solenoids or servos ?

     Hold currents for solenoids are far lower than pull-in currents.
It's common to use pull-in currents that temporarily overload the
solenoids, then drop to a lower holding current.  Pinball machines
and modern piano players do this. You need two power supplies, or chopping
with power MOSFETs.

     There are special solenoid-driver ICs that make this easy to do.
See "http://web.media.mit.edu/~dmerrill/mas863/assn3/solenoid_driver.pdf" .
You just turn the solenoid on with a logic level signal, and the
driver handles the high pull-in current, and the timed power reduction.
Built-in overload protection, even.

                John Nagle

some experience-based answers!

pulling the same amount of weight, friction, same

talking about 2 scenarios: that of momentary

that amount of weight for about 30 seconds or so. I

both.

When you say 2 power supplies are needed, you mean one for the initial "pull-in"
and one for the holding ? I took a *quick* look
at the data sheet you linked to and only saw one power supply. Having seen that
it uses PWM, I wonder if PWM can be used to
control how far in a solenoid is pulled in ?

But what about the comparison between servos vs. solenoids ? Which ones pulls
less current throughout the overall range of
motion - meaning which one would run down a battery first, all other parameters
equal ? For example, lets say we're designing
robotic gripper that uses some sort of spring to compensate for whether it is
gripping something large or small, and just
"hammers" it every time. Which would be better, in your opinion ? solenoid or
servo ?

Thanks!

Sorry, no experienced based advice from me.  But, here is some
thinking on my part... I observe that solenoids are rarely seen in
robotics.  So, I think that is your answer.  The question is, why?

There are obviously some advantages to servos. By the simpicity of
their design, they are inherantly more reliable, and cheaper.

While most solenoids are on/off type devices, there are proportional
types as well. But, most are the "hammering" type you mention.

Some solenoids are latching... a magnetic force holds it in position
without power until a reverse current is applied.

I just thought of a large scale installation of "robots" that uses
solenoids.  I believe the Tiki Hut in Disney uses solenoids for the
birds mouths. This quick motion of a light-weight object is one good
application for solenoids.

And another application comes to mind... actuators for ultra-light
indoor R/C Aircraft.  They should be considered rotary solenoids.
These are often home made coils with a magned mounted to pivot in the
middle.  Not just, on-off, but proportional control by controling the
current provided to the coil.

In the above two examples, the solenoids are fast acting, but don't
use the mechanical advantage of gears, so they are weak.

If I were to take the rotary solenoid used in the R/C Aircraft, and
attach a gear to the magnet, then repeadly turn it on and off, I will
have turned into a brushless gear motor.  Conversely, if i take a
purpose-built brushless gearmotor, then remove the gears, and only
actuate it by applying a DC current, it will function like a rotary
solenoid.

The same thinking can be done with a linear solenoid.  If I stack a
bunch of linear solenoids, and put one plunger in it, and then actuate
the coils in series, I will have made a linear motor.

So, with the above thinking, they two devices are essentially the
same, and therefore are likely to share the same force and efficiency.

The difference is if they are designed to actuate over a large range
of motion, or a short range.

Joe Dunfee

I think you are probably right!

After I started this post, I realized that I really only had one application in
mind (the gripper) and that everything else I was
considering needed some kind of proportional control - more than an
all-or-nothing hammering action.

Thanks!
JCD

Check out: http://users.rsise.anu.edu.au/~roy/SMA/
Larry

Solenoids are mostly applied for ON-OFF operations and SERVOS for
position control of the valves.
The question here is then, do you need position control of the
plunger ?

In robotics, we do apply solenoids for gripper applications which are
mostly ON-OFF and usually on pneumatic power for simplicity reasons.
On the other hand, we do apply servo systems on hydraulic driven
robots where the actuators (piston  in cylinder) are moving the robot
links. Hence, to obtain position control on the actuators, the
pressure is balanced between both sides of cylinders using those
hydraulic distributors which are fitted with servos.

Hope it helps,

LHR

    Position control with solenoids is possible, but rare.
It is sometimes used in proportional hydraulic spool valves set
up for electrical actuation.  See
"http://www.continentalhydraulics.com/proportional_control_valves.aspx" .
It's not easy to do, because standard solenoids have a pull-in force
that varies with the position of the slug in the coil.

(Don't put too much effort into this; the original poster was
more trolling that seeking a useful answer.)

                John Nagle

Thanks!
JCD

   If you want linear position (or more properly, force proportional
to current in which the proportion varies MUCH less with position than
in a solenoid), you may want a voice coil actuator. I think all modern
hard disk drives use rotary versions of this, and I took apart an old
(really old) "full-height" 30MB AT-clone hard drive that had a linear
voice-coil actuator to move the heads.

   This brings to mind the idea of using disk-drive head actuators in
a gripper. It would be light, fast, and easily able to apply a
variable force, but would still need sensors to detect exact position
and force.