Servicing my Hamilton 992B Railway Special - Model 2 - 1947 - Pt 2

Amplitude is the arc from rest position (when the watch is not running) to peak amplitude. The 330 degrees you are referring to is peak to peak arc (so 165 degrees amplitude). Check this out.

There are lot of concepts to learn about watchmaking. I'm currently practicing on an inexpensive pocket watch and Chris (and Al too) have been a terrific help to steer me in the right direction. Set your time machine to 44 degrees lift angle to see if you can get 180 amplitude when the balance wheel makes a full turn on each cycle on the video.

The "slo-mo" video capability of smart phones is a terrific tool to determine lift-angle.

I saw the question around 4 am on my phone this morning when I couldn't sleep, and before anyone had replied, so too late Jim! I didn't watch the video, but I didn't need to as I knew what it was going to show, as it's a common misunderstanding of the fundamentals of what amplitude is.

I've mentioned several times before in my posts here that the amplitude you see displayed on the screen of your timing machine is an average - there is an amplitude in each direction, and the amplitude displayed is the average of those two amplitudes. I guess until you get to a point where you don't have a proper lift angle and are forced to actually look at what's going on do you start to ask these questions, so I suppose me saying that what's displayed is an average likely seems rather obscure until you are face to face with this problem.

If you ever get to the level of adjusting pallet fork jewels, and have a timing machine with a scope function, you can display the 2 amplitudes separately and see how close they are. Ideally you want them to be equal, and this can be affected both by beat error and how those jewels are adjusted in and out in relation to the escape wheel teeth.

The common basic understanding of beat error is that it's the difference between the "tick" and the "tock", but usually people stop there thinking about it, and only know it should be close to zero. Well if you really give a little extra thought to how it is that the tick and tock can be different, you'll pretty quickly realize it's because the amplitude in each direction is different, therefore the time for travel of the balance is different - these ideas are all related to each other and have real practical implications when troubleshooting (for example when you encounter rebanking).

Cheers, Al

Hi Jim,

If the lift angle is right, the balance amplitude is low. Can't recall if you replaced the mainspring, so if this has a blued steel spring in it still, I would replace that with a white alloy spring as the blued steel springs are often very set.

So what checks/adjustments of the balance spring have you done so far? I typically do my initial checks and adjustments right after cleaning, so I lubricate the balance jewels and mount the balance on the main plate - this allows me a clear view all around the balance spring:



I check to make sure that the coils of the spring are all concentric, evenly spaced, that the spring is flat, and that the portion of the spring that goes through the regulating pins is centered between those pins. In a watch with the ETACHRON system, I also set the distance between the regulating pins. These are the basic adjustments I do on every watch, and all this must be done before proceeding. From there the timing machine guides my next steps.

I then look at the readings for any really off measurements, so make sure the amplitudes all look as they should, and I don't have any major differences between the 4 vertical measurements, or between the 2 horizontal. I then look at the timing numbers for any wild readings, but also to note the difference between the average of the 2 horizontal readings, and the average of the 4 vertical readings. If the averages are significantly different, then adjustments to the regulating pins may be required.

Generally speaking, if the average of the vertical readings is slower than the horizontal, as in your case, then the pins are too far apart. If the average of the vertical is faster than the horizontal, the pins are too close together.

Now when adjusting the pins, you need to keep them parallel. Not having the parallel is a common source of dial up to dial down timing errors. This is an extreme example of what I mean from a Tudor I'm working on that was a complete mess - regulating pins as the watch was received:



And after adjusting:



As a rule of thumb I typically want the pins to be spaced so that there is about the thickness of the balance spring clearance on either side, and then adjust from there based on results again.

Cheers, Al

It's expected the lying positions (Dial-up and Dial-down) have better amplitude than the pendant positions because the latter have higher pivot friction.

It's good that lying positions are close in amplitude. What is the amplitude at full wind? Like Al mentioned the values are low. It should be above 270 degrees. In your previous post you mention 270-290 with the 40 degree lift angle.

I would go back and true the wheel again - still has a wobble that is too much IMO.

On the poising, just a few things to mention:

As Chris mentioned, you need to level the tool with the bubble level on the cleaned jaws. I clean the jaws with a piece of pith wood dipped in alcohol, and then blow them dry using your hand blower. If you do the actual poising with your elbows sitting on the work bench, do the levelling with your elbows on the work bench also - if you don't even with a sturdy bench you can affect the level just by resting your arms on the bench.

Hard to see in the video, but my impression is that you have the jaws too close together, and may be touching the radius of the staff - the balance needs to be riding completely on the straight portion of the pivots only.

Going by the video, you need to give the balance more time to settle in your poise checks. You tended to jump right back on the blower when the balance was still doing a bit of a wobble. I have seen balances appear to nearly stop, but under magnification they are doing a little wobble, and sometimes 10 or 15 seconds go by of this, then they start to move again. Poising requires a lot of patience, and doing several tests to confirm that if there is a heavy spot, it stops in the same place each time. Only when it stops in random positions over several tests is the balance in poise.

Note that dynamic poising can often introduce an error in the static poise. For me personally, I probably would have gone right to dynamic poising on this one - I typically only static poise right after a staff change, if the error is very large, or if the balance is missing a screw or something like that. If someone has already done dynamic poising, you may be taking out the adjustment they have already made by doing a static poise. Nothing wrong with that, but it may end up in more work in the end.

Whenever possible adding weight to the balance is preferable to removing it by cutting screw or the balance itself. Of course with a smooth balance you don't have much choice but to remove material directly from the balance.

Cheers, Al