Design Process

First Idea

Our first attempt at converting a natural gas tank to an electrically powered water heater involved removing the sacrificial anode used to prevent corrosion inside the tank due to the galvanic reactions. These anodes are typically made out of a more electrochemical metal than the type used for the walls of the tank. Our hope was that we could take this out and use the space as a port to insert a retrofitted water heating element.
For more details about this process click here

Second Idea

Our second idea consists of removing the drainage pipe and installing a heating element typically found in electrically powered water heaters. This method would consist of a button thermostat that serves as a temperature-controlled switch rated at 90°C. The button will be in series with the heating element and the solar panels; at a temperature of 90°C, the switch will break the circuit and prevent current from flowing to the heating element.
For more details click here

Design 2.jpg
Design 1

Third Idea

Our third design placed the button in indirect contact with the water within the tank. Our heating element is in series with the photovoltaic cells and the button thermostat, but the button is placed in between the inner and outer walls of the water tank. The button will be placed on the outside of the internal tank, but inside the outer shell of the water tank.
Click here for more details

Our final design is similar to our third design but with two different pieces of hardware The first change is the type of feed through we use to insert the heating element into the tank. Having a nipple and reducer connected to the element adds unnecessary length to our design and increases the chance of boiling to occur in the nipple. In an attempt to put more surface area of our heating element in the tank while avoiding the sloping floor and vertical piping, we bent our element into an “L” shape. We then used a swivel and a flanged nipple with rubber washers to screw the element in without moving it from the initial inserted position. Along with teflon tape and pipe dope, we were able to insert our prototype into the tank and found no leakage.

The second change switches the thermostatic button found in designs 1 – 3 with a thermostat regularly found in an electric water heater. While we want our water to reach 200°F, the thermostats typically found on commercial electric tanks don’t go past 150°F. After dissecting some thermostats, we found a way to increase the upper limit with calibration. Once testing begins we will see if this method is valid, but if not, we will return to our initial idea of thermostatic buttons.


o test our design while waiting for construction at The Grange a prototyping space was created at Sangha energy. Called the Schwartz Prototype Space or SPS, this large military lab space was cleared out and with a few changed made into a lab space for the many projects taking place, including water heating. A racking system is being designed by other Cal Poly engineering students for solar panels to go on top of the SPS. The electric water tanks we found were put on the back of the SPS so testing can be done.

SPS- Schwartz Prototype Space at Sangha
Back of SPS with water tanks

To run these tests, we will be using a thermocouple that will feed through a rubber stopper in a T joint. So the thermocouple will be in the water, but the electrical components and the thermocouple temperature reader will not be in the water.

For piping for the prototype we ran into problems with the 30gallon electric tanks we were planning on using. The tanks were discarded as junk before we picked them up. They are obviously old because the nipples on one of the tanks will not come off. Our original piping plan consisted of running water from a hose to both tanks by splitting the input water with a T joint and piping the water to both tanks. We wanted to replace the input and output nipples that were on one of the tanks and put new ones in. This proved to be harder than expected and we ended up deforming the cold water input nipple that was on one tank with a large pipe wrench and brute strength. After multiple attempts and an expert opinion of how to get the nipples off, or how to salvage the tank now that the nipple was bent, we decided to scratch that tank and use only one tank for data collection.

Top of electric water tank

The other tank has a street elbow and nipple coming from the hot water output so we will use that and attach an 8” nipple to that so we can then hook up a valve and T joint to put the thermocouple in, and hose for the hot water output. So we will use one tank to for experimentation instead of two.
The hose will be able to screw to a nipple to screw into the cold water input for this tank.

Feed Through.jpg
Rubber stopper with a hole drilled through the middle used with an electrical conduit to feed through our thermocouple

Hole drilled into side of tank for button thermostat to be inserted

Side of electric tank with holes for element and button to be inserted

We have successfully hooked everything up and stopped all leakage. While waiting to put the panels on top of the SPS, we have put one SunPower panel on the ground and hooked up our element, thermostat, and circuit breaker in series.

Once the solar panel was hooked up we began initial testing to calibrate the thermostat.

Thermocouple reading in F at 10:30am on July 20th

Panel on ground with water tanks on SPS