Unplanned teardown-restoration of unknown device

[TerraHertz]

Today I finished one thing in the morning, and expected to be going out for the rest of the day. But had a short block of spare time. Did a little tidying, and came across an old relic that I'd previously attempted to open out of curiosity, but failed. This time I thought I'd have another go, briefly. Just to have a look.

It's like a hand-cranked megger, but more complicated. I don't have any information on it or how to use it, so didn't even know if it works. But it's old and interesting, and on my 'look into' list. Badly needs restoration.

Many of the case screws are sealed with hammered-in lead plugs. I'd previously removed the lead plugs from the case top, and removed those screws. But the top still didn't come off. Something inside holding it loosely.

This time I opened the bottom. Success!
Then the afternoon's expected activity didn't happen. So I just kept on mucking around with this thing, in general fixing what's broken and starting on cleaning. Still don't know what it is, exactly. But it looks cool.

Because of the random whim start, I didn't take overall photos first. There are some later.
The lead seals are stamped with the manufacturer's logo: SEV with a pi symbol.
Base - nothing special, just a lot of scratched bakelite and the usual completely failed ancient rubber feet.
Inside - Wow. *LOTS* of soldered fine adjustments everywhere, and not a flexible wire in sight. The blue thing is a surprise - bright blue plastic of some kind, in an ancient thing like this?

Screw missing  - One of the contacts of the mode selection switch is hanging loose. Might as well fix it...
Not all missing - I'd thought it had stripped out of a threaded hole in the bakelite. but no. There is the nut, and other half of the screw still in place. It actually sheared in half. Over tightened plus corrosion I guess.

Fixed - This was _such_ a hassle. I have a large assortment of very small screws from all sources, but not one that fit that square nut. So I had to make a square nut that would fit a similar screw I found. Plus drill out the bakelite hole very slightly to fit that screw. Which meant a long hunt for that odd sized small drill bit...

Decades _ seemed like decades later. One thing that most needs repair, is the plastic windows to the numbers on this row of decade resistors. They are disintegrating. Once I had the box open, it was possible to remove this. just unsolder two wires.  Then struggle to jigsaw-maneuver it out, without bumping any of the very fragile looking tiny wires neatly wound in freestanding coils to trim each of the resistances. This thing was NOT made for easy disassembly.

Wipers - to get to the 'windows', everything has to be removed from the front panel. But first I tried measuring the resistance end-to-end while moving the dials. Hopeless... really bad contacts everywhere, and not improving with repeated wiping. So continue disassembling.

Dirty. Oh I see. That black stuff does not wipe off easily. Not even with solvents. Some kind of chemical reaction between a grease and the copper. OK, there's a quick brute force way to deal with that. (Not involving explosives, though tempting.)

Clean. Emory paper and a cork block. Now the contacts are clean. When putting it back together I'll try lithium soap grease, see how that goes. Unless anyone has a better suggestion.

[TerraHertz]

Detents. The knobs have a really nice clicky feel, and here's why. No skimping on detent springs.

Pins - Nor on anything really. Huge brass bearing surfaces molded into the bakelite. The spring provides pressure to the wiper contacts. The other two pins are just to get the knob exactly pinned to the detent wheel, so there's no play. Also an inset felt washer to keep dust out.

Dial. Hmm... the white stuff is some kind of very tough paint. But it has yellowed a bit and has a few spots. I'd _like_ to improve that, but on consideration I can't. Because of the way this was made. The black letters are the underlying bakelite, raised. It was molded with the numbers raised, painted, then machined down in a plane to expose the bakelite numbers. Urrgh. Whatever I try would probably ruin it. Just have to live with the yellow.

Cracked cover - this stuff is probably nitrocelulose. It's very fragile and has to be replaced. But the brass sleeves are press fitted in the bakelite. Probably would be a bad idea trying to remove them. The front rims are painted black to match the bakelite, but I know during cleaning that paint is going to go. Also the ring clips holding in the clear cover are really tough, and once in place the only way to remove them is to bash out the clear cover.

Extracted. Finally. Now, before I try to clean up this cover, I'll practice on the base plate, to be sure I actually can sand back bakelite then polish it. Annd... yep, it works. 240, 400, 600, 1200 grit paper, then cloth polishing wheel with SSX stick. Gets quite a nice finish. The windows now have polished brass bezels, not black. I may decide to make them black later, for historical accuracy.

Unit - after all this it seemed worth taking pics of the overall thing. Whatever it is.
Dial - This is literally the only identifying information anywhere on it. Maybe there was a name and model number of the metal cover that must have protected the glass? Now there are only the hinge pins remaining. If I can find a picture of what this originally looked like, maybe I can make a new cover.
Presumably the 'up arrow' is the British symbol for military item?

Plate - the closest thing to a user manual. But it assumes a lot, and don't really understand it. "Distance of fault from earth terminal in ohms"  OK, I get that it's for locating a cable fault. But what is the distance unit? Ohms? Ha ha.
Anyway, still to black paint re-fill the letters, then polish and coat this plate.

Controls - are two knobs. One has three positions: MEG, VARLEY, and  BRIDGE. The other is a multiplier: /100, /10, x1, x10, x100.

And three terminal: Earth, Line and Varley. Why no, I do not know what a Varley bridge is, and have not yet googled it. Tomorrow.  "Infinity adjuster" is obvious, but still amusing.

[Vgkid]

Very cool. It looks like your decades cover 1,10&100,1000 ohms per step, with some nice beefy silver switches.

[xrunner]

Neat! It's amazing the effort put into making these mechanical devices back in the day. 

[Andy Watson]

Lovely! It's a "bridge megger" with (I believe) optional Varley loop tester built-in.

[CatalinaWOW]

This link may provide more information on your goodie. 

http://www.richardsradios.co.uk/meggers.html

[TerraHertz]

Very cool. It looks like your decades cover 1,10&100,1000 ohms per step, with some nice beefy silver switches.

Yes, it does. Once the contacts were clean I measured stud to stud with a multimeter and they are as you say.

Lovely! It's a "bridge megger" with (I believe) optional Varley loop tester built-in.

Ha. You had me all excited for a moment there, reading the 1st page of the PDF. That's it exactly! But there are no more pages...

@CatalinaWOW thanks for the link. That's an interesting read. This unit is definitely by that company. Will dig more tomorrow.

Btw, for making replacement number windows, can anyone think of something commonly available (or scrap) that is clear, can be cut (say, with a plug punch) and is 0.5mm thick?   All the clear film/sheet materials I've dug up so far are either too thin, or too thick. Unfortunately, from the way the circlip holds the sheet, it has to be very close to 0.5mm.
Diameter: 11.7mm

[Andy Watson]

Ha. You had me all excited for a moment there, reading the 1st page of the PDF. That's it exactly! But there are no more pages...

I have the same model, ex R.A.F.. It works well although I don't use it much I can't find much information on the web about these things.
Here's another snippet.

Edit: Updated pdf with a slightly less cropped version.

[Andy Watson]

Damn it There should be a second page but the forum won't allow it due to size. I'll try a PM.

Edit: Pah!, No can do. PM me your address it you want the file. It's only a couple of pages  from an old telephony book but it is scanned image and taking up about 3MB.

[TerraHertz]

I have the same model, ex R.A.F.. It works well although I don't use it much I can't find much information on the web about these things.
Here's another snippet. (Megger_page_1.pdf )

   How did you find that, and who created a one page pdf with the left and right sides of the text cut off?
I'm constantly amazed by the astoundingly awful things people call 'electronically captured' documents.

Relevant: http://xkcd.com/1683/

Edit: Owp, you scanned it? Well, I'd like it anyway, even if the text sides are cut off. Did you know they are?
My email addr is in my profile here. But, did you try coding the scan as plain PNG, with say grayscale 16 levels? (4 bits/pixel) It will be much smaller. Especially if the whites are all white, as opposed to paper texture.

[Andy Watson]

How did you find that, and who created a one page pdf with the left and right sides of the text cut off?

The megger came from a local electronics shop - about 25 years ago!
I can't remember where the document came from. It is only about three or four pages - I think the text cropping may be due to my bodged attempt at extracting a single page such that it would "fit" fit on the forum.

[Vgkid]

For the plastic sheet, I have seen some  thick plastic sheets used for laminating that would work.

[TerraHertz]

For the plastic sheet, I have seen some  thick plastic sheets used for laminating that would work.

Ah ha.  I foresee a trip to Officeworks in my future Monday.

[BurningTantalum]

TerraHertz- Re: the windows. I have had a lot of success recently cutting circles of thin glass from old microscope slides and picture glass. I used a Dremel type tool with diamond cutting wheels that were sourced for not a lot on eBay from the usual country. I scored lightly around a mark made by a fine pen, then a heavier cut, then did the other side and carefully cracked the disk out by hand with a pair of 'parallel pliers' with insulating tape over the jaws. The diamond disks then trimmed any 'lumps' and squared the edges to size.
Regards, BT

[Andy Watson]

I hadn't registered that you were missing the "door". Apart from identifying the instrument the door also has the instruction manual

[CatalinaWOW]

Another idea for the windows.  If you can't get quite thin enough material, just chamfer the edges were the retention ring fits.

[TerraHertz]

I hadn't registered that you were missing the "door". Apart from identifying the instrument the door also has the instruction manual

Thanks for the pictures. I was going to ask you for those, but you beat me to it.
Yeah, I guessed the instrument name and maker were probably on the missing door, hence unknown on mine.
Now, am I going to try and make a new door? Possibly. But placed a way down the queue.

The number windows are next. Sorry BT, I thought of cutting glass using a diamond hollow cutter, but the tolerance required is very small. Also I think 0.5mm thick glass is too fragile anyway, especially given it's in a metal holder.
So flexible plastic sheet it is. Once I find some the right thickness. I can make a precise hole punch to cut the circles exactly right.
Can't put the resistor decade block back together until the windows are in.
Next: deriving the circuit diagram (can't be too hard, it's all resistors and visible wiring & switches!)
Then removing all the guts, so I can strip & repaint the case.

In the meantime I made some new rubber feet. The baseplate required feet to fit circular recesses, so I machined some standard rubber feet to suit.

[babysitter]

constant 500v pressure. I like that.

[TerraHertz]

constant 500v pressure. I like that.

Me too! That's why it has such a solid hatch on the meter window. All the pressure!
Seeing that makes me really want to replicate the hatch and plates. The rest is easy, but photo-acid etching those aluminium plates will take a bit of experimentation. I still have my old UV exposure tubes for PCB resist, but not sure how to generate a photo mask from a photoshop recreation of the lettering.

As for the numeric windows, today I made it to officeworks, and nope, no luck. All the plastic lamination and cover sheets are only up to "200 microns" aka 0.2mm thick.

On the way home I happened to stop at the hardware store for a few things. Among them batteries. Bought a blister pack of AAAs. Guess what? The blister pack has nice flat, clean areas, and the material is 0.5mm thick. Perfect.

[CatalinaWOW]

Etching your door info plates may be a good application for toner transfer techniques.  I have had good success on aluminum doing this.  Use Hydrochloric Acid (called Muriatic Acid here in the states and sold in hardware stores for concrete cleaning and swimming pool maintenance).  Do it outdoors because the gases generated are nasty. 

As you say it takes a bit of experimentation to get the results just as you want.  You will probably have to preheat the substrate during the transfer process.   In my case I have been generating raised letters, while in your case you want the printing etched into the plate, which will change things a bit.  I would also verify that etching is what you want.  I can't tell from the photos if the lettering is actually etched into the plate.  The toner transfer technique also works well for generating a surface print.  Covered with a clear coat it is quite durable.

[Andy Watson]

I can't tell from the photos if the lettering is actually etched into the plate.

I think it must have been. I can't see any evidence of tooling marks on any of the letters. The letters are crisp and the bottom of the groove appears to be flat - as far as I can tell.

Here is a close-up taken with a crappy USB "microscope".

[CatalinaWOW]

Looking at your USB picture I agree that they were etched.  It is surprising to me the the places where letters are partially scraped off on the plate show so little evidence of this.  Learn something new every day.

[TerraHertz]

Looking at your USB picture I agree that they were etched.  It is surprising to me the the places where letters are partially scraped off on the plate show so little evidence of this.  Learn something new every day.

Well, the brass plaque next to the meter cover is definitely etched and black paint filled. So why wouldn't the meter cover be too?
It seems to be very shallowly etched though. Most of the abrasion gouging is deep enough to completely remove the lettering. You can see spots where it only partially dug down into the letters, in the "LONDON" at bottom.

And I wonder why they used aluminium there, but brass for the other plaque?
Maybe because aluminium was rare then, and they thought it would look swankier?

[CatalinaWOW]

Or because they made a large stock of the brass plates, and still hadn't run out.  The door cover might have been updated with new patent numbers or whatever, allowing the change to what was by then a cheaper material.  Particularly if your megger is post WWII.