Cleaning a plain bearing in a 50uA moving coil meter

[trobbins]

I have a Hickok in-circuit transistor tester with a 50uA hermitically sealed moving coil meter from circa mid-1960's. The meter needle is sometimes sticking, with a gentle tap needed to let it move freely to its final position.

The movement is in a hermitically sealed enclosure with rubber sealing in good condition, and is removeable. It has a plain bearing comprising a rotating shaft with fixed ends that each carry a hairspring. The bushing is a tube fixed in to the cylindrical core which is held in place between the horseshoe magnet.

There is only a very minor gap between each shaft end and the bushing, which is not too easy to access on either end. So I'm just arranging for a needle and syringe kit (as my caffeine fix doesn't need to be intravenous) with the aim to sort of inject IPA in to the top of the shaft access, and hopefully the IPA may seep in to the bearing and flush any debris (assuming it is debris in the bearing and not related to a bent shaft).

One hassle is that the stiction is very slight, so I won't obviously know if I have improved the situation, and I'll have to try and work out some kind of test before I put everything back together again.

Does that seem reasonable?

[Craig Sawyers]

End bearings are usually needle and cup, often with a jewel-type bearing, which might not be immediately visible.

Alternatively sticking is often caused by flakes of magnetic debris getting into the gap.

I've had success by using a thin strip of adhesive tape, very, very carefully inserting down the gap between the coil and the magnet pole pieces. You go fishing for the magnetic crud. This is not without risk though.

Careful use of a magnifying glass might reveal any crud in the magnetic gap before you do anything potentially risky.

Craig

[trobbins]

Luckily or unluckily (depending on the outcome), this is not the more commonly seen type of cup or jewel bearing. Nor is there a likelihood of magnetic attraction of debris in to the bearing. The coil former and hairsprings are clearly separated from any nearby surface.

It is from military spec equipment, for portable use, so no doubt this form of bearing has found favour for possible shock and vibration applications, and its sealed construction has avoided any dust or moisture ingress over the decades.

Ciao, Tim

[Malcolm G6ANZ]

The sticking could be caused by static on the glass/plastic face. When tapped the static is removed. Probably wrong but a thought.

[Bazz4CQJ]

That rear surface of the movement seems to be covered with numerous white particles - some sort of corrosion product? I've seen the exact same thing on a 50uA meter from an Avo VCM. I think that before you start with the bearing, it would be good to get the whole thing (and the housing) as clean as possible. If there seems to be a specific place where those particles are forming, clean it and perhaps smear a little Vaseline over it with a Q-tip.

It may be the case that any success you have in flushing the bearings is short-lived unless you do a good job of general cleaning first. That approach worked with my VCM meter.

B

[The Philpott]

Ferrous debris more likely than a pivot problem -especially if the needle always sticks in the same place. The debris tends to be in the form of flat shards, like slate when it breaks into tiny pieces. This makes it damn near invisible from certain angles..

There's very little room between the coil and the concentrator, and it doesn't take much to make it stick.

Dave

[trobbins]

B, that surface covering is just an imaging effect from a nearby light reflecting off the rougher casting in that region - luckily there is no corrosion or covering on any metal surface - it is pretty much pristine inside the housing due to the hermetic sealing.

Similarly there is no sign of any debris between the horseshoe choke and coil, and between coil and central core - lots of separation to look through and clearly see. The hairsprings are normal with nothing distorted and no overlaping.

[space_charged]

I doubt that the sticking is due to the bearings. Its much more likely that there is something catching between the coils and the poles of the magnet. I work on tuning-fork watches (Bulova 218 and 214 movements) That has to be done under a high power stereo microscope. I always see tiny bits of magnetic material on the fork magnets. They are really tiny, no chance of seeing such things without the microscope. The same sort of things will be seen on the magnets in a meter I'm sure.

As somebody else here said, such bits can sometimes be removed. A trick I've used in the past is to use a small watchmaker's screwdriver that IS magnetically susceptible. When very close to a fragment (supposing you can see it) the fragment often sticks to the tip of the screwdriver. It can then be slowly moved away. Its not for the nervous and its very easy to do damage to the coils.

I have, in the past, removed the coils from a meter. That is seriously not easy. The hair springs have to be disconnected. Then getting the coil out from between the poles of the magnet can be very difficult. To avoid the poles crushing the coil when released, suitable non-magnetic shims have to be placed to hold the poles apart. Then it all has to be cleaned and re-assembled. If you're doing this for the first time the chances of success are low.

I doubt if injecting IPA into the bearings will cure the problem. It will also remove any remaining lubricant. They should probably be re-oiled with a suitable watch oil. If the bearings are dirty, oiling might not help.

[trobbins]

Ta. For this application it seemed that using any kind of lubricant was not good.

The hassle is that the bearing itself is pretty much total enclosed, with only that very small gap between the central tube bushing, and the end fitting on the spindle (at each end), as an access to the bearing surface itself.

The spacing between other moving parts (coil to either pole face or central concentrating core) is relatively wide and clearly visible with low magnifying optics. I am aiming to set up the meter movement out of its case to see if I can replicate the stiction, as it is quite minor and not noticeable with simple movement of the needle through its arc and back. At some stage I will sweep a poly or wood sliver through all that spacing.

[crackle]

I would be surprised if oiling a jewelled bearing would help it swing freely.
Mike

[Uncle Bulgaria]

Jewelled bearings do need oiling periodically. It's standard service practice for watches and most timing problems I've come across have been because of a lack of lubrication. However, if you meant that in this case adding oil wouldn't help, I agree! It does seem more likely that removing particles would be the first port of call. If it turns out that the bearing is running on congealed oil, it will have to be removed to be cleaned and re-lubricated. Dripping IPA down sounds like a tempting bodge that will end in tears. Having done that after losing patience, I recommend not giving in to temptation!

[pmmunro]

The design of pivot and jewel systems was studied by Verney Stott of the National Physical Laboratory in 1931 and seems to have been still considered an authoratitive source several decades later. (N INVESTIGATION OF PROBLEMS RELATING TO THE USE OF PIVOTSAND JEWELS IN INSTRUMENTS AND METERS. Journal of the Institution of Electrical Engineers, Volume 69, Issue 414, June 1931, p. 751 – 756).

The effect of lubrication is described below;

"Initially the frictional torque was about the same (2 dyne-cm) for both the dry and the lubricated pair. For the pivot and jewel which were run dry the torque due to friction remained about 2 dyne-cm for the first half-million revolutions and then rapidly increased, rising to 48 dyne-cm by thetime 1 million revolutions had been completed. A drop of clock oil was placed on the other pivot and jewel at the start of the test, and this was replenished at intervals until 1 million revolutions had been completed. The frictional torque remained practically constant at 2 dyne-cm not only during the first million revolutions but also over a period of 6 months, during which time the total number of revolutions reached 15 millions, and no further oil was added after the first million revolu-tions had been completed. After 15 million revolutions the torque increased, attaining 10 dyne-cm after 16 million revolutions, when the test was stopped. The important results emerge that the presence of oil made no appreciable difference to the initial frictional torque and that the useful life of the pivot and jewel was increased by lubrication from 1/2 to 15 million revolutions".

The implication is that for continuously rotating instruments, lubrication would be beneficial and this would apply to clocks and watches and possibly kilowatt-hour meters. However, it is accepted that clocks and watches need servicing and cleaning because the lubricating oils and greases deteriorate over time. (I don't know what the thinking on kW hour meters was and it is unlikely they would be regularly serviced).

In a moving coil meter, any benefit from lubrication would be outweighed by the difficulty of having to clean the pivots periodically.

Avometers, the instruments I know best, use sprung sapphire jewelled bearings and these are not lubricated. (The adjustement of the jewels is considerably looser than would be used in horology). I think this is true for most other electrical instrument makers.

Many electrical instruments are designed to allow the moving coil assembly to be removed from the magnet complete with the bearings and concentrator. This greatly aids servicing and the moving coil can, with considerable care, be dismounted for servicing. One useful point to bear in mind is that only the outer ends of the hairsprings should be unsoldered unless the springs are to be replaced. Resoldering the inner ends is not at all easy. When soldering hairsprings it is important to guard against any flux deposits remaining as these can migrate on to the coils of the hairspring, causing the turns to stick together.

PMM

[Uncle Bulgaria]

Fascinating!

Of course most meters are not going to be constantly bouncing back and forth like an escapement, so the retention of the initial low friction with possible greater leeway makes perfect sense rather than having to deal with replacing gummed lubrication.