Years past, I worked on many water cooled tube type
amplifiers/oscillators. Compared to those, water cooling a SS amp is
outlandishly simple. Route the Cu water tube through a milled slot in
the heat spreader. The most complicated part is how well you want to
monitor the water flow and temp. Typically a restrictor is placed in the
input line and the pressure is measured on both sides of the restrictor.
Due to the heat spreader being at chassis ground, heat sensord can be
directly attached to the spreader plate.
This has the advantage that a finned cooler can be added, so if the
water flow stops, the amp can still be fan cooled
73, Roger (K8RI)
On 2/14/2017 2:07 AM, John Lyles wrote:
I've been using water cooled pulsed RF amplifiers since 1995 at work.
The first uses were DMOS MRF151G variants running @28 VDC. In the last
3 years we have moved those out and gone to 48 volt LDMOS, far less
devices to get 5500 watts peak at 200 MHz.
I know water-cooling has been discussed here before, but I agree that a
well designed water-cooled SSPA has the potential to gain some
traction in
the amateur market if the $ make sense. That said, I think it is
unrealistic to expect one of our ham radio manufacturers to offer an
"inexpensive" water-cooled SSPA for amateur use. The added complexity
and
potential maintenance might be enough to put the water-cooling
concept on
"ice" from a ham radio manufacturing point of view.
.....
So far, this is about the closest I have seen to getting to where the
water-cooling needs to go (at least from another ham). I suspect in the
ISM/radar/broadcast world, water-cooling SSPA's is fairly
straightforward.
.....
In addition to solid state RF for driver stages, we use a number of
different tubes in plain water cooled and hypovaportron cooled.
The difference is that we have already got a huge deionized water
plant, so air cooling is generally avoided for all stages. The
amplifiers are cleaner inside.
John K5PRO
A big difference between vapor-phase and liquid-water cooling is that
vapor
cooling is much more efficient per volume unit of water passing
through the
system. From the Dick Ehrhorn's description on cooling efficiency:
"In typical closed-loop liquid-water-cooled systems, the maximum outlet
(hot) water temp must be held well below 100 deg C to avoid hot-spot
boiling
on the anode surface, which can and does create steam bubbles, which in
turn "insulate" the hot spot from the water so the hot spot gets even
hotter. This typically creates a temperature runaway and may lead to
destruction of tube and/or cooling components. Typical inlet (cool)
water
from the chiller may be specified as </= say, 45 deg C, and maximum
outlet
water temp as 80 deg C to avoid spot boiling.
Anyway, in this example each gram of cooling water passing through the
tube's water jacket can absorb not more than (80-45) = 35 calories of
heat.
Conversely, each gram of 45 deg C water entering a vapor-cooled tube's
boiler absorbs approximately [540 + (100-45)] = 595 cal/gm while
vaporizing. So, vapor cooling requires passing only about 35/595 =
1/17 as
much water
volume through the system as does water cooling."
Paul, W9AC
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|