Motor selection for rotating table

[Azazello]

Hi everybody,

I am building a rotational positioner for a robotic welding station. I am completely lost on how to tackle the driving problem - never dealt with motors aside from a class that I took 5 years ago (and failed it, the only failed class in my academic career).

I have 3 pivots, two of the loads (welding tables) are about 900 pounds, 36" wide and the main axis is 900x2 + 900, double the width. The tables will be rotated only 180 degrees, back and forth, a lot of start/stop action.

I started by approximating the load distribution at the ends of the table and finding rotational inertia, but not sure how to proceed..

Also, how realistic is it to find a motor to spin the table that would be less than 3.5"x3.5"? What other characteristics should I consider?

Thanks for any help!

 

[7anoter4]

It is difficult to understand the problem. As I understand there are two tables with welding equipment to be rotated [simultaneously?] 180 degrees back and forth. The main supports the above two, but it is not clear if it is rotated also and how.
What could be the velocity of each table and if the welding equipment would get some force [or better torque] as a constant load and for how much time.

 

[desertfox]

Hi Azazello

A sketch speaks a 1000 words

desertfox

 

[Azazello]

Thanks for your replies, gentlemen!

7another4, you got it almost right. The tables don't have the welding equipment, only the positioning clamps and the part to be welded. The two tables rotate independently of each other and the main frame.

I'm sorry, I don't have a picture before Monday, but this video -

http://www.youtube.com/watch?v=UA5mTTEycbs

- shows the concept.

 

[waross]

Try a 1/2 HP gearhead motorwith a brake for the main rotation and a 1/4 HP gear motor with a brake for the work rotation. You may consider running both motions up against a stop for repeatable positioning. You may use a solenoid activated stop or a sliding stop and reverse the rotations every cycle.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Azazello]

Waross, could you show me how to arrive at the HP you specified? I can't seem to be able to wrap my head around the conversion problem, my moment of inertia is 260000 lb*in^2 (one table), and it would be nice to have the table flip 180deg in less than half a minute. I tried three different ways and got three different answers, which don't bring me even close to your suggestion.

I guess I'll call it a temporary (hopefully) mind block, if someone can help - please do!

 

[waross]

I used a careful SWAG analysis. Observation of other positioning devices. Too big is big enough. You can use Cowern's discussion of RMS HP loading to gain a significant fudge factor.

http://www.baldor.com/support/literature_load.asp?LitNumber=PR2525

You may consider 180 degrees of rotation, back and forth and use air cylinders.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[tygerdawg]

If it is similar to the video, then what you are trying to build is a headstock-tailstock positioner. You can buy these things, you know.

If you want to go through the effort of building your own, there is significant engineering required. For a quick summary of the dynamics involved, websearch for "Smart Motion Cheat Sheet". Use that to determine your peak torque requirement which is a function of speeds, accels, decels, gravity effects, and rotational inertias. Your solution may involve transmitting the torque to your mandrel via gearing of different types. Size your motor for that. Depending on your accuracy & speeds required, you have a variety of options like AC gearmotors with encoder feedback, servo motors with encoders, or just dumb simple motors of some sort moving to a switch and/or hard stop.

For robotics, you also should take into consideration any mechanical deflections of your rig and stiffen it up structurally as required.

TygerDawg
Blue Technik LLC
Virtuoso Robotics Engineering

http://www.bluetechnik.com

 

[Azazello]

Thank you very much, these two papers are indeed helpful!

TygerDawg, we're building a very specialized machine, and hoping to make it simpler and cheaper than the head-tailstock positioners available. We need only 2 positions for both tables and the supporting frame, which greatly reduces the control complexity.

The plan is to use a DC motor with PWM control, proximity sensors for determining the end of rotation and locating pins for securing the table in position.

Does anyone know of a similar paper that would cover DC motor-specific design considerations?

Thank you again for your time and knowledge!

 

[7anoter4]

First of all wk^2=weight[lbf]/g[gravity acceleration]*radius^2/2 so 900 lbf and 36" dia wk^2=900/32.18*18^2/2=4530.8 lb*inch^2.
Second: Mass moment of input shaft must be multiplied by the ratio squared and added to the mass moment of the output shaft. The time to reach the rated velocity is about 0.03 sec but the table will continue to be rotated. So, what you need is the table velocity, let's say 30 rpm, and the time of rotating 180 degrees is about 1 sec. [1 rotation=60/30=2 sec, 1 half of rotation 1 sec.].
If the motor rated speed is 1800 you need a reduction of 1800/30= 60 [approximate].
The motor of 1/4 hp, as waross said, is suitable, of course.
The problem could be the number of starting per hour. I think the waross proposal of compressed air cylinder could be better.