I'm doing some casual experimentation at home and would like suggestions on building a
rotary tank to study fluid motion. I have many questions and probably could have gotten a
few of them answered on my own but what the heck. I have no experience in metalwork,
though I have experimented with brushed DC motors and PWM controllers.

Kalliroscope is a type of rheoscopic fluid used to study fluid motion.
For example:
http://www.kalliroscope.com/matisse/index.htm
click on Kalliroscopes / Revolving Kalliroscopes

A circular tank about 6-8" diameter by 3-4" deep, is mounted to a shaft supported
by a suitable bearing block assembly. It will be filled with water and kalliroscope or
other particulate matter, driven by a PC controlled reversible DC motor. The tank could be
a cylinder, but prefer a hemiphere. The problem is finding a hemiphere that can be mounted
accurately to a shaft. Perhaps a metal spinning co. might have some leftover I could work
with, however haven't yet asked around. If I had to have one built, what would be the ~
cost of a hemiphere, aluminum or stainless steel, 6" dia x 3" height with an additional
1" height vertical wall section, with a boss or mating surface at the bottom that would
allow accurate placement on a flange or shaft? I imagine the less TIR, the more expensive.
Wall thickness can be whatever is necessary given it only has to contain water. If it's
stainless steel then it can be similar to a kitchen mixing bowl. Max speed is around 200
rpm or whatever that doesn't cause the water to be thrown out of the tank.

Coupling the motor to the tank. Should it be belt driven, or direct using a flexible
coupler? Thinking of low speed DC servo motors in the ~300-550 rpm range at
www.surpluscenter.com and perhaps they can be direct coupled? Or should I consider a
higher rpm motor and reduce the speed thru a belt drive? Since tank rotation will be
reversed frequently, I'm concerned with backlash of a coupler. I've seen zero backlash
couplers but would then have to adapt shaft sizes, unless there are some available with
different sizes on each end. But I guess it depends on what the motor shaft and tank shaft
diameters happen to be. Now if I had one of those 7x12's...

Another thought (though seems insane) was using something like this:
http://www.etech4sale.com/commerce/c...roduct_id=1308
but the max rpm was I believe 63 rpm for this model. Other models go to about 220 rpm or
so but were not on sale. However I'm not at all familiar with servo actuators, and if it
would be worth it/too much trouble to adapt. The controller is expensive. Seems like
overkill at 25 lbs. and they refer to it as compact, which kinda strikes me funny. Of
course what am I thinking - I'm talking about a 6" diameter tank on my desktop. Will a
servo actuator of that type allow for programming a smooth change in velocity without much
trouble? What other smaller servos with bearings suitable for direct mounting would be
reasonable for this application? A surplus servo and controller would be ideal but I need
help in finding the right one that won't cost an arm. Having one that can take the tank
directly would save a lot of work.

All this just to have the ability to repeat the rotary motions in case a particular fluid
pattern interests me.

I appreciate your help.

Ben

--