1920's Waltham dash clock overhaul

[prewarnut]

    Ok, I need this thread like a hole in the head but I am doing another forum member a favor. I thought I'd show what it takes to do a simple overhaul on a 100 year old auto clock (simple? we'll see). These work like oversized pocket watches and are not well fixed by regular clock repair people unless they understand how to service watches. It also stands to show why every 8-10 years when you take your prized Omega, Rolex, Vacheron, Lange or what-have-you in for servicing it makes the labor rates of our best auto restorers look palatable. It also shows there's more unexpected inside just like some of the engine overhaul threads we have here. Nonetheless this anonymous member is getting the Cadillac treatment (or as Ed would say, the Pierce-Arrow treatment).

   This may get long and laden with photos so if not interested I understand.

 

   Some "do's" and "don'ts":

 

- Don't try to keep winding if it is wound (there's no such thing as an "over-wound" clock or watch) - there's something else jamming it or stealing energy (we'll see that here).

- Don't aggressively shake (same rationale).

- Don't over oil with 3 in 1, motor oil, vegetable oil or really anything without disassembly.

- Don't / never use WD-40. If you do you will be briskly shown the door with multiple 4-letter words and only wish that 7 years of bad luck could be coming your way. This agent will destroy a quart of expensive clock/watch cleaning solution.

 

- Do invest in someone who knows what they are doing - preferably someone who hasn't spent a few hours watching internet videos.

 

OK, here's a '20s Waltham, 8-day, dual barrel lever escapement watch with nickel plates. It is pretty robust. It looks temperature compensated. It has 7 jewels. Not a bad mechanism. Some are a little more decorated but this is not seen inside the dash. The outward appearance is good. No rust. It looks clean. Pivots are dry though and some external matter is visible.

 

 

 

[prewarnut]

Time to disassemble:

A screwdriver us used to loosen the dial feet.

I want to get the balance out of the way early. I find the roller jewel is broken off (ughh). The guard finger on the lever is bent. Also I can easily kick up the exit jewel on the pallet. All of these will need to be addressed.

 

[prewarnut]

Here's why trying to wind, oil, spray won't work. The movement is fully wound and yet the escapement's club tooth should be resting on the exit pallet yet it is hanging in free air (I can't get this photo to upload at present). There's more foreign objects seen on disassembly. Some dried oil is seen at the pivot holes. The winding barrels are over-oiled in attempts to get this running. Balance jewels need to be removed to inspect and clean. A special jeweling tool is needed for this operation. One can see the small size of these (pivot and cap jewels) placed next to a nickel.

More to come...

 

[alsancle]

Fantastic!  Thank you.

[prewarnut]

   I'm back from the morning errands....

   Here's it all disassembled.

   I usually like to take care of fragile things first since I worry parts which are not together may get lost. Also if a pallet jewel falls out it is ridiculously hard to find and re-orient. So, I'm taking care of the loose pallet first. Actually both were loose with age (I almost knocked the other out). To start with I decided to straighten the guard finger with special round plyers which won't score it. This finger's purpose on watches is so if there is a bump or jar the lever doesn't cross midline and when the balance comes back it isn't there to meet it and the balance becomes overbanked. It can't be reset without disassembly. So, I can't ignore this.

  Next it is time to re-secure the pallet jewels. They are shellac'd in place. Shellac is great for fine furniture but we use it here as well. I have 200 lifetimes of shellac flakes - as you'll see a little is needed.

   There is a special holder whereby the anchor holding the pallets is held upside down. A very small piece of shellac is grabbed with tweezers without breaking or dropping and placed on top of the jewel. The whole thing gets heated. The shellac melts towards the heat. Thus, the underside is used as the jewels usually protrude out a little there and cosmetically on watches this side isn't seen. The shellac should be placed on the jewel and it gets drawn off of it, into the recess holding it. Extra can be scraped off. 25 or 30 seconds does it in the flame. The last phot shows the exit jewel done and new shellac awaiting heat on the entrance side.

 

[prewarnut]

   Here's the back of the cover plate which holds back the ratchet wheels for the winding barrels. A lot of dried oil and grease is here requiring hand scrubbing before cleaning. Even after a preliminary clean of the plates I always peg out the holes.

   Here, in the second photo, the center pivot still shows impregnated dirt that comes off on a wooden toothpick or pegwood. This is why a 7 jewel movement will stop when oil dries out or if dirty as the very fine holes are not pristine and there is a huge loss of power. Manually pegging out the holes is helpful.

   The third photo shows clean plates. I don't have a large enough movement holder so these will be held with watch paper as traditionally; these are lint free.

   The last photo shows the application of synthetic grease (sparingly) to areas of the winding apparatus. I am aiming for the square, exposed stem so the winding pinion can slide freely. It is in default setting mode right now. A number of areas are greased including some of the sliding pieces and springs.

   More a little later...

[tom_in_nh]

Fascinating. 

Would you mind telling us what you use for a microscope? And camera?

Keep up the great tutorial. 

Tom

[Mr. Don]

Fascinating!  Tools and techniques! Love it!

 

I have seen the small Brass hand vises before, but never in use. A way to draw heat toward a delicate part! I can use that! Thank you!

[prewarnut]

15 hours ago, tom_in_nh said:

Fascinating. 

Would you mind telling us what you use for a microscope? And camera?

Keep up the great tutorial. 

Tom

Sure, when looking at parts I use a 4x loupe in the right eye for general work. Mine's by Bausch and Lomb but they're made by many companies. When I started at age 15 I could do a lot of this without and it is said that the watch companies hire young staff in their repair departments for that reason. I have a 7X (and even a 10X) to spot check things like trying to rule-out a crack in a jewel or assessing a bend in a hairspring or seeing if I have succeeded in black polishing steel adequately, especially if making a new part. My camera is just my Samsung S21 which will pulls really nice close-ups but generates a lot of memory in image size. Some people use a microscope but I prefer not to as it gets in the way. Trust me filming myself is not the right thing to do. It is like a pilot in the cockpit - I need my hands free and ability to not cast shadows, etc. Some things I don't show here but overall enough steps to give a good idea. Thanks for the interest.

[prewarnut]

12 hours ago, Mr. Don said:

Fascinating!  Tools and techniques! Love it!

 

I have seen the small Brass hand vises before, but never in use. A way to draw heat toward a delicate part! I can use that! Thank you!

Thanks, that will make a reprise soon when I come to the roller jewel - one of the most dreaded operations for me. Talk about self-induced pressure.

[prewarnut]

Good morning, I am probably at hour 4 here. I am spreading the project out over the holiday weekend. I also find there are certain times to do certain operations. For me fine work is only from 7-10 PM, no coffee or alcohol. Obviously last night was not ideal *laugh*. Some of my steps are a little out of my usual order but there doesn't have to be a particular order to all things. Just like auto or boat restoration it is a lot of little sub-assembly or procedures.

Here I have the balance wheel jewels already cleaned and reset and now being screwed down to the front plate. The balance cock was treated the same and reassembled. Note the upper cap jewel is set in a solid gold jewel setting. Fine watches had this and for Waltham to self-impose this on a utilitarian auto clock shows their minimal standards; they were a great company. Their watches from the 1880s are felt to be some of the highest quality and best decorated of any American company. Note I have the regulator dead center. There should be no excuse why this can't be delivered to the customer this way with the timepiece running dead-on accurately. I have two tricks up my sleeve, one Waltham has provided for me.

Edited January 1 by prewarnut (see edit history)

[prewarnut]

Since I have to work on replacing the roller jewel the balance will be further evaluated and cleaned. Often this is left alone and on modern Swiss wristwatches the whole balance assembly is left in place it the ultrasonic (for risk of bending something). But here, like sending the cavalry in, we're going for it all-in. I have the hairspring demounted. It's position needs to be assessed so this can be put in beat later. I usually draw that on paper for reference. I have the balance spinning free in my hand-held calipers to roughly check poise, concentricity and wobble. It actually looks pristine thankfully. No adjustments needed. I will carefully hand polish. I do wonder if it may be missing a screw as many have much more weighted screws on the rims and I won't know until final time rating. The reason to polish is to reduce rust. Rust will alter the weight of the material and also the position of mass and affect timekeeping. While a steel and brass bi-metallic balance was the norm from the 1860s to 1940 the screws would be solid gold on better watches as gold does not tarnish (except perhaps only slightly in alloyed form).

Edited January 1 by prewarnut (see edit history)

[prewarnut]

Ok, so time for the Cadillac treatment. One of those "while-you're in there" moments... I decided to break out my Steiner (Jacot) tool. This is a handheld lathe operated traditionally with a small bow or recoiled spring wound monofilament. The goal is to polish the pivots (axles). Now, since true polishing will reduce the diameter and we don't want that what the operation is is to burnish or work-harden and polish the pivots. The steel is already hardened a bit (not completely glass-hard) and so this can be risky if something goes awry. The pivot size on this balance is 0.11 mm in diameter. To polish it one end is supported on a runner that is 0.10 mm deep leaving a small-exposed portion. The balance is spun and in the opposite direction in my left hand I apply pressure with a burnisher and keep it level on the hardened runner. The burnishers can be steel or sapphire. This is done for 15-20 seconds or what is needed. Sometimes and amazingly one can sometimes feel when it is improved. Technically this is where a microscope could be used but in normal cases (and this movement was not abused and in overall good condition/low miles so to speak) it is not necessary. The over-the-top part of this is I decided to do the same to the escape, 4th and 3rd wheels since they don't run in jeweled bearings on this. Trust me, no one is going to do that normally.

[Mike "Hubbie" Stearns]

I m sure glad we have people like you that knows how to work on things Ike this. You are amazing. Mike

[prewarnut]

    The spring barrels are able to be reinstalled with just a modicum of grease on the ridge point (blue) and synthetic oil to the arbors (red). There is not much friction with these highly polished nickel plates and steel barrels so a little goes a long way.

   Next the gear train is installed. I like to oil from the inside since uncapped bearings have a shoulder, which if oiled externally as everyone seems to do, won't get adequately oiled. Also less oil is present externally. This doesn't have a dust ring but older pocket watches do which makes dust intrusion less possible.

   Note the highly polished screwheads. Steel is usually black polished (it looks invisible unless light is reflecting from it) or heated and blued, both to mitigate rust development.

   Lastly the click, clicksping and coverplate is added. This looks much better tan initially (I need to see if I can find the "before" pic. Truly though showing grime and tarnish is difficult on camera.

[prewarnut]

4 minutes ago, Mike "Hubbie" Stearns said:

I m sure glad we have people like you that knows how to work on things Ike this. You are amazing. Mike

Thank you!

[prewarnut]

Another comment: watches can have 2-3 different greases and three different oils utilized. My oiler in the photo before screwing down the back plate has a clear drop on it as I use a lighter weight for the escape wheel and balance assembly.

[prewarnut]

Ok, so I now need to attack the missing roller jewel. Jewels are of three types. There are hole jewels that act as bearings. There are the anchor or pallet jewels which interlock with the escape wheel and there is the roller jewel that the balance wheel plays off of. In all cases they are highly polished and have a low friction surface. They traditionally were real stones but can be synthetic. Mostly they are rubies but can be sapphires or diamonds. The Brits used diamond endstones in their better watches from the mid 1860s to 1910 or so. In the U.S. some of our railroad grade watches had the same. So, which jewel do I need?

Edited January 1 by prewarnut (see edit history)

[prewarnut]

   So I have a roller jewel feeler gauge but I have found it can be off. I generally "present" the stone to the fork under 7x power and see how it interfaces recognizing I'll need just a slight bit of play as the roller rotates away. I also measure as well with my bench micrometer. Here, it looks like we'll need a 0.50mm wide roller jewel. The length is measured as well but I'm shot-gunning that aspect for now.

[prewarnut]

What follows is the torture. Waltham did at least make this easier in one aspect for me. Usually the jewel is inserted into a round hole. In cross section the jewel is "D" shaped and the flat of the "D" faces outwards. In a round hole it is up to me to make sure it is oriented right, inserted just the right amount, and not tilted in any direction (needs absolute perpendicularity). Here Waltham made the hole "D" shaped thus helping with one aspect of the process. However, the hole is generally about 1/100th of a mm bigger. Any bigger than that and the jewel just falls through. Any smaller and it isn't going to go in anytime soon. Now, on a jewel only 0.75 to 1 or 1.5 mm long, a #2 tweezers is 0.6 mm wide so I'd be covering nearly all the jewel in holding it. Also it will slide through due to the high polish. How do we insert? We have to lick the inside of the tweezers and that will get the jewel to stick upright on one of the two prongs. I then approach from top or bottom. I show top here but I quickly inverted and did from below. Now if I am off by said 1/100th of a mm in getting this lined up not only does it not go in the hole but it slips away never to be found again (if unlucky) , cracks, etc. So basically this took me 8 tries. It was only 7 the last time (probably the Guinness I had this afternoon). What did I say about drinking? Arghh...

Edited January 2 by prewarnut (see edit history)

[prewarnut]

Ok, once done correctly it is turned over and more shellac is placed on top but without disturbing the jewel hanging in the hole upside down with nothing but the grace from above allowing it to be so. This only took me 10 tries. It is then heated to secure.

[prewarnut]

   Then, I had to do this all over again because in my final cleaning the shellac loosened up and the jewel came out (I know what I did wrong). Yes, that's why if this was my day job I'd be eating dog food.

[Laughing Coyote]

All I can say is Wow! This detail work is just amazing. I can't imagine how they could have even made such detailed parts back in the 20's compared to todays high tech tooling. Just incredible.

Edited January 2 by Laughing Coyote (see edit history)

[prewarnut]

It can be an art form in miniature....

A special oil is used for the pallets which has high shock resistance to impact but doesn't migrate like other oils will do. The pallet stone faces are oiled.

Apparently the guard finger which I straightened upright should lean a little. That took me 6 tries to get within spec. I did not photo this agony.

I then mounted the hairspring in my staking set. A special stump with cut-out is used to not squash the roller jewel.

It then was assembled and on just a turn or two of the stem (not shown) it took off running!

There is still more to do. I need to set beat. I need to check locks, I need to check timing in the vertical. I need to check full wind to ensure the springs hold. Amazingly it is running 6 sec slow/day as-is horizontally. Not bad for guess work!

Edited January 2 by prewarnut (see edit history)

[prewarnut]

I probably won't have too much more to add here but can share two more points.

One has to be really careful cleaning dials. Ammonia, as in Windex for example, can take painted numbers off or smear them. It is usually best not to "improve" things. A little trick with Rodico (expensive Swiss silly-putty) is that it can pull a lot of grime off surfaces. A few stains in the dial are noted and much of it is beyond the decorative  "mask" of the dash. You can see how much dirt comes off though. I think I have it 90% better and will then leave it.

[prewarnut]

Lastly one isn't done unless getting the movement on a timegraphing machine. In the old days watches were run in 3-6 positions, in an oven and in cold temps and then at different positions of the wind to assess the rate. Doing this one day each will take a over a week and then after adjustment it is started all over again. The factories would take over a month to release a batch of movements to jewelers. Nowadays machines can pick up the ticking sounds and ascribe the daily rate and also tell if the watch is beating evenly and how much amplitude the balance wheel revolves at - all within seconds. Still it can take a few hours if one wants to finesse that much perfection out of it. This clock is simple in that it will always be mounted upside down in the dash. Yet making sure the upside down and right side up rates are close is important as when the spring winds down these rates can exchange with one another (no one probably cares about the technical graphs on this...). On my photo each tick is a dot on the screen forming a line on simulated graph paper like the paper machines of the '50s or '60s. In the left hand column the lower line of dots lean to the right with the movement 12:00 down meaning it is running nearly 10 seconds fast upside down. The upper column shows it running 10 seconds slow when mounted correctly in a vertical position. So,  they're nearly 20 seconds apart. I can do better, and have adjusted the small screws on the balance to equalize this without touching the regulator on the back. We'll see how the week goes. It isn't a chronometer but I like to see how good one can make it. Now a movement from a SJ Duesenberg or Hispano-Suiza with a second hand is another matter.

Well that's all for this. Thanks for the interest.

[Laughing Coyote]

Just amazing. 

[nsbrassnut]

Thank you for the very educational description of cleaning and repair the Waltham clock.

 

After reading this one, I think that I'll leave the one that I have on the shelf alone. This one has the second hand, and another item which I assume may be a day / night indicator.

 

Perhaps you can provide additional information on the period and model that it may be.

[Luv2Wrench]

Absolutely astonishing!  Makes brain surgery seem simple.

[prewarnut]

On 1/5/2024 at 6:59 PM, nsbrassnut said:

Thank you for the very educational description of cleaning and repair the Waltham clock.

 

After reading this one, I think that I'll leave the one that I have on the shelf alone. This one has the second hand, and another item which I assume may be a day / night indicator.

 

Perhaps you can provide additional information on the period and model that it may be.

This is a similar version of their standard 8 day movement. Per the serial number, using known databases of their production it looks to have been made about 1917 (give/take a year).

[nsbrassnut]

 I found and checked a website with Waltham numbers today and got the same results estimating 1917 for the watch.

 

I'm wondering now if the circle might have a dial underneath that shows roughly how long the watch would last. Such as green for fully wound, yellow part way and the red for needs winding. Its not presently running and I don't want to wind it too much after reading about the work required to repair one above. So, I'll leave it along and on the shelf where it looks good.