Setting up Bevel Gears on an Elliott 9-tube mechanism

One of my hard and fast rules is that I remove as many screws as possible before putting clock parts in a clock cleaning solution. This has always served me well - it eliminates the risk that a screw will be corroded in place, making it very hard to remove down the road. Of course, I know that I must take careful notes so I know which screws go where. As an aside, this is especially important with earlier clocks, the ones with hand made screws: It is best if each screw goes back in the hole from which it came – visually identical screws oft times are not actually all that identical.

But, in the case of a 9-tube Elliott mechanism that I had in for repair, this hard and fast rule gave me a chance to do a bit more work.

The screws in question are those which adjust the fitment of the bevel gears that drive the pinned drum in a 9 tube mechanism. These screws, as well as the two bevel gears in question, are shown in the first two pictures in this gallery:

So, assuming the shortest duration between strikes reflected the best location for the bevel gears, I decided to go with 1 turn in. <br />
<br />
But, being blessed (or, some would probably say cursed) with an inquisitive mind, I wondered why it was that the mechanism came to a virtual halt on several of the strikes for the 6 quarter turns in trial. My only conclusion was that the teeth on each of the bevel gears had enough variation that they did not always mesh as well as one might hope. So, to test this theory I tried a number of different arrangements for the two bevel gears – removing the upper bevel gear and reinstalling it in a different rotational position. And, low and behold, using the timer as above, I found that there was a set-up that significantly reduced the very long pauses with the screw turned in 6 quarters of a turn. So, using the best gear arrangement, and the best screw position, I ended up with the last point on the above plot – the “1- Final” data point.<br />
<br />
I suppose the bottom line is that one should just not remove the adjusting screws discussed above, or one should at least count the turns to remove them, or precisely measure their position. Hopefully I will remember this bit of wisdom next time I work with bevel gears.<br />
<br />
This experience also reminds me of a visit I made to a local clock shop many many years ago. The shop had a similar mechanism in for repair, and were not having any luck getting the quarter strike train to spin. They had managed to get things adjusted such that, with an additional 30 pounds, the train would spin, but didn’t seem to be able to do much better. As I was looking at the mechanism I noticed a small pile of what looked like brass cuttings below the bevel gears. Clearly they had a bit of a problem with the settings for those gears! And, yes, I checked, there was no such pile of brass below the bevel gears on the Elliott I discussed above. Thank goodness.

5 / 9

So, assuming the shortest duration between strikes reflected the best location for the bevel gears, I decided to go with 1 turn in.

But, being blessed (or, some would probably say cursed) with an inquisitive mind, I wondered why it was that the mechanism came to a virtual halt on several of the strikes for the 6 quarter turns in trial. My only conclusion was that the teeth on each of the bevel gears had enough variation that they did not always mesh as well as one might hope. So, to test this theory I tried a number of different arrangements for the two bevel gears – removing the upper bevel gear and reinstalling it in a different rotational position. And, low and behold, using the timer as above, I found that there was a set-up that significantly reduced the very long pauses with the screw turned in 6 quarters of a turn. So, using the best gear arrangement, and the best screw position, I ended up with the last point on the above plot – the “1- Final” data point.

I suppose the bottom line is that one should just not remove the adjusting screws discussed above, or one should at least count the turns to remove them, or precisely measure their position. Hopefully I will remember this bit of wisdom next time I work with bevel gears.

This experience also reminds me of a visit I made to a local clock shop many many years ago. The shop had a similar mechanism in for repair, and were not having any luck getting the quarter strike train to spin. They had managed to get things adjusted such that, with an additional 30 pounds, the train would spin, but didn’t seem to be able to do much better. As I was looking at the mechanism I noticed a small pile of what looked like brass cuttings below the bevel gears. Clearly they had a bit of a problem with the settings for those gears! And, yes, I checked, there was no such pile of brass below the bevel gears on the Elliott I discussed above. Thank goodness.

Average