We seek to understand and document all radio transmissions, legal and otherwise, as part of the radio listening hobby. We do not encourage any radio operations contrary to regulations. Always consult with the appropriate authorities if you have questions concerning what is permissible in your locale.

Author Topic: Liquid cooling solid state  (Read 5407 times)

Offline RedMercuryLabs

  • Jr. Member
  • **
  • Posts: 82
  • Earth
    • Yahoo Instant Messenger - redmercurylabs
    • View Profile
    • Red Mercury Labs
Liquid cooling solid state
« on: August 02, 2013, 1646 UTC »
We're working on a liquid cooled solid state mobile rig.  Anyone ever liquid cool?  Just throwing it out there...chum in the water.
On a clear blue sea...
Yoshi 5000 Transmitter
Doublet hanging from the yardarm
Saxton 8322  RG 223/U  Feed
RV Red Mercury
"I don't post much, but when I do, I prefer HFU"

Offline jFarley

  • Marconi Class DXer
  • ********
  • Posts: 5922
  • near Chicago
    • View Profile
    • Email
Re: Liquid cooling solid state
« Reply #1 on: August 02, 2013, 1935 UTC »
RML;

I work in a small shop which uses extensive RF Induction Heating in the manufacture of product.  We have about 75kW of total RF available in a about a dozen distributed systems which range from 2-5kW output, 10-500kHz.  Since the end of life for one of these units is generally catastrophic, we use very cheap Chinese imports which are really intended for low duty cycle brazing and welding operations.  We run them flat out in 6 hour long continuous runs, 18 hours a day, and because of this, we generally have to beef up the cooling before they go into service.

Solid State RF Induction Heating units share many of the problems a transmitter or linear have, and many components need to be cooled including tuning caps, large ferrite load transformers, work coils, and power supply.  The real killer is keeping the switching transistors cool.  Most of these units come with very flimsy liquid cooled heat sinks, and I have been replacing them with this:

http://www.mcmaster.com/#catalog/119/535/=nw4vig

This is a simple off-the-shelf solution for liquid cooling solid state components.  It's McM so it is overpriced, but I have no doubt that similar units can be found online for cheaper, or you could DIY for cheaper still.

It's pretty easy to liquid cool anything; the problem and expense come from getting rid of the waste heat.  We use a number of methods at work:

1)  Water to air (pump, radiator, fan)
2)  Water to water (pump, heat exchanger, process chiller)
3)  Pump 'n Dump (city water through the units then down the drain) in emergencies

Water to air is the easiest for portable operation, and would likely work if you carefully shop components (the costs can mount quickly) or if you have a shotgun full of nickels you can shoot at the problem.  It all depends on the device temperature you need to maintain at anticipated power output and the ambient temperature.

A couple of years ago I was looking at using Peltier devices for cooling parts of the product that were getting too hot in the RF field.  (I have since dropped that ball.)  They have come a long way, and I was impressed by the Qmax figures (max heat transfer for the device) for some of these guys; up to 120 Watts in a small package.  Maybe an alternative?
Joe Farley, Near Chicago
SDR-IQ / R8 / R7
Remote Resonant Loops for HF and LF / ALA 1530
Active 60" Whip / PA0RDT
QSLS appreciated to:    jfarley44@att.net

Offline redhat

  • DX Legend
  • ******
  • Posts: 1585
  • USA
  • Music is my drug.
    • View Profile
    • Email
Re: Liquid cooling solid state
« Reply #2 on: August 02, 2013, 1938 UTC »
They've been doing it in Europe for many years.  Ever tightening EMI restrictions mean that in many cases, liquid is the only option.  Many of the products from Bext and its many clones (RVR, OMB...et al) have had the option.

For small projects quoted in some datasheets, you have a conventional heatspreader (a 1/4" plate of copper) with tubing soldered to the back of it.  Many of the RF Plasma sources used in semiconductor manufacturing and wafer fabs use this to keep a small footprint.  Imagine a 3KW 13.56 MHz transmitter with three phase input about the size of a shoebox.

FWIW, a local TV station had a water cooled dummy load they used cictern water to cool.  When they were done with the water, they just dumped it back into the hole.  I was thinking of doing the same here.  From time to time there are a few KW's that need to go.. somewhere.

+-RH
« Last Edit: August 02, 2013, 1947 UTC by redhat »
Somewhere under the stars...
Airspy HF+, MLA-30/Mini-whip/Chi-Town Loop
Please send QSL's and reception reports to xfmshortwave [at] proton [d0t] me

Offline Pigmeat

  • Marconi Class DXer
  • ********
  • Posts: 6684
    • View Profile
Re: Liquid cooling solid state
« Reply #3 on: August 07, 2013, 1855 UTC »
I saw part of the former amp of a notorious high powered pirate years ago. Solid state device mounted on a copper plate with with a copper tube attached to the back, as stated above.

I was told there were several of the devices in the final, all water cooled. When the guy quit pirating he parted sections of the amp out.

Offline ChrisSmolinski

  • Administrator
  • Marconi Class DXer
  • *****
  • Posts: 31106
  • Westminster, MD USA
    • View Profile
    • Black Cat Systems
Re: Liquid cooling solid state
« Reply #4 on: August 07, 2013, 2010 UTC »
A former co-worker of mine was an RF engineer in Egypt a few decades ago. He said their transmitters, Soviet supplied, were steam cooled. This would likely have been megawatt class MW transmitters.
Chris Smolinski
Westminster, MD
eQSLs appreciated! csmolinski@blackcatsystems.com
netSDR / AFE822x / AirSpy HF+ / KiwiSDR / 900 ft Horz skyloop / 500 ft NE beverage / 250 ft V Beam / 58 ft T2FD / 120 ft T2FD / 400 ft south beverage / 43m, 20m, 10m  dipoles / Crossed Parallel Loop / Discone in a tree

cmradio

  • Guest
Re: Liquid cooling solid state
« Reply #5 on: August 07, 2013, 2021 UTC »
Yep! Used a cooling block from a bartender's station used to cool syrup, then a scavenged evaporator from an air conditioner, added a pump, distilled water and away I went! ;D

Peace!

Offline radiogaga

  • Sr. Member
  • ****
  • Posts: 344
    • View Profile
Re: Liquid cooling solid state
« Reply #6 on: August 08, 2013, 0013 UTC »
ice ice baby  ;)

Offline Rockpicker

  • Hero Member
  • *****
  • Posts: 565
  • Middle of No Where Montana
    • View Profile
    • Email
Re: Liquid cooling solid state
« Reply #7 on: September 10, 2013, 0437 UTC »
There are more than a few "All in One" closed system "water" cooling systems for CPU's these days.
For example
http://www.newegg.com/Product/Product.aspx?Item=N82E16835103181
Better Examples
http://www.xoxide.com/watcoolkit.html
http://www.frozencpu.com/cat/l1/g30/Liquid_Cooling.html
Or maybe a Peltier
http://www.frozencpu.com/products/2409/exp-02/80W_Potted_Peltier.html?tl=g30c105
One or more of these, maybe some copper bar stock from Storm Copper, and some good thermal paste?
http://www.stormcopper.com/?gclid=CJyy7ZOBwLkCFWxyQgod2TEAYg
Lots of stuff out there for the DIY crowd.
In reference to RML's question for a mobile app., "desktop" cooling components generally require a nominal ~12 DC (~ 9 to 13.6 volts)
As far a noise - Acoustic, I used Noctua brand case fans (and heat sinks) in my last to builds and love them, very quiet and they come with a selection of resistors
to slow the fan down. (downside - pitiful ugly colored plastic)
As far as electrical noise, I would avoid any of the available add on fan speed controllers as they are PMW
units and would likely introduce "noise" in your project.

« Last Edit: September 10, 2013, 2042 UTC by Rockpicker »
Central Montana
Winradio G31DDC Excalibur, YB 400PE,  Satellit 750
RF Systems MLBA with 85' of copper flex weave

cmradio

  • Guest
Re: Liquid cooling solid state
« Reply #8 on: September 10, 2013, 0747 UTC »
There are more than a few "All in One" closed system "water" cooling systems for CPU's these days.
For example
http://www.newegg.com/Product/Product.aspx?Item=N82E16835103181

Hmmm, I wonder how efficient that thing is and what there is for electrical isolation.... can I slap this straight to the anode block for a conduction cooled transmitter tube and not get spitzensparken?

Peace!