I need a motor with enough torque to do something like hit a cymbal or snare drum with a drumstick. Speed is an issue as it does have to be fast to respond and it will also have to have enough torque to put enough force on the cymbal to make it sing. Any suggestions? I have very little experience with motors.
The great advantage of forums like these is that you effectively brain-storm the concept across a wide audience.
Some ideas will be good and some will be rubbish but somewhere out there there might be a realistic answer
Have no experience of what you wish to do but you might want to consider the following.
Rather than a straight motor driving the "stick" or "brush" you could perhaps use a spring loaded pivoted lever to do the striking. A stepper motor then drives a cam which simply raises the stick off the cymbal/drum against spring action. The advantage of a stepper is that you can stop it at predetermined points and get reasonably fast response.
Thank you for response. These are exactly the kind of replies I need. I'l be looking into your suggestion.
Someone suggested a solenoid. I did a bit of research and it seems like it would be reasonable for what I'm trying to do. Although I'd still love to here other suggestions or thoughts on using a solenoid.
I'm not musical, I can just about play a CD.
I think when a drum or piano string etc is struck the drumstick (etc) doesn't stay in contact with the surface. Would this mean that a mechanical device would have to have a rest position just above the surface and drive a "strike" down to impact and back up. I hope this makes sense. It seems to me some sort of cam drive might be needed.
I also suspect that a harder blow drives the drumstick further into the surface of the drum - it's not the same process as banging a hammer on a cast iron anvil.
Some drummers can be very hard, very fast and at other times very gentle.
I suspect a spring mounted drumstick would be good at some frequencies and lousy at others.
Maybe a mechanical system could use much heavier drumsticks than a human would - perhaps make from steel. That might allow the use of a very stiff spring which might help.
Personally I would go for a drum synth ...
...R
Yeah, I use vst's for drums. Really I just have an overly ambitious idea too have a machine read in midi data (that same thing a vst does) and play mechanical drums. You have the right idea. It needs to have a snapping movement that gives enough to come off of the drum but has enough torque to sound the drum. The cymbals I imagine could go straight through the hit. This of course would have to be done with extreme elegance.
I can imagine how a cam could work. I had to do a little research as to what a cam actually is as I am much more a programmer than a mechanical engineer and this makes your suggestions extremely helpful. A cam seems like a decent fit. My concern is that it may not allow the stick to "bounce" off of the drum.
Imagine a heavy drumstick is suspended some distance above the drum surface by a suitable spring and a cam is used to push the drumstick downwards. If the cam pushes quickly the drumstick will gain enough momentum to cause it to continue moving downwards away from the cam - it will hit the drum and then rebound back to its idle position waiting for the next cam movement. If the cam rotated faster the drumstick would hit the drum harder. The problems would be the need to have a damper to prevent the drumstick hitting the drum a second time and the possibility that it would not return fast enough for fast drum beats.
All this thought makes you realize how sophisticated a drummer's brain is. 
...R
I think I'm able to visualize what your talking about. Something like this?
So I could use a stepper motor to bring the cam to a resting point? Would I be able to fire off one rotation on the cam fast enough to drive the stick into the drum, or will it take a few revolutions to get up to speed?
For damping on the stick I could use another spring, like the pedal to a bass drum. I think I'm beginning to get a decent visual of how this will work. Thank you for your responses.
That's a lovely animation - and there seem to be several more which might inspire you.
Imagine if the cam in the animation turned much faster the spring would not be able to keep the follower in contact with the cam.
What I suggested in my previous post is to cause that situation deliberately so that the drumstick flies down and hits the drum. The idea in my head is more like in the attached image. The red bit is the spring.
It would be essential for it to do one drumbeat for every revolution of the cam.
This idea may be total rubbish.
Another thing to keep in mind is the need for the mechanism to be quiet enough not to be audible within the music.
...R
That's more elegant than what I was thinking. Thanks. So I've done a bit of brainstorming: Supposing I can get the cam to fire off and drive the stick down, then return to rest fairly quickly, I could use another spring for damping and include a guide for the stick. I added this to the image for clarity.
1* this would be the guide for the stick and could handle the rebounding in both directions. It would swivel at point 2. The stick would be able to move freely inside the guide. However, this seems problematic too me, mostly because of the spring. The direction of it's opposing force seems odd.
A better design may be to have the square portion of the guide constrained to move up and down and have it rotate with the stick, then I would not need a swivel at point 2. So for the guide, the longer vertical piece would stay straight up. The square portion that holds the stick would be connected to the vertical piece so that it could rotate and slide up and down along the vertical piece. I'm a kind of brainstorming as I type this up. And I'll remind you, I am no mechanical engineer. Any mechanical skills I've gained have rusted up quite a bit since college physics. I fear I may be over-complicating things.
"It would be essential for it to do one drumbeat for every revolution of the cam."
In that case I may have to figure out some other mechanism to use. As the grand idea it to read bytes from a midi file and signal the hit. The timing would be somewhat random, but not totally as there is a pattern. The stick may need to hit the drum twice in one second or just once in once second.
I know this can be done and so far I believe this post has been truly useful. Any more suggestions to new mechanisms or alterations of the one above would be greatly appreciated.
There may be some confusion. I didn't mean that the cam would rotate continuously and cause a drumbeat on every revolution. What I meant is that you can cause a single revolution whenever you want and each revolution will cause a drumbeat. I was trying to respond to your question about needing a few revs to get up to speed.
I doubt if there is anywhere to tell you how fast you could get a motor to do a single revolution or what power it would need. Stepper motors typically work at a few hundred rpm. Many cheap DC motors will run at 10,000 to 20,000 rpm, but they would need a time to accelerate so I don't know how fast they could do 1 revolution from a standing start. You could probably supply a high voltage for a very short time to speed things up. I think that's what I would try.
I can't emphasize too much that you will need to to a ton of experimenting before you get this to work no matter what mechanism you choose. 6 months might not be enough, and it still may not work.
On the other hand, it's interesting.
I've just realized that there have been mechanical musical automata that play drums for well over a hundred of years. I suspect they rely on gravity to cause the drumstick to fall at the right moment - released by a trigger of some sort - and then wind it up ready for the next beat in the silent period. I suspect suitable music is chosen not to make unreasonable demands on the mechanism. I'm sure this would be a fruitful area for a Google search.
...R
Thanks, it's nice to hear that this project is interesting to others as well. And I do agree about the amount of experimentation I'll have to do. Maybe this would be a simple task for some but not so for me. I have little experience with mechanics, but I'm confident I can learn enough to get this working eventually.
Anyways, this inspired some ideas.
Using their very own solenoid, these guys were able to achieve rapid hits on the bars. The solenoid approach seems to better simulate an ON/OFF scenario, I hope that makes sense. Using a lever, a high voltage signal to the solenoid could push the back of the stick up and drive the head down. A counter weight attached to the back end could help bring the stick off of the drum and prevent it from bouncing back.
I think you have arrived at the "suck it and see" stage.
...R
Great video, by the way. So many of them are c**p.
Hi,
Interesting ideas and discussion...reminded me of a couple of similar things I've seen which may be useful?
There are full details, plans and instructions on how to build the mechanisms for this automated drum kit on the Instructables website.
Also, Felix Thorn's machines are amazing!...here is an example Felix's Machines - HD - YouTube - there are more videos on YouTube and info on his website.
Hope these are interesting/useful!
Best of luck with your project 
I would go for a solenoid. But there are some issues with that to.
In order for music to sound good the notes should be timed acuurate to less than 10ms relative to notes from an other instrument or the predicted sequence of note from the instrument itself. This means that the delay from sending the pulse to the stick hitting the object must be fairly predictable and constant.