DIY Heat Exchanger for Automated Water Change System

I change out a lot (100+ gallons/day) of water. It has always bothered me how much warm water I have been dumping down the drain. Recently I started thinking about a DIY heat exchanger to help reduce the cost of heating all the fresh water. I have a well, so water cost and chlorine are not a problem. The water comes out between 56 and 60 degrees, which really makes my heaters work.

I run four tanks off of a central sump located in my crawl space. The sump is a 300 gallon Rubbermaid stock tank. There is a 210 gallon, heavily-stocked New World Cichlid tank in my living room. There are three tanks in the garage - a 135 with two big cats, a 25 isolation tank, and a 55 grow-out.

The cores of the heat exchanger are made of two 26" x 1" pieces of thin-wall PVC (for incoming fresh water) and about 70' of 1/4" copper refrigeration tubing (for the waste water). I wanted to use brass instead of PVC for better heat exchange, but at $340.00 for a 20' piece I decided against it.

After wrapping the PVC pieces with the copper tubing I joined the PVC with two elbows and the copper with a compression coupling. I added compression adapters and 3/8" PEX to the ends of the PVC. I connected 3/8" I.D. vinyl tubing to the copper for the waste water.

The next step was insulating. For this to be efficient it needs to be very well insulated. I started with foam pipe insulation and then wrapped the whole unit in two layers of aluminized "bubble-wrap" type insulation.

The first thing I did was mount the unit on the front of my sump.

Then I tapped into my return line (off my Iwaki MD70 pump) and put a 1/2" stop valve with a 3/8" adapter.

From there I ran a 3/8" vinyl line to an inline strainer. I don't want to clog up the relatively small copper, so hopefully this works.

The waste water then goes into the heat exchanger.

The fresh water is regulated by a float valve in the sump.

The outflow of waste water from the heat exchanger dumps into a glass vessel (some sort of chemistry/industrial device I picked up somewhere) that allows me to visually monitor flow. The valve on the drain line allows me to shut off the drain and accurately measure flow - I have one pint marked and I time how many seconds it takes to fill.

A 5/8" I.D. vinyl tube takes the waste water to the sewer line. I drilled and tapped an adapter into the pipe (not pictured).

I am still playing around with flow rates, but this is what I have figured out so far:

I have about 20 degrees of difference between the outgoing/incoming water.

At 10 GPH (240 GPD) the water leaving the heat exchanger is 76 degrees, five degrees cooler than the water entering it. This indicates an efficiency of about 25% - not great, but still significant.

I dropped the flow rate to 5.6 GPH (135 GPD) and the temperature immediately started to drop. I will check on it after a while and continue to fiddle with the flow. Obviously lower flow will give greater efficiency, but I would like to see how it changes.

fg4608;5079338; said:

I was just wondering:

Did you do any calculation as to flow rates between the two paths? I´ll admit that I am no schooled engineer but my intuition is questioning the rate at which the water flows through the PVC pipe vs the copper tubing. Is there enough length of exposure for the heat transfer to take place?
Also, would it not be a more efficient set-up if both pipes were of the same material? I was thinking about using PEX tubing in both paths...

Either way, I applaud your effort and it certainly is taking things to another level. I have pondered making my own geothermal heat exchanger, as I anticipate wanting to address the same issues.

And, oh by the way, yes I do find it somewhat disappointing that you have not gotten more feed back than you have.

Click to expand...

Thanks for your feedback!

I think you are right that the dissimilar materials and flow rates through the system are not ideal. The truth is, the design was dictated largely by the materials at hand - I had the copper, PVC, and most of the fittings just laying around.

My suspicion is that the copper is heating up well but the PVC is not transferring the heat efficiently to the incoming fresh water. I am trying to stay away from copper on the freshwater side, or I would have used 1" copper instead of thin-wall PVC.

I have played around with the flow and got the following results:

10 GPH = 5∘difference (~25%)
5.6 GPH = 8∘difference (~40%)
3.75 GPH = 11∘difference (~55%)

While these numbers may not seem like much, I have already noticed a big difference in how often my heaters cycle on.

I would like to maximize flow and heat return, so I plan on adding another heat exchange unit. The next one is going to be a different design, probably an all PVC or PVC/PEX tube-in-a-tube design.

I, too, am a little surprised there hasn't been more feedback. To me the issue of heat loss in the form of waste water is a big deal, but I live in a cold area and change a whole lot of water every day. Anyone else?

BMcAfee;5079584; said:

I love the idea of a heat exchanger smart move. Have you thought of aluminum instead of pvc for a better exchange rate?

Click to expand...

FishFlake;5080067; said:

Just curious, why are you staying away from copper for the 1" pipe? You can buy similar systems for residential use that are made of copper & copper. They "claim" 50-60% efficiency. I think if you went with copper, you'd know you had the best heat transfer and then you could focus on other variables like flow rate etc.

Great job, by the way.

Click to expand...

nfored;5080486; said:

Would it have worked better if the pvc was 2" and the copper element was inside the pvc so the water directly touched the heat exchanger. Then run it through some hypersorb to remove any excess copper from the water. The hypersorb can be recharged very easily, and at a flow rate of 5gph you should have little fear of copper poisoning.

Click to expand...

Aluminum or copper would definitely conduct heat a lot better than the PVC, but I am trying to stay away from metals. I might be overly paranoid, but I don't have any copper in my plumbing (all PEX/Brass) and I really don't want to have to worry about it. Believe me, it would be much easier (and more effective) to run the freshwater through the copper coil immersed in wastewater as nfored mentioned, but I just can't bring myself to do it.

I already have a pretty good idea of what my next unit (which will be added to the present unit) will look like - 12' of thin-wall 1" PVC (I have a pile next to the barn) inside a 4" PVC shell. I will post photos as I start construction.