Separate names with a comma.
Discussion in 'Clock Repair' started by darita, Jul 28, 2017.
How many days?
So do you attribute your run-time success partly to your mainspring servicing, Willie?
As I recall, you're pretty aggressive when it comes to cleaning and polishing the existing springs. Much more so than I am. I normally use WD-40, a synthetic scouring pad (like Scotch-Brite) and elbow grease to remove dirt and degraded lubricant..followed by removal with Mineral Spirits drying and lubrication. Lately I've also followed the manual cleaning steps up with a U.S. Bath and quick drying with a heated air flow. That gets into the inner coils where I can't. Manually cleaning first keeps a lot of gunk out of my U.S. Bath. I think it does a pretty good job but it doesn't always remove "stains" and it certainly can't remove rust or rough metal. I usually only have trouble locating a suitable replacement spring when I'm working on imported movements. It's' always good to have options and If I can save money while achieving good results, that's a no-brainer.
Willie, while my reasoning may be conventional, it is also partially based on my own experiences and observations in the field...albeit much less than your own... I know that experiences with one type of movement (the ST 113 always comes to mind) are not universally applicable, but when you see the results of a mainspring failure (from catastrophic to falling torque curve), you can't help but be cautious. I'm here to learn and to help if I can.
I use steelwool and kerosene to remove old dried oil. Steelwool will also easily remove rust without damaging the surface of the spring. Steelwool can also reach into the small centre coils and remove the dried oil, but it would be difficult to remove any rust in the tight centre coils. Afterwards I dunk it in benzene and allow to dry. After replacing in the barrel I lube it with thin oil and using the spring winder, wind it up and unwind it about 3 times to ensure the oil gets distributed.
Okay. Sounds pretty thorough. Steel Wool sheds but it sounds like you're pretty careful to remove all traces of it. I know some folks go as far as to use various grades of sandpaper too.
I guess I jumped into the thread because there seemed to be some question as to whether or not a "Set Spring" is a mythical beast. I find it pretty incredible that we even have to argue that point. There is a place for Conventional Wisdom and isn't the human race fortunate that every new generation doesn't have to re-invent the wheel?
I've said plenty so I'll go back into Lurking/Listening Mode now.
TAT and Kilo,
Our methods or similar and should have similar results. My materials vary according to the condition of the spring. Some springs have such a tough coating of scale, I start with 150 sandpaper, then a few strokes with 220, 320, 400, then steel wool. I find this takes much much less time than using a finer paper from start to finish. One thing for sure about this method is that you can easily spot any defects in your spring and it only takes about 5 to 10 minutes per spring, once you get a routine worked out.
Best wishes, Willie
Hey Willie X. Just curious, what is your record for the run time on American 8 day clock?
(also do you attribute it to mainspring care?)
One hour short of 18 days. And no I can't take much of the credit. This was just another day at the shop for me. The attribute goes to early American engeneering and know-how. Also, this is a good testimony for keeping old mainsprings, if they are still in good shape.
[Click-Listening Mode Off]
Willie, wasn't that a Loop-End Open Spring set-up? If so, while still a pretty impressive job with a nicely engineered clock, a whole different animal from Barreled Spring Set-ups
[Click-Listening Mode On]
Not that much difference. The spring turns against a load for a certain number of turns and then stops when it hits something, or can no longer move the load.
I respectfully disagree. I think there is much more sliding friction in a barrel. While there is sliding friction in an open spring, I think it is much less as the coils are free to expand until the outer coils hit a limiting pin (ideally) or an arbor in a less ideally designed movement. Less sliding means less friction, no? A good argument for your sanding/polishing service treatment, especially in a barreled spring setup.
Also, for run times, you're going to get a much longer run time out of an open loop-end spring. The clock may not keep good time much past it's designed wind cycle but it can and will keep going until the spring is finally restricted by something or its torque is no longer sufficient to drive the going train.
How long did your record holder keep good time? I want to say 10 days, (which is great for an 8-day movement), but I'm not sure.
Some folks like to ramble on, I don't. That's why I don't usually respond ti a thread after about 20 hits, unless there is still a genuine helping or learning process going on between the OP and the regular listers. Duno why I'm off on a tangent on this one when the OP is long gone! Ha
Has anybody actually measure the torque vs rotation for springs, either new, old, or refurbed?
I ask, because I'm building a torque ripple test setup for wooden gears**, and it is distinctly possible to modify it to actually test and exercise mainsprings.
I'm currently learning the Arduino mega2560r3, so it will be right in the general vicinity.
** The wooden gear clock (tempo, Boyer) in my dining room is unstable vs humidity, and I suspect gear meshing is a primary culprit.. I have a 5 lb weight (80 oz) on a 1/2 inch radius spool, 40 in-oz... and at the second arbor, it reduces by a 64/8 ratio, or 5 in-oz. At the escape wheel, reduced by 60/8 to 2/3rd inch-oz torque. Max.
I plan on measuring delivered torque at all three arbors. Then I'll work on spring measure.
Just call me the Rambler Willie. Maybe the OP is still around and just not commenting. I only came back in to point out that there is a distinct and significant difference between loop-end and hole-end spring set ups. Obviously you don't agree Okay. I'm out.
Both an open spring and a barreled spring slip some
as they wind around the arbor.
The outer winding of a barreled spring do little if any
slipping as the windings lay against the barrel.
On an open spring, the outer turns being weaker per turn
are in slipping contact for longer distance of the spring
than the inner turns of of the barreled spring.
In other words, there is greater slipping action on an open
spring than a barreled spring.
How much difference that makes to run time I could not guess.
So is "consensus" here that heat treating (hot-cold-then temper) can rejuvenate a mainspring, yes or no?
I believe that would be true under ideal circumstances. Most of us don't have a hearth with enough controlled heat to make it feasible. I believe that is the problem with the India made springs.
I see guys melting aluminum all the time in small home made metal foundries on youtube. That or a kiln might be a perfect tool to heat up a mainspring enough. But I was thinking that maybe if the mainspring was put in a circle about 7 or 8 inches wide. Not fully unwound, wrapped by tie wire, so a propane torch could heat the whole thing red hot and then dunk it...
I like the idea of a kiln. Imagine what can be done if we pour our own molds... Ha...! I bought me a used welding machine too. Uses flux core wire. Like a Mig but I am learning. Got it cheap at the flea market. Things are going to get interesting now. Soo many times I wish I could weld.
I bet the problem with the India mainsprings is they ain't real metal no more. Probably some mix to save a few bucks. Who cares if it last...
Psssssssssssssssssssssssstttt ttttt tttt t t t ttt t t,,,,, in the oil.
Well its not my 'consensus' although I believe there is some general validity in the concept. For a reasonable chance of success we need to know what the steel alloy is and what the hardening and tempering temperatures for that alloy are for the specific degree of temper we desire. It also assumes that the alloy is uniform through out the entire spring and that one has a precise means to control the temperature and maintain it long enough for the process to work. I suspect that many springs, especially from India, may have failed specifically because the spring metal was defective to begin with. My take away from all this is that seldom become set to a degree that causes a problem for an otherwise healthy clock, and fooling with attempting to re-temper a defective spring is not practical from the time/cost perspective and there is no assurance of a good outcome while a bad outcome could cause further damage.
I agree -mostly.
Ref Tinker at #64
Wouldn't the power curve have a longer tail off with an open spring? Then the run time depends on other factors like friction and stiction surely and the set-up of the escapement. Wouldn't that be so?
Many people seem to think that open springs wind down in a big spiral, with the outer coils against the sides of the case. In reality they only expand about 1/4 inch after an 7 day run.
For a typical .018" x 3/4" x 8' spring, it's
1.940" OD wound up tight and 2.190" OD after a 7 day run. Willie X
Willie X Did you read the pages in this book?
A 100 Years ago they just didn't replace just because.
They reworked, repaired, made from scratch even.
I am only saying this because of the reply's of #2, #20, #35, you wrote in this thread.
Not trying to be an ass.
I haven't come across a 'set' mainspring. Broken, yes. Rusty, yes. Congealed with porcupine grease, yes. But as long as a mainspring is reasonably intact, it ought to run the clock, and if it doesn't the difficulty is probably not in the mainspring anyway.
One exception may or may not be 400-day clocks. I'm unable to do much experimentation in this regard because I buy suspension springs directly from the Horolovar Company, and ever since Chris Nimon told me that these clocks run lots better with a new mainspring I'm always afraid I'll receive a gaze of disapproval if I don't buy a new one for each 400-day clock I fix.
One observation concerning typical American open spring clocks is that an 8-day clock usually takes 8 full turns of the key to rewind after an 8-day run. When a disassembled clock is reassembled and the main spring is still restrained with a "C" clip or wire loop the diameter of the relaxed spring is somewhat less than the diameter of the great wheel. It takes a lot more than 8 full turns to fully wind it. The other point that isn't often mentioned is that the torque required to turn the winding key is the same torque the going train receives as it runs with the spring wound to that point. It stands to reason that if a spring is so set as to not be able to run the clock, one should notice that it feels "weak" as it is being wound. Once can fit a winding key to a torque screwdriver and compare the force to wind a new spring and an old one of the same size.
A 100 years ago what would you have done with a 100 year old spring ??
Again, it's a matter of opinion, which is heavily influenced by the privilege (or lack thereof) of having replacement parts available. Unless the Royal Swiss Horological Bureau of Enforcement and Terror weighs in on it, I should think we could all make our own decisions. I personally wouldn't think that a mid-spring riveted repair would be all that sinful provided that the spring was reasonably well annealed right there.
As it is, it's never been clear just where, when, and why a mainspring will break. It's entirely possible that nobody knows, and at the present pace of research concerning century-old cheap horological specimens it's entirely possible that nobody ever will. If anyone has paid any attention to this sort of thing it would be in the field of fracture mechanics applied to steel cables, which bear a certain resemblance to mainsprings in heat-treatment and tension.
I am just saying don't replace cause you can.
An old friend said once, "just because its new doesn't mean that it good"
What in HELL are you guys fighting about?
The usual repair vs. replace, I suppose. I think it's been this way ever since clock repair changed from a slightly blue-collar trade into the curatorial staff of the British Museum.
If the spring is actually busted the logical choice is replace if a replacement is available. Otherwise the determining factor isn't how far the relaxed spring expands but will it power the clock for the specified time. For me, the more difficult decision is, on a time and strike and/or chime clock, if one original spring is busted, should the others be replaced as a preemptive strike?
There was a time when replacing springs was pretty common, and even encouraged. Then with the appearance of poorly made springs from India becoming common in most market places and the bad experiences that occurred because of them, a shift was made in the thinking of many, and we started to really analyze whether a replacement was wise. In time it became obvious that main springs were almost never the problem, so the thinking became "why replace them then?". I almost never replace unless the old one is broken. I haven't had any issues from that approach.
Is there any evidence that an old mainspring is more likely to break than a new one? The reason I ask is that one superstition of many repairmen--of appliances, cars, clocks, and electronics--is that you're taking more chances with a new part that might have been poorly manufactured than with an old part that's been doing its job for years. I tend to be of that opinion (though not always.)
Very interesting thread, I enjoyed reading. I'm still relatively new at repair but something Willie X wrote years ago will stick with me forever, paraphrasing; "unless the spring is broken in the middle or has several breaks, the best spring for your clock is the one that's in it now". I've practiced this since.
I was facing this just this week. The chime and time springs on a chiming Mauthe I am working on were rusty and each broken in about 15 pieces, I hadn't come across that before. The strike spring was fine, I cleaned and examined it thoroughly looking for fatigue. I decided to just order the two springs and keep the other original, I'll get to see how that works out over the long haul.
Maybe the same movement, but the strike side spring "on the one I am working on"was very rusted and broken
It seems that spring took a bath in an aqueous cleaning solution while assembled. Probably a case of dunk n' swish.
Its only the strike barrel.
If a spring is rusted inside a barrel the clock has been soaked and all should probably be replaced if possible. Keep the old ones to show the customer.
Multiple breaks in a clock spring are relatively rare and rather mysterious, at least to me. I've had only one case of this, in a 400-day clock. It apparently happens routinely in a wind-up phonograph, and those springs are in barrels as well. I've never seen a multiple break in an open spring, so I wonder if barrels have something to do with that.
I've seen a few alarm clock springs with multiple breaks. How much of this is due to their width or initial low quality is anyone's guess.
I held the record for a while at 41 pieces but someone bested that by about double; this was several years ago. Willie
But 400-day mainsprings and those used in wind-up phonographs were presumably the best quality available.
The physics of mainsprings get rather complex very quickly, which is generally the case with what's called 'materials science.' For example, when an object breaks, something called 'surface energy' is released and cannot be recovered, if I have it right. I think I'm going to re-read "The New Science of Strong Materials," by J.E..Gordon, which is a thorougly splendid book anyway.
My phonograph mainspring probably yielded ten to twenty pieces, but the smallest shards were less than an inch long. Willie X's 41 piece record is amazing. What sort of clock was it?
Ha! Ha! I am dead keen and all for getting the Brits a mention Mark. We're a little bit fossilised here I must admit. Time for us has stretched a long way. We have picked up some bad habits along the way. BjG
That 41 pieces record has been 'shattered' some time ago ... Ha
What I meant is that clock repair was an honorable enough profession 50 years ago, but was a lot like appliance or automobile repair as an art form. Since I've come back to it, clock repair--or at least a lot of what I see here--seems to have become heavy with historical gravitas and technical perfection. Maybe it always was, and I just missed it.
I used the British Museum as an example of historical gravitas and technical perfection.
Yeah I was only smiling Mark. We Brits can take it. I think I see what you're driving at. I suppose nearly everything is talked up at first then as the word spreads and the reality of earning a living sinks in, those involved find ways to get it done without being very indulgent. Now, with electronic substitutes which work well and maybe very clever but out of reach to any kind of interesting repair, the scene has changed. Now we realise just how difficult it used to be to get such reliability and accuracy out of mechanical science. That's why I feel in awe of those old craftsmen and their struggles on a day to day basis which must at times been very heavy going indeed. I like reviving the mechanical stuff that by now many of them probably hate as much as I hate electronics! I give all the credit due to the traditional watch & clock people whether they want that or not! In everyday reality I am very given to a "good-enough" philosophy even if I don't always say so! You could say, I suppose, that I can manage without a minute hand. :<))
Best regards BerryG