I would be much more worried about microscopic (or larger) metal particles being rubbed from either meshing gear through any mismatch of tooth form, and then travelling around the engine. With aluminium as one of the gears you need to be alert to microscopic particles becoming embedded in its teeth, and those then forming a lap which wears the mating gear.

Deviation from specification is not always a sound idea unless you have a very firm grasp of the underlying design criteria and the likely consequences of changes.

Your leaving the forum would be a significant loss to the helpful and well thought through contributions to solving the problems of others. At least stay online for a while to monitor the evolution of the forum. There are many of us who would not wish the forum to degenerate to the level of the lowest common denominator as seems to happen in so many other aspects of our modern society.

Pat, looking at your photos you are going to have to dismantle the armature in order to disconnect the old capacitor (condenser). If that is not done it is likely to be a pointless exercise in mounting a 'Brightspark' capacitor at the points plate. You need to follow the 'complete works' chapter on the Brightspark website in order to repair the magneto that you are working on. Any defects in the old capacitor, other than a complete open circuit, will continue to influence the working of the magneto. Even if the capacitor is currently open circuit, it may well re-establish a connection and hence a fault condition, when in service.

I hope that you followed the advice to provide magnetic 'keepers' for the magneto body before removing the armature.

Pat, as you will have now seen, the capacitor (condenser) is contained within the brass body at the contact breaker end of the armature. The centre screw that I said yesterday should not be insulated, actually screws into the body of the condenser. As a consequence you need to retain some of its physical presence - if not its electrical connections. The Brightspark tutorial is good but you need to be very careful removing the slip ring when dismantling the armature. The HT wire from the armature coil usually becomes corroded where it enters the slip ring so clean it well when putting things back together. The wire is just a push fit into the hole and establishing a good contact is a bit hit and miss.

For the WICO style of rotating magnet magneto Martin Perman is correct in that the capacitor (condenser) is mounted in an easily accessible position. The magneto currently in place on this engine would appear to be a rotating coil magneto and with those it is common practice (but not universal) to embed the capacitor in the armature.

The capacitor is embedded within the armature. Unfortunately it can only be accessed by a complete strip-down of the magneto.

Edited By Swarf Maker on 26/06/2019 01:00:14

The resistance that you measure when the points are open is that of the primary coil on the armature. That is as it should be. If there were an open circuit that would be a fault condition!

The fact that there is resistance present is not a positive test however. Shorted turns on the primary coil of the armature will not show up. Also the capacitor is connected across the points and thus across the coil as well. These capacitors are paper/foil/mica constructions and sometimes (but rarely) fail showing a low resistance. More likely is that dampness and corrosion degrades the connections between foil layers and their capacitance value falls dramatically. This failure mode often occurs when a damp/degraded capacitor is exercised after a long period out of service and is the most likely cause of your symptoms. You started with a strong spark, the energy of which the capacitor couldn't cope, and then the capacitor failed, spoiling the tuned circuit and hence reducing the spark.

All repairable/replaceable but not for zero outlay.

That screw is not meant to be insulated. It forms one of the connections to the inner workings of the armature.

Yes, F360 files can be stored locally. A recent upgrade also means that you can export them to 11 other CAD file formats for use in other programmes.

You can also import quite a wide range of file formats as well. Both import and export translation require you to be connected to the internet and the computational power is then provided by 'the other end'.

I often use the simple QCAD Pro software to generate dxf files and then import them to F360 to start the creation of a 3D model. I also use TurboCAD 2016 Pro Platinum for some of the things that it does best, like gear and thread generation. Files are then transferred to F360 in either dwg or STEP format.

Before you get too far into the internal working it would be worth rotating the phase connections to the lathe. Under some conditions the generated phases from the converter will drop in voltage and any contactors will drop out until the voltage comes back up and then the motor current draw causes the voltage to drop once again. A self perpetuating 'fault' condition with associated cacophony!

Move all 3 input phase wires round in the same direction once and try again. If that doesn't work then continue to move the wires round one further step in the same direction. If one of these setups works then all well and good. If none do, then a deeper investigation is required.

Rather like the specification of an

"8X( Vintage Stainless Steel Ruler Rule Scale Machinist Tools 20CM Z9N8)"

on ebay at the moment:- " The measurement allowed error is +/- 1-3cm "

I think I could make most things fit within that sort of tolerance!

As a bye-the-bye on CAD, I use QCAD Pro for straight-forward quick 2D drawings. These sometimes get imported (dxf files) into Fusion 360 as the basis for 3D models.

Fusion360 is now my preferred 3D modeling software. The 2D drawing generation from the 3D model is adequate for most purposes and the whole package has great utility, and I really love the timeline features of F360 that make changes and error correction a breeze.

Because F360 is still evolving there are some features that are lacking. For drafting where F360 has shortfalls I use TurboCAD Pro Platinum. Both 3D modeling and 2D drawing generation are fully featured but a bit more hard work.

Saving TurboCAD files as DWG allows the work to be easily uploaded to F360 so the best of all worlds. Similarly the DXF 2D sketches from QCAD can be imported into both T'CAD and F360.

Put 'Paul the CAD' into your browsers search engine. UK agent for all versions of TurboCAD.

I have posted this several times before but if you are concerned about the cloud 'losing' your design you can select 'Save As' and place the design wherever on your own computer you wish to keep it. Treat it as a backup strategy as although you can work with that file, it won't 'up-version' in line with the copy on the cloud.

An advantage of your design being 'in the cloud' is that you can access it from any of your computers and even your smart-phone - good for showing off your CAD modelling expertise at the club! I do most of my CAD work late at night in the comfort of home, but the next day I can fire up the workshop computer and immediately access any of my projects to work from - on screen. If during the workshop session I identify a needed change I can do that then and there without having to make notes and carry out the changes when back in the house.

The other advantage of cloud storage is collaborative working. As the owner of a project you can invite/allow your colleague/friend etc full access to any specific design or element of it. They can then follow/modify or outright re-design against other aspects that you yourself might do. It is a great way of getting minds together to achieve a model or real world item of tooling and it would be great to see it being used on projects in this forum.

In response to DrDave and to add to what he says viz; "Your work is stored in "the cloud", it is important to know that your work is also stored locally in a location of your choice.

The programme is also resident as 'stand-alone' on your computer and runs quite happily without an internet connection. It does require you to sync with the cloud ebery 30 days or so.

One other added value of Fusion360 is that you can choose to share any files or projects with other colleagues to the extent that they can edit/work with those projects to the extent that you as the project owner allow.

Edited By Swarf Maker on 26/12/2018 22:43:23

Here we go then Sebastian; Having looked at your videos I haven't seen anything that would give me cause for concern. That is it would be without knowledge of the problems that you have been experiencing. Your manipulation of the bearing cage shows nothing untoward and what I had interpreted as rollers flopping around is clearly nothing of the sort. You are having to apply some force to skew them. As that test was done with everything cold then the expected end float may well be commensurate with that degree of available movement.

What needs to happen is a re-examination of the salient facts and first amongst those is the degree of eccentricity that you identify as randomly and periodically occurs. You have described it as visible on both the work piece and the chucks I believe (too late at night to re-read the whole thread). We really need some numbers for that.

The next crucial point to examine is the running temperature in the vicinity of the front bearing. From your description and the helpful input from other forum members, it would appear that your spindle is getting much hotter than anybody elses. That heat can only be coming from the dissipation of energy. Discussion has tended toward the bearing pre-load being too low as a consequence of our interpretation of your descriptions with respect to the rear bearing, but that is inconsistent with a hot spindle which is much more likely to occur if the pre-load is too high. So do the torque tests as already identified and lets see what that tells us.

As all of the measurements that you have taken are essentially nominal we are left with only subjective assesments to deal with viz. hot spindle, visible wobble. Difficult to ensure that we are working to the same criteria on that basis as witnessed by the previous mis-understandings of the symptoms.

Perhaps the best thing is for you to put it in a large 'jiffy' bag and post it up to me for a good looking at!

The videos certainly tell a lot more than the prior words convey as the rollers seem to be captured correctly in the bearing cage. The vision I had conjured up from your description of the rollers falling down when the cage rotates past 45 degrees was akin to the cage being non-existent. "Some of them actually drop just through gravity as they rotate past the TDC position through 45 degrees into 90 degrees rotation..." Perhaps my over active imagination! Goes to show the value of pictures cf. words.

Tony's lathe site declares that "The spindle, which runs in Gamet Super-precision opposed-thrust roller bearings."

So adjusting the end float might well cure things although there will likely be the usual problem of the bearing not wanting to move in response to the adjustment after this amount of time. Not withstanding that, I think that there has to be some damage to that rear bearing for the rollers to have that amount, and degree of freedom, to move as they are. The bearing cage should still support the rollers in their correct place and orientation, no matter how slack the bearing adjustment is so my money is still on a broken bearing cage.

Go no further - including running the lathe under power - until you have resolved the issue with the rear bearing - that is most clearly wrong!

First time in 55 years of motoring that I've been hit with that one. It was the main agents testing it, so it MUST be right.

Yes, they probably are. Brake fluid is hygroscopic and the water content that it attracts leads to internal corrosion of the hydraulic parts. It's a good and useful test to carry out and should be done fairly regularly. Surprised that you have not come across it before!