I did a test a while back on the performance of the 300 ft coil of PEX as a heat exchanger.
It was pointed out that the test did not really show what happens with long water draws at somewhat larger flow rates, so this is another test that shows a 22 minute long hot water draw with a flow rate of 2 gpm -- 44 gallons in all.
The heat exchanger is a 300 ft coil of one inch diameter PEX that is simply immersed in the solar heat storage tank. The incoming cold water makes a single pass through the heat exchanger before it goes to your backup hot water tank. The up side of this arrangement is that it offers a relatively cheap and very simple to build heat exchanger with no pumps or controls. Another up side is that the coil stores about 9 gallons of water right in the coil that is heated right up to tank temperature -- so, for the first 9 gallons, the exchanger is 100% efficient. The not so up side is that the performance of the PEX coil as a heat exchanger on long draws suffers a little from the lower conductivity of the PEX tubing wall. The test below shows how this comes out.
Two gpm was chosen because it represents a fairly heavy duty (non-lowflow) shower, or a reasonable speed to fill a bathtub (typically with some mix of hot and cold).
The idea was to maintain a flow of 2 gpm for long enough to make sure that the 9 gallons in the coil was completely used and long enough beyond that to see the performance of the PEX coil as a heat exchanger. Ten minutes is probably long enough to do that, but I ran it for 22 minutes to be sure.
The tank temperatures came off the display for the Caleffi controller for my solar water heating system, and the heat exchanger outlet temperature was measured with a thermistor clamped to the copper pipe that the HX ties into and then insulated over --this was logged with an Onset Computer logger. Sensors placed this way never quite get up to the fluid temperature in the pipe. To account for this, I corrected this logger temperatures up about 2F to make the initial part of the draw (which is known to be at full tank temperature) match the tank temperature.
I measured the flow rate by timing the filling of a 1 quart container and trying to hit a 7.5 second fill time (2 gpm). I tried to maintain the flow rate at 2 gpm, but our water supply is from a well, and the pressure changes as the well pressure tank goes up and down. I checked flow rate quite often using a quart container and stopwatch, and adjusted the flow as necessary. I believe that the average flow is close to the 2 gpm.
The green and red lines show the bottom and top of tank temperatures (not much difference).
The Purple line shows the temperature of the water leaving the PEX heat exchanger with a steady 2 gpm flow rate.
To get the total volume at any given minute, just multiply the minutes by 2 -- e.g. the total flow at minute 10 is 20 gallons.
The pipe coil is 0.863 inches in inner diameter, and 300 ft of it holds 9.1 gallons. So, for the first 9.1 gallons (about 5 minutes), the temperature of the water leaving the heat exchanger is the same as the tank temperature.
After 5 minutes, the heat exchanger outlet temperature rapidly drops to about 20F below the tank temperature.
This is a bit more than I expected, and indicates that for large water draws (over 10 gallons) that there is a bit of a hit compared to (say) a big copper coil heat exchanger.
My heat exchanger coil is basically just plopped into the storage tank without much attempt to space out the coils to allow the tank water better access to the coils and expose more surface area. I would guess that some of the 20F drop could be regained if the PEX were recoiled, or if the coils were separated with spacers in some fashion. Here is one example on Ken's system...
Whether its worth it to make the heat exchanger more efficient for large draws depends on your usage. For us, it makes essentially zero difference as we just don't normally do large draws. In addition, if the tank is 20F or more above the temperature you want, there is no loss as the heat exchanger delivers water above your target temperature. Making the heat exchanger more efficient could be done by changing to a copper coil (which could be shorter), or (maybe) by spacing the coils out carefully to improve heat exchange with the tank water.
Note: I forgot to measure the incoming water temperature during the test, but the day after it was 46.9F -- I'd guess it was quite close to this for the test as the water comes from a well.
Gary March 30, 2012