Thursday, August 18, 2016

American Wire Gong Repair (Tutorial)

I happened to notice the other day that one of my clocks (a beautiful William S. Johnson Ogee from around 1845) was simply making a "clunk" sound rather than striking. I opened the door to find the gong wire sitting in the bottom of the case. I've never had this happen on any of my other clocks, but this isn't the first time I've had to repair a wire gong.

Since this is an easy and simple repair, I thought I would post a quick tutorial.

Spiral wire gongs are simple devices used in clocks, and they have been around since the late 1700s, I believe. They vary widely in quality from thick, heavy Victorian ones (found largely in English and German bracket clocks) to thin wiry ones found in Vienna clocks and American clocks. The sound they produce depends on the wire gauge, as well as the length, quality and hardness of the steel wire. The wire can be square, rectangle, oval, or round. Some wire gongs sound incredibly beautiful and mellow, while others sound absolutely terrible.

The gong on this particular clock has a very nice sound. The link at the beginning includes a sound file on the web page.

Unfortunately I didn't have enough lighting in this photo, so it's blurry, but you get the picture. The gong wire snapped off the brass hub right at the edge. This could have been caused by bending or adjusting the gong too much (in the past), and then eventual metal fatigue at the joint from all the thousands of hammer blows inflicted over the past 11 years since I've owned the clock (the clock was bought in 2005), and all the ones before that. For fun, here's the quick math: 156 hammer blows per day, x 365 x 11 = approximately 626,340 hammer blows over the past 11 years. Unrelated, but fun to think about: this math also means that I've wound this clock daily approximately 4015 times.

To repair a gong, you CANNOT and SHOULD NOT ever use solder. Solder will deaden and potentially ruin the sound. The wire needs to attach to the hub by purely physical means. This is done very simply by a friction fit, and a punch mark.

To do the repair, simply mount the hub into a small vise, and drill a hole into it for the gong wire to pass. The hole doesn't need to be tight, but you want the fit to be as close as possible. My wire was 0.055" diameter, and the hole I drilled was 0.054". With the wire coaxed into the hole, align it into a parallel plane, and on the back side make a small punch mark with a centre punch. That's it. Done. This is about a 5 minute repair.

You can see the old wire stub here. If it hadn't broken cleanly to the surface, any protruding stub can be filed down flat to the surface, or if you're able, the stub can be punched through the centre, and removed, and a brass pin can be used to plug the hole. Be warned, however, that these wires are usually set into the hubs very securely, so you may do more harm than good trying to remove the old steel stub.

Here you can see the punch mark on the reverse side. Nothing too crazy. This is extremely secure, and the gong wire will not slip out or lose its position.

Sounds exactly like it did before, the only difference is that I've lost about 1/8" on the length.

Wednesday, August 10, 2016

English Longcase Clock - W. Routledge Brampton Ca. 1770-1780

Wow, so even before I begin, I've just noticed that Google has changed EVERYTHING for the photo albums, so this is now going to take me about an extra half hour to an hour to re-learn all the album links, menus, and all that other crud.

Alright, so after about an HOUR, and a LOT of frustration and hair-pulling, I've figured out a way to add my photos to my posts. With that nightmare out of the way, I can start to tell you about this clock (from one nightmare to another, I suppose).

My antiques restorer friend Enrique dropped by my workplace one day last month and asked me if I could have a look at a clock he needed to have repaired.

He had the case (in pieces) in his van, which he was going to repair, and he pulled out this large box of loose parts which had once been the clock's movement. He could fix the case easily enough, but couldn't do anything with the movement.

What an absolute disaster:

Note that both bottom pillars are torn right out of the backplate.

The loose wheels were put in a baggie.

The rating assembly is totally mangled.

Backplate, with torn-out pillars and bent bell stand.

More views of the pendulum and rating assembly.

The suspension spring was torn out (and lost).

Loose parts. Some of the wheels with bent pivots. Crutch broken (previous repair).

Two of the dial feet were also torn out of the dial.

The dial is quite beautiful, but all the silvering is gone, it's been lacquered, and both the second hand and calendar hand (and wheel) are gone.

The style and layout of the dial suggest a date of around 1770-1780. The Arabic "1" numerals appear very early in style (such as those on lantern clocks), but the sheer size of the dial (about 14 x 20 inches - huge!) indicates a later period brass dial. The hands are incorrect (1820s), and I suspect this would have had black Serpentine hands originally. As with all brass dials of this type, the chapter ring and name boss would have been silvered.

Unusual half hour marks, no inner hour hand track or half quarter marks (in the outer ring) also point to a later dial. The lack of a rolling moon dial or function in the arch (like a strike/silent feature) seems to point to an earlier clock, but could have also been done this way for economy.

I was not able to find any information about W. Routledge. Note the wheat border, which is also a fairly early feature. Wheat border decoration went though a craze around 1740 (IIRC) when it was very popular and used profusely on dials (around the entire dial border, on bosses, sometimes in the inner circle or around spandrels and seconds dials as well, etc).

Note the torn-out dial pillar in the lower left. The two bottom pillars pass through the engravings. The third pillar is just above the seconds dial and you can see that the brass is distorted there.

The spandrels are a variant of the "large cherub head" pattern, but they don't match any of the ones in my books. The key difference is the centre bib or scarf below the head. The castings on these are fairly rough, and the edges are not chased or cleaned up very much. Lots of craggy edges.

Here's the reverse side of the dial. All the decorations are fixed in place with flat topped machine screws, with the exception of the boss, which is pinned in place. Earlier dials always had pinned chapter rings, and spandrels were usually held with large chunky screws (often square).

I was reluctant to take on this project, simply because of the sheer level of damage to the movement, and the added possibility of finding all kinds of other serious problems with the movement once it was back together. To add to all this, Enrique wanted all of the work done within a few weeks! Normally I would estimate a job like this to take over a month, since I largely do clock repair in my spare time. Let's just say that with all this on my mind, I wasn't afraid to charge for my work. I was also told that this was likely going to be covered by insurance, since the clock was moved with a moving company. I later found out that it was shipped through UPS (which are the absolute WORST company to deal with - everything I've had mailed with them has been destroyed), and that the clock was not taken apart at all before being packed for shipping.

Overall, the movement was not too difficult to repair, but it was done in multiple small stages. First, the simplest of the repairs were done, including re-riveting the movement pillars in place, and straightening the bell stand. Other small jobs like straightening the rating assembly, which is never an easy job were also done in small work sessions. To avoid breaking it, or mangling the threads, the threaded portion had to be heated to red hot, and then carefully unbent. This makes a decent repair, but it's never going to be perfectly straight again unless the entire threaded portion is replaced. In this case, the assembly was kept all original, partly because someone in the past decided to solder the pieces to the rod (the rating assembly is soldered to the pendulum rod). I kept everything exactly like it was.

Most of the movement parts were in good shape (no broken teeth). Two slightly bent pivots were repaired, and then everything was thoroughly cleaned and polished.

The back plate was re-polished to sand-down the marks from the riveting/peening process on the pillars.

Most of the taper pins were replaced, and once the movement was reassembled and oiled, the clock ran beautifully!

I did, however, have problems with the badly damaged/grooved seatboard. The deep grooves had to be repaired (see farther down). I did make sure that the movement plates were perfectly squared when the pillars were reattached.

By far, the hardest part of this restoration was repairing the dial. Making things even harder were the client's demands that the dial NOT be touched or refinished in any way. Not only did they not want to pay for any additional work, they wanted the dial to stay exactly as-is.

Normally, that kind of request is easy to accommodate, but if this were my clock I would have re-riveted the feet, then sanded and polished the area smooth, and redone (touched-up) the carvings and wax filling for a perfect and invisible repair. But that would have meant refinishing at least the entire centre of the dial, which was a no-go.

Basically, I was told "do whatever, bang it together, I don't care, just make it work". This is not how I like to do things, but in the end, I did the best I could to minimally damage the dial, while still making it as sturdy as possible.

One thing you need to realize is that this particular dial is insanely heavy. The sheet and chapter ring are around 1/16" thich, and with the ornaments added, the dial weighs about NINE pounds! All of that weight is carried by just the three small dial feet. Two of them were torn out, and one was slightly loose.

Because of the specifications laid out above, this is how the worst of the pillars turned out. Not great, but what can I say...

This one got just a few light taps to secure it.

The top pillar, which is completely hidden got the most aggressive treatment, since it carries the bulk of the weight. I hammered it down, and further deformed the brass outward with a centre punch.

Here's the dial with all the decorations removed. I removed them just so I could work on the pillars/feet. Nothing was cleaned or polished as per client's wishes.

Here are the repairs done to the seatboard.


The curved pendulum notch is an interesting detail. Normally these have just a crude trapezoid shape.


Small areas were carefully cut away for new wooden patches.

The patches were trimmed, sanded, and puttied smooth with the surface.

Paints and stains were applied to hide the repairs.

While reinstalling the weight lines (which were sadly steel cable) I took the time to remove centuries' worth of old gut knots left by prior repairmen. I am not a fan of "fast and dirty" repair jobs like these. It's really not hard to pull out the knotted end, but most prefer to just cut the gut line and leave the knots permanently trapped in the barrel to rattle around and ultimately cause additional interior corrosion to the brass when they get wet from cleaning.

Here is the completed clock. The time side works SO WELL that it will practically run on 1 pound. As I was packing the clock, I used two small chunks of foam between the pulleys and the seatboard (to avoid the steel cables unraveling into a big messy tangle), and the mere pressure of the foam on the line (which was not very tight) was enough to make the crutch swing back and forth vigorously.

The new suspension spring had to be entirely hand made, since the pre-fabricated ones I had on hand were fitted only for a square bottom profile. This clock had a rounded notch for the suspension spring support. The top is nothing more than a thin rectangle of brass, folded over, shaped to a circle, drilled, and pinned through the steel. The pin is trimmed and then hammered flat. There is a close-up of it farther down.

Rear view of weights and bob.

Repaired movement pillar. You can just barely make out a bit of sanding lines. The sort of waffle print mark on the edge, from a pair of pliers, was already there (not me!)

I cleaned up the edges of the screw slots as well, which is standard practice on the clocks I repair.

Here is the repaired bob. I left the dent at the base, but repaired the others around the perimeter. You can see that the rating assembly (threaded rod) looks fairly good, but still just a tad "wiggly".

There is a scratched inscription or name on the back of the lead bob, but it was impossible to make out.

The weights that were with the clock are interesting, so I photographed them. They each weigh 12lbs (roughly), and are made from cast iron. They are quite old, but I strongly suspect that they are replacements. I think the original weights would be closer to 10lbs.

The most notable feature of these weights is that they are sort of oval, rather than round. This means that they have much less of a chance of hitting the pendulum in a snug case.

Here is a video of the clock striking. The bell is actually not very loud, despite what it sounds like in the video.