Ball Bearing Clock

[Krypto]

Good to hear!  Glad you got all the bugs worked out of that big clock.  I'm sure it's very satisfying to watch it run correctly, especially when you had to work for it.

My moon clock is full assembled as well, minus a bit of extra-credit artwork for the face.  While I've been seeing clocks with moon dials my entire life, I finally learned how to interpret them: full moon is right at the top dead-center.  Handily, we just had a full moon here so it was easy to set the dial correctly and it seems to be working as it should.

Printed the next to the smallest weight shell and used lead ingots with some BB's added as filler.  Ran out of BB's with more room inside the shell, but the clock is running strongly at just 4 pounds 10 ounces compared to Steve's 7.6 pounds.  I don't know why the weight is so much lighter.  I am seeing a small amount of escapement bounce with the right pallet, but from what I read more weight will probably make that worse.  It doesn't seem to be a problem and I'll just run the clock and see how it goes.  I'm already within a minute a day with more tweaking to go.

[Kim]

I'm happy to report that after 4 or 5 days of on / off fiddling, and the reprinting of the long gear #8, the clock is running well and keeping good time. I will do some fine tuning over the next while but the clock is a runner.

 

Tom

Congratulations, Tom!   
That is very exciting!  Well done!

I'm finding that my clock just stops every once in a while.  It will run for over a week, then it just suddenly stops.  The first time it happened (it's happened 3-4 times) I just turned the ball bearing gears, and it started again.  But with subsequent events I've done a better job investigating.  When it just stops like that, there is NO torque on the escape wheel. None whatsoever.  The pendulum will swing back and forth between the teeth of the escape wheel and it just doesn't move.  I have traced it back several wheel sets, and it seems that some of the wheels are rubbing against each other - the faces of the gears are actually in contact, not just the teeth. There is a lot of shake in the gear train, and sometimes, it seems that two sets of gears shift such that the face of one gear comes into contact with the face of a near by gear.  This causes enough friction that it stops.  The wheels down stream from that are free to turn, but since they are getting no torque applied from up stream, the whole thing just stops.

I'm working on devising a plan that will incorporate some spacers in the right place to keep these gear faces from touching, even at the extreme ends of their movement on the arbor.

I think I read about this in Steve's troubleshooting guide, but it is called 'friction between gears,' and that seemed like friction between gear teeth to me.  But now I think it is friction between the faces of gears! A very different thing.  I clearly didn't understand what he was saying when I read it.  It took me re-diagnosing the problem to understand what he said!   

Glad you have it working now, Tom!
Kim

[Kim]

Good to hear!  Glad you got all the bugs worked out of that big clock.  I'm sure it's very satisfying to watch it run correctly, especially when you had to work for it.

My moon clock is full assembled as well, minus a bit of extra-credit artwork for the face.  While I've been seeing clocks with moon dials my entire life, I finally learned how to interpret them: full moon is right at the top dead-center.  Handily, we just had a full moon here so it was easy to set the dial correctly and it seems to be working as it should.

Printed the next to the smallest weight shell and used lead ingots with some BB's added as filler.  Ran out of BB's with more room inside the shell, but the clock is running strongly at just 4 pounds 10 ounces compared to Steve's 7.6 pounds.  I don't know why the weight is so much lighter.  I am seeing a small amount of escapement bounce with the right pallet, but from what I read more weight will probably make that worse.  It doesn't seem to be a problem and I'll just run the clock and see how it goes.  I'm already within a minute a day with more tweaking to go.

Glad it's working so well for you! 

Can you make an adjustment to the pallets to get rid of some of the bounce?  I was able to do that on the ball bearing clock.  But I don't know if Steve built in the same kind of adjustments into the moon clock.

Kim

[wagnmkr]

Kim, the drive train shake was why I reprinted and replaced the long gear #8. The end that connects with the ball wheel is where the problem was. It was quite rough from the printing. When I printed the second one I slowed the printer down a bit so that the 8" tall print didn't shake so much.

I also had your other problem with the gear faces touching and causing friction. I made some spacers out of thin wire and bent them in the shape of a horseshoe. I just sat them over the appropriate axles between the surfaces that were touching. It was meant to be a temporary thing but they seem at home where they are and are doing the job. A drop of sewing machine oil made them very slick.

Cheers

Tom

ETA: The wires are in the black circle...

[Krypto]

Glad it's working so well for you! 

Can you make an adjustment to the pallets to get rid of some of the bounce?  I was able to do that on the ball bearing clock.  But I don't know if Steve built in the same kind of adjustments into the moon clock.

Nah, there's no adjustment for that on this clock.  It's one of those things that unless you're a true clock nerd you would never notice: just a little vibration in the escapement after hitting one of the pallets.  It doesn't seem to really effect the clock in any manner, it just that I won't add a more weight because of it.  Of course, I guess I really don't need more weight as it's been clicking away merrily as it is.  Not sure how I managed to build a clock that uses almost 3 pounds less than Steve's. 

[Kim]

You're just that good! 

Kim

[Kim]

Kim, the drive train shake was why I reprinted and replaced the long gear #8. The end that connects with the ball wheel is where the problem was. It was quite rough from the printing. When I printed the second one I slowed the printer down a bit so that the 8" tall print didn't shake so much.

I also had your other problem with the gear faces touching and causing friction. I made some spacers out of thin wire and bent them in the shape of a horseshoe. I just sat them over the appropriate axles between the surfaces that were touching. It was meant to be a temporary thing but they seem at home where they are and are doing the job. A drop of sewing machine oil made them very slick.

Cheers

Tom

ETA: The wires are in the black circle...

Interesting solution!   Nice and simple 

Kim

[Krypto]

You're just that good! 

Probably just luck, but I would rather be lucky than good any day of the week. 

[wagnmkr]

Kim, it was a matter of seeing what was going on and looking around the shelves for something that would just provide a bit of space between the gears. I didn't want to take the clock apart ... again ... so I tried the bent wire. A bit lazy on my part really, a quick and dirty experiment, but it is still going.

Krypto, these clocks seem to be either instant runners or a constant source for tinkering. Steve does mention that the weights he suggests are just a starting point. A couple of the clocks I have done needed more than he said and one needed less, but not that much less. Your printer must have been printing just perfect when you did those gears. The next one might be totally different.

I have one of his clocks that will run perfect in direct mode on 4 lbs. It has a near perfect beat and keeps very accurate time ... in direct mode. Just the minute I put on a pulley and double the string and the weight, the clock stops and won't run! Nothing else gets changed! So, I let that one run in direct mode!

Cheers

[wagnmkr]

Krypto, I was reading some messages on Steve's site this morning and there was a gentleman that had a moon clock running on 2.5 kilos accurately so that is not too far from what you are using. You are not alone!!

Tom

[Krypto]

Yeah, I remember seeing a moon clock with very little weight, I think less that this one, although the pendulum amplitude was very weak.  I find it hard to navigate for old posts with Steve's new group forum.  Not sure why his provider thinks this setup is new and improved.

Looks like the moon clock is very close on the pendulum bob setting.  Down to just weekly adjustments and it might not need another one.

[wagnmkr]

Krypto, I agree that the clock forum format is not all that easy to get around in. I don't go there that much any more because of it.

I am happy to report that the Roller Ball clock has now been running for a week with no problems. Once I changed the badly printed gear it was off and ticking. It lives in the downstairs coffee room where I spend a lot of time. It a bit noisy but I find that comforting for some reason. It reminds me of an old machine that makes noises from time to time but basically just sits and does it's job.

Cheers

[Krypto]

Hey Kim, if you are still having trouble with your balls dropping, ahhh with your clock, you might want to check out this last post from Steve:

https://www.stevesclocks.com/group/debug-hints/discussion

Looks like you weren't the only one as he's soon adding a throttling mechanism to prevent that.  I didn't check but the extra STL files should soon be uploaded to myminifactory.

[Kim]

Interesting!  I wonder if that was my problem?  I was blaming it on gear faces rubbing against each other, but it certainly could have been this issue.  I'll have to make this addition when he releases it.

I'm still waiting for another one of those 'lockups' to happen.  But it's still running strong!

Kim

[wagnmkr]

I had that issue as well but a little sanding around the opening the ball bearing goes through solved it for me. That is an interesting mod that Steve has made but like Kim, my clock is going strong and keeping excellent time so no need at the moment.

Thanks for the link Krypto,

Tom