I've been thinking about what Paul said in the above post, and I have come to the conclusion that he is quite correct in saying that refrigeration design is very complex. As he said, it calls for "extensive back ground including the laws of thermodynamics, physics, chemistry and electricity theory".
He is quite right. But if we start with an already-designed unit and utilize as much of the previous design work for our 'modified' purposes, I think we can still get some very useful work done.
My dad used to tell me the story of the Bumble Bee... that aeronautical engineers had carefully studied that creature and had concluded that because of many factors, especially the weight to wing-area ratio, the Bumble Bee is incapable of flight. But the Bumble Bee didn't know this and he flew anyway!
So, I think that we're all a bit like that Bumble Bee...
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> >
I've been seriously scouring the Internet for information on selecting a brazed plate heat exchanger for hacking purposes.
I have learned that brazed plate heat exchangers were designed by the Germans, and are being made by several companies, of essentially the same materials, and have essentially the same performance.
So I was able to dig up a selection program here:
http://www.flatplate.com/downloads/f...ateSELECTw.exe
...this one doesn't seem to be supported anymore, but it is for the same product line as the newer one. This one is down-loadable, which I like because you can try various scenarios quickly. Also note that this program has been picked up by Internet Archive "wayback Machine"(
www.archive.org), so it should be available for a long time.
The newer one is available here:
Login
...you'll need to register for this one, but there doesn't seem to be any 'fact checking' in the registration.
I have also Googled my tail off hunting down <"brazed plate" refrigeration selection> documents, of which I found quite a few. Mind you, that when you are seeking this information you need to look for info for "refrigeration" and not water-to-water.
Here are a few documents:
http://www.eptec.no/images/Marketing...20NA08583B.pdf
http://www.flatplate.com/pdf/refrig/Cbrochure7-98.pdf
...this should get you pretty close to the size you want to use.
I noticed that the selection program seemed to be selecting larger heat exchangers than the product literature suggested.
Since the product literature came closer to my experience so far, I prefer it.
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
So we need to get used to the idea that we can think in terms of BTU/hr and also watts.
For instance, to convert 1000 watts to BTU/hr, we multiply by 3.412:
1000 x 3.412 = 3412 BTU/hr
To convert 8000 BTU/hr to watts, divide by 3.412:
8000 / 3.412 = 2345 watts
So for instance, if you have a compressor that draws about 800 watts, as measured by a watt meter (kill-a-watt), you don't choose a heat exchanger that has an 800 watt capacity (2729.6 BTU/hr). Instead you look at how much heat it can move, which can be 5 times the watts it draws. (Amazing, yes?)
So for instance:
800 x 3.412 x 5 = 13648 BTU/hr
If you are looking on ebay for an exchanger, they almost always give you heat transfer numbers that are either for water-to-water purposes like solar or hydronic use, or else they are completely 'plucked from the sky'.
My advice is to really study the brazed-plate refrigeration selection charts, as many as you can find.
I did that, and noted what the brazed-plate refrigeration heat transfer was for several sizes of product from several manufacturers. Then I came up with a figure for the heat transfer divided by the total area of the various units. I came up with different numbers, but they were in the range of 5000 BTU/square foot.
You should do your own research on this since it's your money you're going to spend.
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
So, having done all that, I came up with a small brazed plate unit that is about 7.5 inches by 2.88 inches by 10 plates thick.
From my previous post:
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
So the formula might look something like:
Heat = ((width) x (length) x (# of plates - 2)) x 5000
...where width & length are expressed in feet
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
width = 2.88 inches = 0.24 ft
length = 7.5 inches = 0.625 ft
so:
Heat (exchanger) = (0.24) x (0.625) x (10 -2) x 5000 = 1.25 x 5000 = 6000 BTU/hr
Next, I'll check to see what my compressor can move by calculating the heat I anticipate that I will be able to move with my compressor:
Watts I measured when I ran my compressor = 380
COP I can expect = 5
Heat (compressor) = (380) x (3.412) x (5) = 6482.8 BTU/hr
CONCLUSION: According to these calculations, the compressor can move more heat that my heat exchanger can. Normally this is NOT such a good situation. In most cases, I would prefer that my exchangers exceed my compressor, for greatest efficiency. However, I built a heat exchanger last summer with exchangers that were twice the size I am now using, with a compressor that was just about equal in size to this 380 watt unit. I ran great, but I never knew if my guesswork put me right on the edge of good performance, or if I had a wide margin or a very wide margin. So I've decided to go ahead and build this unit to see if my performance is reduced somewhat.
So, I'm fact-checking my calculations
Can't wait to start cutting & brazing...
Best Regards,
-AC_Hacker