More Koma Midget woes

[Nathan S]

I have read a few things on here about the lack of power to keep Koma midgets running. What I think I have learned is that the earlier 20mm wide ones benefit by a different spring- 12x24. Mine is the 21mm wide version so I think that was done to increase the power so presumably no spring upgrade available? I have stripped examined and cleaned and reassembled and minutely oiled. It has about 120º swing and runs with the hands loose, but tightening and setting time seems to slow it to a halt.Is there anything suggested that will give it more energy,?

 

[sjaffe]

There are two spring parameters that can provide more power: increased width and increased thickness. If you can find a thicker spring, that should provide more power, but it will not run as long. I would try to find a thicker spring and just wind it more often. You may need to modify a replacement spring to suit your needs. If you need to either shorten or punch a new hole in the replacement spring, heat the portion to be modified to red hot and let it cool slowly. This will anneal that portion of the spring so you can cut/punch it.

Stan

 

[KurtinSA]

I'm pretty sure that the midget plates are pretty thin. Wear may have set in over the years and possibly holes need bushing. It should be clear that the clock left the factory with good performance, so any subsequent wear will reduce that. Not to mention, the minis and midgets are notorious for being difficult to get running well.

Kurt

 

[Nathan S]

There are two spring parameters that can provide more power: increased width and increased thickness. If you can find a thicker spring, that should provide more power, but it will not run as long. I would try to find a thicker spring and just wind it more often. You may need to modify a replacement spring to suit your needs. If you need to either shorten or punch a new hole in the replacement spring, heat the portion to be modified to red hot and let it cool slowly. This will anneal that portion of the spring so you can cut/punch it.

Stan

I have got the Koma Midget running but it seems down on power. I have mentioned it is the 21mm wide later clock, which has a more powerful spring. I wonder if anyone knows of a better spring for this model? UnfortunatelyThe Book only gives width and diameter and not thickness which is probably the information needed. Any advice gratefully received.

 

[KurtinSA]

I would doubt that it is the main spring. More than likely, it's the movement or escapement where power is being lost.

I have a Kundo narrow plate clock that I have been unable to get to run with any power. I tried numerous things. I finally gave up and put a new main spring in it...didn't make a difference. I finally took it to a friend who has looked it over. He says that the locks and drops are not right, that power is being wasted because of that. And here along I thought things were set up well. I'm anxious to get with him so he can show me what I was missing.

Kurt

 

[Nathan S]

I would doubt that it is the main spring. More than likely, it's the movement or escapement where power is being lost.

I have a Kundo narrow plate clock that I have been unable to get to run with any power. I tried numerous things. I finally gave up and put a new main spring in it...didn't make a difference. I finally took it to a friend who has looked it over. He says that the locks and drops are not right, that power is being wasted because of that. And here along I thought things were set up well. I'm anxious to get with him so he can show me what I was missing.

Kurt

Looking at it is pin pallets and I can see no means of adjusting them. I have stripped and cleaned and examined closely and the pins look good and are not bent. Pivots look good and the ‘lean’ on the arbors is less than recommended on larger clocks that I have made. Everyone says these midgets are the devil’s children so I am determined to beat it! I was fairly pleased it ran for over a week. Thanks for sharing your experience.
Al.

 

[KurtinSA]

But you can adjust the eccentric, right? The locks and drops still have to be right. Does the anchor pin move evenly from the left to the right side, about 4 degrees or so each side? If not, then your impulse from the pendulum, suspension spring, and fork is not optimal.

I don't have a lot of experience with pin pallets...I'll work on them when I run out of the others...which at this rate is going to be a long time!

Kurt

 

[Wayne A]

400day midget pin pallets can be difficult due to marginal power and pin pallet friction. I'm running 6 of them but have been avoiding additional ones unless its spectacular. I like to coat the pin's with a dry film teflon, light oil works too, really helps allot.

Currently working on one of my jum/7's. Just put the barrel back together after bushing it, lids arbor hole was 12 thou off center plus the arbor was very rough with a loose fit. This was not from wear, it's how it was made and had to be causing power loss. Going to bush every pivot as there sloppy. This clock took two turns on the mainspring to run before I started, can't wait to test again when done.

Wayne

 

[Nathan S]

But you can adjust the eccentric, right? The locks and drops still have to be right. Does the anchor pin move evenly from the left to the right side, about 4 degrees or so each side? If not, then your impulse from the pendulum, suspension spring, and fork is not optimal.

I don't have a lot of experience with pin pallets...I'll work on them when I run out of the others...which at this rate is going to be a long time!

Kurt

There is no eccentric to adjust pallet action. Clocks I have made have this feature in both plates, and I cringe when I see it in one plate as it is moving the arbor out of line. After over half a century in precision engineering and working on prototype jet fighters etc I started working on clocks thinking they were the ultimate delicate precision machine, but som of it is quite crude,.
I think I've read of people deliberately bending stuff to adjust pallets!

 

[KurtinSA]

Well, technically you can't really adjust pallets if they are non-adjustable...at least as far as I can think. You can regrind them, etc., but that's different. I typically don't like to bend the anchor pin. If the anchor pin isn't in the right position...say too far left...during the escapement, I would prefer to adjust the pallets but lengthening one and shortening the other by equal amounts. But if the pallets are fixed, and the anchor pin

 

[sjaffe]

The manufacturer knew they had a problem with this clock and hence why they increased the size of the mainspring. Insufficient power can result in a pendulum with no overswing/supplemental arc (clock will not run) or insufficient overswing. There is a certain amount of "noise" in the pendulum overswing amount (I believe variation in escape wheel teeth uniformity for one, temperature, other environmental factors). So if there is just a very small amount of overswing, it is possible that one or more of these parameters can vary to cause a decrease in overswing to the point it goes to zero. Once that occurs, the clock stops. You can increase the overswing by changing the fork position, but if you have insufficient power, you can vary the fork height indefinitely and it won't help. I believe this particular model was a very marginal design with just barely enough power to run. Introduce the previously cited degrading parameters and the clock will eventually stop. A thicker mainspring will provide more power at the expense of a shorter run time. So if you're OK with a 200 Day clock...

Stan

 

[kinsler33]

There is no eccentric to adjust pallet action. Clocks I have made have this feature in both plates, and I cringe when I see it in one plate as it is moving the arbor out of line. After over half a century in precision engineering and working on prototype jet fighters etc I started working on clocks thinking they were the ultimate delicate precision machine, but som of it is quite crude,.
I think I've read of people deliberately bending stuff to adjust pallets!

Your aircraft parts exhibit precision that's orders of magnitude greater than that found in any mechanical clock. Recall that clockwork was originally designed to be manufactured by people working with files, and if you analyze it you'll discover that precision is unnecessary: gear teeth can have almost any shape, there are crude plain bearings used throughout, and if you shake a movement that's in good working you'll hear it rattle. I rather doubt that a fuel-controller for one of your jet engines does that. We routinely bend parts to make fine adjustments, and if we need a bit of brass to be a bit longer we will beat upon it until it is.

 

[Nathan S]

Your aircraft parts exhibit precision that's orders of magnitude greater than that found in any mechanical clock. Recall that clockwork was originally designed to be manufactured by people working with files, and if you analyze it you'll discover that precision is unnecessary: gear teeth can have almost any shape, there are crude plain bearings used throughout, and if you shake a movement that's in good working you'll hear it rattle. I rather doubt that a fuel-controller for one of your jet engines does that. We routinely bend parts to make fine adjustments, and if we need a bit of brass to be a bit longer we will beat upon it until it is.

Of course that is right! I'm currently making a reproduction of a sixteenth century clock using techniques as you describe as a nice piece for my cottage of around that age.I must get real and adjust, I did not wish to appear a smart-arse! I am just an amateur with a few years of experience.

 

[kinsler33]

Of course that is right! I'm currently making a reproduction of a sixteenth century clock using techniques as you describe as a nice piece for my cottage of around that age.I must get real and adjust, I did not wish to appear a smart-arse! I am just an amateur with a few years of experience.

And mechanical watches are really no better.

 

[etmb61]

I find that torsion clocks need to be a little messed up to work right. They are, after all, a compromise design.

Eric

 

[kinsler33]

400 day clocks are exceptionally clever but were never robust enough to do much business with. I'm surprised that nobody ever tried to improve on the escapement and suspension spring arrangement.

 

[Nathan S]

Since starting this thread the Koma Midget ran until today. Now it won't restart.Has anybody any suggestions as to why it would run for over a week and then stop comprehensively? It is fully wound.

 

[KurtinSA]

What was the exact state of the movement when it stopped? You might be able to learn something by examining the movement closely. Also check for imperfections on the teeth or pinions. Lately when I take a clock apart for overhaul, checking each wheel with a loupe is on my checklist.

Kurt

 

[Nathan S]

My latest thinking is that because it ran for several days there may be a problem near the mainspring on the slowly rotating parts as nearer the escapement the wheels would have turned numerous times I think. I will put Sharpie marks and take photos.
I did once calculate a normal clock gearing and it was 360,000 to 1, I seem to recall it was a 30 day movement.

 

[KurtinSA]

I did some calculations for a number of clocks one being a Reiner miniature. My numbers were in revolutions per month (or 30 days):

- main spring -- 0.39
- 1st arbor -- 2.8
- 2nd arbor -- 15
- 3rd arbor -- 96
- minute arbor -- 720
- escape wheel -- 8640

But that doesn't necessarily help. If you have a flaw somewhere and it was only a few teeth away from engaging a pinion, the clock would stop quickly. But if the flaw was further around the wheel, then the time to stoppage would be a different time. It's best to just look over everything carefully to know for sure.

Kurt

 

[Wayne A]

Don't know which Koma you have but the one Koma I have the gearing recorded for is a plate 1392. First arbor turns once in 13.65 days, 2nd arbor 2.13 days. Anyway I like to do allot of testing before final assembly. Put one arbor in at a time and check for smooth free spinning, then add additional arbors checking at each addition. Finally without EW in the clock run it with the mainspring for a couple rotations of the mainspring by only adding a few clicks of power at a time. Sure its not loaded exactly as in use but glaring problems should be evident.

Wayne

 

[kinsler33]

I assume that the movement is under some sort of dome, for a puff of the evening breeze will stop the thing dead.