[Amps] Getting rid of blower noise - an alternate cooling scheme using the phase transition of liquid carbon dioxide

[Paul Marbourg]

Dear Amplifier Group,

Another possible approach that could be used to control or almost eliminate
the noise generated by air cooling an amplifier is to eliminate air as the
working fluid used to transport heat away from the thermal envelope of the
vacuum device.  The most obvious commercially-utilized alternate approach
for cooling vacuum tubes is liquid or vapor phase cooling.  

The preferred approach is to utilize a change in phase during thermal
transport, as the vapor phase transition heat capacity of a working fluid
such as water undergoing a phase change is VASTLY higher than the heat
capacity water with no change in phase.

Given:

Water takes 1 calorie/gram-degree C in a liquid state

Water takes 540 calories/gram-degree C for heat of vaporization

Obviously a state change is desirable for efficient thermal transport.

Okay, since most amateur radio amplifiers are designed to utilize air as the
thermal transport working fluid, changing the design over to water-based
vapor phase cooling may be considered impractical or very technically
challenging.  I therefore propose the use of an alternate working fluid for
vapor phase transition cooling.

The alternate working fluid must posses several important properties:

1.  Little or no electrical conductivity (dielectric constant)
2.  Non-corrosive to working environment
3.  Non-toxic to humans
4.  Ease of implementation to modify existing design
5.  Low cost of material

(to name a few properties).

There is one substance which immediately comes to mind here: carbon dioxide.

Carbon dioxide is a cheap, widely available liquid at 300 PSI and room
temperature.  It is not something you would want to allow to accumulate in a
closed environment with humans present, but is non-toxic in low
concentrations.  The material is easily contained by operating the amplifier
in a fume hood or a sealed envelope environment or by using a fan external
to operating shack to extract the spent gas.  

To convert an air cooled amplifier to cooling with carbon dioxide:

Obtain a large supply of liquid carbon dioxide.  This material is commonly
used for carbonating beer and other beverages and is available in
conveniently sized cylinders for this purpose.  Install a small
solenoid-controlled valve on the amplifier, and mount the CO2 cylinder in an
inverted position to allow liquid CO2 to flow out of the cylinder to the
solenoid valve. From the output side of the solenoid, install a spray nozzle
to the underside of the tube mounting socket. A phase-transition diffusion
plate may be necessary to allow for an even flow of CO2 around the vacuum
tube.  Install a infrared "spot" transducer in the amplifier to measure the
tube's anode temperature by remote IR beam.  Build a small
microprocessor-based control circuit to actuate the solenoid valve
controlling the CO2 flow rate. Several IR transducers could be incorporated
for microprocessor data polling to insure fail-safe operation of the cooling
system.   A simple analog differential amplifier control circuit would
probably work here also for CO2 injection cooling with proper control
hysteresis designed in place.  It will be necessary to adjust the
flow/expansion rate of the CO2 working fluid by selection of the proper
spray nozzle geometry to prevent/minimize formation of solid phase CO2 (dry
ice) around the vacuum tube device.  

Advanced implementation of this design could incorporate a closed-cycle loop
for vapor phase cooling with carbon dioxide as the working fluid.
Obviously, if transitioning the design to a closed-loop cycle where
consumption of the working fluid is avoided (as well as atmospheric
contamination), a vapor phase cooling medium such as one of the
halo-methanes could be implemented.  Any associated compressors required for
a closed loop system could be located outside the operating environment of
the amplifier to minimize noise exposure to the operator.

Because the consumption of CO2 in an open cycle cooling system for a high
power linear amplifier would be directly based on the thermal cycle loading
of the tube in use, intermittent amateur radio power cycling of the
amplifier would allow even a small cylinder of liquid CO2 to last for
perhaps many hundreds or even thousands of hours of use in cooling the
vacuum device.  This is due to the storage of CO2 as liquid under pressure
in a common cylinder and the CO2's excellent heat capacity (136 Cal/gram
heat of vaporization) when undergoing a phase transition from liquid to gas
around the amplifier's vacuum tube.

At a heat of vaporization of 136 calories per gram of CO2, CO2 has only 25%
of the heat of vaporization as compared to water, but this is still a very
significant increase in thermal transportation capacity over any means of
liquid phase cooling.  For this application, it is helpful to envision a
vacuum tube surrounded by a gently flowing supercritical cloud of liquid/gas
carbon dioxide throttled by the microprocessor controlled solenoid valve to
provide efficient and almost silent cooling in manner tailored to maintain
the tube within its safe operating temperature range.  As a side benefit,
one's beer could always be properly carbonated as well.

One caveat here, adequate ventilation is essential to the proper use of this
cooling scheme.  At NO TIME should the percent of atmospheric CO2 be allowed
to exceed 10% in the operating position.

73

Paul WN7T