Newsgroups: comp.robotics
Path: brunix!news.Brown.EDU!agate!howland.reston.ans.net!vixen.cso.uiuc.edu!uchinews!cdsmail!timbuk.cray.com!hemlock.cray.com!kilian
From: kilian@cray.com (Alan Kilian)
Subject: Re: Help Needed w/ unknown servo
Message-ID: <1994Feb21.125039.6622@hemlock.cray.com>
Lines: 125
Nntp-Posting-Host: poplar12
Organization: Cray Research, Inc.
Date: 21 Feb 94 12:50:38 CST

> From: mdb0201@TAMUTS.TAMU.EDU (Michael David Bigby)
> Subject: Help Needed w/ unknown servo
> Date: 21 Feb 1994 01:53:54 -0600
> Organization: Texas A&M University, College Station
> 
> 	I have about 6 big servos that I bought at a surplus store in Dallas
> for about $5 each, and I can't do anything with them so I thought I would post
> in hopes of suggestions or references or something. When I bought these servos
> I thought I was getting a really good deal on some big steppers.  They weigh 
> about 5lbs each and they "clicked" when I turned the shaft like a stepper 
> would.                   ^^^^^^^^^ 
                           This is called "magnetic cogging" and steppers
                           do it more than regular DC motors, but regular
                           DC motors also do it. As do brushless motors.
                           Special motors are designed to minimize this if
                           the motor is designed for low speed operation.

> However on closer inspection at home they have a bunch of control 
> circuitry inside the case and they don't look like any stepper I've ever 
> gutted.  
> There are 3 heavy gauge wires (white, blue, red) and 6 not so heavy
> gauge wires (white, red, black, blue, brown, yellow) running into the circuit.

I think you have a "Three-phase brushless DC motor" with shaft position
and velocity sensing hardware.

It could be either a 3 or 6 phase motor since you have a group of three
wires of one thickness and six of the other. I think it's three phase
since the group of three is the thicker group.

Now going on that assumption, it could be either "Y" connected or "Delta"
connected. Imagine three line segments which represent three coils of
wire in the STATOR of the motor. To make a "Y" connected motor, connect
one end of all three line segments together to form a "Y" shape and from
the other three ends of the coils, connect three thick wires and run them
to the outside of the motor.

To make a "Delta" connected motor, connect the three lines in a triangle
and run three wires from the verticies of the triangle to the outside world.

Wye "Y" connected motors are more common.

I tried to figure out how to determine which you have by resistance
measurements, but they come out identical if you don't know the
per-winding resistance.
For a Wye motor: Any two leads reads 2*R
                 Short two leads together and measure to the third lead
                 reads 1.5*R   (R + R/2)

For a Delta    : Any two leads reads                  1
                                               -----------------
                                                   1      1
                                                  --- + -----
                                                   R     2*R

                 Short two leads together and measure to the third lead
                 Reads R/2

Now the problem with this is let's say you measure 200 Ohms and 150 Ohms
Which one do you have? 

If it's a Wye, the coil resistance is 200/2 = 100 and everything is fine.
If it's a Delta, the coil resistance is 150 * 2 = 300 and the first
                 equation says you should have gotten 200 Ohms so you
                 can't tell if it's a Delta or Wye

Now MAYBE you can tell by the voltage and current ratings which coil
resistance is correct, but I don't think so.

Since you have them open, you can probably tell by looking or probing
with an Ohm meter.

> Inside they have an reflected IR incremental encoder and an array of
> 3 hall-effect sersors (i think) spaced about 15 degrees apart on one side of
> the rotor.  

These are probably the other six wires. The hall sensors are for determining
the position of the rotor and the incremental encoder is for measuring velocity
(and maybe position since your hall sensors are all on one side of the motor)

> The case says Kokusan Denki Co., TL0108, DC Brushless Servomotor and some
> info about the wattage and voltage and current draw.  

Look them up in a phone book if you can. Or call information and ask.
Tell them what you are trying to do and they can help find out which state
the company is in.

Here are some books that wil explain things:
_Brushless DC Motors Electronics commutation and controls_ TAB #3267  1990
_Direct drive robots Theory and practice_ MIT press 1987

Other companies that make brushless DC motors are:
   Shin Meiwa industry Co. Japan
   Motornetics Inc.
   International Scientific Inc.

   Motors in general:

   PortEscap (I think) 516-234-3990
   Pittman 215-256-6601
   Airtronics 714-830-8769
   Copal 213-618-1225
   Globe motors 513-228-5000
   Inland Motors 703-639-9045
   Mabuchi (Maybe) 212-686-3622
   Micro Mo Electronics 800-635-3555
   Namiki 201-368-0123_ 
   

   So get on the phone and try to track down a manual for these big guys.

> 	Even if I can't find info about how to control these they are worth 
> what I paid for them (just the bearing in the motors are worth $5).  But I
> have a feeling that these are neat stuff and would love to use them.

Now that's the attitude I like!




     -Alan Kilian
-- 
 -Alan Kilian           kilian@cray.com 612.683.5499 (Work) 612.721.3990 (Home)
 "Seeing is much more than the perception of images. 
  It is the application of knowledge" Some Visual Agnosia expert on NOVA
