[sci.electronics] Mechanical actuators

rsteele@XN.LL.MIT.EDU (Rob Steele) (06/20/89)

I need a mechanical actuator to operate under microprocessor 
control.  Ideally it would be a solenoid that works off of 
5 VDC and not much current.  The thing will be pulsed at
around 7 Hz. with a 50% duty cycle and need not exert much 
mechanical force.  Am I dreaming?  What alternatives should
I know about?

larry@kitty.UUCP (Larry Lippman) (06/20/89)

In article <1436@xn.LL.MIT.EDU>, rsteele@XN.LL.MIT.EDU (Rob Steele) writes:
> I need a mechanical actuator to operate under microprocessor 
> control.  Ideally it would be a solenoid that works off of 
> 5 VDC and not much current.  The thing will be pulsed at
> around 7 Hz. with a 50% duty cycle and need not exert much 
> mechanical force.  Am I dreaming?  What alternatives should
> I know about?

	Try Ledex as a possible vendor for your solenoid.  7 Hz is a pretty
fast actuation rate, and I would be concerned about mechanical wear and
overheating if this rate will continue for any length of time.

	Also, I would consider a higher voltage - like 12 or 24 - for the
solenoid, since it will vastly reduce the current which your solenoid
driver circuit needs to handle.  5 volt DC solenoids tend to be rather
ineffiecient, and draw a lot of current, in addition to being more affected
by voltage drop across the driver circuit.

<>  Larry Lippman @ Recognition Research Corp. - Uniquex Corp. - Viatran Corp.
<>  UUCP   {allegra|boulder|decvax|rutgers|watmath}!sunybcs!kitty!larry
<>  TEL  716/688-1231 | 716/773-1700  {hplabs|utzoo|uunet}!/      \uniquex!larry
<>  FAX  716/741-9635 | 716/773-2488     "Have you hugged your cat today?" 

mmm@cup.portal.com (Mark Robert Thorson) (06/21/89)

rsteele@XN.LL.MIT.EDU (Rob Steele) says:

> I need a mechanical actuator to operate under microprocessor
> control.  Ideally it would be a solenoid that works off of
> 5 VDC and not much current.  The thing will be pulsed at
> around 7 Hz. with a 50% duty cycle and need not exert much
> mechanical force.  Am I dreaming?  What alternatives should
> I know about?

You don't mention the travel distance or the amount of force required.
A piezoceramic element might satisfy the actuation rate, wear, and power
consumption problems mentioned by Larry.  Perhaps he could suggest a
vendor for piezoceramic actuators?  (Maybe you could disassemble one of
those piezoceramic fans often sold as add-on equipment for Macintoshes
to cool the CPU chip.)

Another possibility is to use an audio speaker as your actuator (travel
distance and force would have to be very low).

Another possibility is to buy an old chart recorder from a junk shop
and use the actuator for the pen.

mmm@cup.portal.com (Mark Robert Thorson) (06/22/89)

As Larry suggested, Ledex is a good source for solenoid actuators.  Their 
product line is very broad.  Their address is:

Ledex Inc.
P.O. Box 427
801 Scholz Dr.
Vandalia, OH
45377

(513) 898-3621

mattern@ncoast.ORG (Duane L. Mattern) (06/26/89)

rsteele@LL-XN.ARPA.UUCP (Rob Steele) writes:
>I need a mechanical actuator to operate under microprocessor 
>control.  Ideally it would be a solenoid that works off of
>5 VDC and not much current.  The thing will be pulsed at 
>around 7 Hz. with a

Some questions...                      0 1 0 1 0 1 0
- Pulsed?  Like in a 50/50 square wave _|-|_|-|_|-|_ ?
- Do you need or will you want in the future, a finite resolution
  in the movement or just bang-bang?
- Have you consider using a stepper motor and a rod-crank mechanism
  to generate straight line movement, or do you need specific
  velocity, acceleration characteristics of the movement?
- How accurate do you need the pulse frequency, 7.000 Hz, or
  +/- 1 Hz.  If you don't need high accuracy, you could use a
  toy DC motor to drive you rod-crank mechanism.  Even the rod
  and crank could be obtained from a toy.

You could build a solenoid yourself.  It would just take some 
windings.  We did that at Purdue for one of those ME land 
vehicle races.  We used a two speed gear box which shifted 
gears automatically on a slope which was detected by a mecury switch.
The solenoid shifted the transmission by force meshing a new gear.
The solenoid force will be proportional to the current which depends
on the voltage supply and the number of windings squared (I think).
You can find this in any physics book (mines not handy). 
Our solenoid was driven by batteries with a hand wound solenoid.
Frequency response will of course be determined by the magnitude
of the maximum possible force generated and the size of the load,
which also includes the mandrel.
---------------------------------------------------------------------
|  Duane Mattern (216)433-8186         (mattern@ncoast.uucp)        |
|  Sverdrup Technology, Inc.           at NASA Lewis Research Center|
|  21000 Brookpark Rd, M/S 77-1        Cleveland, Ohio  44135       |
---------------------------------------------------------------------

johno@tekirl.LABS.TEK.COM (John Ollis) (06/27/89)

In article <1436@xn.LL.MIT.EDU> rsteele@LL-XN.ARPA.UUCP (Rob Steele) writes:
>I need a mechanical actuator to operate under microprocessor 
>control.  Ideally it would be a solenoid that works off of 
>5 VDC and not much current.  The thing will be pulsed at
>around 7 Hz. with a 50% duty cycle and need not exert much 
>mechanical force.  Am I dreaming?  What alternatives should
>I know about?

Several years ago, I worked on a driver for a solenoid actuated Selectric
typewriter running at as much as about 15 Hz, so I'd say the speed you want
is possible.  I don't know what your background is, so I'll warn you about
something that you may already know about.

The coil of a solenoid is an inductor and doesn't like to have the amount
of current flowing through it change.  So when you attempt to turn the coil
off, it produces a whole lot of voltage and probably burns out any circuit
you are trying to drive it with.  A common technique to prevent this is to
place a diode across the coil such that the diode is reverse biased while the
coil is turned on.  Now, when you turn the coil off, the current will flow
through the diode until the energy is dissipated.

Depending on the type of solenoid you select, you may want to look at a driver
chip such as something from the TI SN754XX family of drivers, some of which
already have the diode (they call it a clamp) built in.  See the TI Interface
Circuits Data Book.  You can also build a driver yourself with transistors and
stuff, but it seems easier to buy one that's already together.

I hope this rambling will be some help to you.

John Ollis
johno@tekirl.LABS.TEK.COM