Servo load question

[Youkai]

I'm working on a project where I'm having a servo move a few 3d printed parts back and forth. The servo is holding the parts in their position. I.e. the part is on a freely moving scissor jack and the servo is holding it somewhere in the normal range.

Currently I have a working version using an HS-311 servo. Ideally I'd like to use a smaller (physically) servo. I have a couple of HS-55 on hand and I think I might be able to get the parts light enough that the servo could move them appropriately.

How problematic is having a servo maintain a position using somewhat near it's maximum force? Do servo's overheat if they are encountering too much resistance? If the servo is fighting against gravity to hold the piece up for a long period of time (minutes or longer) would that break it?

With the HS-311 I'm pretty well within it's capabilities so I don't think there's any concern, but I think (with no actual practical experience) that I'm probably very close to the limit of what the HS-55 can do with this use case.

Thanks for your support.

[Benta]

Electrically, they are normally protected internally. But if you manhandle it so the the gears break, you have a problem.

[Psi]

Yeah, I would never buy a plastic gear servo for anything that actually matters. Metal gear is far more reliable especially when you are coupling direct mechanical force against the servo (as opposed to like a plane aileron where the air is the force)

One other thing to consider is digital vs analog servos
at least with RC servos
- A digital servo can put full power on the motor to move even when it's a fraction of a degree off where it wants to be.
- A analog servo will reduce the force it delivers the closer it gets to the position it wants to be at.

Digital servos are quite aggressive but they also need higher peak current and they tend to produce a lot of noise/emi.
They also tend to be louder as you can hear them applying full power to 'hunt' and maintain position

[Youkai]

So is it correct to say that the biggest risks for overloading it is that A) the gears would break, and B) it's going to make some noise as it fights to hold the position?

If that's the case it might be worth a few tests to see if it can actually hold the weight I'm trying to get it to. It's for a prop peice I'm making, for fun, so if one of the servo's breaks itself it's not the end of the world. I could always re-engineer it with stronger servos in the future.

[Nominal Animal]

I could always re-engineer it with stronger servos in the future.

I'd consider using a small stepper with a threaded shaft and a matching nut as a linear actuator.  They range from very small ones (6mm diameter) to as large as you need.  With a suitable thread pitch, it is self-holding; you only need to apply power when changing the position.  Take a look at e.g. eBay with keywords "thread" "stepper" for examples.

[Youkai]

I'd consider using a small stepper with a threaded shaft and a matching nut as a linear actuator.

I appreciate the advice. From what I know of how steppers work: if I did this I'd need to wire up some switches at the "bottom" of the travel so that when I power it on I can get to a known position. Correct?

[Benta]

Why go over the top?
If you want a lead-screw solution, just choose a sail-winch servo. Off-the shelf and easy to use. Works exactly as normal servos, but with several revolutions on the output shaft instead of just a limited angle. But it'll be slow in comparison.

[Nominal Animal]

Why go over the top?
If you want a lead-screw solution, just choose a sail-winch servo. Off-the shelf and easy to use. Works exactly as normal servos, but with several revolutions on the output shaft instead of just a limited angle. But it'll be slow in comparison.

Good suggestion!  Instead of support arms, the line –– say, strong but transparent fishing line –– is easier to hide, too.

[Youkai]

Good news! I did a test with the HS-55 and it seems like it will be able to handle the load I need it to. I think I'm gonna try printing my parts a little thinner to save some weight also.

[Doctorandus_P]

Why go over the top?

Why would a stepper be over the top?
There are plenty of models available for small prices, and they are often of "industrial" quality, unlike most hobby servo's.
Quite often you don't need an end switch. You just run it one way for more then the total travel length to get it into a calibrated position.

In general I don't like the hobby / RC servo's at all. It starts with a lousy protocol which has an inherent low resolution and slow update rate. On top of that there is no speed control at all. There are a few brands of those servo's with a serial protocol and settable PID parameters (and more) but these cost a lot more. Some of these also have a HALL sensor for the position (so no potentiometers that wear out) and these can be programmed to do multiple revolutions.

[Nominal Animal]

Why go over the top?

Why would a stepper be over the top?

For the stated use case, because a servo will perform acceptably, be easier to use and control, and probably more robust than a stepper (due to inherent rough absolute positioning ability, which steppers lack).  A stepper would require quite a lot of additional parts for this particular use case, including end stops.

In general, you have a point; I initially suggested a stepper for similar reasons.