Really small motors

[fat_cat]

Hi, I'm working on a project where I need to lift and lower hundreds of small pins. The best solution I could think of was to use small electric motors. Figured magnets would cause interference with each other and hydraulics wouldn't allow the same amount of control without a lot of calibration.

The issue is the smallest motors I can find are too big, and cost at least a dollar each, would add up way too fast. Does anyone have an idea for how to use one motor to control more than one pin? or know where you can find small motors for less than a dollar each? Any ideas are greatly appreciated.

The pins are going to be small, 2 mm diameter and 2-3 cm tall. The material isn't important and they all need to be moving independent of each other and ideally have different speeds that they move at. They don't need to be tightly packed either.

Thanks for reading, and for any help you can give

 

[Gryd3]

If you would like to expand on the intended purpose of the pins it would help us out... do you want the pins to be simply be in motion... or would you like to set the position of each pin? Would you like analogue control of the pins? or is a simple in or out good enough?

My first though is a solenoid... You dont need a magnet, just a spring and a metal pin that will get drawn inside the coil when power is applied... (You said they did not need to be tightly packed...)
Simple construction would require you to explore some air core inductors to use, you you would need to wind your own. Just make sure you use a metal pin that is magnetic

Here are two projects that came to mind... it is a bit of a stretch but may give you some insight into alternative methods.
http://hackaday.com/2013/12/12/fail-of-the-week-physical-pixel-display/
http://hackaday.com/2014/04/03/lightbyte-animated-shutters/
*Edit: Look at flip-dot displays as well... I am unsure on overall budget... or if your pins can be mechanically multiplexed or if they need individual simultaneous (or close to) control

 

[KrisBlueNZ]

I'd imagine electromagnets (solenoids) would be the way to do. Also you might get some ideas if you dismantle the printhead from an old dot matrix printer (if you can find one - try a technology museum

 

[fat_cat]

Thanks for the replies! The 1st link from Gryd3 and the inkjet printer are a lot like what I'm trying to do. Where could I buy solenoids that small, cause I couldn't find that. And the pins will ultimately be controlled digitally and on a massive scale, 100X1000 so the cost is pretty essential. I have the programming side down but i just cant figure out a way to do the mechanical part economically. The pins also need to be variable in speed. I'll research the old dot matrix printer more, but that sounds like it can only move each pin and a fixed speed which would end up being problematic, can solenoids move a pin at different speeds? i know they have push pull solenoids but i thought you couldn't control the speed. Thanks again, really helps me a lot.

 

[Gryd3]

This sounds like quite the project you have on the burner...
As far as I know, push/pull solenoids require magnets, where as a 'pull' only solenoid could use a ferrous metal. (but would require a spring or other method to return the pin to it's starting position once power was taken away)
As far as variable speed is concerned, a speaker pops in to mind. That type of control would require an analogue signal or PWM...

You can try to buy pre-made solenoids, or you could construct your own. (Have you ever made a small coil gun as a kid? Wrap some wire around a pen case and connect the wire to a camera capacitor...)
Your home-made coils wont need the 300V from a camera flash

 

[gorgon]

Is this a kind of art display, with a moving picture? Is there any load on the pins? what are the direction of the pins, horisontal or vertical? Can the movement be in steps?

100,000 pins sounds like a major installation.

 

[Laplace]

I'm thinking of a spring-loaded solenoid with a small coil driven by a multiplexed sample-and-hold circuit. Let's say there is an emitter-follower with a high gain transistor (MPSA18?) and tantalum capacitor at the base so the voltage on the capacitor appears across the coil, and the emitter current through the coil is proportional to the base voltage. The capacitor is charged through an analog switch multiplexed to a DAC, so the position of the solenoid core is controlled by the voltage of the DAC.

Now let's say for each pin you can bulk purchase the coil, transistor, capacitor, & analog switch for $0.50. For 100,000 pins that will total $50,000. Now let's guess that each coil will take a maximum of 10 mA current for maximum pin displacement, and on average only 5 mA. So the average current consumption for the pin coils is 5x10^-3 x 10^5 = 5x10^2 = 500 amps. How big is this box that will consume 500 amps?

Will the pins be in a protected environment so the solenoids can be optimized for low-friction operation? Or will the pins need to be individually sealed? If so, then multiply the above estimates by a factor N. I can't guess what N might be.

 

[shumifan50]

I would tend to not want to drive each pin independently all the time. Depending on the speed with which each pin position needs to change, it might be possible to set up a 'transport' that does a group of pins, one at a time and then repeats the cycle continuously. You can then group the number of pins/group to suit a reasonable number of groups and have only one actuator per group. Each row should have the same pins/group so you will need only one transport mechanism that moves all the vertical groups at the same time. That way you can have a matrix of much fewer actuators to drive all the pins.

Following the above, the number of actuators needed are much fewer and therefore more expensive 'motors' can be considered, e.g servos or stepper motors. Each ;motor; can also be bigger as they dont have to be one per pin. For example a servo is about 5mm thick(micro servo) and 2cm long, so if a group can be a minimum of 10 pins, you will have enough space for the servo with 2mm between rows of pins. However this can only be done if the movement does not have to instantaneous and/or simultaneous for all the pins.

To retain the pins in their position, cams be used, driven by the servos/steppers with the pins being spring loaded to push back against the cams.

 

[fat_cat]

Thank you for the help, I've been busy lately and am going to come back to this later. Just wanted to make it clear the posts are appreciated.