PDA

View Full Version : Servo



DynamoBen
04-08-2008, 10:30 PM
It’s still pretty early but I've been designing some props for my Halloween show. I've been researching servos and while I'm able to control them with no problem I'm a bit confused about how a servo’s torque specs translate to actual lifting/pulling power.

I have a set of Parallax servos kicking around, according to the specs they are 47oz-in. Am I correct to assume that these can lift/pull 2.9375 lbs?

rlilly
04-09-2008, 07:07 AM
By exactly one inch.

ppohlman
04-09-2008, 03:55 PM
The amount of weight that the servo can push/pull is directly related to the lever arm length and direction of force.
The attached pictures are simple diagrams to hopefully help clarify this concept.

The Torque (T) of the servo is equal to the perpendicular Force (F) it can push/pull multiplied by the distance (D) from the pivot arm. T = F x D

This means that a servo with a torque capacity of 47oz-in (2.937 in-lbs) can lift/pull something 2.937 lbs if the arm is 1 in. long (like rlilly mentioned above).
If the lever arm is changed to 12 in. long, it could only lift/pull something with a force 0.245 lbs.

Now if the direction of force is not perpendicular to the arm, then you need to determine that force.
This is done by multiplying the angled force (f) by the cosine of the angle (q).
This results in the perpendicular force (F).

Hope this helps someone out there.

DynamoBen
04-09-2008, 04:11 PM
Let me explain my idea and see if it’s feasible with servos.

I want to create 4 tombstone "pop up" heads. The heads will be plastic with a light bulb inside so they light up (not very heavy). I looked into pneumatics but ruled it out because of expense, complexity, and lack of granular control. Then I got thinking about using a stepper motors or servos. Servos are easy to control, inexpensive, and only require three wires for control which are all wins.

What I'm envisioning is a box (12"x12"x24" or smaller) behind the tombstone. The heads up/down motion (height) would be controlled by a motor. This motor would be controlled with Vixen; I then can coordinate the height and speed of the heads individually in each sequence. Think Whack-a-mole.

rlilly
04-09-2008, 04:21 PM
That certainly sounds feasible.

Assume you need to raise the skull 12". The weight of the skull (& light) would need to be < 4 oz. if you use the servo you mention. You could raise a greater weight by a less distance.

If you are raising straight up, then you need to reduce the amount of weight by the cosine of the maximum angle between the servo arm and the rising skull.

ppohlman
04-09-2008, 04:27 PM
That's one of the biggest problems with using a servo motor - they don't give you a lot of linear travel.

If you're wanting to move something 12", then you're going to have to use some sort of linkage system. How much travel are you wanting on the heads?

DynamoBen
04-09-2008, 04:41 PM
Travel would need to be 8-12"

I guess I could do steppers if need be.

DynamoBen
04-09-2008, 08:33 PM
What if instead of a linkage I go with a cam and follower. I wonder if that might work just as well and help keep weight down. :?

ErnieHorning
04-09-2008, 09:17 PM
You should be able to do what you want with a surplus wiper motor. Make you own double linkage and you should be able to pick whatever distance you want. Momentarily supply run power and let the motor park itself. If done right, the head should pop up and back down.

DynamoBen
04-09-2008, 09:31 PM
I'm after more control than just up and back down. I would like to be able to do position control. From Vixen 0-100% would equal down-up. I could then ramp the head up and down, snap it up and down, or even make it go 1/2 way. I would always know where the head was based on the percentage in Vixen.

The more I think about it the more it seems that a stepper might be a better fit. I'm pretty sure I have at least four lying around that would work.

ppohlman
04-10-2008, 01:59 PM
How much angular rotation do you get from the servos that you currently own?

DynamoBen
04-10-2008, 02:02 PM
180 degrees for the two I have lying here.

cgeek
04-10-2008, 02:16 PM
Ben there is a much easier way to do this...

I'm pretty sure that Parallax is a Futaba 3003. You can google up a continuous rotation mod for it. Then you can use a pulley (available from the hobby store) or threaded rod to move the skulls.

It can drift but in general if you time how long it takes to move it up and down, you should be able to use the timing in Vixen instead of limit switches or other complex methods to control the movement.

You can also control the speed of movement too. This is a real good link to start with.

http://www.seattlerobotics.org/encoder/200009/S3003C.html

Good Luck

ppohlman
04-10-2008, 02:41 PM
If you go that route mentioned by cgeek, the pulleys would be better. Using a threaded rod as a screw drive would be EXTREMELY slow to move it 12".

I just crunched a few numbers and put together a few diagrams of possible solutions.

6" long servo arm (Light Blue):
- 12" travel with 180º rotation
- You can lift something approx. 0.489 lbs. using the 47 in-oz servo (I'd stay below that weight)

4" long servo arm:
- 8" of linear travel
- Can now lift something approx. 0.734 lbs. (once again, I'd stay below)

If you go with the cam idea, a simple cam would be a half circle. To get 12" of travel, you would need the diameter to be 12" as well. Depending on the friction of the follower and the weight of the cam itself, you might be able to lift something even heavier than the above options.


------- EDIT ---------
A couple of ways to be able to lift more weight is to use gravity in your favor.
You could either:
1) Add a counter weight on the opposite side of the servo arm, or
B) Throw a spring in there to help offset the weight of the head.
Just a couple of thoughts to throw out there.

cgeek
04-10-2008, 05:50 PM
Both of those are excellent idea.

BTW if that is a 3003 and he runs it off 6v it is rated at 57 not 47 so he's got a nice cushion.

ppohlman
04-10-2008, 06:17 PM
In my example, increasing the torque to 57 in-oz would increase the capability of the 6" servo arm to 0.594 lbs. and the 4" arm to 0.891 lbs.

Rocketman4
05-08-2008, 08:26 PM
The simplest way to fix this problem with the servos it to create a counter weight. This will make the lift 0. The only thing you have to deal with would be drag which done right would also be very small. I didnt label the rope in yellow, the pulley black. The head is now on a dowel through 1 or 2 washers to hold it straight and fishing weights are added to balance it out. Add a servo a little oil and there you go.

DynamoBen
05-10-2008, 11:07 AM
Interesting thought, I may have to play around with these options in the coming months and see which one works best. I now have access to both servos and steppers so I may try both.

Elmo2resc
07-27-2008, 10:57 PM
Ben, What are you using to control your servos?
Jerome

DynamoBen
07-28-2008, 12:20 AM
I ended up moving to steppers but I had the servos connected to a basic stamp for testing.

pete
07-30-2008, 03:12 PM
I did a project at work where I had to make a prox sensor adjustable so the operator could change its reference point. I set it up like a Genie garage door opener, ball screw setup. but instead of using a ball screw I used an acme thread and nut. I programmed the controller I used to run the motor a given time which would equal the distance I wanted it to cover. I believe i used a gearmotor which operated at 17 rpms . Not sure it this Idea is any help but I thought Id throw it out there.

Pete