Today I decided to hook up a pump to the ground loop that was previously put in.
I have several diferent pumps laying about, so I decided to start with the smallest, cheapest one to see what the performance was like.
The pump is actually sold as an aquarium pump. it draws 35 watts and has a magnetically-driven impellor. I've misplaced the specifications for the volume and head curves of the pump.
I made an adapter to connect the 1/2 inch MPT output of the pump to a 3/4 barbed connector and connected the pump to the input of the loop field. The pump went into the bottom of a barrel and the output from the loop field went into the barrel also.
When I plugged the pump in, I wasn't really sure if it would even have enough power to circulate the water at all through the loop field. I was also very corious what the temperature of the loop field would be.
I tested regular tap water and it was stabilizing at 49F to 50F.
The loop flow from the aquarium pump was really pathetic, I measured the time it took to fill a gallon jug, and it took 3 minutes to fill it up.
After the water in the loop had flowed (slowly) for maybe 45 minutes I took another reading and it was a 51 degrees. I had expected it to be higher, but I'm more interested in what actually is.
The test I ran two summers ago yielded this graph:
... The black line, labeled "chill" was temperature if the barrel I was cooling... like the incoming water in the loop field. The light blue line, labeled "COP*10", that is bouncing all over the place like a drunken flea is the momentary, calculated COP. If I imagine a trend-line through the COP line (by thinking like a sober flea), it looks like it will be about COP = 4 when the incoming water temperature is about 50F. Not as good as I had hoped for, but good enough to keep on with the testing...
I Googled about for "how to purge a ground source loop field" and
up popped this little ruby.
This document is calling for specific pieces of equipment, but the basic proceedure is there for me.
I noticed small bits of debris drifting in during the hour or so that I ran the pump. The proceedure calls for a linear velocity of at least 2 feet per second to flush out debris. I haven't calculated the velocity of the water driveling through the pipe, but it was way, way below 2 FPS.
Next step, hook up the next larger pump and continue flushing debris & purging air.