# Thread: High precission axis

1. Abacus
Join Date
Nov 2008
Posts
3
Rep Power
0

## High precission axis

Heya!
I'm new to robotics but my task is simple and I know some programming. I need to control an axis rotating in two dimensions (like a camera that can tilt and pan) from my PC software but there's a lot of pressure on the rotational speed and precission, the more the better.

Any help much apreciated!

Thanks!

2. ## Re: High precission axis

Welcome to the forum!

I'm afraid you don't give us a lot to go on here. How much weight will you be mounting on this pan-tilt head? What's your minimum desired rotational speed and precision? Space constraints? Budget?

The more information we have on your project, the more help we can all be.

Again - welcome to the forum!

3. Abacus
Join Date
Nov 2008
Posts
3
Rep Power
0

## Re: High precission axis

Hey A:
Fair enough. What I want to do is to point a laser around. I'll probably use a heavier type of laser pen. It's for a video installation where it needs to follow movements synced to music so we're talking about a fairly rapid pace. I then also want to sync this to a video projection. I would guess that from a distance of a few meters I need to point within a radius of something like a centimeter. Is that doable? Money is flexible but I would prefer to end below 150\$.

Thanks again.
Last edited by bhenriksson; 11-11-2008 at 04:30 PM.

4. ## Re: High precission axis

Thanks, Bhenriksson, that helps a bit.

Your budget may be a little low, but within your budget you could get 0.09deg pretty repeatably. You weight is really low which helps a bunch.

High-accuracy servos are pretty expensive. I'd probably start off with something like:

2x HS5645 Servos (60 degrees in 0.18sec, should be repeatable to something close to the 0.09deg) ~\$110 total
1x SSC-32 Servo Controller \$40
1x Pan/Tilt kit, no servos \$10

Probably the "right" way to do it is with stepper motors. You'd be building your own brackets, programming is a lot tougher, and it'd probably blow your budget.

5. Abacus
Join Date
Nov 2008
Posts
3
Rep Power
0

## Re: High precission axis

Thanks a lot! That sounds like a very good start. One more thing though, how do I know what different speeds these motors can be driven in? It will have to create fairly smooth motion and not something I want to compromise with.

6. ## Re: High precission axis

Servos are a little "different".

They're continuous duty cycle, and you tell them where to go by sending pulses to them.

A 1500uSec pulse is "center". A 900uSec pulse is 45deg left. A 2500uSec pulse is 90deg right.

It's worth noting that they don't go 360 degrees by default. They range from 90deg to 180deg total, give or take depending upon the specific. When you first test them be careful when you get over about 60deg left and right of neutral. Approach it slowly and when it starts to chatter against the stop, immediately back it off. The servo controller will cheerfully destroy them trying to force them past their maximum range if you ask it to...

The servos I quoted typically only do 145deg total. Check the spec sheets when you get them and approach those limits initially with caution.

You can tell the servo controller how long to take getting to the end point. You can say "#1 P1300 T1100<cr>", and it will move servo #1 to position 1300uS over a time of 1.1sec (1100ms). If you command it to move faster than the servo can move, the servo will go as fast as its widdle legs will carry it and no faster.
Last edited by Adrenalynn; 11-11-2008 at 06:19 PM.

7. ## Re: High precission axis

In terms of syncing the servos, I've found that it is better to sink on the slower end, and have the servos "snap" to the beat. Depending on how well you need to sync the movement I've found that you can make robots "dance" by interfacing them with midi. That way you can set up a sequencer and then tweek the motions by changing which notes you send.

(by snap I'm suggesting you set the end points of the motion, and then have the not trigger the servo moving either on or just before the beat depending.

In any case, sounds like a really fun project.

DB

8. ## Re: High precission axis

One question: since it involves a laser pointer, why not use the tried and true method of moving mirrors with a fixed laser, instead of waving the entire laser pointer around?
It may take a bit more tweaking of the mirror angles but it would mean you can use stepper motors from a scrapped printer/scanner/floppy drive, increasing the budget for anything else.

9. ## Re: High precission axis

Scud...your smrt...I mean smart.

DB

10. ## Re: High precission axis

Going with scuds idea, you could use a car mirror assembly if you don't have to move the mirror that far, and if you do, then the original pan/tilt assembly should do the trick. Not only that, but I was thinking in terms of the MIDI thing, depending on how fast you want things to happen, take midi notes

Note Name = x value Note Velocity = Y value, and Note length = amount of time for the move. You would always have to be one note ahead, but you could even set it up so that not including a time made the servos attempt to "snap" to the desired position.

DB

##### Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts
•