My new Knox clock, two problems

[Bernhard J.]

As reported

https://mb.nawcc.org/threads/robert-knox-beith-scotland.188775/

I recently added a longcase clock of about 1760 to my things, which nobody really needs (but which I like).

It does run nicely, aside the fact that it is about 8 minutes fast per day. I estimate that the bob would need to go about 5 to 10 mm lower. It is at the lowest possible point already. I will have to think about what to do, simply a longer pendulum spring is, of course, not the solution. But that is not my instant problem.

I was wondering that the change of the date did not always occur at the same time, but with a constant time shift. Having a closer look and making a tooth count revealed the following. The hour wheel pinion, which drives the date change wheel, has 30 teeth. The date change wheel has 56 teeth . So no surprise that it does not shift the calandar wheel at the same time each day, because it would have to have 60 teeth for doing the job properly. Looking at the date change wheel it becomes further evident that it does not belong to this clock. The finger for driving the date wheel has be soldered to the date change wheel in a rather crude manner. And the way it has be riveted to the (presumably original) collet does not look very fancy either.

First I wonder, which takes such a silly approach for whatever reason. Any purchaser or repair shop customer will notice after a few days that something is wrong with the date mechanism and throw eggs and tomatos at the seller or clock "repairer". I cannot throw that far, unfortunately, and would not anyway, since I cannot really complain in the view of the price paid.

Now my first question is: Are there suppliers of wheels, wherein one can choose diameter and teeth number? If not, could you recommend someone making wheels to specifications? And what would the latter alternative cost, roughly estimated?

My second question relates to how the seconds hand is correctly fixed to the escape arbor. My seconds hand is quite sure not original. Very sure the method of fixing is not original. Someone has cut a thread to the end of the escape wheel arbor and fixed a small nut to the backside of the seconds hand. By this the seconds hand is simply screwed to the escape arbor. It took me a bit to figure that out when disassembling the dial for getting the date change wheel out. However, this is not an all too stupid approach in case of a vintage clock that is probably considered rather "affordable". But I would nevertheless like to know how the seconds hand is correctly fixed to the escape arbor originally. Because I will be looking for a more or less correct seconds hand anyway.

Thanks in advance for your comments and information,
Bernhard

 

[Bernhard J.]

Here are photos of the wheel in question. It should have 60 teeth.

 

[Uhralt]

To your first question: yes, there are people out there that make custom wheels. I found somebody on ebay years ago but i think others can give you more recent recommendation. With regard to your second question, usually the second hand is attached to slotted pipe that is just pushed on the escape wheel arbor. The slot is there to provide a little flexibility and friction.

Uhralt

 

[Bernhard J.]

I today just mailed Patrick Woodward, whom I know for brilliant watchmaking work. Perhaps he does clock work also and has access to a wheel cutting machine of the required size. My wheel cutting machine works for watch wheels only.

If not, he might have a recommendation.

Cheers, Bernhard

 

[Bernhard J.]

With regard to your second question, usually the second hand is attached to slotted pipe that is just pushed on the escape wheel arbor. The slot is there to provide a little flexibility and friction

Thanks for confirmation of what I had thought. But since I know little about vintage British longcase movements in detail, there might have been some other interesting design.

In the meanwhile I have found a reproduction seconds hand of appropriate lenghth and presumably at least better matching style. What do you think? It will, of course, need to be blackened. And not be fixed with a screw, but a pipe, which I will make.

 

[gmorse]

Hi Bernhard,

Perhaps he does clock work also and has access to a wheel cutting machine of the required size.

Well, he works for Geoff Allnutt in Midhurst, whose team certainly does have the facilities to cut wheels.

Regards,

Graham

 

[RJSoftware]

Just curious, the hour wheel pinion is typical motion works gear, or is it friction fitted to hour cannon tube? Those way back when had different methods. You're probably right just curious if possible backwards. Where pinion should be 28 making date advance gear correct at 56.
I'm just stretching out of the box thinking. Probably I'm wrong

 

[Bernhard J.]

The hour wheel actually has three components, which are fixed to each other on the hour cannon tube.

First, the large diameter wheel, the actual hour wheel, driven by the motionwork.

Second a "pinion", rather a broad second wheel, which drives the wheel in the photos of the calendar work. This "pinion" drives nothing else and is not driven by anything else. It consequently rotates once in 12 hours together with the actual hour wheel and has 30 teeth.

Third the stricking snail.

The time of day, wherein the date changes, has to be selected by the rotational position of the finger of the wheel in the photos when assembling the hour wheel together with the wheel of the photos (called calendar change wheel, by me, is there a correct term for this?). If you do not think of this upon assembly, you will likely have the date change at a daytime not desired. The shift of the date wheel takes something between 2 and 3 hours to complete and I set it to beginn at about 12 (24).

 

[Uhralt]

Thanks for confirmation of what I had thought. But since I know little about vintage British longcase movements in detail, there might have been some other interesting design.

In the meanwhile I have found a reproduction seconds hand of appropriate lenghth and presumably at least better matching style. What do you think? It will, of course, need to be blackened. And not be fixed with a screw, but a pipe, which I will make.

View attachment 715630

The seconds hand looks very nice to me. Usually they come with multiple slotted pipes to accommodate the escape wheel arbors of multiple clocks. The pipes have a little dimple on top to rivet to the seconds hand.

When I said "pipe" I actually meant a piece of brass rod, partially center drilled (making it the "pipe"), then slotted, and a solid piece with the dimple on top. Easy to make on a lathe if your hand doesn't come with an assortment of pipes.

Uhralt

 

[Bernhard J.]

Patrick answered and he or one of his colleagues will attend to making the correct calendar change wheel. I will send him the old one, the hour wheel, and the center distance between the two axes, so that not the whole movement needs to sent back and forth, which would presumably result in piles of customs forms and correspondence. Brexit really is a pain (imho).

Cheers, Bernhard

 

[Dells]

In the meanwhile I have found a reproduction seconds hand of appropriate lenghth and presumably at least better matching style. What do you think? It will, of course, need to be blackened. And not be fixed with a screw, but a pipe, which I will make.

View attachment 715630

I think that hand is from the same person I purchased a pair of hands from for the skeleton clock I am repairing/ restoring very good quality.
Dell

 

[Bernhard J.]

Another detail may be interesting and this concerns the pin seen above on the calendar change wheel (correctly called 24 h wheel, a term I saw upon researching?). This pin is bent and that is not correct, of course, it should be straight and fitted at a larger radius instead.

Placing this pin on the 24 h wheel would seem to be easy at a first glance, but it is not, if one wants the day number to appear exactly in the middle of the aperture of the dial. Something often seen not achieved in clocks and watches operating with calendar rings. Best will be trial and error on the movement. Getting this right just with measurements will presumably not be too successful, since four dimensions are relevant. First the innermost diameter of the calendar teeth ring, second the radial position of the pin, third the diameter of the pin, and fourth the rotational position of the pin relatively to the teeth of the 24 h wheel.

For that reason I asked Patrick to make the blank wheel without any holes and I will make the pin and place it correctly with the whole movement at hand.

To do this, I am thinking of making a short arc-shaped sloted hole in the wheel at the appropriate radius, threading the pin, and bolting it to the wheel with a nut on the rear of the wheel (there is enough space). This is not original, of course, but could avoid me loosing nerves when positioning the pin. After all, the wheel is not original anyway, so this does not screw up any original substance. What do you think?

 

[Jevan]

Assuming you get the first calendar wheel back why not use it to find the correct position, drill as many trial holes as necessary, whack in a pin and test on the clock... then transfer the measurement to the new wheel.

Sometimes the advance pin is not a simple fit as it often has to have clearance with the snail, make sure your new pin position operates clear of the snail throughout a twelve hour period before drilling the hole in the new blank, of course I am assuming your centre seconds clock has the snail mounted on the hour hand pipe.
Some earlier clocks have "L" shaped brass flags instead of straight pins for this very reason.

The engraved date ring seems slightly unusual, it appears to be running on the dial pillars held in position with pins through the pillars as opposed to more traditional calendar pulleys or flags.

It is also not a good idea to automatically assume the calendar ring teeth are equally spaced or the engraving is accurately positioned.

Added later.
The clearance with the snail comment relates to the adjustment required allowing the date to change at twelve o'clock.

 

[Micam100]

Perhaps this?

Turn one end of the pin to a point, Friction fit the pointed end of the pin to a piece of brass and super glue the brass "foot" (temporarily) to the wheel and test. A little heat will release the super glue so you can re-position and test again. When you find the right spot, tap on the pin to mark the spot where you will fit the pin. If you measure the tested radial positions of the pin, you may be able to come up with a formula that will guide you to the final position.

Michael

 

[Bernhard J.]

So, here are a few photos of the dissassembly. Nothing special and self-explaining in most cases.

Cheers, Bernhard















The spring on the left lower side obviously is not original. But for example the often missing screws fixing the movement to the wooden support seem to be original.

 

[Bernhard J.]

With respect to reassembly I have three questions to the knowledgable.

First, I note that the date ring is a close fit in the dial pillars, which have a collar at the dial (not well visible in the photo above. Obviously there never were intended rollers on the two lower pillars, which do actually make sense. Would it be a bad or good idea to apply some grease (purpose made for watches/clocks, I have that stuff) at the circumference of the date ring for reducing friction between the date ring and the lower pillars? If so, would some other lubrication material make more sense?

Second, I wonder about the spring below, which I thought non-original. But I begann troubling now. The springs force needs to be very feeble for the striking rack lever to work properly. I do not think that anything other than a thin wire spring, as seen in the photo, would work properly. What do you think? If another kind of spring would have been fitted originally, how would this look like?

Third, I have ordered a bunch of tapered steel pins. Since these will likely have a stronger taper than original, if applied, would it be appropriate to give the thinner end a slight bend in order to avoid it getting loose? The holes in the pillars are oriented horizontly, otherwise it would be simple to apply from the upper side. Or would one in any case rather opt for making appropriate pins with a very low taper?

Cheers, Bernhard

 

[RJSoftware]

A bit I can say about the spring. If the brass wire currently in place has become too elastic you can order brass spring wire from Timesavers.

For brass to establish rigidity of a spring is not the same as other metals where heat + cold is used. Instead hammering brass, compressing it creates spring effect.

Another option is to find the old style click writing pen. The spring inside is very desirable for it's qualities for clock levers etc. Although it may be too thin & obviously not correct style. Timesavers sells brass stock in variety pack, but might not have correct spring tension or might work. Brass stock has enough spring potential anyway.

As to taper pins I avoid bending them, just makes removing them later harder. Usually the post can be turned so pins go vertical. Also I try to leave them long. The more pin surface contacts plate the more grip. The trick is to order the correct size. If you make the pins, holding a spinning grinding wheel by hand, to spinning stock works.

 

[Bernhard J.]

Usually the post can be turned so pins go vertical.

This is, unfortunately, not possible in this kind of movement, because the posts are rivetted to the front plate and at least in my case even slightly squared where they are in the plate. So, no chance to rotate them, even if they were loosened a bit. By the way the holes in all posts for the pins are oriented horizontally. The pins are stuck through the holes from the outside and the thought of the horizontal design presumably was that this makes it most easy to grab and pull the pins. If they are not a too tight fit, at least.

The spring wire is functionally perfect like this and definitely not too elastic. I tend to come to the conclusion that nothing else had been used upon designing the movement. But perhaps someone knows better?

Now I have a possibly ignorant question of an amateur. I wonder whether it would not make most sense to use grease rather than oil for the arbors of the most wheels, except of the escape and the anchor? Please correct me and call it rubbish, if it is so . My thought is, that grease will not to such extend as oil tend to spread out and rather remain where it is.

Over the weekend I will clean the movement, correct what might need to be corrected (in the moment I see a slightly bent post of the motionwork only), reassemble it, lubricate it, set it up again, and then wait for the new 24 hour wheel. The latter is no hurry and I will manage to survive some time without the date wheel showing the correct date.

By the way, I have decided to leave the dial as it is, at least for the time being. Silvering may then be the option of some other owner after I have gone to the watch and clock heaven

Cheers, Bernhard

 

[Bernhard J.]

Second, I wonder about the spring below, which I thought non-original. But I begann troubling now. The springs force needs to be very feeble for the striking rack lever to work properly. I do not think that anything other than a thin wire spring, as seen in the photo, would work properly. What do you think? If another kind of spring would have been fitted originally, how would this look like?

I looked around a little bit and it seems that one would expect the rack spring to have a so-called "comma foot" made of brass, wherein the spring wire is fixed. I will make one, fix a spring wire in it, and figure out, how the original spring wire might have been arranged/bent.

P.S.: lazy as I am, I ordered a new old stock comma foot and brass spring wire (provided separately) in the UK via Ebay. Several vendors there offer rack springs, not a single offer from an EU country. Great, shipping and duties will be about double the price of the actual good, not to speak of the slow processing by parcel services and customs. Never mind, the new rack spring can be fitted later, together with the new 24 hour wheel. Until then the instant spring will do service.

 

[Bernhard J.]

Today I disassembled the movement, cleaned all components, attended to various detail flaws, mainly bent posts and levers, made a new pendulum spring, tested the movement. Everything is fine, I believe that the time train will run with a significantly lighter weight. I tested by applying force to the barrel by hand and this was sufficient over minutes (I tested for time keeping with the new pendulum spring for a couple of minutes using a chronograph).

One interesting detail is that all holes for pins appear to be tapered, see example photos. I do not believe that this was done by a clockmaker upon servicing at a later juncture.

Everything seems to be original except the rack spring, no mystery holes, no repositioning of components. In the following a few impressions of the dissassembly and the finished movement. One hint, perhaps, it makes most sense to remove the plate with the pillars. The other way around at least some wheels will probably topple around and let you try out which belong where.

 

[Bernhard J.]

With respect to the rack spring I would appreciate the thoughts of the experts.

First, I have begun wondering whether the spring arrangement with just a wire might be original like that. Because in the book about English country clocks a very similar movement is shown, where the rack spring is mounted just as a wire with a loop for the screw as well. However, this might be non-orginal in this photo also and the text does not put light on this. Thus, what do you think, should I make a new rack spring with a "comma foot" of brass, or should I stay with the simple wire with loop for the screw (perhaps differently formed, see below)?

Second, normally I would bend the rack spring to a form similar to that shown below, i.e. a tall "U" form. However, a pillar end with pin is in the way. What form of the spring would you suggest? Perhaps a wide "U" around the pillar and pin? An alternative would be to form it as an upsidedown "U" going over the winding arbor. There would be sufficient place. What would you think would be rather original? Or just leave the present form (it works fine)?

Cheers, Bernhard

 

[Bernhard J.]

So, I have now received the "new" seconds hand and the comma foot (nos) and brass wire for the rack spring, but have decided to nevertheless leave these parts as found. They are part of the clocks history and since these two issues (aside the the wrong 24 hour wheel) seem to be the only non-original parts (if so, I am not 100% sure), that is OK for me. I will put the "new" parts in a small bag and place this in the case. Any other person to which this clock might be passed at some future juncture can then decide to use them or also leave this as it is.

In summary, the only new part will be the new 24 hour wheel, which will replace the 24 hour wheel as found having a wrong number of teeth. So this will continue to be non original, but then with correct technical function.

I regard the new made pendulum spring as a service "consumable".

I really liked taking the movement apart and admiring the nice workmanship. And that the wrong 24 hour wheel was the only real issue with need of attention (aside the pendulum spring).

As soon as the new wheel is obtained, I will put the clock in service again and try out with what minimum weights of the weights the trains will consistently run. And then start looking out for appropriate weights of contemporary style. If anyone has contemporary (ca. 1760) lead weights with brass mantle, I would be happy to get in touch.

 

[Bernhard J.]

Ha, I now have found a source for brass cased weights as they would seem to be correct for this clock. This source is in .... the UK, where else (I love Brexit *irony*)

https://www.johnwardle.co.uk/longcase-un-polished-brass-weight-case-only-79-p.asp

These look nice and I have separately ordered 10 kg of lead bars (costs about 50.--, including shipping, withhin Germany). I can now match the amount of lead to the needs of the movement trains such that the trains work properly, but the load is as low as possible. I will start with a base mass and melt such base amount into the cases. And can then continue to add lead by throwing more lead into the cases, until the needed weight is achieved.

What would you suggest for the going train and the chiming train as masses to beginn with?

Cheers, Bernhard