Gear relation in clock won't work

iain_downs

I'm modelling a simple clock here
 https://cad.onshape.https//cad.onshape.com/documents/f99aa55da6675b042c4856be/w/23e1ba9aefb1ab2e6d82f886/e/72ff353f0125745c60e45552com/documents/f99aa55da6675b042c4856be/w/23e1ba9aefb1ab2e6d82f886/e/72ff353f0125745c60e45552 .

All the gears are in place and linked together with gear relations, for the main part they seem correct.

The one which is giving me trouble is the reverse minute to hour relationship.  The ration should be 12 to 1.  If I set it to that I get the hour wheel turning at about the same speed as the minute.  If I set it to 1:1 the hour does not move.  If I set it to 1:12, then the hour moves MUCH faster than then minute (12 times?).

I've tried breaking it down and building it back up, but the same thing happens.  I have no idea what could cause this.  Any help appreciated.

Iain

Evan_Reese

The link isn't working for me.

iain_downs

Not sure what happened there.  Does this work?  https://cad.onshape.com/documents/f99aa55da6675b042c4856be/w/23e1ba9aefb1ab2e6d82f886/e/72ff353f0125745c60e45552

Iain

iain_downs

Can I just ping this question in the hope that someone can address it, please?

Evan_Reese

I poked around on this a little bit and wasn't able to get it working, but I didn't have enough time to really dig into it.. If you end up breaking it back down and trying again I'd suggest not treating the mates so literally. You don't have to mate a shaft to a bushing to a gear etc. Consider adding only revolute mates to the gears and getting that working before mating anything else to anything else. Use offsets if needed. Not sure if it's something to do with that or what.

The model is really really awesome though. good work! way more in depth than my pseudo-clock.

S1mon

I spent a little time with a copy of this too, and gave up. If I really needed to solve this, I would definitely try breaking it down (maybe just suppressing parts/mates) and then build it back up. I'd also think about adding some sort of temporary markers (color a tooth, add a hole, something...) to see the rotation better. I found myself doubting the direction that things are moving because of the frame rate and the number of teeth.

I also echo Evan's comments about the number of mates for all the sub-assemblies.

I noticed that there are a few places where the gear teeth look like they are going to crash into other shafts, so you'll need to check your clearances carefully.

iain_downs

Thank you both.  I was wondering if I should make the hour revolute against the frame rather than the bush (with an offset).  A revolute on a revolute with a gear relationship does add some complexity.  I will try those things.

S1mon - The original design is metric and in Model Engineering magazine some years ago.  I found it very difficult to follow the details in the article, so I wanted to redraw it in metric and use the process to understand how it was meant to work.  There are some odd things (for example making the intermediate wheel a little smaller than it should be to avoid interfering with the winding arbor.  I didn't like this, so may end up increasing the teath from 110 to 120 (or whatever) to give more space.  VERY easy to do in OnShape.  Hell with pencil and paper.

So there are still things to understand and sort out.  I do note, by the way, that the pinions and wheels don't mess right and I should be able to fix this by rotating one or the other, but It's not high on my todo list.

Thanks again

iain_downs

For information, what I've done is to change the rotation mate for the hour bush from the minute bush to the pivot hole for the minute pivot, with an offset to bring the bush (and wheel and hand) back up to the right place.  This seems to have worked.  

Iain