An Escapement Problem
Escapements can be frustrating, especially when they have been “adjusted” by a wanna-be clockmaker. I recently worked on a mechanism that has been on the shelf for over a year now – one that I had committed to repair for a customer. He indicated that the mechanism had always had a problem running – he thought perhaps the problem was “in the escapement”.
Here is the list of problems I found on initial inspection:
-Anchor, arbor and crutch did not come from same clock.
-Poor solder job on crutch hub
-Crutch hub fit poorly on arbor – so someone crushed the crutch hub to make it fit tightly on arbor
-Anchor drilled from back – creating a tapered hole such that the anchor could slide farther onto the square on the arbor – and the square on arbor had been filed apparently in an effort to make the anchor fit – overall poor fit and finish, as well as poor alignment of anchor on escape wheel.
-Washer between the pin that holds the anchor on the arbor and the anchor - trying to hold the anchor firmly in place on the arbor
-Pallet faces rounded and grooved from filing
-Balance bridge badly bushed, bushing offset from original location, and there were brass shards in the pivot hole.
-Back pivot on anchor arbor bent
-Front pivot for anchor arbor bushed off center, and new pivot in arbor oversized, not centered in arbor, and too long
-Pulley pivot screw badly grooved and rough on wear surface
-Grease and particulates in teeth of motion works gears
-All gears poorly cleaned – deposits on teeth
-Pallet faces not perpendicular to arms – tilted so the anchor would be pushing the anchor against arbor shoulder
-Pallets set too far apart – in recoil mode.
-Pivots in train look to be sanded, not sure how flat they were, definitely not burnished.
Sort of hard telling just which problem would keep it from running – I can see 4 or 5 that would be sufficient.
I then set to work trying to get the escapement to operate in dead-beat mode.
First thing I did was to regrind the pallets’ impulse faces – making the impulse faces “square” to the arms, flattening the rounded pallet faces, and removing the file marks…
Next I pulled one of the pins locating the anchor arbor bridge – such that I could leave out the screw associated with the removed pin. I was then able to manually rotate the bridge up and down to adjust the depthing of the anchor.
I then brought the pallets together a bit (roughly 0.13 mm). By adjusting the depthing (up roughly 1.5 mm) I was able to get the escapement to run in half dead-beat mode – the escape wheel teeth fell on the impulse face on one pallet, and on the dead-beat face on the other. If I brought the pallets any closer together the pallets fouled the escape-wheel teeth.
Next I changed the angle of the pallet that was in dead-beat – flattening the pallet angle. And then steepened the angle on the pallet that was in recoil mode. Whilst it only took perhaps 20 seconds to type these two sentences, in truth this was a trial and error effort – fortunately (amazingly actually), by making small changes, and then adjusting the depthing I was able to get both pallets in dead-beat mode, and then adjust the pallet faces further while working with the depthing until I achieved equal drop and lock.
In the end the anchor needed to come up about a millimeter – but, looking closer at the anchor arbor bridge – it had been bent down about a millimeter – so reversing the previous “fix” brought it back into line.
I then went through the rest of the mechanism. Found that both pivots on the motion works idler gear were bent, as was one of the pivots on the escape wheel. A couple of the pivots were in decent shape, the rest pretty rough.
When all done I shot the pictures in this gallery – showing how the mechanism turned out. I am especially proud of the short movie of the escapement, ticking away happily on a 1.5 pound weight (compared with the 2.5 pound weight that came with the clock). The pictures clearly show an excellent example of a Viennese mechanism – the Geneva Stop is very rare in Viennese pieces – very gratifying to be able to bring back to life such a great little mechanism!
While I was working on this mechanism a gent in Australia – Paul - sent me an e-mail asking for guidance in working on a mechanism he had just bought – here is his story:
“The reason that I am contacting you is of course clock related. I am a jeweler by trade and in the last few years I have been helping friends and family with minor clock repairs and of course this has grown into a sort of major part of my life.
In these few years I have tackled all sorts of clock repairs and restorations and I have of course I accumulated many clocks of my own. Which leads me to my most recent purchase - a Vienna Regulator. After looking on the net I found a close picture on Wiener Uhren's site which of course led me to you.
The clock itself is a timepiece, single weight, with a white porcelain face, inner and outer brass rings, with a matching porcelain disc in the bob bearing the makers name Wilh Kollmer.
The movement is in great condition it has no stamped numbers or identification marks, it is without any great wear and the train spins well and clean. The only problem is that I can't get it working for more than 5 minutes. I think it is a 30 day duration movement as when the escapement is removed the weight takes 25 minutes to finish the cable. The movement is very similar to VR-332 on your website. The weight is 1.9 kg and the pendulum is195 grams although the pendulum has been shortened at the top by 10mm allowing for a shorter suspension spring of about 5mm. I think the spring may be to stiff.
There is no maintaining power on the barrel and the winding drum surface is smooth, no cable grooves.
The thing that worries me the most is the escapement. It is a single piece, V shape Graham which has had some major pallet adjustment. I don't think that it is original.”
This leads me to my question, is it possible to maybe replace the escapement with an original or to manufacture a substitute to the original specs?
My response included guiding Paul to a number of tid bits on the http://www.SNClocks.com site – discussing donuts on winding drums, bushing Vienna Regulator mechanisms, preparing hard Arkansas stones for working pivots, making burnishers, my gear-holding techniques, and finally my techniques for grinding pallets, and adjusting pallets. I suggested that he work with the existing anchor and see if he could make it work.
A couple of days later I got this e-mail from Paul:
“Great day today. I left the workshop at 6 pm and the clock had been running for an hour without stopping. The lock is very small about 0.3 or 0.4 mm. I have flattened the pallets as per your pictures and closed them in; thickness is all the same at about 0.9mm. I refaced the impulse pallets; don’t know about the exact angle, though it looks about 45 deg. But both are the same and lapped flat - almost as good as your reflection picture. Very difficult to obtain 66 deg without the right stuff.
The main adjustment was to bend the pallets downwards to match the radius of the rest of the anchor. This seemed to be the biggest help. Whatever I did it seemed to work.
I tried to increase the lock by closing the pallets more but the escape teeth started to grab.
I have also closely checked the pallet arbor hole in the bridge, it has been bushed, the pivot is 0.5mm and I can only fit a 0.6mm drill bit in the hole, so I think it is ok.
I am really happy to see it running as of course I brought it as not working.”
But, then, the next morning Paul wrote… “It ran all night which was great. But a bit hit and miss today. The lock is right on the corner of impulse and dead-beat face, missed a couple of times. I am going to try and get the 66 deg, will it help to reduce the thickness - is 0.9mm to thick? The tooth pitch is 2.4 mm.”
I suggested to Paul that the pallets were not too thick and should not be reduced in thickness.
Then, a couple of days later Paul sent: “I have had a busy week, but today I had time to play with my clock. Had great results, got the impulse faces to 66 deg. I made up a jig for lapping; I think it is pretty close. The drop was way out - lowering the arbor did the trick, really happy! I gave it a push and away she went, doesn't even look close to stopping. One thing, the locks are different, entrance is slight about 0.2 mm, exit is double that. I played with the beat but only for a short time. Could the entire crutch be misaligned with the pendulum through the centre line?”
To which I responded:
“Unequal lock typically equals a slight depthing problem, or a problem with the pallet angles. Regarding misalignment - when you think about it, if there was a slight misalignment of the anchor and the escape wheel – well, you would just move the anchor over a bit relative to the crutch and all would be ok.
I do not typically sweat a slight difference in lock, especially if the mechanism is running strong and has good over-swing. In fact, it is not uncommon to find that the locks are not equal on mechanisms that have no apparent evidence of having been worked on before.”
Paul spent a couple more days working on the mechanism, and just sent me the following:
“I must say that in the last couple of weeks I have learnt so much, which I thank you for. The movement is finished, pivots burnished, plates and wheels polished, screws and hands blued, looks magnificent! The clock has now been running for 7 days in the case on the wall - I think this clock owes me a few favors!”
He then sent me a couple of shots of the pallets after doing all the work. They look great - as you can see in the last two pictures in this gallery.
Read MoreHere is the list of problems I found on initial inspection:
-Anchor, arbor and crutch did not come from same clock.
-Poor solder job on crutch hub
-Crutch hub fit poorly on arbor – so someone crushed the crutch hub to make it fit tightly on arbor
-Anchor drilled from back – creating a tapered hole such that the anchor could slide farther onto the square on the arbor – and the square on arbor had been filed apparently in an effort to make the anchor fit – overall poor fit and finish, as well as poor alignment of anchor on escape wheel.
-Washer between the pin that holds the anchor on the arbor and the anchor - trying to hold the anchor firmly in place on the arbor
-Pallet faces rounded and grooved from filing
-Balance bridge badly bushed, bushing offset from original location, and there were brass shards in the pivot hole.
-Back pivot on anchor arbor bent
-Front pivot for anchor arbor bushed off center, and new pivot in arbor oversized, not centered in arbor, and too long
-Pulley pivot screw badly grooved and rough on wear surface
-Grease and particulates in teeth of motion works gears
-All gears poorly cleaned – deposits on teeth
-Pallet faces not perpendicular to arms – tilted so the anchor would be pushing the anchor against arbor shoulder
-Pallets set too far apart – in recoil mode.
-Pivots in train look to be sanded, not sure how flat they were, definitely not burnished.
Sort of hard telling just which problem would keep it from running – I can see 4 or 5 that would be sufficient.
I then set to work trying to get the escapement to operate in dead-beat mode.
First thing I did was to regrind the pallets’ impulse faces – making the impulse faces “square” to the arms, flattening the rounded pallet faces, and removing the file marks…
Next I pulled one of the pins locating the anchor arbor bridge – such that I could leave out the screw associated with the removed pin. I was then able to manually rotate the bridge up and down to adjust the depthing of the anchor.
I then brought the pallets together a bit (roughly 0.13 mm). By adjusting the depthing (up roughly 1.5 mm) I was able to get the escapement to run in half dead-beat mode – the escape wheel teeth fell on the impulse face on one pallet, and on the dead-beat face on the other. If I brought the pallets any closer together the pallets fouled the escape-wheel teeth.
Next I changed the angle of the pallet that was in dead-beat – flattening the pallet angle. And then steepened the angle on the pallet that was in recoil mode. Whilst it only took perhaps 20 seconds to type these two sentences, in truth this was a trial and error effort – fortunately (amazingly actually), by making small changes, and then adjusting the depthing I was able to get both pallets in dead-beat mode, and then adjust the pallet faces further while working with the depthing until I achieved equal drop and lock.
In the end the anchor needed to come up about a millimeter – but, looking closer at the anchor arbor bridge – it had been bent down about a millimeter – so reversing the previous “fix” brought it back into line.
I then went through the rest of the mechanism. Found that both pivots on the motion works idler gear were bent, as was one of the pivots on the escape wheel. A couple of the pivots were in decent shape, the rest pretty rough.
When all done I shot the pictures in this gallery – showing how the mechanism turned out. I am especially proud of the short movie of the escapement, ticking away happily on a 1.5 pound weight (compared with the 2.5 pound weight that came with the clock). The pictures clearly show an excellent example of a Viennese mechanism – the Geneva Stop is very rare in Viennese pieces – very gratifying to be able to bring back to life such a great little mechanism!
While I was working on this mechanism a gent in Australia – Paul - sent me an e-mail asking for guidance in working on a mechanism he had just bought – here is his story:
“The reason that I am contacting you is of course clock related. I am a jeweler by trade and in the last few years I have been helping friends and family with minor clock repairs and of course this has grown into a sort of major part of my life.
In these few years I have tackled all sorts of clock repairs and restorations and I have of course I accumulated many clocks of my own. Which leads me to my most recent purchase - a Vienna Regulator. After looking on the net I found a close picture on Wiener Uhren's site which of course led me to you.
The clock itself is a timepiece, single weight, with a white porcelain face, inner and outer brass rings, with a matching porcelain disc in the bob bearing the makers name Wilh Kollmer.
The movement is in great condition it has no stamped numbers or identification marks, it is without any great wear and the train spins well and clean. The only problem is that I can't get it working for more than 5 minutes. I think it is a 30 day duration movement as when the escapement is removed the weight takes 25 minutes to finish the cable. The movement is very similar to VR-332 on your website. The weight is 1.9 kg and the pendulum is195 grams although the pendulum has been shortened at the top by 10mm allowing for a shorter suspension spring of about 5mm. I think the spring may be to stiff.
There is no maintaining power on the barrel and the winding drum surface is smooth, no cable grooves.
The thing that worries me the most is the escapement. It is a single piece, V shape Graham which has had some major pallet adjustment. I don't think that it is original.”
This leads me to my question, is it possible to maybe replace the escapement with an original or to manufacture a substitute to the original specs?
My response included guiding Paul to a number of tid bits on the http://www.SNClocks.com site – discussing donuts on winding drums, bushing Vienna Regulator mechanisms, preparing hard Arkansas stones for working pivots, making burnishers, my gear-holding techniques, and finally my techniques for grinding pallets, and adjusting pallets. I suggested that he work with the existing anchor and see if he could make it work.
A couple of days later I got this e-mail from Paul:
“Great day today. I left the workshop at 6 pm and the clock had been running for an hour without stopping. The lock is very small about 0.3 or 0.4 mm. I have flattened the pallets as per your pictures and closed them in; thickness is all the same at about 0.9mm. I refaced the impulse pallets; don’t know about the exact angle, though it looks about 45 deg. But both are the same and lapped flat - almost as good as your reflection picture. Very difficult to obtain 66 deg without the right stuff.
The main adjustment was to bend the pallets downwards to match the radius of the rest of the anchor. This seemed to be the biggest help. Whatever I did it seemed to work.
I tried to increase the lock by closing the pallets more but the escape teeth started to grab.
I have also closely checked the pallet arbor hole in the bridge, it has been bushed, the pivot is 0.5mm and I can only fit a 0.6mm drill bit in the hole, so I think it is ok.
I am really happy to see it running as of course I brought it as not working.”
But, then, the next morning Paul wrote… “It ran all night which was great. But a bit hit and miss today. The lock is right on the corner of impulse and dead-beat face, missed a couple of times. I am going to try and get the 66 deg, will it help to reduce the thickness - is 0.9mm to thick? The tooth pitch is 2.4 mm.”
I suggested to Paul that the pallets were not too thick and should not be reduced in thickness.
Then, a couple of days later Paul sent: “I have had a busy week, but today I had time to play with my clock. Had great results, got the impulse faces to 66 deg. I made up a jig for lapping; I think it is pretty close. The drop was way out - lowering the arbor did the trick, really happy! I gave it a push and away she went, doesn't even look close to stopping. One thing, the locks are different, entrance is slight about 0.2 mm, exit is double that. I played with the beat but only for a short time. Could the entire crutch be misaligned with the pendulum through the centre line?”
To which I responded:
“Unequal lock typically equals a slight depthing problem, or a problem with the pallet angles. Regarding misalignment - when you think about it, if there was a slight misalignment of the anchor and the escape wheel – well, you would just move the anchor over a bit relative to the crutch and all would be ok.
I do not typically sweat a slight difference in lock, especially if the mechanism is running strong and has good over-swing. In fact, it is not uncommon to find that the locks are not equal on mechanisms that have no apparent evidence of having been worked on before.”
Paul spent a couple more days working on the mechanism, and just sent me the following:
“I must say that in the last couple of weeks I have learnt so much, which I thank you for. The movement is finished, pivots burnished, plates and wheels polished, screws and hands blued, looks magnificent! The clock has now been running for 7 days in the case on the wall - I think this clock owes me a few favors!”
He then sent me a couple of shots of the pallets after doing all the work. They look great - as you can see in the last two pictures in this gallery.
