Member postings for Macolm

It could be a voltage reference or a logic pattern feeding the wire loop, and a check that the received voltage or signal is the same. This could all operate at minimal current both in the wire and in the circuitry, so allowing a long battery life.

With a pre assembled single length of wire with resistors incorporated it would indeed be tedious. With a better detection system though, I envisaged that short lengths with central resistors, and plugs and sockets, would be procured. Each item owner would fit their wire, then simply a case of plugging together. Easy, also to remove one item, and set up the protection of the remaining items.

It seems to me that a simple wire loop alarm is none too secure these days. It would be all too easy to bypass a section, for example using a length of wire with pins attached at each end, with these simply pushed though the insulation of the loop so as to contact the wire inside. Then cut the loop wire without triggering the alarm.

A possibility to improve security would be to include a resistor at each item to be protected. Thus a length of wire, each with the resistor in the middle, for each item, would be looped through it to conceal the resistors. All would then be connected in series as before.

Now an electronic circuit would be required with a comparator against a narrow window of total resistance, so that shorting out any resistor would trigger the alarm. It would be simple to have a potentiometer that could set the window thresholds to cope with the actual number of items being protected. This addition circuit would only require one integrated circuit with ancillary components, as well as whatever is already needed for the alarm function.

It can be easy enough to bypass a centrifugal start switch. Assuming the terminals are accessible, it needs a momentary push to make mains capable switch connected across the start contacts. While the motor is powered and trying to start, press the switch for a couple of seconds. If it gets up to speed, it is bad contacts. If not, it is something else. Make sure to release promptly in either case.

I used to have an E Lutz (before ELU) sander with an induction motor, and a double toggle switch. One side was power to the run winding, the other was sping biased to off for the start winding. Simply press both to on, then release, leaving the motor running.

The disadvantage of water as a thermal store is that the maximum temperature must be less than 100C (unless the store is constructed like a steam boiler with all that entails). If it were a good choice, electrical storage heaters would use it. Instead, the material they use stores much more heat by exploiting a large temperature range, even though the specific heat capacity is less than water.

The Sunamp idea is good as a hot water store because water taken from storage is delivered at a constant temperature until the capacity is exhaused. The inverse, "wax coolers" using latent heat have been used for decades for rocket and missile electronics, since there is no air for cooling once out of the atmosphere.

A quick search found this:

**LINK**

Thanks for better information, I was merely assuming common design practice. That probably means that there is no end float adjustment available. In that case, a fault in the angular contact pair seems possible. I suppose the good news is that replacements will be somewhat cheaper than taper roller bearings, provided they are not unusual imperial sizes.

A further thought, for the lower bearing next to the arbor, partial failure of the roller cage might allow one or two of the rollers to skew out of position, resulting in the observed small amount of eccentricity.

Regarding adjustment, agreed the optimum end float is zero. However, roller taper generates an end force on the rollers, and the ends run against a flange to keep them in position. This is why there is an observable increase in drag when the end float has been adjusted out. Unfortunately with a double nut, the correct end float set using the first nut is displaced when the second is locked against it. So you need to start with enough slack so get to the correct final position. Unless you are happy that you can distinguish this correct friction from the amount when too tight, the safe solution is to set almost negligible slack, commonly 0.0025mm or 0.005mm (0.0001” or 0.0002&rdquo.

That would seem the kind of value to expect on a well used machine, and normally slight adjustment of the bearings would reduce it. Unfortunately, in conjuction with the runout, it suggests either a problem with one of the bearings, or the spindle has been crashed into something solid and bent. I am afraid I know nothing of the Beaver Mill, so I can only make general comments.

Unless anyone has an alternative suggestion, it looks like the spindle needs to be removed. I think it was suggested the quill can be detached, and this would allow a better inspection. With it on the bench, a quick test would be that the splined drive shaft is closely concentric when rotated. If it is not, it either points to a bend lower down, or to a partial failure of the upper bearing which can be due to breakup of the cage holding the rollers or balls so that they clump together rather than being evenly spaced. Such failure of the lower bearing would presumably lead to more eccentricity than you have measured.

Good luck with investigating this, and keep us informed. This sort of problem often turns out to be something entirely different from preconceptions, so do look out for the unexpected.

I have made quite a few such thumbscrews using aluminium for the outer. I just use a tight press fit, ensuring there is a smooth lead radius on the cap screw. The aluminium will easilly expand if the fit is over tight, though a smear of WD40 or thin oil is a good idea to prevent seizing. The design in the first post, with the hex retained, is sensible, but none of mine so far has come slack. If necessary, step the bore so that the cap screw knurling will grip.

Bear in mind that a given solar panel in December will produce roughly one sixth of the mid summer output.

By the way, an excess of solar generation sounds like an excellent way to destabilise the electricity grid, since the plan seems to be to mandate heat pumps, whose COP will be a minimum at the same time as the solar output is also a minimum.

Ultra high molecular weight polythene (UHMWPE) has notably low permiability to water and other vapours. Many plastics have quite high absorption and transmission of vapour. However, silicones including RTV are particularly bad, an order of magnitude worse than common plastics.

Edited By Macolm on 22/09/2023 15:01:10

RE cooker hoods etc. Induction motors driving fans are a special case. Because the drag of the fan is proportional to speed squared, it falls off rapidly as speed reduces. Within a range of "slip v torque" of the motor characteristic, reducing the voltage does indeed allow a range of speed control..

This obviously does not apply to other types of load, where for more than a certain drag, the motor will stall (or not start).

There are two likely situations, assuming the run out is real and not due to a misunderstanding. Either there is slack in the bearing setup, or less likely, the spindle has been bent in some previous serious crash.

To test for bearing slack, fit an indicator which bears on the end of the spindle (ie vertically), and try by hand to see if there is vertical play, that is try to lift the spindle. Any vertical movement by hand should be less than perhaps 0.005mm. In the event that adjustment is required, look for the makers recommended procedure, and proceed with extreme care.

Testing for a bent spindle is less easy, and would probably need the spindle removed then the bearing inners removed (assuming ball or roller races), and supported and rotated on V blocks on the bearing fitting diameters. It would be best to involve someone with appropriate machine tool experience.

If I understand the situation, my approach would be minimally sophisticated. Assuming I wanted it mounted on a pole, I would get a broom handle of slightly greater diameter than the thread OD, then plane or sand a shallow taper on the appropriate end to a bit less then the thread ID. Then screw on the feeder to get a witness of the thread, trimming the length back if necessary. This may achieve sufficient retention, but if not, resort to a triangular file to form a deaper and longer thread until it is judged adequate.

If the feeder is plastic, be careful not to split it, so perhaps resort immediately to filing the thread. Araldite enthusiasts could use glue instead of making a thread, assuming a full H&S assessment was completed first.

A long shot if desperate might be to contrive friction welding a steel rod, assuming you have direct axial access to the broken part. As you may know, the parts to be joined are rotated with pressure until the interface melts. There must be provision to stop the rotation dead by a clutch and to hold in position till the interface solidifies. Results are impressive if carried out properly, with nearly full strength due to the inherent complete and clean metal to metal contact.

But you would indeed need to be desperate, a botch would not succeed.

Correction - C1 to N

Just to be clear, I suggest checking the winding resistances to be quite clear about the circuit of the windings.

For example, this circuit is a possibility. Connecting the capacitor from C2 to N would produce symptoms similar to those reported.

Edited By Macolm on 27/08/2023 15:25:29

A further thing to check. Some older motors had the second (start, or phase shifted) winding as two halves with the capacitor in the middle. If the capacitor has been incorrectly replaced, the motor will be unhappy!

The simple solution to flicker is three co-located lamps, each fed from a different phase of a three phase supply!