hole to centrally locate the end of the armature in the fixture, and slit the fixture on the mill. Send questions or comments to [email protected]

A SIDEBAR ABOUT SCREW THREADS:
Although pitch gauges were indispensible for identifying the threads used in this magneto, so were three types of microscopes. Because of the broken screws I had to spend a few hours researching threads before I could repair them, and what I found was that this magneto is filled with bastard threads, some of which are badly made. The interesting thing is that, notwithstanding the BA screw, this seems to be how it came from the factory.

Although the magneto was designed in Germany, it was made in the U.S. and was worked on in the UK so there is a mix of Fasteners on it. There are components on the magneto that are definitely of German ancestry, but others that are American. For example, the diameters of both ends of three armatures are 2.000" +/-0.003". I would have to get my hands on a ZEV that I knew was made in Germany to be sure but, presumably, it would be 50.0mm = 1.969". Also, since it was made nearly 100 years ago, I can't assume the Fasteners were manufactured perfectly in the first place, so both the diameters and pitches could vary slightly from what they should be.

There are at least 13 different functions performed by screws on this magneto (not 13 different thread forms, though). I've analyzed 7 of them, requiring three microscopes and several precision thread gauges to come to the following conclusions. It wasn't straightforward getting there, and I doubt many, if any, other people know what I'm about to tell you. First, though, a tiny bit of history.

What we now know as "metric" threads is an invention of c1965, with various national metric standards used before then. In some cases the Fasteners made to the old standards are a good fit with modern ISO metric, but it some cases they are not. It happens that c1890 the Germans adopted the Loewenherz standard for small screws for use in instruments. This is the same function numbered screws serve for the U.S. (e.g. 6-32) and BA for the British, the latter of which was a standard adopted straight from the Swiss Thury system with one trivial modification.

Anyway, the smallest screw I measured was one that holds the earth/ground lead for the condenser in one of my spare ZEV armatures (an unknown screw was in this hole in my friend's armature). Although it is 2.50 mm in diameter, it has a pitch of exactly 0.45 mm, making it a slightly inaccurate Loewenherz 2.6x0.45 (the next size smaller in the series is an L2.3x0.40, whose pitch makes it an impossible match).

Next up in size are the screws for the HT pickups. I started with these, and they caused me no end of grief before I came to a conclusion. Their diameter is 0.159", whereas 4.0 mm would be trivially smaller at 0.1575". However, if they are L4 they should have a pitch of 0.75 mm. Unfortunately, as I finally concluded, pre-War factory standards must have slowly degraded by the time this magneto was made nearly a decade after Bosch's U.S. assets had been seized, resulting in a bastard version of the L4. The pitch is slightly finer than 0.75 mm (which would be 33.86 tpi), and it is also finer than 34 tpi, but coarser than 36 tpi. I could find no reference in an old edition of Machinery's Handbook to any standard U.S. screw ever being made with 35 tpi. But, since less than 1/4" of thread engages in the housing, even if the holes they go into were made with a perfect L4x0.75 tap, and the pitch of the screws were as far off as far as being 35 tpi, the total mismatch would be only ~1/4 thread, so they would go in with only a little difficulty at the end of their travel. However the fact the factory didn't just adopt the nearest American screw for the armatures indicates that after WWI they instead continued working with ever-more-worn German tooling and using ever-more-out-of-spec replacements.

Next up in size are the armature screws, with diameter 0.1345". Here both 3.5 mm (0.1378") and the old U.S. 9/64" (0.1406") are potential candidates. An L3.5 mm screw has a pitch of 0.6 mm (42.33 tpi), but also should have flat roots and crests each of width 3/8 of the pitch, and a 53-deg. 8 min. included angle (vs. 60-deg. for the American thread). Here, the (in)accuracy of pitch gauges were not up to the task of distinguishing between threads differing by less than 1%. Because these screws were broken and I needed to find a suitable tap to repair the armature, I had to make careful measurements using three types of microscopes. The illumination is from below in the following micrograph taken in a metallurgical microscope (i.e. the screw at the bottom, appearing as black).



Although the 55-deg. angle of these screws is close to that of the Loewenherz thread form, the roots and crests are not flat as they should be if properly made. To accurately determine the pitch I used a traveling microscope as shown in the next photograph. This instrument uses a precision micrometer thread to translate a 21x microscope over the specimen, allowing lengths up to 10 cm to be measured to a precision of 0.01 mm (0.0004"). This is shown in the next two photographs using a 6-32 screw for scale.





My measurements determined these screws are 42.0+/-0.1 tpi, not the almost identical 0.6 mm (42.33 tpi). However, the engagement in the armature is only 1/2" so the mismatch in this length would only be 0.16 of a thread even if the factory had used a perfect L3.5 tap. So, like the previous example, my conclusion is that this screw is a poor tolerance imitation of the German Loewenherz 3.5x0.6 thread, as opposed to being an American substitute.

Next, larger in size are the screws holding the covers. These are definitely 7/32"-24, which was an American standard at the time. The nearest Loewenherz screw has a diameter close enough, but the pitch differs by too much (28.22 tpi).

Finally, the four holes in the base that are used to mount it to the motorcycle are American 5/16"-18.

What the above shows is that when this magneto was made in the 1920s it left the factory with a mix of American and metric Loewenherz Fasteners of dubious quality. Because of this, I couldn't make any assumptions about any thread without carefully measuring it. Send questions or comments to [email protected]

REPAIRING DAMAGE DONE TO ARMATURE BY PREVIOUS RESTORER (CONTINUED):

After determining the correct size of the armature screws (Loewenherz L3.5x0.6), I mounted the armature vertically in the mill's vise using the fixture described in a previous post that I had made for this purpose (shown in the third photograph, below). I then used a centering microscope to locate the positions of the three broken screws to +/-0.001" and recorded the x,y coordinates of the mill's Digital Read Out (DRO) so I could return to each of them after replacing the microscope with a carbide center drill. Meanwhile, I had ordered an M3.5x0.6 tap because this is an uncommon size and wasn't in my set of metric taps and dies (more on the difference between an M tap and an L thread below). There was plenty to do, so waiting a few days for the tap to arrive didn't cause any delay.

I didn't take photographs at the time when I was using the centering microscope, so I set it up again a few days later using a 6-32 screw, which has nearly the same diameter as the L3.5 mm.





Each of the Concentric rings on the microscope's reticule differs from the next by 0.010", so it should be apparent that it is easy to use this microscope to center the screw directly under the spindle of the mill to quite a bit better than this (specifically, to within 0.001"). Depending on how the photograph is displayed on your screen, you may be able to see the 0.001" marks on the horizontal and vertical bars.

Having determined the screw sizes, and having located their centers to 0.001", I then drilled out the three broken screws. To do this I used a carbide center drill to start the holes on the jagged surfaces followed by a 7/64" drill, which is 5 thou. smaller than the tapping drill for M3.5x0.6. I used this drill because it's easier to make a too-small hole larger, if necessary, than vice versa.



As can be seen from the next photograph, enough of the threads were exposed by the drill, followed by use of a dental pick, that a tap could properly engage the old threads and clean them out. At this point I could have taken the time to modify a 60-deg. carbide threading insert to the 53-deg. 8 min. of a Lowenhertz thread and used it to make my own "perfect" L3.5 tap. However, from what I had found thus far it was likely the actual internal threads were an imperfect variation of L3.5 anyway. So, although the small difference in thread form between L3.5 and M3.5 means a tiny amount of material from the armature itself (rather than just the fragments of broken screws) might have been removed, in the overall scheme of things I think my use of an M3.5 tap should count as a perfect fix of these holes.



To answer a question asked by several people offline, at this point we're less than halfway through this magneto restoration, so there is quite a bit more to come. Send questions or comments to [email protected]

PREPARING THE END CAP:

Removing the Bad Condenser

Even though the condenser the previous restorer used already had failed when I received the magneto, the failure mode of many condensers is to develop a relatively low resistance but to retain the capacitance value. The condenser in this magneto was 0.22 uF, while the original ones still in my two armatures are 0.12 uF. So, not only did the restorer use one that was inadequate to survive the high current pulses (which is a very common problem with the condensers many restorers use in magnetos, accounting for a high percentage of the failures of rebuilt magnetos), it had a capacitance that was too high.

As an aside, the reason the condensers in my two spare Bosch ZEV armatures are still operational after nearly 100 years is they are made of mica. This material is a very stable dielectric (the resistances of both condensers are still over 20 MOhms), but today mica is used only for specialty applications because of the high expense of fabrication as well as it having a relatively low dielectric constant (so the capacitors are large). However, in addition to it being difficult to remove these mica condensers from their cavities in the end caps in order to transfer them to another armature, I can't send my friend out on the road with a condenser that is nearly a century old. Although the mica is very stable, the thin metal electrodes do corrode. So, it was time to come to terms with the epoxy. An hour with my lathe and mill dealt with it, although creating quite a mess.





Before anyone mentions it, normally using a Jacobs chuck to hold an end mill is like using a Crescent wrench instead of a proper spanner. But, because epoxy is soft it only imposes a small side load on the chuck, and precise concentricity with the spindle wasn't important, so there was no reason to use proper collets. Also, I knew I would be using end mills of several sizes to remove the epoxy, and a Jacobs chuck made this messy work go faster.

Testing the Armature's End Cap



It is not immediately obvious, although it is somewhat betrayed by a narrow moat of red epoxy, but the lower 2/3 of the base of the cavity in the photograph is a separate platform sitting on a thin sheet of insulating material. This platform is electrically connected to one side of the contact breaker points via the central mounting bolt that is threaded into the platform. Since the insulator has to withstand ~200 Volt pulses across the condenser every time the points open, I tested it with a high voltage megohmmeter to be sure all is well. As can be seen in the next photograph, the resistance between the platform at the bottom of the condenser cavity and the rest of the armature cap was 8.9 GOhm at 1000 Volts. This is 4000x higher resistance than is needed, and at 5x higher voltage than it will experience in operation.



Although I've described quite a bit of work in my posts up to this point, keep in mind that everything I've done thus far has been just to undo the damage done by the "professional restorer." This is why I cringe every time I open a magneto and discover it already has been "restored." It is also why I concluded some years ago I had no choice but to restore them myself. Only after having undone the damage can I now start the work to make this magneto functional again. Send questions or comments to [email protected]

--------- Sidebar (not in chronological sequence) ---------
Spoiler Alert! -- don't read this sidebar if you want to wait until the final post to find out if this story has a happy ending.
The Bosch ZEV whose restoration I'm describing in this series of posts just finished the 2012 cross-country Cannonball Motorcycle Run. Unfortunately, the bike's rebuilder was able to finish it only a few days before it had to be transported to the start in New York so there was no time for my friend to give it a proper shakedown. The engine suffered a variety of problems that would have been easy to fix under different circumstances, but that resulted in the bike covering only ~800 miles over the course of the Run. However, the restored magneto was trouble free. Added to the ~1000 simulated miles I subjected it to before shipping it back to the engine builder (to be described in a later post), this restored 90-year old magneto has "travelled" nearly 1800 miles without problem thus far. I have every reason to expect it to be good for many thousands more.
------------- end of sidebar ---------------------

TESTS AND REPAIRS OF THE ELECTRICAL COMPONENTS:

At this point everything that needed to be taken apart has been, all the damage done by the restorer repaired, and the parts required to rebuild/restore are in hand. What remains is to test the coil separately, and then rebuild the armature with a proper condenser. After that I'll put it back together, subject the full magneto to a lengthy series of stress tests, partially disassemble it so I can see if any problem areas had developed (e.g. excessive wear of the carbon brushes), and put it back together again.

The Slip Ring



To make sure the material used for the aftermarket slip ring isn't conductive I wrapped a number of turns of stainless steel wire around it and measured the resistance between the wire and the armature housing. I used this wire because I have a large spool of it and it isn't insulated, so it made contact with the metal on the slip ring. Because the slip ring in turn is connected to the coil, this test also measured whether there was any leakage from the coil to the armature. The resistance was over 11 GOhms with a 2500 volt potential applied, which is comparable (i.e. about half) to the voltage it will be subjected to in actual operation. This amount of leakage is completely negligible, so the material in the slip ring is good, as is the insulation between the coil and the armature.

The Coil

I use an Eisemann tester for extended tests of coils because it uses a simple DC motor and set of points to produce the spark (plus I use a modern power supply instead of a 6V battery). If anything in it wears out it will be much easier to replace than it would be to repair the circuitry of a tester like one of my Merc-o-tronics. I ran the coil for an hour on this tester with the gap set to 5 mm (6 kV). The coil works fine so I will use it instead of winding one of my own. Although it is important to test the coil at elevated temperature, because of time constraints I will wait to do this test on the finished magneto. I do have a concern about whether the coil was vacuum impregnated or not, but I am running out of time. Even if it was not vacuum impregnated, that does not mean the coil will fail, but this is an uncertainty I will have to accept given the limited time available.



The Condenser

----------- Sidebar About Replacement Magneto Condensers -----------
For at least 30 years there have been recommendations for a variety of replacement condensers and, figuratively speaking, for just as long the sides of roads have been littered with motorcycles whose magnetos have failed because of using them. Part of the reason for this is that manufacturers typically do not design capacitors for applications where they will be subjected to high current, high voltage pulses while being repeatedly cycled over a wide temperature range in the presence of moisture, organic vapors, ozone, and vibration, so they do not test them under these severe conditions.

Unfortunately, relying on recommendations -- from motorcyclists, suppliers, or professional restorers -- in selecting a replacement is quite risky. For example, several club magazines have recommended completely inappropriate capacitors from stores like Radio Shack that would fail within the first few moments of operation. For a number of years one well-known supplier sold replacement condensers with the claim "Modern substitute, very high specification, zero failure." Despite this claim, they failed in service. Still, it took a number of years before enough motorcyclists complained about failures for the supplier to cease selling them. Currently another supplier is advertising replacement condensers that are of a type not rated for pulsed high current applications, and whose dielectric layers are made of a porous oxide. Even if they survive the current pulses of a stressed magneto, no extended environmental testing has been done of them, so there is no way to even guess how long they might last in the ozone-rich atmosphere near the points before delamination or breakdown of the oxide might set in. For quite a while one well-known magneto rebuilder sold condensers to other well-known rebuilders at a high price that he claimed were custom manufactured to his rigorous specifications. Not only were they just inexpensive capacitors from which he had removed the markings, they also failed in service. A number of magnetos used in the 2010 cross-country Cannonball Motorcycle Run, all rebuilt by the same well-known restorer, failed due to bad condensers.

Despite the examples given above, and many others, condensers that later failed in service always were claimed at the time to function quite well. To paraphrase Bruce Springsteen's 'Magic', as far as replacement condensers are concerned, you are well advised to 'trust none of what you hear, and less of what is claimed.'

Condensers are not mysterious components that are doomed to failure, but a proper replacement definitely does have to have the right combination of electrical, mechanical, and materials properties in order to survive in a magneto. The Fall and Winter 2011 issues of The Antique Motorcycle magazine contain a two-part article by physicist Dr. Charles Falco explaining the science of why the wax paper condensers that Lucas used in their post-WWII magnetos have a limited lifetime, and detailing extensive electrical and environmental tests he made of particular Panasonic film/foil capacitors, concluding from those tests they would last under the harsh operating conditions of a magneto for the equivalent of at least 40 years or 140,000 miles. As well as Dr. Falco having no connection with any supplier that possibly might affect his recommendation, to the best of my knowledge these are the only replacement capacitors that have been subjected to any such battery of tests designed specifically to mimic actual operation. Because they have the proper electrical specifications and survived a year-long set of accelerated electrical and environmental tests conducted by an independent party, these are the ones I use. Luckily, I have a large stock of them, because they are now no longer in production. Until someone like him conducts tests to identify another, in-production replacement, if you use any capacitor other than these Panasonics it is at your peril.
----------- End of Sidebar -----------

The next photograph shows a complete condenser pack that is ready to install in a Lucas armature, consisting of two 0.082 uF Panasonic film/foil capacitors soldered in parallel to give 0.16 uF total. My notation on it shows I tested the pack after assembling it and found it had the correct capacitance and that its leakage resistance was greater than 20 GOhm when measured with 500 Volts (i.e. 1000x more than it needs to be, at a voltage at least 2x higher than it will experience in use). The pack I used in the Bosch was the same as this one, minus the mounting bracket, and I installed it in the end cap using the minimum amount of 3000 psi epoxy necessary to insure it would stay in place.



Unlike BTH and Lucas magnetos, the recess in the end cap of the Bosch is deep enough that I could have used a single 0.18 uF capacitor from the same Panasonic family instead. They are sufficiently fatter than the 0.082 uF that they can't be used in BTH and Lucas magnetos, so two of the smaller ones have to be connected in parallel instead. However, when I reached this point in restoring the Bosch it was more convenient to use one of the pre-soldered pairs that I already had on hand.

Although earlier I said that the original Bosch mica condensers have a slightly smaller value, at 0.12 uF, than the 0.16 uF I used, that's a bit too low for the properties of this rewound coil. The inductance of the original ZEV coil I have is such that 0.12 uF was not optimum for them, either. However, the mica condensers completely fill the cavity so it would have been impossible for Bosch to make their capacitance any higher than this.

Send questions or comments to [email protected]

TESTS AND REPAIRS OF THE ELECTRICAL COMPONENTS (CONTINUED):

Reassembling the Armature

Although I have every reason to expect this magneto to function for at least a few decades with only minor maintenance (e.g. new points and brushes at ~10,000 miles), it is good practice to avoid doing anything during any restoration that could not be easily undone in the future. Although I could use Loctite on the armature screws without technically violating this, instead I use a torque wrench followed by pinning the heads of the screws using a small punch on the surrounding brass, as was originally done to secure them. This is illustrated in the next two photographs, using as an example the head of a screw in one of my spare Bosch ZEV armatures, with the original pinning that was done by the factory.



The torque specifications for Fasteners depend not only on the quality of the material used to make them, but also on the shapes of the heads. Because of this there is a fair amount of variation in the values quoted in handbooks for 3.5 mm steel screws, and in the end I decided on 10 in.-lb. as reasonable for these flathead screws of unknown quality steel. This value falls nicely in the range of one of my smaller torque wrenches.



Testing the Reassembled Armature

After reassembling the armature with the new capacitors I checked the output using one of my Merc-o-Tronic 9800 testers. The blue arc in the window of the tester shows the operation of the coil is reliable at a test current of just over 1 Amp supplied to the primary. This is as it should be, so the completed armature passes this test.



Brushes

As I mentioned in an earlier post, many aftermarket brushes in circulation are too hard, and many are too soft. Since I don't know the origin of the brushes that already were in this magneto, I made a note during my initial inspection to replace them with NOS Lucas brushes if possible. Luckily, the Lucas brushes (0.193"-0.194" diameter) fit without problem in place of the 0.185"-dia. brushes that were in it, easily taking care of this issue.

Pursuing this a little further, I subsequently conducted a test to see how soft the brushes were. Pressing down with the pressure I would use to write with a pencil, I made a half-dozen lines side-by-side using a pencil and four brushes, as shown in the next photograph. Since I don't know how well the subtleties will reproduce in image, following it are my observations:



-- No. 2 pencil (approximately the same as an HB pencil): This serves as a convenient hardness standard others could use for comparison with their own measurements.
-- "Bosch ZEV": Aftermarket brushes that were in the ZEV. These lines are perhaps a tiny bit lighter than the pencil, but comparable.
-- Lucas 451260 HT pickup brush: Significantly lighter than the aftermarket "ZEV" (note: the lines were so light that I pressed harder when making the middle line, which is why it is darker than the ones on either side of it).
-- Lucas 200737 dyno/generator brush: Same as aftermarket "ZEV."
-- Lucas 451260 magneto earthing brush: Same as aftermarket "ZEV."

The differences might be more easily seen in the following composite micrograph, where from the left the three lines are the No. 2 pencil, the aftermarket "ZEV" brush, and the Lucas HT brush. Although it may not be apparent in the photograph, in a stereomicroscope it even can be seen that the Lucas brush was hard enough to tear small fibers from the paper that are standing above the surface.



Although the above test isn't quantitative, it definitely shows that the brushes the "professional restorer" had installed were significantly softer than the ones Lucas supplied for this application. Also, although the other two Lucas brushes I tested are about the same hardness as those that came in the ZEV, they are meant to run on metal surfaces, not phenolic. My speculation is the reason many aftermarket HT brushes are soft is they were manufactured using the (incorrect) specifications for brushes intended to be used on copper or brass. Although phenolic is softer than these metals, it is more abrasive (because of this, carbide rather than high speed steel is commonly recommended for machining it). Send questions or comments to [email protected]

Overall, a very interesting and informative series of posts, Magnetoman. A few comments/queries:-


Could you explain this peculiarity of the Bosch ZEV? In all other models of rotating-coil magneto I've examined, that black cylinder (the earth brush) is part of the secondary circuit, and it's doing its business primarily when the points are open to provide continuity for the plug current from the magneto body (and therefore the sparking plug body) back to the armature body bypassing the armature bearings and their insulators.


I'm not so sure you should draw the conclusion from this test that the slip ring is good. 11 gigaohms at 2500 volts can certainly break down to a far lower resistance at the higher operating voltage of the slip ring. Breakdown of aftermarket slip rings is indeed a problem, but in my experience it often only rears its head at elevated temperature. A slip ring may test good at room temperature, but start breaking down at operating temperature.

I am also concerned that, with this test, you are applying the 2500 volts not only to the slip ring conductor and the HT tail of the HT winding, but also to the whole of the HT winding and more worryingly the LT winding. In operation, the insulation around the LT winding sees voltages of perhaps 250 V, and it might be fair enough to test it to three times that voltage, but testing it to ten times that voltage is, IMHO, asking for trouble. Even if it passes the test, the test may itself trigger the onset of a later breakdown. If you were a professional restorer and if you were to do that, I think you ought, as you have said in the past, to give your customers their money back.


The Eisemann tester looks like a nice piece of kit. But does it get the armature up to operating temperature, or do you heat the armature elsewhere and then test it on the Eisemann while it's still warm?


If you are going to have yet another tedious go about the Brightspark EasyCap, it might help people if you named names instead of leaving them wondering. The fact is EasyCaps have undergone extended environmental testing; they carry on working; they are guaranteed; if one were to fail, it is a ten minute job to replace it without the need to pull the armature apart and mill out the old one as you have described in detail above; and last but not least the customers and their bikes are happy.


Yes, that was a very interesting article. One thing I did find puzzling about it was Dr Falco's reasoning that a breakdown of the waxed paper dielectric caused an increase in the equivalent series resistance of the capacitor, when schoolboy physics would suggest it causes a decrease in the equivalent parallel resistance of the capacitor. Did you understand that?


Crikey, that's not much help then, since they're out of production. Do you know whether Dr. Falco has any plans to test any capacitors which are in production?


Yes, if securing a capacitor in the end cap with resin, it is important to use the minimum necessary. We have seen some where the resin totally fills the end cap, bridges the gap between it and the windings, and encapsulates the tails of the winding. Removing a dud capacitor without also damaging the winding insulation or the tails is then, ummm, difficult to say the least. Some people use a blob of bathroom sealant, which seems to work quite well.


Again, do you recommend a room temperature test or a hot test?

But as I said at the start, a very interesting and informative series of posts. Keep up the good work!

Ken
Brightspark Magnetos

Thanks for catching the miswording.

Your experience certainly is different than mine. I've used both an IR thermometer and a thermocouple to measure the external and internal temperatures of a number of magnetos, and my measurements found the highest temperature a slip ring will see under the worst of conditions is well under 50oC. There is simply no physical mechanism I am aware of that could increase the electrical conductivity of whatever impurities are in a phenolic-like plastic by many orders of magnitude over that limited range. And, even if there were, the failure mechanism for a slip ring is completely different than that of a condenser that is warm. In the case of a slip ring it would be dielectric breakdown of the material, not increased electrical conductivity of the material itself.

I don't know what measurements you made that found slip rings that are good at room temperature break down 20-30oC higher. All I can guess is that either the slip rings you measured started out with very much lower resistances than 10 GOhm, or you misinterpreted whatever observations you made that led you to this conclusion. Perhaps arcing through a track of carbon that you hadn't noticed. This is why I test them at 2.5 kV, to reveal any such issue.

You have a fundamental misunderstanding of the electrical circuit of a magneto. When the 2.5kV was applied to the coil in that test all of the wires of the primary and secondary were elevated to that potential. There was no voltage whatever between one layer of windings and the next, so the insulation on the LT (and HT) wires is irrelevant.

The only places where there was a large electric field during this test was between the operating surface of the slip ring (where I had wound the stainless wire) and the nearest piece of metal on the armature, and between the bottom winding of the primary and the core of the armature over which it was wound. Had the person who made this coil wound the first layer of the primary directly on the armature, rather than properly having a layer of insulation there first, the primary certainly would have broken down since whatever insulation is on the LT wire itself would have failed at less than 1 kV maximum. Had the 2.5 kV induced arcing of the primary to the armature, it would have been a good thing, since it would have revealed the fact the coil had been defectively wound. This is why this is an important test to conduct and, rather than giving someone's money back for having conducted this test, quite the opposite.

Good question. I should have written in that section that the only way to be sure the coil would not fail in operation would be to test it while at elevated temperature. This is what I would have done at this point in process of rebuilding one of my own magnetos. However, in the case of this Bosch, I knew I would later test the entire assembled magneto for an extended period at elevated temperature (to be described in a future post), so I bypassed testing the coil alone at elevated temperature. In several places in my posts I explicitly point out where I took shortcuts because of time constraints forced by the deadline, and I should have mentioned that here as well.

I had written all that I planned to about condensers, so if you consider the following text tedious, you're responsible for prompting it:

Your website is not the easiest to navigate, but the only information I could find on "extended environmental testing" is a page in your FAQ section where you describe finding no degradation after one of your capacitors spent a day in a pan of hot water. That might be a fine test for sugar cubes, but says nearly nothing about the long term survival of your capacitors in a magneto. I can only infer that the fact you devote a page by itself to this test, as if it were meaningful, that it indicates you actually think it is meaningful, and that it somehow counts as "extended environmental testing." It does not. Further, your capacitors are located in the points housing, where there is an elevated level of highly reactive ozone created by the sparks from the points. I could find nothing on your site about testing whether the porous oxide of your capacitors degrades in the presence of ozone.

Given the long and checkered history of inappropriate capacitors being sold for use in magnetos, the onus falls on the supplier to conduct appropriate tests to indicate yet another new capacitor isn't going to fail. As far as I could find on your website, you have not conducted such tests, which are needed independent of whether or not you offer a free replacement if/when they fail. There is also the issue of the degraded performance of the magneto itself as a result of removing the armature to install any condenser, including yours, which I will deal with in a quantitative fashion when I come to the point of magnetizing the Bosch in a future post.

I searched the archives and can't find that I've ever made a single negative comment about your EasyCap. The sidebar about capacitors in my previous post is the first time I've said anything about your capacitors, and even then not by name. Unless I've missed something, if you say people are having a "tedious go" about the product you are selling, it means people in addition to me have issues with these capacitors.

Schoolboy physics doesn't have much to say about the breakdown of capacitors. But, if it did, it would say that the use of an equivalent series resistance (ESR) is a very widely used approximate model found to be quite useful in describing the electrical behavior of capacitors when used in a circuit. If you Google "equivalent series resistance" (in quotes) you will get over 800,000 hits, indicating how widely used this model is. On just the first page of hits you will find good explanations for why it is so useful to lump together all the complex internal physical phenomena into a single ESR

The same criticism certainly could leveled about the Lucas HT brushes I suggest using, since they've probably been out of production for 40 years. And yet, it's possible to find them.

Yes, this is one of the reasons I cringe when I discover a magneto has been "professionally restored," since this is one of the atrocities I sometimes find that has been committed. However, although it is a headache, I've found it to be not all that much worse than the one I had to deal with on this Bosch. It's easy to tell where the coil ends, so the epoxy holding the coil to the endcap can be milled away by mounting the armature tilted slightly from horizontal (or by carefully using a Dremel).

But, the broader issue is whether there is an easy way out. One that avoids having to deal with whatever condenser is currently in the magneto, but that still restores the performance to the as-new level. I would love it if there were. If there were, I wouldn't have had to waste an hour of my life milling away all the epoxy the guy used who restored the Bosch. However, the fact I went to all that trouble isn't because I enjoy cleaning up wads of epoxy chips. It's because my experience tells me there is no proven-reliable capacitor I could use other than the Panasonic. I wish there was an easier way to restore proper performance, but I am unaware of any such easy way out.

The use of bathroom RTV sealant is definitely the sign of someone who doesn't know what he is doing. If the RTV is still in place, it only means the electrical leads on the inappropriate replacement capacitor were strong enough to hold it in place for the short time the magneto was used before the capacitor failed, because the shear strength of RTV is too low to have done any good. Do not use bathroom sealant.

To briefly repeat the answer I gave earlier, about the test on the Eisemann tester, I "always" (except for this Bosch) run this test at elevated temperature. A future post will describe the extended test I ran of the assembled magneto at elevated temperature, which it passed.

TESTS AND REPAIRS OF THE ELECTRICAL COMPONENTS (CONTINUED):

Brush Spring Pressure

In addition to hardness, spring pressure is the other important parameter for a brush. I determined from measurements of the depth of the recess in the HT pickup that compressing the spring until the bottom end of the brush is 5 mm above the surface is a good value to use to simulate the average pressure on a brush when it is partially worn. Since Pressure = Force/Area, if a spring has the incorrect spring constant it will exert either too much or too little force on the rubbing surface of the brush (Force = -spring constant x displacement). The next photograph shows how I made this measurement using an electronic scale with a resolution of 0.01 g:



What I found was the brushes that were in this magneto had springs that are ~25% stiffer than those on Lucas HT brushes. This means that not only were the "KEV" brushes the restorer used significantly softer than proper ones (so they would have worn away faster), they were subjected to 25% greater pressure on the slip ring (wearing the brush away even faster still). Had I not replaced these brushes with proper ones the slip ring soon would have been covered with a conductive coating of carbon that likely would have resulted in an internal short, bringing the motorcycle to a halt.

Contact Breaker Points

Although the points in this magneto were fine, had they not been, and if there were no replacement readily available (e.g. cannibalized from my ZE1 magneto), I would have had to rebuild them. This can be done by silver soldering short lengths of pure tungsten TIG electrode rod over the ends of the worn points. Tungsten wets easily with certain silver solders, and a single 7" piece of 4 mm-dia. pure tungsten rod is enough to restore a lifetime's worth of magnetos. The end of one of these TIG electrodes is shown in the next photograph using my unrestored Bosch ZE1 for scale.



Note that if you rebuild your own points you need to use pure tungsten TIG electrodes, which are painted green on one end, as in the photograph. Do not use rods with either red or yellow ends because they contain 1-2% radioactive thorium. Because of the nature of the radioactivity, in solid form this isn't an issue, but you don't want to breathe the dust and have it permanently settle on the surface of your lungs. Obviously, it is difficult to form the W pieces into the necessary shape without creating dust. Also, the reason for thorium is that it makes it easier for the TIG electrode to emit electrons, and that is the opposite of what you want in contact breaker points. When the points separate, you would like it to be as difficult as possible for an arc to form.

Because platinum oxidizes only very slowly, Pt points (actually, Pt alloyed with Ir, Ru, and/or Os) were used in the earliest magnetos before it was realized laminated armatures significantly improved the low speed operation for which it had been felt Pt provided an advantage. Perhaps because of this, and citing a "lower surface resistance," shop manuals often say Pt should be used for the contacts in magnetos intended for higher performance motorcycles. However, despite what is commonly written, my measurements show that the difference in surface resistance between the two metals is negligible unless the magneto sits unused for some months in a humid garage, in which case tungsten can benefit from a light burnishing to remove the resistive oxide.

The next photograph shows that the resistance across the tungsten points in a Lucas KNC1 magneto is only 9.4 milliOhm (actual values varied between 2 and 11 mOhm when measured after successive make/breaks when rotating the armature by hand). I found the same range for the W points in the Lucas KVF beside it. An important fact is that both of these magnetos were given new tungsten points 16 years ago and have been sitting unused inside the house ever since. Even after this extended time, the oxidation of the W was negligible.



Since 10 mOhm is less than 2% of the resistance of the primary coil itself, the contact resistance of these points will only reduce the primary current, and hence the output of the magneto, by ~2%. What this means is, even if the resistance of Pt points were zero, the output of two otherwise identical magnetos with Pt and W points in good condition would differ by a negligible amount.

Although Pt will resist oxidation in a humid environment better than W, it reacts with organic vapors to form PtC, which is both resistive and rapidly erodes the contacts. The choice of materials for magneto points is addressed in Chapter V 'The Magneto-Type Ignition Systems' of Electrical Contacts Data Book: Materials, Designs, and Applications (P.R. Mallory & Co., 1945): "… because of the location of the magneto and its design, the contacts may be subjected to oil mist. In such cases it has been the practice to use tungsten for the contacts, because platinum and its alloys have a tendency to become brittle… [causing] high contact resistance and a corresponding high rate of electrical erosion." Since organic vapors typically waft about on old motorcycles, my conclusion from such reference books as well as my own measurements is that tungsten is the better choice for motorcycle magnetos in general, and for this Bosch ZEV in particular. Send questions or comments to [email protected]

Hello,

Just an observation here. I'm not a magneto expert, but I do work with electric motors for a living, and I'm having trouble understanding the above statement in your previous post. In my experience, if a coil of wire is conducting a current, there will be a difference of potential between the wires in different layers of that coil, because any length of wire will have resistance. This is why a coil can short circuit to itself- a turn to turn short. In my industry, a turn-to turn short is one of the more difficult faults to catch, and is usually caused by overhandling the wire during coil winding. Vacuum impregnation of potting varnish helps, but if two wires are in contact with each other, and their varnish coating is degraded, it will prove to you quickly that there is a definately a voltage between the layers of a coil.

Perhaps I'm not understanding a subtle difference between a magneto and an electric motor.

This is not meant to detract from a thoroughly enjoyable and informative thread. I'm looking forward to the next update.

Kevin

This is a very reasonable question. In the context of your question, there is no difference between an electric motor and a magneto (although, it actually is a transformer). However, this particular test is different than the one you normally would run on an electric motor. In the test I ran there is no current being conducted by the wire.

First, assume all the insulation on the wires themselves are fine, and a layer of insulation thick enough to withstand 2.5 kV was applied to the armature before the primary was wound on it. That is, assume the coil was properly wound. Since none of the leads from the coil are attached to the body of the armature during this test, applying 2.5 kV between the slip ring (which in turn is connected to the HT lead) and the armature body will cause a tiny current to flow for a moment until the potential of all of the windings are elevated to 2.5 kV with respect to the armature. Since all of the wires in the coil are connected together, once this potential is established there is 0 V between any two layers of the coil. Therefore, even if the insulation were missing from some of the internal coils, and two adjacent coils were shorted together, no current would flow between them -- in this test -- because there is no voltage difference to drive the current.

What may be confusing you is this isn't quite the test you would normally do for an electric motor. In an electric motor you would actively drive current through the coils by connecting your ohmmeter to the leads. In this case, current would flow between coils that were shorted together.

What the test I did checks is if the various insulators between the coil and the armature body are good. It says nothing about whether there are internal shorts between the coils which, as you say, is something of great concern. To test for internal shorts requires a dynamic tester, like the Eisemannn or Merc-o-tronic. If there were internal shorts, but not ones so bad as to make the coil totally inoperable, that would be revealed by needing to apply a larger AC current to the primary. You can't do this dynamic test for shorts on an electric motor because it only has a "primary," so you can't measure the AC current required to create a spark from the "secondary."

Although it gets a bit off topic, for testing for internal shorts on a dynamo (which is completely equivalent to your electric motor), I have an Evoibi coil tester. It measures the losses in the dynamo at two frequencies. Since the frequency dependence of the losses from internal shorts is different from the basic coil itself, this reveals whether there are any such shorted turns. By the way, these testers were developed for testing electric motors, where the market is a bit bigger for diagnostic equipment than it is for old motorcycle dynamos. Unfortunately, the large core losses of a magneto's armature makes it impossible to use such testers to determine if there are shorted coils in them. When working on motorcycle electrics, one can't have too many testers...

There is no need to get your back up. If ever I answer a different question than the one you intended to ask, all you have to do is to point that out. The wording of what you wrote in your previous post is consistent with the same misunderstanding expressed by someone else in a question in a subsequent post, the answer to which I provided. Fine that you say this isn't what you meant.

As I've written in a few places, I had no way of knowing if this coil had been properly wound, or if it had been vacuum impregnated. If it had been properly wound, there would be a layer of insulation around the core of the armature that easily can withstand 2.5 kV. If not, my test would have destroyed the coil and I would have had to spend a day winding a new one.

Had the coil been directly wound on the armature, and even if it had been vacuum impregnated, not all of that resin might have made its way as far as the first layer of windings. In operation, those windings are subjected to a cyclic electromagnetic force as the armature rotates, and relative motion caused by this force can and does abrade insulation from windings. Even if the coil passed my other tests, there still would be the non-negligible possibility it would fail due to this abrasion after some unknown number of miles.

Even though the coil already had passed one dynamic test with a Merc-o-tronic tester, that test cannot determine if the primary was improperly wound directly on the armature. Again, the reason for doing this test is a properly wound coil will pass it, and an improperly wound one will fail.

My experience is that no slip ring has passed this test but failed in operation. The resistance was 10 GOhm (i.e. ten-thousand-million Ohms, to emphasize the point) at that voltage, and either the resistance would have had to fall by many, many orders of magnitude between 2.5 and 6 kV to fail (which is physically implausible), or something in the slip ring that survived 2.5 kV would have to suffer dielectric breakdown at 6 kV. Amongst the many things I might worry about happening on a magneto, this is way down on the list. This is why I concluded it was good.

I do it this way because it simultaneously tests whether there is insulation on the armature and whether the operating surface of the slip ring has a sufficiently high resistance so as not to be a concern. Attaching the tester to either of the primary coil leads wouldn't test the latter.

Already answered.

As I wrote previously, there is a long and checkered history of condensers that have been sold with seller's written assurances as strong as the following: "Modern substitute, very high specification, zero failure." Despite that seller's unqualified assurances of not a single failure, an unreasonable percentage of them failed in service. I have no way to confirm your own claim that there hasn't been "even a single failure" of yours, nor do I even care. What I do know is I have enough legitimate concerns about your capacitors, some of which I have expressed, not to have any interest in relying on them myself.

However, an important point is, it is not up to someone else to demonstrate your capacitors do have problems in service. It is up to you to provide credible test data to indicate they do not. Having them sit in a pan of hot water for a day, or telling us that they have had zero failures, is not sufficient.

I'm not going to take the time to look for the quote, but at least one treatise on magnetos explicitly comments on the importance of not sealing the points housing because of the buildup of ozone levels. That's why even a Lucas Wader magneto has a way for water to get in -- in order to let the ozone out. When a capacitor is placed in the cavity at the other end of the armature, the reactive ozone created by the points, or any of the brushes, has to make at least a few right angle turns in contact with metal to get as far as the condenser, which itself is covered by epoxy as well as its own coating. In contrast, your condenser is directly adjacent to the points, where the ozone level is highest, and not protected by anything other than itself.

The simplified, not "simplistic," ESR model was used in an article in a motorcycle magazine to explain to that audience the essence of why Lucas wax paper condensers failed with time. This present thread is about restoring a c1920 Bosch magneto, and has nothing to do with failure modes of various condensers. However, since so many types of replacement condensers touted by so many people over the years have failed, I included a sidebar about capacitors because I believe it is important for people reading this to understand why I used the one I did in order to have high confidence it would survive long term. You have not reported any tests on the capacitors you sell that would be needed to give me the necessary confidence to even consider using them.

At this point I think we have covered capacitors in far more detail than anyone reading these posts cares to know about them, so please direct any future questions to any aspects of this restoration other than capacitors. I'll be happy to discuss any and all issues, other than why I don't use the capacitors you are promoting.

MECHANICAL COMPONENTS:

Since the electrical issues seem to be under control, it's time to turn to the mechanical components.

The Bearings

The bearings and races were new and of reasonable quality, as shown in the next photograph.



The next photograph is of one of the races, taken through the eyepiece of a stereomicroscope. The track for the balls that runs vertically is approximately 0.1" wide. Because of physical interference, it was impossible to place either race in my metallurgical microscope without removing it from the armature, so I couldn't make a higher magnification micrograph (since the races were new, I didn't want to remove them since the stereomicroscope already had revealed what I needed to know). The important thing is, the bearings pass this microscopic inspection.



Had I needed to remove the races from the armature, I made brass clamps for the two common sizes of magneto races. Although it may not be easy to tell from the photograph, the inner surface of each clamp is machined with a convex profile to make contact within the track of the race around the entire circumference to ensure there can be no damage caused whatever. Once clamped on the race I use a two-jaw puller along with a brass plug at the end of the armature for the screw to push against.



In addition, for "emergency" use away from home I machined a standard bearing puller to give it an inner surface that matches the track of the race of the larger of the two bearings. When used on the smaller race it does not make full contact everywhere on the circumference, but I calculated that the maximum pressure it exerts is far below the yield strength of the bearing steel so it still is fine in spite of the imperfect fit.



Pullers are sold for magneto races that use four radial screws to clamp on the groove of the race. I definitely do not recommend using them since the tips of their screws subject four small regions of the race to very high pressures when removing the race. Even a small mark made in a track can seriously reduce the lifetime of the bearing.

Another use of the modified bearing puller is to remove slip rings. Slip rings are made from brittle and fairly fragile materials (e.g. Bakelite, Tufnol, or similar phenolic). Further, they are tapered, so the outer edge that one is tempted to pull on is the weakest. However, by loosely clamping this bearing puller inside a slip ring I can gently apply uniform pressure to the stronger inner circumference using a two-jaw puller. While the modified bearing puller serves double-duty as a slip ring puller when away from home, the next photograph shows an aluminum puller I made specifically for this purpose.



Reinstalling races can be done with a socket of an appropriate size and an arbor press, but I made a "tool" specifically for doing this. I haven't included a photograph because it looks pretty much like a socket. For "emergency" use a hammer can be used instead of an arbor press, but I much prefer to use a press whenever possible.

Both of the outer races are insulated from the housing by thin paper "cups" to eliminate them as paths for the current that flows between the coil and earth during the spark. Ideally, the resistance of the cups would be infinite so no current at all could flow. Such a current would electro-etch the bearings over time. However, as long as the resistance of the cups is significantly larger than the 10 mOhm or so of the earth brush, only a small fraction of the current will take the path through the bearings. For example, even if the resistance were as little as 100 Ohm;, only ~0.01/100 = 0.01% would flow through the bearings. The resistance of the paper cups under both races was greater than 10 kOhm;, so I did not have to remove either outer race of this Bosch ZEV to replace the insulators.

Often races are held quite firmly in the housing, which is a problem when it is necessary to remove them. A blind bearing puller is perfect for this, except one of the races is located deep inside the main housing where it is difficult to reach. These pullers work by expansion, which requires holding the "nut" that is in the middle of the puller while turning the "bolt" at the outer end to expand the tapered mandrel onto the inner surface of the race. Since the "nut" is otherwise inaccessible inside the housing, I fabricated a clamp small enough to hold it while still fitting inside the housing. This puller and clamp is shown on top of a Lucas Magdyno housing in the next photograph, with the expanding end positioned at approximately the depth of the inner race. A socket and 3/8"-to-1/4" reducer is on the "bolt". Send questions or comments to [email protected]

MECHANICAL COMPONENTS (CONTINUED):

The Cam

Although a Harley-Davidson is a "45-degree V-twin," the HT leads of the Bosch ZEV magneto don't fire 45-degrees apart. The configuration of this engine is equivalent to a vertical twin (e.g. BSA 650) whose cylinders have been split to be at 45-deg. rather than 0-deg. with respect to each other. As a result, lobes one and two on the points cam need to be 157.5 degrees apart. The reason for this is, since the magneto operates at half engine speed, 157.5-deg. corresponds to 315 degrees of rotation of the crankshaft. Thus, if one cylinder fires at 0-deg., the second fires after the engine turns 315-deg. further, i.e. 45-degrees less than another full revolution (360-315=45). It then takes another 45-deg. for the crankshaft to get back to the first cylinder, which is now on its exhaust stroke, plus another full rotation before that cylinder is ready for another spark, which is 405-deg total (45+360=405). So, the angular separation between the second lobe and the first needs to be half that, i.e. 202.5-deg. Since 157.5 + 202.5 = 360, this all works out.



I used my Talyrond roundness tester to check the contour of the cam itself. The resolution of this instrument is overkill for what is needed, since it lets me center and level components to within 10 micro-inches, measure heights to 10 micro-inches, and determine angles to arc-seconds . However, if I didn't have the Talyrond, the 15 arc-sec. resolution of my mill's rotary table, along with a mechanical dial test indicator, would have been fine for these measurements (although not as fast or convenient to align as the Talyrond). Given how bad other aspects of the restoration were, I was afraid I might find the cam to be a poor tolerance attempt to be 45-deg., or even a 42-deg. cam for an Indian. If this were the case, I either would have to quickly find a replacement (that might turn out to be no better), modify it, or machine one from scratch. However, I was pleasantly surprised to find the angular separation between the lobes at the positions where the ramps would open the points by 0.012" to be precisely 45.0 degrees (or, rather, 157.5 and 202.5 degrees).

Unfortunately, the contour of the cam alone does not determine where this magneto will fire, since the OD of the cam ring locates it in the ID of an adjustable housing which in turn is a sliding fit in an annular slot in the magneto. Because of this, there is no way to know without measurement how the various tolerances add up to affect the final timing. So, once I knew that at least I was starting with a good cam profile, I assembled it in the magneto, centered the axis of rotation of the armature precisely on the axis of rotation of the Talyrond, and again measured the cam profile with respect to that axis.



I found when the cam assembly is installed on the body of the magneto the tolerances do add up to cause some deviation of the timing from the 157.5/202.5 values. However, most or all of the problem is intrinsic to the design of the magneto itself, which allows a few thou. side to side motion of the cam because of the need for a sliding fit of the assembly for advancing and retarding the ignition. Because of the slope of the ramps on the cam, a sideways displacement of a few thou. has a significant effect on the timing. However, the deadline for getting this magneto finished and shipped back to the guy who is rebuilding the engine is rapidly approaching, so there isn't time to re-engineer the design and fabricate something with tighter tolerance than it had when it left the factory. I will proceed with it as it is but, if I find the timing too far off in my later dynamic test, I will have no choice but to devise a solution.

Concentricity of the Armature Components

To be sure the components of the armature are properly Concentric after I assembled it, I used a precision bench center with four dial test indicators of resolution 0.0001" (ten-thou.). I reconstructed the setup for the next photograph a few days later using a different armature and only three of the four indicators.



In fact, this is not how I hold armatures in place for these measurements, since the races determine the axis of rotation, not the ends of the shafts. However, the fixtures I designed to locate armatures with respect to the races in this bench center would completely obscure what is going on, so what I show here is more of a schematic to illustrate the important points. The indicator at the left is shown measuring one of the races. In this setup, if the race were perfectly Concentric with the dimple in the end of the shaft, the indicator would not waver as the armature was rotated. However, what is important are the races, not the dimple, so the fixtures not shown here are designed to use the races as the reference surfaces. Similarly, the middle indicator checks for the run out of the slip ring with respect to the race, the third indicator would check for the run out of the earth ring if this were a more modern BTH or Lucas, and the four indicator (not shown) would check for the run out of the second race. The Bosch uses a different surface for the earth brush, so actually I used an indicator on the fixture located on the second race rather than the armature housing.

Another subtle alignment issue that also needs to be checked is if the shafts at each end are coaxial with each other. As an example of why this could be a problem, imagine if while reattaching the end cap that I caught a metal chip between it and the body. This would cause one shaft to be tilted with respect to the other. While such an alignment problem may not be likely when carefully assembling an armature that left the factory as one piece, it certainly cannot be relied on when dealing with a rebuilt armature that might be a mongrel made up of components from several others. The operational consequences of this would be that it would misalign the bearing races (causing premature wear).

Since the shaft at one end of the armature is stubby, checking alignment accurately presents a measurement problem. Although I used the Talyrond for this measurement, I could have done it more precisely than the way I did, albeit only if I spent more time setting it up. Instead, it was faster to check if the shafts were coaxial by installing the points block in the taper of the armature's shaft and measuring the run out of the back face of the block near the outer edge. When I did this I found the run out was several thou., but it also varied by a few thou. each time I removed and reinstalled the block. In any case, this is small enough not to be a concern, so finding the source(s) of the small run out and reducing it further wasn't necessary.

Although the armature itself was in good alignment, I determined that there was almost 0.01" of run out of the surface of the slip ring, which would cause unwanted motion of the brushes in their holders. So, I mounted the armature in the lathe using a collet to hold the race at one end (again, because the race is the surface that determines the axis of rotation), and a high precision live center at the other end, as laid out schematically in the next photograph using a different armature. The total indicated run out (TIR) of the spindle of my lathe is less than 0.0002", and the TIR of this precision center is 0.0001", so this machining operation results in an armature that is at least as good as it was when it left the factory c1920. I then polished the track to achieve a measured roughness of just under 1.5 um.

A critical observer will notice the lathe collet clamps on the outer surface of the race, not the track the balls run on. However, I've verified with several races on the Talyrond that these two surfaces are Concentric to at least 0.00002" (20 millionths), and that on this particular race they are to at least 0.0001". The lower precision is because I measured it using a mechanical indicator in the assembled configuration on the bench center rather than with the electronic indicator of the Talyrond. Also, it would have been more precise if the live center somehow could have been fixed to the race at that end of the armature rather than in the inner taper of the shaft. However, to do this would have required machining another precision fixture, and any added precision from going to this effort would have been negligible anyway. That race is ~4" from the center of the slip ring, which in turn is only ~0.5" from the race held in the collet. Because of this ~8:1 leverage, the ~1 thou. run out I measured for the race at the tailstock end with respect to the ID of the shaft as held by the precision center added a "wobble" of only a ten thou. to the slip ring when mounted on the lathe. This is comparable to the TIR of the lathe's spindle, and is at least 10x better than it needs to be.



Send questions or comments to [email protected]

Agreed.

Why would the rubbing block wander in and out? (apart of course from the effect of the cam.)