Jaggy said:
The air pump I will have to use is a 2 to 3 watt aquarium pump which indeed
uses a small solenoid which vibrates the bellows.
I tried to run one of these things off of a regular $5 dimmer and it worked
fine, but as I said 'momentary' performance does not impress me when failure
means having to fly to Europe to fix it, that is where my wont for a
reliable solution comes in.
Using the aquarium pump continously and bleeding off the excess is an
interesting thought.
Controls like this, either a valve or the pot of a dimmer, see an incredible
amount of abuse in a science museum environment and the final solution
should be the one which results in the most rugged 'visitor interface', and,
in spite of its perceived ruggedness, still the one the maintenance people
can easily fix.
If needed I can get around the minimum load requirement by using an
additional dummy load resistor in conjunction with the regular dimmer.
I am aware of PWM but that doesn't seem to be a solution in my case.
What is the variable frequency drive Chris is referring to??
Thanks for the postings so far, this is helpful to me
Uwe
Hi, Uwe. A couple of points:
* Sorry -- the 20 watt variable frequency drive business was a bit of
a joke (note the "winky" ;-) ). VFDs are frequently used for speed
control of multi-horsepower AC motors.
* You've got a point -- if the museum patrons are gooig to be tweaking
this continuously, a small needle valve might not be a good idea. They
wear out fairly quickly. In addition, they can be ruined easily by
cranking them past full off.
* I did a quick Google scan, and saw a couple of manufacturers of
aquarium air pumps that have models with electronic air speed controls.
These could possibly be the R+pot-C/DIAC/triac controls in the cheapie
lamp dimmers, and one description mentioned that it had a bellows-type
motor. So, in fact, what you're talking about with a dimmer-type
controller might be possible.
* One issue with these lamp dimmers I didn't mention has to do with
the asymmetry of the DIAC triggering voltage. This results in a DC
component in the load voltage. While this isn't a problem with lamps,
it can cause issues with motors or inductive loads like solenoids. It
heats the coil, and can result in damage. It's something that has to
be taken into account.
* I believe a power resistor in parallel with your pump is a good
idea, but you shouldn't need more than a 680 ohm 25 watt resistor for
120V, or 1500 ohm, 50 watt for 240V. Since lamp dimmers are in series
with the AC load, this should provide enough load current to make sure
your dimmer works properly.
* You mentioned that this business might be shipped to another
country. You should be aware of voltage differences (if you get a
120VAC pump and send it to a country with 240VAC, the pump will have a
very short, warm and exciting life. Also, if you're building it in a
60 Hz environment, and you ship it to a 50Hz environment, you'll have
less air volume. Remember, bellows stroke frequency will be dependent
on line frequency).
* Given that this is going to a museum, you've got a couple of more
issues, if the museum patrons are going to be playing with this
control. In many countries, exhibits with electrical controls may be
classified by the local regulatory agencies as machines. If so, you
may have to deal with the requirement for low voltage (24V max.) panel
controls. This would obviously mean a line voltage lamp dimmer with
line voltage on the pot would be out.
* You also have to deal with the potential for patrons to be downright
malicious, at least by accident. Can you imagine what would happen if
a 3rd grade urchin spills a Juicy Juice on a typical dimmer control?
Bzzzt. Ouch.
* From a practical standpoint, too, you have to look at the life cycle
of the pot or dimmer control you're using. Most dimmers are basically
made to be tweaked a couple of times a day for 10 years or so, meaning
about 50,000 cycles. One busload of 3rd graders could run through that
in one day, especially if your exhibit is fun and exciting to more than
one of the little urchins. Imagine the delight of the museum tech
having to change the dimmer every month. You might want to look at an
AC control with remote pot, and use one of those Allen-Bradley machine
panel potentiometers with washdown capability. The pot itself is
mounted behind the panel mounted shaft, and joined with a flexible
coupling, which eliminates the side-to-side torque that usually results
in premature pot failure. Since the only moving part is on the front
panel, it should be fairly easy for the museum tech to replace. I
guarantee that this will be a predominant failure mode here. Get the
one with the small terminal block under the pot for the three wires.
You can buy or borrow a lamp dimmer and actually try your idea.
Measure the DC resistance of the solenoid coil. Then, after you hook
up the parallel load resistor, run it straight off line voltage for a
while with the cover off. Get an idea how hot the coil becomes at full
voltage. An educated (but careful) finger is probably good enough
here. Then hook up the lamp dimmer in series, and see what happens.
Measure the DC component by putting a DVM across the load, set for
highest range DC. Measure maximum DC voltage (this will usually occur
between 5% and 40% of maximum control voltage). Figure out how much DC
power is going to be dissipated. Try leaving the dimmer at that
setting for twenty minutes, and then see if the coil gets hot. Try
several power settings, and see. See if yuo can make the thing stall
and get stuck at lower voltage. It's a start, anyway. After you've
done this, you can look at different types of AC controls that will
work for you. There are some which work with inductive loads, some
which have forgiving minimum current requirements, and can be used with
an external speed pot. There are also others which provide an isolated
low voltage for a remote control pot which complies with EU strictures
on low voltage controls. I'd need a lot more information before
specifying something myself.
<dreamland>Of course, if I wanted to nuke this problem, and money was
no object, only reliability, I'd use a 24VDC motor driving a
peristaltic pump, and use a PWM motor controller with the Allen Bradley
washdown machine panel pot as the user control. You can get everything
you need except the Alen Bradley pot, including the speed control, from
the Cole-Parmer catalog. Make sure you use the most expensive, most
durable tubing available.</dreamland> (This is very expensive, and
should be viewed as humor, unless you've got a lot of cash to spend.)
Where I live, the local science and industry museum has a tech
department, and has people who are paid to do this kind of thing. I
know if I had a problem of this type, I'd want to call them up and ask
a few good questions. They also may be able to come up with more
gotchas off the top of their heads. If you're a member of your local
science museum patron support group and you ask nicely, you might be
able to buy one of the techs or engineers a good lunch, and in
exchange, be able to ask a few questions and pick their brain about
your project.
Good luck
Chris