0
\$\begingroup\$

Ok, here's my problem:

I need to create a needle that can rotate continously in both directions and an infinite amount of "loops". I guess the ideal is to use a servomotor, isn't it? I couls use steppers but I want to keep the price as low as possible.

The problem is tht the most important thing is the precission of where the needle is pointing. I have found already several links that explain how to modify servos so they can run continously, but I lose that precision I'm looking for. I am asking whether there exists a way to achieve this. The only thing I can think of is using some kind of sensor that tells me where the needle is pointing so I can stop the motor when I want it to. Inexpensive sensor if possible...

Does anybody know a solution for my problem? As my last choice, I would use a cheap stepper of about 7.5 degrees of min. rotation and a gear to get a better resolution, although even those cheap steppers are about 8$ while I can get cheap servos for a third of that.

\$\endgroup\$
8
  • \$\begingroup\$ How much precision do you need? Have you considered gearing a motor down to achieve this? Is the needle light enough that you could use a galvonomic (magnetic) system to move it? \$\endgroup\$
    – pjc50
    Commented Jun 18, 2013 at 12:10
  • \$\begingroup\$ @pjc50 The needle could be made of paper or a clip, like 5 cm long. Vero light \$\endgroup\$
    – MyUserIsThis
    Commented Jun 18, 2013 at 13:06
  • \$\begingroup\$ Steppers are cheaper and easier. Within reasonable parameters they can be operated open-loop (no feedback, or just one basic "end stop" sensor) which greatly reduces the task. If this is a one-off or very-few-off then just cannibalise CD/DVD drives or printers for their steppers. \$\endgroup\$
    – John U
    Commented Jun 18, 2013 at 14:57
  • \$\begingroup\$ @JohnU I will try to get them from an old machine. Thanks. Do you know what kind of sensors are used for this? \$\endgroup\$
    – MyUserIsThis
    Commented Jun 18, 2013 at 17:03
  • \$\begingroup\$ Usually a small optical sensor, they're usually a U-shape with an infra-red LED one side and a photo-diode the other to detect the beam being broken by some part of the mechanism. \$\endgroup\$
    – John U
    Commented Jun 18, 2013 at 17:06

2 Answers 2

Reset to default
3
\$\begingroup\$

I have worked on a airplane simulator that also had this problem. We used a stepper motor, which I think is also a good answer for your case.

A stepper works well here since you know it will need to deliver very little torque and therefore can be counted on to not loose absolute position once it has been established. I don't know why you think a stepper motor will be more expensive than any other small motor, but I don't think that's true, especially when adding the cost of position feedback and gearing. One thing you do need with the stepper is add some way to know you are at a fixed position. A opto-interruptor is usually a good solution for that. At startup, you hunt for the known position, then the firmware tracks it after that.

A servo-controlled motor will be more complicated because you need to add position feedback somehow. That could be a pot with two wipers, or more likely, a shaft encoder. The bare ordinary (as apposed to stepper) motor will need to be geared down, so that adds more cost. This just sounds like more trouble to get working and to keep working, and I doubt there will be a cost savings after you consider all the other stuff this method will need beyond the bare motor.

\$\endgroup\$
3
  • \$\begingroup\$ Thank you very much for the answer. I think I'll finally go for the stepper motors as I actually found some really cheap ones. By the way, I'm also building an airplane simulator and this problem arised when trying to build all the neddles that move continuosly, like the altimeter, or heading indicator, do you have any links to that particular projects. Anyway, thank you very much and I will look for the opto interruptors \$\endgroup\$
    – MyUserIsThis
    Commented Jun 18, 2013 at 13:10
  • \$\begingroup\$ @MyUse: Unfortunately that project never made it to production. That company had ownership problems, so the project was abruptly cancelled and everything put in a warehouse maybe 15 years ago. I don't know what happened to it after that. It's amazing how many little companies die for stupid reasons, having nothing to do with the product not working, poorly targeted, unproduceable, etc. I've seen everything from single owner with analisys-paralisys never deciding what to invest in until nothing was left, to VCs deciding to sacrifice one company to shore up others in their portfolio. \$\endgroup\$
    – Olin Lathrop
    Commented Jun 18, 2013 at 13:35
  • \$\begingroup\$ that's bad to hear. Anyway thank you for your help. \$\endgroup\$
    – MyUserIsThis
    Commented Jun 18, 2013 at 17:01
0
\$\begingroup\$

An optical encoder would work well for this. Mice use them in their wheel (or axes in the case of a ball mouse). simply pull the encoder and photointerrupter from the mouse and attach it to the shaft with a belt or gears. You can then detect and count pulses from the photointerrupter with a micro, and with a bit of calibration and proper gearing you can get high precision and accuracy.

\$\endgroup\$

Not the answer you're looking for? Browse other questions tagged or ask your own question.