Maker Pro
Maker Pro

Seeking a simple binary linear actuator

J

Julian Vrieslander

Jan 1, 1970
0
My workgroup is interfacing some equipment on an optical bench. We have
three small elements (lenses and filters) that need to be moved, on
command from a computer, in or out of the optical path. The elements
are on ball-bearing linear slides, movement is horizontal, and it only
takes a few ounces of force to move them. All the off-the-shelf
actuators that I have found so far are very pricy, and overkill for what
we need. Our requirements are minimal:

a) IN position can be set by a hard stop, and must be repeatable to
0.02". OUT position is not critical, as long as it is about 3" away
from IN.

b) No need for proportional control or stopping at intermediate
positions between IN and OUT.

c) Transit time is not important: 10 seconds or less.

d) Ideally, each actuator would be controlled by one or two TTL control
lines from our interface.

After some googling and brainstorming, I can think of a couple of
designs. A motor and leadscrew is the obvious starting point. A small
DC motor could be driven in both directions by an H-bridge, plus a
couple of transistors or logic gates to handle the polarity switching,
limit switches to stop the motor at each end, maybe a flip-flop to keep
track of the current state.

Or I could use a stepper instead of a DC motor. HSI has small steppers
with integral leadscrews:

<http://www.hsimotors.com/linear-actuators/26000-Ncap.htm>

I found a nice stepper driver chip (Allegro 3967):

<http://www.allegromicro.com/sf/3967/>

It needs only two TTL control lines: one selects direction, the other
counts pulses. Perhaps with a stepper, I would not need the limit
switches.

Other solutions were considered. Solenoids might be too violent,
pneumatic gizmos too complicated. An R/C hobby servo uses proportional
control, but it could be rewired as a bang-bang. I even thought about
scavenging the tray actuator from a compact disk drive.

Any other ideas? Thanks in advance.
 
D

Dave Garnett

Jan 1, 1970
0
Use a spring to bias against the inner stop. Pull the carriage away from the
stop with a piece of string attached to a servo (look at servos intended for
sail control for model yachts). Make sure that the string is slack at the IN
position. Control the servo with one of the servo drive units attached to a
serial port (or parallel if you want).

Dave

message
....
 
T

Terry Pinnell

Jan 1, 1970
0
Any other ideas? Thanks in advance.

I'd go for the simple servo approach too, as Dave recommends. That's
essentially the approach I used for my electronic shed lock. Here's a
circuit you can use for the servo.
http://www.terrypin.dial.pipex.com/Images/ServoUpDown.gif
The SPST 'Up/Down' switch would of course be replaced by your PC
signal.


Another method, one you're already considering I think, would be to
use microswitches at either extreme position and a cheap, suitably
geared DC motor, again with a spring tensioning the IN position. In
that case you could adapt the circuit I've used for various projects,
such as an automatic bedroom window controller.
http://www.terrypin.dial.pipex.com/Images/AutoMotorExtremes.gif
That uses a DPDT switch (C1,N1,T1 etc refers to 'Common, Normal
Transferred), and a couple of normally closed SPST microswitches.
You would again replace my manual on/off switch with a PC control
signal, either directly or via a relay.
 
B

Bill Sloman

Jan 1, 1970
0
Terry Pinnell said:
I'd go for the simple servo approach too, as Dave recommends. That's
essentially the approach I used for my electronic shed lock. Here's a
circuit you can use for the servo.
http://www.terrypin.dial.pipex.com/Images/ServoUpDown.gif
The SPST 'Up/Down' switch would of course be replaced by your PC
signal.


Another method, one you're already considering I think, would be to
use microswitches at either extreme position and a cheap, suitably
geared DC motor, again with a spring tensioning the IN position. In
that case you could adapt the circuit I've used for various projects,
such as an automatic bedroom window controller.
http://www.terrypin.dial.pipex.com/Images/AutoMotorExtremes.gif
That uses a DPDT switch (C1,N1,T1 etc refers to 'Common, Normal
Transferred), and a couple of normally closed SPST microswitches.
You would again replace my manual on/off switch with a PC control
signal, either directly or via a relay.

You need to remember that microswitches don't last forever.

In the U.K., my central heating system relied on a microswitch in its
electrically controlled valves.

The cheap-skates who designed the valve assembly used microswitches
rated for only 500 operations, which lasted for about a year. I
eventually got sick of replacing them with the original parts, and
found something that was good for 100,000 operations.

Optical forks don't wear out in the same way, but the LED emitters
lose brightness over time, and have been known to fill up with dust.

I tend to stick a magnet on the moving bit and use a stationary digtal
Hall effect sensor to detect when it gets within range.
 
S

Sam Goldwasser

Jan 1, 1970
0
You need to remember that microswitches don't last forever.
In the U.K., my central heating system relied on a microswitch in its
electrically controlled valves.

The cheap-skates who designed the valve assembly used microswitches
rated for only 500 operations, which lasted for about a year. I
eventually got sick of replacing them with the original parts, and
found something that was good for 100,000 operations.

The designers must have looked long and hard to find a microswitch
that only lasted 500 operations! :)
Optical forks don't wear out in the same way, but the LED emitters
lose brightness over time, and have been known to fill up with dust.

Yeah, maybe in 10 years of continuous operation! These are used ina
all sorts of consumer and industrical equipment.
I tend to stick a magnet on the moving bit and use a stationary digtal
Hall effect sensor to detect when it gets within range.

Magnets may loose strength over time. ;-)

--- sam | Sci.Electronics.Repair FAQ Home Page: http://www.repairfaq.org/
Repair | Main Table of Contents: http://www.repairfaq.org/REPAIR/
+Lasers | Sam's Laser FAQ: http://www.repairfaq.org/sam/lasersam.htm
| Mirror Site Info: http://www.repairfaq.org/REPAIR/F_mirror.html

Important: The email address in this message header may no longer work. To
contact me, please use the feedback form on the S.E.R FAQ Web sites.
 
J

Julian Vrieslander

Jan 1, 1970
0
Dave Garnett said:
Use a spring to bias against the inner stop. Pull the carriage away from the
stop with a piece of string attached to a servo (look at servos intended for
sail control for model yachts). Make sure that the string is slack at the IN
position. Control the servo with one of the servo drive units attached to a
serial port (or parallel if you want).

Thanks for the suggestions. The spring and string combo is appealing
for simplicity and ease of construction. I would need to make sure that
the spring does not cause the slider to creep back if the servo motor is
depowered.

I had already considered a sail winch servo. But (since we don't need
proportional control) I'm not sure this offers any real advantage over a
bang-bang controller with limit switches and a (non-servo) motor. Of
course, the servo solves the spring creep issue - as long as it is
powered on.
 
J

Julian Vrieslander

Jan 1, 1970
0
You need to remember that microswitches don't last forever.
....

The cheap-skates who designed the valve assembly used microswitches
rated for only 500 operations, which lasted for about a year

Yow. The switch in a cheap flashlight probably lasts longer than that.
Optical forks don't wear out in the same way, but the LED emitters
lose brightness over time, and have been known to fill up with dust.

I tend to stick a magnet on the moving bit and use a stationary digtal
Hall effect sensor to detect when it gets within range.

For this project I am not too worried about longevity - the slides will
be moved only a few times per week.

With a limit switch I am more worried about on/off oscillation at the
flipover point. But a good microswitch should have a crisp overcenter
transition. And I also would need to make sure that the limit switch at
the IN position caused the slider to stop at the same spot (within our
0.020" tolerance). I could do this by using a spring and string (as
suggested by Dave), or by spring-loading the nut on a lead screw. Then
I could move the slider to a hard stop, overdriving the transport
mechanism a bit before turning off the motor.
 
M

Mitch Berkson

Jan 1, 1970
0
Julian said:
Thanks for the suggestions. The spring and string combo is appealing
for simplicity and ease of construction. I would need to make sure
that the spring does not cause the slider to creep back if the servo
motor is depowered.

You could use a magnet instead of the spring. Position it so that the force
pulls the moving part in at the approprate distance. When it is farther
away the force will be negligible.

Mitch Berkson
 
S

Si Ballenger

Jan 1, 1970
0
My workgroup is interfacing some equipment on an optical bench. We have
three small elements (lenses and filters) that need to be moved, on
command from a computer, in or out of the optical path. The elements
are on ball-bearing linear slides, movement is horizontal, and it only
takes a few ounces of force to move them. All the off-the-shelf
actuators that I have found so far are very pricy, and overkill for what
we need. Our requirements are minimal:

a) IN position can be set by a hard stop, and must be repeatable to
0.02". OUT position is not critical, as long as it is about 3" away
from IN.

b) No need for proportional control or stopping at intermediate
positions between IN and OUT.

c) Transit time is not important: 10 seconds or less.

d) Ideally, each actuator would be controlled by one or two TTL control
lines from our interface.

After some googling and brainstorming, I can think of a couple of
designs. A motor and leadscrew is the obvious starting point. A small
DC motor could be driven in both directions by an H-bridge, plus a
couple of transistors or logic gates to handle the polarity switching,
limit switches to stop the motor at each end, maybe a flip-flop to keep
track of the current state.

Or I could use a stepper instead of a DC motor. HSI has small steppers
with integral leadscrews:

<http://www.hsimotors.com/linear-actuators/26000-Ncap.htm>

I found a nice stepper driver chip (Allegro 3967):

<http://www.allegromicro.com/sf/3967/>

It needs only two TTL control lines: one selects direction, the other
counts pulses. Perhaps with a stepper, I would not need the limit
switches.

Other solutions were considered. Solenoids might be too violent,
pneumatic gizmos too complicated. An R/C hobby servo uses proportional
control, but it could be rewired as a bang-bang. I even thought about
scavenging the tray actuator from a compact disk drive.

Any other ideas? Thanks in advance.

WalMart did have a $10 Black&Decker 6v cordless screwdriver that
might be useful. You could make a setup using two SPDT relays and
limit switches to operate it (simple schematic below). If you
need something strong the screwdriver could spin a threaded rod
with nuts on it attached to the sliding device. For less
strength, a plastic jar lid could be put on the end and be used
as a pully to pull a string attached to the sliding device.

http://www.geocities.com/zoomkat/switch.htm
 
B

Bill Sloman

Jan 1, 1970
0
Sam Goldwasser said:
The designers must have looked long and hard to find a microswitch
that only lasted 500 operations! :)

Having gone to some trouble to replace the first failed microswitch
with an identical part, I was a bit pissed off when I realised how
easy it was to find an interchangeable part with a better
specification.
Yeah, maybe in 10 years of continuous operation! These are used in
all sorts of consumer and industrical equipment.

When Hewlett-Packard published their first application notes on LED
life-times, back around 1975, I seem to remember being told that a LED
run at its rated current could lose half its brightness in a year.
Dust build-up depends on the environment.
Magnets may lose strength over time. ;-)

Know any application notes that talk about mechanisms and rates of
loss? I've yet to hear about any problems, but I've always used
neodinium-iron and samarium-cobalt magnets which are difficult to
demagnetise, short of heating them above the Curie point.
 
F

Frithiof Andreas Jensen

Jan 1, 1970
0
message
....

actuators that I have found so far are very pricy, and overkill for what
we need. Our requirements are minimal:

Car door actuators - about as cheap and nasty as it gets ;-)
 
F

Frithiof Andreas Jensen

Jan 1, 1970
0
The cheap-skates who designed the valve assembly used microswitches
rated for only 500 operations, which lasted for about a year.

It *could be* that you need an RC network across the switch points to
prevent arcing from destroying the switch - there are some standard "black
boxes" for that purpose, in case it is mains voltage, safety, etc.
 
J

Julian Vrieslander

Jan 1, 1970
0
message
...



Car door actuators - about as cheap and nasty as it gets ;-)

I found something even cheaper and nastier: a motorized linear
potentiometer.

<http://www.allelectronics.com/cgi-bin/category.cgi?category=search&item=
MSP-10K&type=store>

For $7, I could not resist. We have a couple on order.

We really don't need the potentiometer part of this thing - only the
motor and slide mechanism. It looks like the motor does not have gear
reduction. So I will need to see if it has enough torque to pull a
couple of ounces of (linear) friction force, and without slamming our
optical components too hard. Maybe a PWM drive signal will help with
that.
 
M

Marc

Jan 1, 1970
0
Dear sir,

We can quickly find the actuator you need at the best price on the
market.

This is how we work at PhotonicSourcing (our services are FREE for a
limited time).

Try us, you will be convince:

1) We will search for you across the 15,000 suppliers in
PhotonicSourcing database, send the request for quotation to some
potential suppliers and place it in your account on our website.
2) You will receive an email when someone is answering your request
(quotation or asking you some questions). You will need to access your
account on PhotonicSourcing. Feel free to answer their questions. All
info you need to contact them will be available to you.
3) During this process, your organization and contact name will be
available to the potential suppliers. Your email, tel and fax will NOT
be visible unless you made it available.

Let me know if you want us to move forward.

Waiting for your reply,

Sincerely yours,

Marc Duquette

[email protected]
www.photonicsourcing.com
We source any photonics, opto-electronics, imaging or laser products.
 
S

Si Ballenger

Jan 1, 1970
0
I found something even cheaper and nastier: a motorized linear
potentiometer.

How about a cordless screwdriver turning a threaded rod with nuts
on it, which push/pull some object.
 
J

Julian Vrieslander

Jan 1, 1970
0
How about a cordless screwdriver turning a threaded rod with nuts
on it, which push/pull some object.

Someone else suggested that. I took a quick look at cheap cordless
screwdrivers at Sears and Home Depot. But the ones that I saw were too
big, too expensive, and/or the chucks were designed to accept hex-ended
bits, etc. We could probably cobble up one of these things and make it
work. But for the amount of work involved, it would probably be easier
just to buy a naked motor with gear-reduction.
 
J

JeffM

Jan 1, 1970
0
The designers must have looked long and hard
to find a microswitch that only lasted 500 operations! :)
Sam Goldwasser

Maybe it's one of those companies
where Purchasing gets to overrule Engineering
--and they found such a bargain.
 
H

Hugh W. Gleaves

Jan 1, 1970
0
Dear sir,

We can quickly find the actuator you need at the best price on the
market.

This is how we work at PhotonicSourcing (our services are FREE for a
limited time).

Try us, you will be convince:

1) We will search for you across the 15,000 suppliers in
PhotonicSourcing database, send the request for quotation to some
potential suppliers and place it in your account on our website.
2) You will receive an email when someone is answering your request
(quotation or asking you some questions). You will need to access your
account on PhotonicSourcing. Feel free to answer their questions. All
info you need to contact them will be available to you.
3) During this process, your organization and contact name will be
available to the potential suppliers. Your email, tel and fax will NOT
be visible unless you made it available.

Let me know if you want us to move forward.

Waiting for your reply,

Sincerely yours,

Marc Duquette

[email protected]
www.photonicsourcing.com
We source any photonics, opto-electronics, imaging or laser products.

I may have missed soemthing, but why isn't a simple solenoid suitable
for this application?

There are only two positions (from what I can gather).

The solenoid of course has an "unpowered" position so when power is
off it returns (via spring pressure) to its unactivated state. But I
see no reason why this wouldnt do.

As for the 0.02" (approx 0.5mm) tolerance required for repeatability,
this too should present no problem, the solenoid will always be in one
of two extremes and these will always be identical no matter how many
times its operated.

Unglamorous perhaps, but suitable, I think.

Hugh
 
D

Detector195

Jan 1, 1970
0
I may have missed soemthing, but why isn't a simple solenoid suitable
for this application?

There are only two positions (from what I can gather).

The solenoid of course has an "unpowered" position so when power is
off it returns (via spring pressure) to its unactivated state. But I
see no reason why this wouldnt do.

As for the 0.02" (approx 0.5mm) tolerance required for repeatability,
this too should present no problem, the solenoid will always be in one
of two extremes and these will always be identical no matter how many
times its operated.

Unglamorous perhaps, but suitable, I think.

Hugh

They tend to move rather abruptly, which could be tough on the
delicate optics. All of the standard solutions provide the advantage
of smooth and predictable motion speed.
 
J

Julian Vrieslander

Jan 1, 1970
0
I may have missed soemthing, but why isn't a simple solenoid suitable
for this application?

A solenoid with a 3" throw might be a bit too slam-bam violent. The
actuators will be used to move components on an optical bench. There is
a biological prep with microelectrodes nearby. Impacts and vibrations
could cause problems.
 
Top