Mike Swift writes:
> To make a simple vaporizer you could use a refrigeration
> condenser or just a long pipe with fins welded to it then use a
> pressure switch to open a solenoid valve allowing liquid CO2 into the
> vaporizer when the pressure gets below say 100 psi. A second solenoid
> valve could be placed to take gas from the storage tank if it gets
> above say 125 psi.
I don’t think it is that easy. Nothing meters the liquid. It’s kind of
like saying, if someone is thirsty, put a fire hose in their mouth, with
a button controlling a solenoid valve.
The volume of 20 ft of 1/2" ACR tubing is 572 cc. Liquid CO2 expands to
gas at about 500 to 1 volume ratio. So a full gas charge of the tube is
a mere few cc’s of liquid vaporized. The make-up volume of liquid for a
bolus through the solenoid would be a fraction of a cc. If you admit,
say, 2 cc instead of 1/2 cc, you’ll get a wicked blow-off through the
overpressure relief of over a liter of gas.
In control system terms, there is a nasty hysteresis in such a system,
and what is to prevent the system from being underdamped? Something has
to meter the precise amount of liquid to restore the gas pressure when
the pressure drops. You can’t depend on the pressure differential to
meter liquid because that changes depending on the phase-change of the
liquid to gas, and the phase change lags in time because of the time it
takes to transfer enough heat to vaporize the liquid.
A naive setup will just oscillate, admitting pulses of liquid into the
vaporizer, with an overshoot of the proper amount jetting in, and gas
then burping out the overpressure relief as it vaporizes with some time
lag for heat transfer.
In principle you need a restriction on the liquid input (orifice or
capillary tube) to limit ("pinch") the admittance of the liquid, such
that the liquid flow is slow enough to follow the time lag of
vaporization without too much hysteresis. You also need ample expansion
volume (an "accumulator" in refrigeration terms) to make the system less
responsive ("brittle") to liquid input. In control terms, the correct
restriction slightly overdamps the system, but everything is still sized
to maintain the maximum design delivery.
So now I am thinking, the liquid feeds a capillary tube before or after
a constant-pressure regulator (if pressure drop is desired, not needed
if equal liquid/vapor pressure is OK), followed by a check valve and
overpressure relief, followed by evaporator tubing in a constant
temperature heat source, with some amount of accumulator volume
communicating with the evaporator tubing. Accumulator volume can be
assembled from empty high-pressure cylinder(s) with the valves wide open
or replaced with tubing fittings.
The capillary tube is sized to pass liquid at the design vapor output
flow. The evaporator tubing is sized to warm the vapor to the design
output temperature at the design output flow for the design heat source.
Accumulator volume is added as needed to bring the system into
stability.