# Constant torque spring - damping question

1. Dec 7, 2015

### DaanW

Hi all,

I have a constant torque spring that rotates the drive wheel of a Geneva mechanism. I am facing the following problem. I would like a very slow rotation of the drive wheel whenever the drive wheel is not driving the Geneva wheel of the Geneva mechanism (which is 270 degrees of a rotation). However, when the pin of the drive wheel rotates the Geneva wheel by 90 degrees, I need a lot of torque because there is a load connected to the Geneva wheel. Therefore, I need to use a high torque constant torque spring (7.50 in-lbs.).

I think that a rotational damper can help me slow down the rotation of the drive wheel (it is a timing mechanism, and the rotational velocity should be ~1 rpm), but when getting to the point of rotating the Geneva wheel, I would need the damping to be gone (because there, I need full torque).

Does anyone have a suggestion for a rotary damper that is capable of doing this? Or are there other ideas/designs that can solve the same problem in your opinion?

Thanks in advance!

2. Dec 8, 2015

### BvU

Hallo Daan, !

How about a kind of flywheel to even out the irregularity of the angular momentum required ?

3. Dec 8, 2015

### JBA

Do you need a change in rotation speed; if not, a liquid rotary damper will not prevent the mechanism from delivering its full torque, it will only limit the rate at which the torque can be delivered.

4. Dec 8, 2015

### BvU

It's not logical to dampen the drive wheel for 3/4 of the time: waste of energy.
Still in favour of the flywheel idea - but even that has air resistance; perhaps it's better to use a heavy drive wheel that slows down during 1/4 turn of work and revs up 3/4 turn ?

5. Dec 8, 2015

### JBA

If you are using a constant force spring for the full rotation then there is no waste of energy if you slow the rotation for the 3/4 portion.

6. Dec 9, 2015

### DaanW

Thanks for the replys. An additional constraint is that the component can't be any bigger than 1 inch diameter.. I think I can solve the problem by using a rotary damper, only problem now is to find one that can resist a ~1 N*m torque and bring the rotational speed down to ~1-2 rpm. Let me know if you have a design/manufacturer for such a component. Many thanks!

Share this great discussion with others via Reddit, Google+, Twitter, or Facebook