Asymmetric Pendulum Swing on Long Duration Vienna

[Dr. Jon]

I am repairing a long duration Vienna. By my count, based on drum and gears, it looks to be about 50 days. I now have it running but I am having to set the beat adjust far off center and more curiously when I set it up sting with the pendulum at the center the swing is asymmetric, not a lot, its a low amplitude clock but it is easily noticeable . When I re-adjust the case for a symmetric swing the pendulum comes to rest, when it stops, off center.

It is very sensitive to beat error.

The pendulum has a wood rod with a brass plate over its slot. The bob is lens shaped and filled with small bits of lead.

 

[bruce linde]

can you post a video of the escapement in action so we can see what you’re seeing?

 

[shutterbug]

Those gauges at the bottom of the case are often not centered. See if you can eyeball it at case center. Also check your movement - centered in the case and level at the mounting board.

 

[Ralph]

Are you sure the crutch has not moved on it’s arbor. Originally, I would have expected the clockmaker to position the crutch relative to the anchor, so with the movement mounted in a plumb case, the beat adjustment would be in the center or very near center.

Over the years, someone may have moved the crutch. On a Vienna, they are usually pretty tight and not intended to be moved, but….,

Do not trust the beat plate center as being accurately placed, unless you verify it.

Regards, Ralph

 

[Dr. Jon]

Difficult at the moment, since it is now running and set to swing symmetricly. Not much to see, the pendulum swings further to the right, locking the entry tooth when the pendulum rests at center.

 

[Dr. Jon]

The crutch is long and complex. I have moved it several times trying to get the beat right with the peg near the center. I have not been able to do this accurately. It ends up either too far left or right.

It seems to me that with the pendulum hanging from a suspension spring case and scale inaccuracy does not matter except the the movement would be tilted. The tilt to the movement should come out when the beat is set.

I bought the clock recently sold as a 20th century reproduction. Here are some photos of the movement, which I beleive is old.

First this is movement and pendulum mount.

I took this with the clock lying on the floor so the pendulum is not hanging. I have not seen teh wedge holding arrangement before, but I am more a watch guy.

This is the beat setting as received and it still the best place for the clock to run. The clock shows seconds on 72 ticks per minute (36 tooth escape wheel)

The screws are chewed up and I have been re-finishing and blueing them as I go.

One of the limiter screws has ben filed to move the range of motion. The marking is "1804"

Here is a side view. I have moved the anchor to have it engage closer to the center.

 

[R. Croswell]

Difficult at the moment, since it is now running and set to swing symmetricly. Not much to see, the pendulum swings further to the right, locking the entry tooth when the pendulum rests at center.

I would begin by setting the case plumb with the floor (base of the case level). Adjust the beat adjuster, or crutch until it is in beat, remove weights and allowe the pendulum to come to rest. It IS now on center. Remove the beat scale and reattach it to indicate center.

If the pendulum amplitude is less than normal it will be sensitive to being out of beat. Friction and wear are the main cause of low pendulum amplitude.

RC

 

[Dr. Jon]

I have no idea what normal amplitude is for this clock but is it very similar to a 30 day Dachluhr running happily now. it is stable at 1.5 of the 2 marks on the scale.

I begin each session with the pendulum still and set it to the center of the scale, as I wrote. This should be the equivalent of leveling the case and moving the scale to line up with the stationary pendulum.

My physics background leads me to expect that a pendulum should swing symmetricly about its rest point and this one it not doing this. I cannot imagine how an escapement or a small rotation of the movement or case can make this occur. I welcome ideas.


Each time I get it running the swing goes further to the right (facing the clock). The anchor is very firmly attached to the crutch so it is not moving differently from the rest of the crutch. I find its amplitude picks up a bit when I the shift the case to make the swing symmetrical.

My idea is that it is the pendulum bob may be the issue.

This shot also shows the amplitude scale.

The top of the pendulum has enough gap that small bit is lead shot and other bits fall out so the weight distribution in the bob is not stable. It is not that a few bits fell out, but rather that most of the bob weight is very small bits of lead. I doubt that these move when the clock is running normally, but ithe "pile" may be offset enough that the pendulum center of mass is not on the centerline of the rod. The pendulum does not look off when still but this may be the issue.

The pendulum moment is the sum of the weight elements time the square of their distances from the pivot point. It may be that this second moment does not line up with the first moment. It happens with watches and tires; that static balance is not the same a dynamic balance.

If so will redistributing this weight make a difference and get some more margin for operation?

 

[R. Croswell]

In this case, remove the movement, make sure the case is level ( because that is how it is supposed to be. If the shot in the pendulum is slushing around you are chasing a moving target. I wonder if the shot is original or something someone added? I think I would look for some innovative way to keep it in place, or Remove it altogether and replace it with a solid piece that can be fixed in place. With the case level, and the pendulum at rest, with a evenly balanced pendulum, it should point to the zero on the beat strip. Your physics is correct, it will swing equally both sides of zero when swung. ( unless the shot shifts) 1 1/2 on the scale should be good.

RC

 

[Dr. Jon]

I think I know what is going on. It's another example of my mantra "It's the pendulum, stupid!'

The weight is assymetric enough that the pendulum rod is not quite vertical at rest. When I rotate the case to mael the swing symmetric about the scale center I am centering it.

 

[Ralph]

Are you going to show off the case?

Ralph

 

[Peter John]

The lead bits are probably not original. They were added to make the clock think the pendulum is longer. I would remove them and try things without them.

 

[bruce linde]

is it possible it's getting more impulse on one side of the verge? have you double-checked angles?

i would also swap in a new suspension spring just to see if that affects anything... easy enough.

 

[Dr. Jon]

Hard to beleive asymmetric impulse causes the asymmetry. Once the pendulum motions i stable it is very nearly symmetric. I have a Sattler clock which has a Graham like dead beat but only impulses on on side. Its swing is symmetric, albeit it may be off but too little to see.

I don't like current suspension spring for several reasons but making a new one is a major nuisance and teh clock seems to be working well.

The lead bits seem very old. There are very corroded and very small much smaller than BB's they are ababout 1 mm in diameter. Either they have been in place a long time or were very old when put in.

BTW here is the case

the top bonnet slides off and also has a hinged front.

The bottom front glass lifts out. I am concerned it may fall out and am thinking of drilling two small holes to pin it.

I bought it as a recent reproduction so these is not an originality to maintain, but I think this may older.

I love any ideas on who made it and when. The screws show a lot of service or one person who was excruciatingly inept. All fasteners were very tight and all the screw slots are marred. I have been refinishing as I go.

I don't have photos of the dial but its very nice. Its a two piece porcelain with nice rim. It has a bronze or dark brass fitting over the winding hole. This is secured from the read by a sheet metal clamp that goes into a slot in the fitting. It has no signature.

 

[Ralph]

An original movement would probably have had a one piece enamel dial, or possibly a milk glass or metal dial. A metal dial might have had engine turning. The dials would have been surrounded with a cast bezel, maybe engined turned or a finely cast pattern.

Lovely case.

Regards, Ralph

 

[bruce linde]

I don't like current suspension spring for several reasons but making a new one is a major nuisance and teh clock seems to be working well.

just trying to help... the suspension spring looks wonky to me, you already have issues with it, and the clock kind of isn't exactly 'working well'... it would at 30 seconds to swap in a generic fusee or banjo clock suspension spring and rule out the existing spring as the cause of the problem.... it doesn't even have to be the right length... just a test.

nice clock... and certainly could a space for it on one of my walls... but after reviewing the photos it looks to me like the pendulum rod has been replaced... original hardware, but methinks the original rod would have been a little wider (note the fit at the top of the bob) and thinner (front to back).

speaking of the rod/bob fit, it's possible the bob is rocking/moving a bit because of the extra space left/right where the rod comes into the top of it... or at least not perfectly centered.

 

[Allan Wolff]

speaking of the rod/bob fit, it's possible the bob is rocking/moving a bit because of the extra space left/right where the rod comes into the top of it... or at least not perfectly centered.

I was thinking the same thing. If the bob is moving on the pendulum stick just a little bit, it can make a big difference in the beat setting.

 

[Willie X]

Off center adjuster, wonkie suspension spring, replacement stick, loose stuff in the pendulum, loose pendulum bob.

That sounds like a good to do list for starters.

For new things, does the escapement have equal drops and minimal but definite locks? How heavy are the weights? Do you see any wear in the movement, bent pivots, etc.

BTW, that wonkie suspension spring looks to thick but it may just be the photo ...

Willie X

 

[Dr. Jon]

I'll get some better views of the suspension spring. It's attachment at both ends is my issue.

I have no clue when this clock was made. One reason for posting was to learn this.

Because of that I doubt that any parts have been replaced in the sense that an old clock had key parts swapped out. I suspect this was made up from old parts. I suspect the brass plate on the pendulum rod is repurposed because it has J W engraved and partially cut by the slot. These are my initials so perhaps I am in the hands of higher power on this.

My temptation is not to mess with this pendulum. but replace it completely with a grid iron and perhaps go to a knife edge, especially when I become certain this is not a an old clock. Does anyone make parts for this?

I did replace the suspension on my 30 day Biedermeir with a thin piece of shim steel. That was simple. As of now the "new" clock is running.

It mounts on brass brackets with stops to I can remove and replace it very accurately.

 

[bruce linde]

pretty sure it would have had a wood pendulum rod... or possibly a flat silver metal one. i have a vr with metal pendulum rod, and it needs additional weight in the bob (thinking about the stuff dining out of yours)

 

[Dr. Jon]

1) I took a look and some photos pf the suspension spring. It attachements are easy to remove and move to another suspension but I do nto see any point. The steel looks fine and no worse than what I woudl be able to make. It has a touch of spot rust but no wrinkles.
2) I removed the pendulum and shook it enough to redistribute the lead bits. It now swings symeteicaly about is static center.

I have done same watch lever adjustment but am not confident about tehse clocks so I followed the tech tip and resurface the impulse faces to about the same angles as received. For that I used a burnishing file and them diamond slops up to 3000, the sand paper to 8000 and them a smooth burnish. I also did this on the locking faces. I then used 9415 to lubricate.

The clcok seems to be very close to running at corrrect time.

 

[bruce linde]

1) I took a look and some photos pf the suspension spring. It attachements are easy to remove and move to another suspension but I do nto see any point. The steel looks fine and no worse than what I woudl be able to make. It has a touch of spot rust but no wrinkles.
2) I removed the pendulum and shook it enough to redistribute the lead bits. It now swings symeteicaly about is static center.

I have done same watch lever adjustment but am not confident about tehse clocks so I followed the tech tip and resurface the impulse faces to about the same angles as received. For that I used a burnishing file and them diamond slops up to 3000, the sand paper to 8000 and them a smooth burnish. I also did this on the locking faces. I then used 9415 to lubricate.

The clcok seems to be very close to running at corrrect time.

glad it's better...

i wasn't suggesting rebuilding a new suspension spring with the old attachments... i was suggesting swapping in one of these just to confirm the issue wasn't suspension-spring related (i keep several on hand for my various banjo clocks and vr's... and it would have required only unscrewing and rescrewing one little screw):



in my experience, the fact that the bob has that much play where the rod enters and could move means that it one day probably will. my vr's with wood pendulums are wide and snug-fit where they enter the bobs, and narrow a bit to the width of the top pieces where the bottoms of the suspension springs screw on. i personally favor the more typical flatter vr (and black) pendulum rods, but if it works for you it's your clock.... enjoy!

 

[Dr. Jon]

I did have a concern in regard to the suspension spring. The movement has limiters. In several efforts toget the clock going the beat position so so far out the the crutch was hitting the stop and and causing the pendulum to rotate about its length.

I had thought this was due to the suspension spring.

At this stage I still had not been able to get the clock to run more than a few minutes. As it was stopping it would occasionally spontaneously restart and run for a few more cycles. I found this mystifying until I realized the the pendulum twist was moving the crutch enough to get off a few more ticks. This in turn was the result of the pendulum crutch hittin the limiter and the pin out enough to induce a twist.

The pendulum motion is very slow and low amplitude. The bon even with the over size opeing doews not have much play and the motion is not enough to cause any bob movement.

 

[Willie X]

The crutch should not come close to the banking pins unless something hits the pendulum. Willie X

 

[bruce linde]

if when in beat the crutch is hitting one of the banking pins, the verge has moved on its arbor. the safety margin on either side of the swinging crutch to the banking pins should be the same.

we still have not seen the escapement in action... can you post video? upload to youtube and then embed here?

 

[Willie X]

Another fine point.

The crutch drive pin should contact the wear plate only, with about .002" to .004" side clerance. The pin should have no contact at all with the wooden stick. Willie X

 

[bruce linde]

Another fine point.

The crutch drive pin should contact the wear plate only, with about .002" to .004" side clerance. The pin should have no contact at all with the wooden stick. Willie X

and... if the crutch plate slot is worn, too much clearance/slop will rob power... and allow the crutch pin to bounce around a little... which could also contribute to pendulum swing oddities or inconsistencies.

 

[Dr. Jon]

As I wrote this twist only occured when I had the crutch mis-aligned such that it hit the limiter as it unlocked. I was able to reposition, re-phase the such that the crutch now stays well clear of the limiter. Once the pendulum settles, it does not twist. As far as I can tell there is no wear on the clutch plate.

I had mentioned this to help someone seeing this odd behavior.