I know I am bumping a very old thread but I just wanted to make sure this info is archived for the future experimenters out there.
First of all let me start out by saying I tried cooling of the LNB and it does work!
Unfortunately it is totally impractical to try this on the hobby level! And ended up costing me about 50 hours of work over a two week period and was one of the most frustrating endeavors of my life.
About 5 years ago I dumped about $1000 USD into this DIY project based on this threads info in order to see if I could bring up the signal on some transponders that would drop out in the summer months.
My first test involved cooling the LNB with DRY ice and I had some really good results. QC levels on those dead transponders rose up from zero to levels that made them usable, but of course the joy only lasted for a short period and as the ice went down the signals went, but it was proof of concept that gave me a lot of excitement and made me go to unbelievable lengths to make something that would work on demand.
I started by ripping off the plastic shroud around an LNB and wrapping the metal body with copper tubing used in AC work, I also used thermal paste to enhance the contact of the metals. I installed a small temperature probe near the coil and ran the wire out. I then covered the whole thing over with insulation tape so as to keep the cold in. I was actually able to Dremel out enough of the plastic LNB case holder bits that I could get the case back on and close it back to it's original form and sealed it with epoxy. The two copper tubes came out of the LNB, one was the full size tubing and the other was the very fine tubing used as an expansion valve to carry the liquid in an let it expand into a gas in the tubing wound around the LNB.
The next thing was to build an AC compressor system and using I think it was a 1/4 HP compressor designed for R22 gas and bought the condenser assembly along with a cooling fan and wiring, dryer etc. I made a case out of sheet steel and the platform base was drilled and brackets etc installed. I then welded everything together and then using about 20ft of copper tubing sheathed in insulation I welded it onto the tubes coming out of the LNB.
First test resulted in mixed results. I could only get down the temperature to about -5C deg with an outside ambient temp of about 29C. I consulted a friend in the AC business and he said the evaporator coil was way to small in the LNB and that I would need either much more tubing or I could use a "expansion tank" at least I think that's what it was called. Basically a canister that held about a quart in volume that was inline with the gas to hold more of the gas.
I added the tank into the system and was able to get down to about -15 deg and started to see some of the better transponders start to tickle the receivers meter, but still no real success. and I aborted after 15 -20 minutes.
I talked to my friend again and he said that I could use a hotter AC gas but it would damage the compressor in a few weeks as it was not designed for that Gas. I do not remember the number of the gas but at the time I thought why not, if it works I will just buy the right compressor for the job.
The third test:
I fired up the unit with the new AC Gas and the temperature came down to -30C and the transponders all started to kick in! Myself and a friend who was helping got out the beers and started to celebrate. Then about 20 minutes into the operation the signal started to drop even though the temperature was still at -30C !! We shut her down and and started to check a few transponders that normally had a good signal and we saw Zeros across the board!
The Autopsy:
During the time the LNB was operating we did see water dripping off of the case but figured that all was well as the LNB's electronics were sealed tight. So I pulled apart the LNB and noticed removed the insulation, the tubing and cracked her open only to find water condensed all over the circuit board.
The Fourth test:
Using a brand new LNB I modified the hell out of the sealing system. I then packed drying bead pouches into any free space I could find and then I buttoned her up and wrapped back the coil and put everything else back into place.
We fired her up and we got a signal and it lasted for about 30 minutes and then it slowly collapsed again.
The Fifth test:
I decided to go Nuclear with solving this condensation problem. My first thought was filling the LNB with nitrogen gas and having a line that kept some positive pressure from a small nitrogen cylinder on the LNB so as small amounts of gas escaped over time the tank would keep a small amount of pressure to keep air from getting in.
I drilled a hole into a new LNB's metal case and threaded it and added a line connector that could be run to with a very thin 2mm tube to a regulator on a gas cylinder. This all ended up being for naught as it was impossible for me to regulate the pressure down that small. I kept blowing off the LNB's front cap and eventually abandoned the idea.
I moved onto phase two which was to use an expensive variable fish tank pump to move air through about a quart of drying beads in a clear plastic cylinder. I used a very accurate electronic pressure switch to turn on/off the pump. The output from the drying beads would go to the same Nitrogen line and push dry air into the LNB as small amounts of air leaked out. I left the pump system running and every 30 minutes or so the pump would click on for a second or two. I left this going for about 2 days to make sure the air inside the LNB was as dry as a bone!
The Sixth test:
I fired up the system and low and behold with a few minutes the signal came up as expected. I sat there watching a small TV next to the dish monitoring all the parameters. I got past the 30 minute mark and things were looking good, I got to the one hour mark and I was ecstatic! Then I started to notice that water was dripping off the LNB's body at an accelerated rate. Soon water was all over the front dome and at the same time I watched in horror as the signal just started to drop and drop. I would wipe the front and the signal would come back up but within minutes condensation was everywhere and the signal would drop again.
The Seventh and final Test:
At this point I came up with the idea of using a hooded 2" high speed muffin fan to blow air over the LNB's plastic dome in order to keep the water from condensing and to hopefully blow it off and dry it off faster than it could gather.
I fired her up again and it worked
One hour passed then two and she kept on going. Then after about 4 hours the signal slowly started to go down
After about 15 more minutes it winked out of existence and my heart sank.
The Final Autopsy:
I pulled the LNB apart and as expected I found small amounts of water had condensed on the circuit board. At this point I gave up and shelved the project
Conclusion:
Anybody who is going to try this I want you to know that it can work but your going to have to use a Nitrogen filled LNB and your going to have to make sure that there is no way for the gas to escape or your going to have to keep it pressurized with a much more complicated system than I used. Also the humidity is going to cause a crap load of water appear everywhere, even things like the RG6 connectors need to be properly sealed with silicone etc.
I know they have commercial units that do this kind of thing but they use wave guides and the LNB's are indoors in sealed compartments. I hope this helps someone out there that might want to try this, you can learn from my mistakes and hopefully get around the problems.
RJG