I just purchased a 4 glass clock with the movement represented by Plate 1419 in the Horolovar book. I cleaned the movement and went to the book to get the spring size and length. I'm missing the bottom block and tried to find out what shape/size to get. The description on the Plate1419 was a disc or 4 ball pendulum. I have a temperature compensating chronometer disc. Is this a problem? On page 196 plate 19 shows a disc associated with Plate 1419. Mine has a similar center piece but a two metal ring. I don't have the capability to photograph the movement. When I do I'll post them. Any hep would be appreciated.
Plate 1419 is for a clock made by Phillip Hauck, and the pendulum you describe should look like #6 on Page 195, which is also by Hauck. The bottom block should be a snug fit into the hook on the pendulum, and anything that fits will work. We would love to see photos of the clock as well as the movement - Hauck four-glass cases are very attractive, in my opinion. Is there a serial number inscribed in pencil or ink anywhere on the pendulum? If so, does it match the number on the movement - that would help determine that the pendulum is original to the clock. Best regards, Mark
Hi Barry, The spring should be long enough that the pendulum hangs about 1/2 inch above the base of clock. Any chance you could post a few pictures? Eric
Measure the length with the pendulum mocked up sitting on the base...then cut the suspension spring longer than your measurement. This gives you a chance to slowly cut the spring back a bit at a time until you get the length you need. Krt
Here are a few pictures taken with my phone during the movement cleaning process. They're not very good, but, they're all I have for now. Will post better pictures in the near future.
Here's a few more pictures. As you can see the case is in rough shape. I'm not sure what I will doing with the case?
Expending a lot of elbow grease, if you don't want to keep it heavily tarnished. To do a thorough job, it would be necessary to dismantle the case into its various components, not a job for the faint of heart! It will be a beautiful clock when finished; this case and pendulum are somewhat uncommon.
If you examine the protected areas on the case you should find what the original finish looked like. For your case you should be able to unscrew the four feet and remove the bottom plate, or two screws inside the case that hold the top plate. If you do take the case apart you should make sure you have a way of keeping track of what part goes where so it goes back in the same place. Brass will turn copper colored as the zinc is leached from the alloy. It also causes pitting on the surface. Eventually it will turn green as the copper oxidizes. Eric
Who would of thought......What do most of you use when dealing with this issue (I used a product named MAAS for the patch that I cleaned)
Another issue has come up with cleaning the case. When I finally remove the corrosion and get to the brass metal, the areas that were heavily choroid came up as a rose-gold colour versus the less choroid areas yellow-gold. Any ideas on how to colour the rose-gold to yellow-gold? I've heard of a lacquer spray with gold in it. Would that work?
Hi Barry, That color is where the brass has lost zinc. The only way to remove it is mechanically. Polish through it. You may need something more aggressive than what you're using. I use Simichrome which is very effective but it even fails if the corrosion extends too deep. Your case is pretty heavy so you shouldn't have to worry about damaging it. Eric
note that the brass used for the case is of lower quality than that of the movement. In my experiance a few discolored spots remain even with an "aggressive" polish.Take care not to mix up frames and crystals ,they are not interchangeable! Make notes or build seperate piles of parts while taking appart. A nice and collectable clock when it ´s done , especially with that compensating pendulum! Burkhard
Decided to stop polishing leaving it with some imperfections. I want it to look it's age. Not new out of the box. What do you think?
Beautiful job! I'd be happy to have it on my shelf! I don't try to get rid of all the spots either. Even when I think I have them all they still show up in the pictures. As I tell my wife, I only work in percentages. I like them clean and bright and with a history. Eric
This is the first torsion clock that I've worked on. I'm having trouble keeping it running. The palate is not moving smoothly in and out of the escapement. There seems to be a pause before it snaps. Is this normal or is there a fix for this?
Hi Barry, The action of the escapement isn't smooth. You should be able to see the impulse at the fork, and often the anchor pin will recoil after the pallet drops off the tooth. Here is a video of a well adjusted escapement for reference: It's much easier to see with the movement out of the case on a test stand. Eric
Mine looks identical to the video. I watch it run for ten minutes, leave, come back in a couple of hours and it has stopped.
Do you have a sense of the lock and drop measurements? If they're not set right and are even with each direction, then you will have uneven power and ultimately a non-running clock. Also, one thing that I look for has to do with the position of the anchor pin when the escape wheel tooth has moved off the lock face on the pallet and moved about 1/3 down the impulse face. The anchor pin should be vertical when the tooth is in that position. If not, or if the deviation is too great, power delivery is weak and the anchor pin might bind during the extremes of the travel. Have you reported on the amount of over swing you're seeing? You may need to lower the fork to ensure better over swing. Kurt
Barry, You made a good choice for your first torsion clock. I didn't make that connection earlier. Hauck clocks are "normally" easy to get going if they are in good condition. If all else is good, a short run time can be an indication that it's out of beat. After the escape wheel tooth drops off the pallet, the amount of continued rotation of the pendulum (over swing) must be equal in both directions. Setting the beat is one of the basic skills of working with torsion clocks. You will also want to eliminate lost motion at the fork as much as possible. The gap between the fork tines and the anchor pin should be about the thickness of a strip of writing paper. Too much gap and you loose the impulse power to the pendulum. Too little an the fork binds on the anchor pin. Eric
How close does the over swing have to be? 0,5,10,15 degrees to be in beat? If less than 10 degrees, any tricks that might help making the fine adjustment using the set screw on the top block? I think that I'm more than 5 degrees but less than 10 degrees.
I think you need to be significantly better than 10 degrees...probably down in the 5 degrees or even better. Admittedly, sometimes I find this adjustment to be very sensitive and it has to do with the arrangement of the top block/saddle and what provides the tension or "locking" feature. Some are friction type, but I find the friction to be both loose and tight...tighter is better. On say Schatz clocks, there's a screw holding the top saddle which lets you control the friction but the friction tightens up when turning one direction but loosens going to the other...I find that very frustrating. And when you do get it into beat, trying to tighten that screw will change the beat. Rats! There is a beat setting tool...see link. I've never felt I could use that...especially on the set ups with heavy friction as I can't see how it could grab the upper saddle feature tight enough to make a difference without scarring the saddle. On some clocks, like the Schatz, I can slip in an ice pick or maybe a special tool like a dental pick to give me some leverage to make small adjustments. It extends outward maybe 5-6 inches so that I can move the far end a far amount, but have that translate to a smaller movement at the saddle. On some clocks, this is not possible, but I look to use this on Schatz clocks...I have one on the bench going through my test phases. Beat Setting Tool Kurt
Ideally you should be able to get the beat equal from drop to drop with no over swing. Slowly rotated the pendulum by hand just until the pallet drops from the escape wheel tooth and release it. If the clock is in beat, the pendulum will rotate in the opposite direction until the other pallet drops and reverses directions with no over swing. If the pendulum rotates past the drop, rotate the top block opposite the pendulum. If the pallet doesn't drop, rotate the top block with the pendulum. Sometimes case mounted clocks have a short arm added to the top block to facilitate setting the beat. Not a bad idea. Eric
As you can see of my pendulum disc, the two weights aren't center between the time adjustment points. What will do to the clock performance?
You posted a picture of your pendulum in #7 above. I don't see any issues with the adjustment weights. The two large "lumps" are exactly opposite each other on the circumference...it looks a little odd they are not 90 degrees from the other adjustment weights, but that's the way it comes. The two smaller weights at the ends of the threaded rod seem to be the same distance from the center. Kurt
Theoretically it's a temperature compensating pendulum. None of them actually worked because they were chasing the wrong problem. The outer rings are bi-metallic and will expand/contract with temperature changes. The closer the two weights are to the free ends of the rings, the greater their effect on time keeping. Some of these pendulums have four weights, but like I said, they didn't solve the timekeeping problem. Eric
Eric - I had always thought that the idea of the bi-metallic rings were somewhat of a good idea, but that the problem was that the earlier suspension spring varied wildly in their modulus due to temperature changes. So, there were two things that were essentially working in counter directions, thus resulting in poor regulation. Is that the "wrong problem" you are suggesting? Seems like even today with better suspension springs which have a consistent modulus of torsion over nominal temperature ranges, now with the pendulum going through its own "compensation", it still works to defeat the purpose of the new suspension springs. That said, my one clock, a Hauck, with this pendulum is actually keeping good time. I just checked it for the first time in months and it's only off around 15 minutes...it's out of my line of sight so I don't usually check it. Maybe it's been up/down in regulation but over the long period the clock has run fine. I have no info as to how sensitive to temperature the old suspension springs were or this unique pendulum, but how much of a temperature change do people's house go through? With thermostats set, wouldn't the inside temperature remain within a tight band? Maybe changing during the switch over from summer to winter, etc. Anyway, just some random thoughts! Kurt
I found some info that was posted when I showed the clock after I purchased it. John indicated that the bimetallic rings do work, but it can be tricky trying to regulate it. He suggested that using something hard (piece of wood) as a wedge in the gap on each ring to prevent the ring from contracting can eliminate this adjustment. The wedge needs to push out on the rings enough so they don't expand further when warmed up to normal indoor temperatures. Kurt
Back when this pendulum hit the market there were articles written about it. One that I've found in the Leipziger Uhrmacher-Zietung translates roughly: "The purpose of the compensation pendulum is to compensate for the expansion of the pendulum spring, which causes the pendulum vibrations to accelerate or slow down in the event of temperature fluctuations, by automatically shifting the center of gravity of the pendulum. Since this expansion is not always the same for every pendulum spring, the shifting of the center of gravity must also be adjustable, and my pendulum has this advantage." We know now that the modulus of elasticity was the problem, but they thought it was physical expansion of the spring. The irony is there were temperature compensating spring materials available and used in watches of the time that would have solved much of the problem. Huber had a licence for one of them but I believe they never applied it to torsion clocks. Eric
Yes, I understand that C.E. Guillaume from Switzerland had invented a temperature compensating suspension spring using Invar in 1904. Several other options were offered by others in the first half of the 20th century. Interesting in the quote you provide...they said "shifting the center of gravity" if that is how it is translated. The center of gravity remains the same...it's the rotational moment of inertia that changes. Kurt
It seems that is the translation of that word. Still, the CG doesn't really change IMO. I heard from some German friends that moment of inertia would be the word "Trägheitsmoment". Guess we beat that one to death! Kurt
The LUZ article also throws some shade at the other compensating pendulum designs. Two of the claims made were: "As a result of the simple construction, the compensation device can be produced very precisely and solidly and is therefore not subjected to any bending during transport, so that the oscillations of the watch remain correct after the pendulum is re-attached." No doubt this is a reference to the loops on the Huber pendulum. "The hanging and unhooking of the pendulum is easier to carry out than with other devices and requires no special knowledge, such as in those where the compensation body is connected to the spring." This one is obviously a reference to both the Huber and Wille designs that grip the spring. Guess you needed to be a rocket scientist to figure them out. I think it would be interesting to run some side by side tests over a range of temps to see who actually came closest to achieving their timekeeping goals. Eric
Regarding the arm weight position on this pendulum, the illustration of the patent drawing found in the DUZ shows the two weights at exactly 90 degrees to the regulating weight adjusting shaft. A substantial majority of the ones I have documented have those weights in that position; on this one and some others for whatever reason someone has moved the weights further out on each arm possibly trying to increase the effect of change in rotational mass with temperature(?). In any event, it's simple to stop the "normal" expansion/contraction of the bimetallic arms by inserting chocks in the openings at the end of each arm and place these weights at 90 deg. I've used wood as Kurt mentioned above, also have used other materials.
Eric, I have substantial evidence that Huber used Elinvar springs in the 1920s; he did hold a license from Guillaume for the use of this material that was patented in 1918. I have found a number of Huber movement clocks from the post-WWI period that appear to have had their original suspension springs, that operated with very little variation in rate over several months. These springs were wider and thinner than the equivalent strength Horolovar spring but "if" they were Elinvar the good timekeeping performance would be expected. I don't have any simple way to confirm, unfortunately. The 1904 Guillaume patent mentioned by Kurt was only licensed to Claude Grivolas to the best of my knowledge. It was a two-piece spring with about half using steel and half using Invar, the material Guillaume invented in 1896. Also this clock case is the Hauck equivalent of the Jahres-Uhr Catalog model No. 222 we discussed separately. I haven't been where I can get the dimensions of the one I have but no question this is the case.
Re positioning the outer weights is a feature of the pendulum. According to the write up in the 1906 LUZ: "2. ist die Ausgleichung derart regulierbar, daß sie auf "Stärker" oder "Schwächer" eingestellt werden kann, ohne die regulierten Schwingungen des Pendels zu verandern." The compensation is adjustable so that it can be set to "stronger" or "weaker" without changing the regulated oscillations of the pendulum. [google translate] The way to accomplish that would be to move the outer weights. Eric
John, Similar yes, but definitely not the same style case. Side by side the differences should be readily apparent. Please note the pictures are not to scale. Eric
But as the rings expand, wouldn't the weights stay in the same position relative to each other, and have the same effect as they would if they were at the 90 degree position? I would assume that the expansion would be uniform across the rings and the gap would be only there to allow for uniform expansion. And second question is why would you try to prevent the expansion of the rings by blocking them? Is the expansion so great that it effects the timing of the clock?
Hi Barry, If the weights are closer to the free ends of the rings, a smaller temperature change will have a larger effect on the speed of rotation. If close to the fixed end they will have little or no effect. They would be the same relative to each other, but their effect on the moment of inertia of the pendulum is very different. While were at it, could you please give the outside measurements of your case? Eric
Eric - I'm not following your logic on movement of the larger weights. Seems to me that no matter where they are located (keeping them exactly opposite each other), the rotational moment of inertia is maintained. That inertia is all about the weight and distance from the rotational center...which seems to remain the same no matter the position. Now maybe you're considering how the bimetallic rings actually bend. It certainly makes sense that the parts of the rings that are near their fixed points move much less than the free end of the rings. How much less remains to be determined. Kurt
Hi Kurt, Lets just say for instance that a temperature change to the suspension spring causes the clock to run fast. To compensate for that temperature change, our bimetallic rings would need to move the weights (mass) away from the axis of rotation of the pendulum. The product of the mass and the square of its perpendicular distance from the axis of rotation (moment of inertia) would then increase. If the weights are mounted at the free ends of the bimetallic rings, a smaller change in temperature would produce a greater movement of the mass than if the weights were at the fixed end of the rings. None of this makes much of a difference if the clock is located in a building with central heating and cooling because you minimize the temperature change problem. Eric
I still have the movement on the test stand. Every time I make a change (adding the the dial plate and dial, mounting the hands etc.) I remove the disc and top-block/spring/bottom-block assembly and make my change. After the change I find the clock runs only a short time and then stops. I have to look for the issues that cause the stoppage and re-adjust. Am I doing this wrong? By removing the spring assembly am I forced into re-adjusting the beat every time? The assembly comes off and goes onto the cradle without any changes to the blocks or fork. .
I don't about being "forced" to reset the beat, but you should at least confirm the beat is still good. Also be sure the fork hasn't closed up on the anchor pin. Does the escape wheel moved with "vigor" when it drops off a pallet? Kurt
It went out of beat by a lot. I've re-adjusted the beat and it has been running for a couple of hours. I still have the hands to put on, and make any adjustments to the spring for time as necessary. The last time I had it all together, it ran 10 minutes fast per day (that's why I took the movement out of the case). I'll just keep checking the beat any time I make a changes and adjust accordingly.
I feel I'm getting close to putting the movement back into the case. But I hope to get a couple of informed suggestions. The clock is running about 11 minutes fast per day with the time adjustment weights in their current position. Should I move these weights into a neutral position (leaving a half adjustment each way (on the pendulum))? This would mean the the clock would speed up from its current rate. The next question is reducing the spring size. I was going to use emery paper to reduce the spring size. If I use moving the emery paper down the length of the spring in one direction as a count of 1. How many times would you suggest that I move the emery paper? I will remove the spring fork and bottom block. Should I only emery one direction (top to bottom) or should I go both ways? Thanks for all your help. I really appreciate it.
As for thinning the spring, I've seen some suggestions for how many strokes but in the end, I just give it a number of strokes...say 10...and check how the rotation rate has changed. Do this as a way to "creep up" on the desired results. I suspect you're after 8 beats per minute. You don't need to put the clock completely back together, but find a way to secure the top block to a fixed point and then put the pendulum on the bottom block...leave the fork off. Give the pendulum a twist and then time what it takes to get 8 beats. Continue this until you get the proper rotational rate. The best way to thin the spring is to leave the bottom block on and remove the fork and top block. Sand the spring from bottom to the top. The part of the spring between the fork and the top block are the most important, so you should do this from bottom to the top. I think you should put the weight adjustors in the middle of the range and do the thinning process. This will allow some under shoot or over shoot as you thin. If you miss the exact thickness, you should have enough range in the adjustors to cover it. Kurt
