With cooling systems like the MOSFETs (tubes in a copper spreader)
electrical conductivity isn't a problem.
Unless your radiator is outside and the plumbing is exposed to freezing
temperatures, straight distilled water is your best bet. Plain water
has better thermal transfer characteristics than the glycol/water mix.
As long as you are using nonconductive plumbing (rubber/plastic) hose
to connect the thermal "plate" to the radiator and pump, there should
be little or no electrolysis even if one is using a copper thermal plate
and aluminum radiator or cast pump.
If the radiator (and/or sump) is outside and exposed to freezing temps,
by all means use 50% water and 50% ethylene glycol.
Other than in the south - particularly where the transmitters were run
24/7 - every water cooled broadcast transmitter I've ever seen (and I
installed, operated, serviced and/or consulted on a substantial number
of transmitters/sites over the years) uses 50/50 distilled water/glycol.
In fact, the only transmitter that did not use glycol did after I fixed
the cooling system <Doh>!
73,
... Joe, W4TV
On 2019-10-08 1:26 PM, Jim Barber wrote:
Hi Bob, thanks for the reply.
In this particular case, the electrical conductivity of the coolant doesn't
matter as long as the cooling efficiency stays up and
the corrosion doesn't become an issue for the radiator, the baffle block or
the pump. (I assume you meant electrical conductivity, not thermal)
Would that change the equation?
Anything that contributes to the conversation and gets people thinking
contributes to the SNR... I'm no expert when
it comes to these kinds of cooling systems, so can use all the help I can
get.
Thanks!
Jim N7CXI
On Tue, Oct 8, 2019 at 9:58 AM Bob Darlington <rdarlington@gmail.com> wrote:
Jim,
Plain ol' distilled water will remove heat a lot better than diluting it
with glycol, however the chemistry may dictate using it. Deionized water
is much more corrosive than distilled even if it starts life as a very poor
conductor, but will quickly leech metal ions into the solution (although
not much volume!) and will likely become more conductive faster than
starting with plain distilled, but it might be a photo finish as to which
needs to get replaced first. Just an FYI, and I hope this has a slightly
beneficial SNR. If not, I apologize for taking up the space.
-Bob N3XKB
On Tue, Oct 8, 2019 at 10:22 AM Jim Barber <barberaudio@gmail.com> wrote:
I'm laying out a 2-module MOSFET amp, and intend to use liquid cooling.
The last time I visited the topic, ethylene glycol antifreeze diluted with
the traditional 50% deionized water was the coolant of choice.
Anything new under the sun? I'm using propylene glycol in my laser tube
and CNC spindle coolers, but those don't have to be very efficient. The
cooling system for the amp is going to have to move some heat.
Thanks,
Jim N7CXI
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|