Damaged vintage variac

[WaveyDipole]

Originally Posted by ajgriff

Looking at your photos again it looks like you've already withdrawn the shaft so I hope I haven't confused matters. The central insulating sleeve seems to be still in place and this is the component which is secured by circlips at both ends on the smaller version. Hope the diagram helps a bit anyway. I suspect that you will end up with a very worthwhile working Variac.

Alan

Not at all. Yes I had withdrawn the shaft and knob, which was the only way to remove it. The knob is still held fast on the shaft as I can't budge the screws. It was I who probably confused things by referring to the insulating sleeve as 'the shaft'. I appreciate your input and did check just to make sure that there were no circlips or anything other that might be holding it together. I didn't find anything and the guy in the video does not appear to remove anything further so I proceeded with another strategy.

This was to use a large socket to push the insulating sleeve out of the base. A spark plug socket seemed just the right size and allowed me to use a little more force. Sure enough after a handful of taps the shaft started to move. I proceeded steadily and several taps later it popped out and the two halves separated. Photos attached. In excellent working condition? Tested before dispatch? It doesn't look like the wiper has run along those tracks in a very long time! I am not at all surprised that I was getting a high resistance reading to the wiper connection. I would definitely not have applied power to that for any testing until it had been thoroguhly cleaned!.

Well now I have got this far, the next step will be to remove the coil from the base and see whether the base can be straightened. For now, though, I will get a brush and clear out the debris!

[Station X]

Debris aside, it doesn't look too bad to me. Whether there are shorted turns is another matter. Close inspection is needed.

I'd have a go at straightening that base. Clamp it down to something solid and gently hit the distorted part with a soft hammer. I'd use a press myself, but of course not everyone has one.

[ajgriff]

Excellent progress. Makes you wonder if the control shaft seized long before the unit was dropped. Assuming that the windings are ok what will you use for the brush? Presumably it will have been a chunky slice of graphite. I meant to type 'Duratrak' (not Duratek) earlier by the way.

Alan

[Alistair D]

Originally Posted by Station X

I'd use a press myself, but of course not everyone has one.

Even a flat surface and some G clamps would do. The main thing is not to try and straighten the base all in one go. Go slowly and a bit at a time. It is more likely to crack if you try to hammer it back into shape.

Al

[WaveyDipole]

While there was still daylight I decided to have a go at straightening out the base. I have a large vice and a flat surface and a variety of hammers but no g clamps. Initially used the vice and proceeded as advised, bit by bit, heating the metal with a blowtorch (warm to hot, but not glowing!) and gently straightening bit by bit on alternating sides. This got me probably two thirds to three quarters of the of the way. For the final third I had to resort to a large hammer on a near enough flat concrete surface. The metal was kept warm and firm but not too heavy taps were used with my foot and hand holding down the opposite side to keep it from bouncing. The result is as can be seen in the attached. It is almost there, but not quite. This was about as far as I could go and I am hoping that it will be good enough. In any case, with the light fading, I decided to quit while I was ahead. The key, as has been recommended, was to go slowly and it must have taken me about an hour in total.

I haven't really thought about the brush yet. The rotator mechanism and brush assembly seem intact and the brush was resting on firmly on the winding, although while the unit is disassembled it would make sense to replace the brush. I think the reason for the high resistance reading is oxidisation and dirt or maybe corrosion of the bridge contact to the centre of the rotator. Although intact, the rotator is seized and dirty and I need to figure out how to remove it for a thorough clean. The socket trick is not going to work here. I have slackened the two screws that hold the rotator to the shaft and it does moves slightly, but at present I can't move it enough to slide it down the insulating material. I have reviewed the video I linked earlier, but he does disasseble that far, so any ideas would be appreciated. This will now be a job for another day.

Having brushed off the dirt, the coil looks OK, but it has had a couple of minor dints and at least externally there do not appear to be any shorted turns. Not sure about the layers underneath. Can I clean the track travelled by the brush with some contact cleaner, e.g Servisol? I do want to clean the exterior of the winding properly as well, but I don't want to strip off any laquer!

[WaveyDipole]

Originally Posted by ajgriff

Excellent progress. Makes you wonder if the control shaft seized long before the unit was dropped.

It was stuck really well but it came out without any damage. There was dirt and plenty of white powdery stuff in the "bearing" which I assume to be oxidised metal, so yes, I imagine that it has been stuck for a very long time.

[Station X]

Originally Posted by WaveyDipole

While there was still daylight I decided to have a go at straightening out the base. I have a large vice and a flat surface and a variety of hammers but no g clamps. Initially used the vice and proceeded as advised, bit by bit, heating the metal with a blowtorch (warm to hot, but not glowing!) and gently straightening bit by bit on alternating sides. This got me probably two thirds to three quarters of the of the way. For the final third I had to resort to a large hammer on a near enough flat concrete surface. The metal was kept warm and firm but not too heavy taps were used with my foot and hand holding down the opposite side to keep it from bouncing. The result is as can be seen in the attached. It is almost there, but not quite. This was about as far as I could go and I am hoping that it will be good enough. In any case, with the light fading, I decided to quit while I was ahead. The key, as has been recommended, was to go slowly and it must have taken me about an hour in total.

You're doing well. You might have to put some temporary packing under the "good" bosses to allow for spring back when doing the final straightening,

[Refugee]

I would be tempted to leave the bend as long as it visibly clears the copper.
Beyond that is just a vanity thing and could end up with the metal casting cracking.

[TrevorG3VLF]

A shorted turn can be found by testing with a lamp limiter. I think there will be only one winding layer.
I would not change the brush unless it is too short. Sometime ago I read that the brush consisted of two brushes glued together so that there is considerable reistance from one turn to the next but the two parts are in parallel for the load current. Also anisotropic carbon is used.
I have a little Variac which had a broken brush. I replaced this with the lead from a 6B pencil. I use it little so wear of the noble metal track should not be a problem.

[joebog1]

I think you have done really well!! As mentioned, leave what remains of the "bend" there.
A new brush shouldnt really be necessary!! Unless its been abused, they dont wear that much. There is also a lapping in to seat a new brush.

Cleaning the bare copper track, servisol or equivalent is OK as would be a clean with a toothbrush dipped in WD40 or RP7. You brush with the turns!! not around the travel as the carbon brush travels. On the sides of the winding compressed air is recommended at about 25 or 30 PSI. DONT hit it with straight from the tank 125 PSI.

Looks like you will have a very nice Variac at the end of the job!!.

Cheers
Joe

[ajgriff]

Originally Posted by TrevorG3VLF

I would not change the brush unless it is too short. Sometime ago I read that the brush consisted of two brushes glued together so that there is considerable reistance from one turn to the next but the two parts are in parallel for the load current. Also anisotropic carbon is used.

The attached blurry image shows an exploded view of the brushes cropped from the second photo in post #21. The holders are protruding and the brushes are quite short so I think it's probably wise to replace them at the same time as the rest of the work. Although they are electrically connected the brushes move up and down independently in their respective spring loaded holders (see JW's video mentioned earlier in the thread). Unfortunately the video doesn't seem to give much information about sourcing suitable anisotropic brushes or how to go about fitting them. Having replaced my own Variac's (Lyons V-5) brush not so long ago I might be able to help with some options and will dig out my notes later in the morning.

Alan

[Refugee]

Those brushes look OK to me.
I used a trimmed down washing machine brush on a Variac of Indian manufacture with no problem.

[ajgriff]

A matter of judgement of course and I'm wrong about the protruding holders as the brushes aren't bearing down on anything in the photo which isn't exactly a definitive image anyway. In view of the fact that this a high current Variac it would be sensible to use the correct type of brush material if the brushes are to be replaced.

Alan

[Station X]

Originally Posted by Refugee

I would be tempted to leave the bend as long as it visibly clears the copper.
Beyond that is just a vanity thing and could end up with the metal casting cracking.

Are three of the bosses redundant then?

[ajgriff]

Three of the the six bosses are used to bolt the whole unit together. The other three are available for mounting the Variac in an enclosure or attaching it to some other piece of machinery.

Alan

[WaveyDipole]

Spent an hour this morning trying to get that rotator off so that I can clean it properly. After an hour of twisting and tugging and running out of energy, only managed to get it 5mm or so down the insulator. Tried WD40 as well as soap to lubricate but neither helped. One of the difficulties is getting sufficient leverage. I didn't use anything as I don't want to damage the metal parts. I am open to ideas as to how to get the thing all the way down the insulator. Only 6in or so to go.....

I have attached a couple of clearer pictures of the brushes.

[winston_1]

Originally Posted by WaveyDipole

Spent an hour this morning trying to get that rotator off so that I can clean it properly. After an hour of twisting and tugging and running out of energy, only managed to get it 5mm or so down the insulator. Tried WD40 as well as soap to lubricate but neither helped. One of the difficulties is getting sufficient leverage. I didn't use anything as I don't want to damage the metal parts. I am open to ideas as to how to get the thing all the way down the insulator. Only 6in or so to go.....

I have attached a couple of clearer pictures of the brushes.

WD 40 is not a lubricant. It is a water displacement fluid (version 40). Try penetrating oil.

[paulsherwin]

WD40 is better than nothing (it is mostly white spirit) but a proper dismantling fluid like Plus-Gas will be much more effective. You only need a tiny amount, so a can will last for many years.

[ajgriff]

The brushes have definitely got plenty of life left in them and look like they are the originals so I agree with the advice offered by others - leave well alone.

Any chance of removing the rotor in the other direction? It's definitely a job for some proper penetrating oil one way or other. Might help with the control knob's screws too.

Alan

[Goldie99]

Heat might be an option as well - not sure what the materials are here, but running it through the kitchen oven (after the roast's out !) might help.