Getting into hot water: water heating ways

Jen Gow's evacuated tube array on the roof.
Five reader stories and five different systems that illustrate there’s more than one way to get into hot water!

A tale of two solar hot water systems

Jen Gow has tried out both flat plate and evacuated tube solar hot water systems, and discusses the differences.

Flat plate thermosiphon solar

The first time we used solar hot water was in 1999, when we lived outside of Maitland in the Hunter Valley in NSW. We installed an Edwards flat plate thermosiphon unit with a 200-litre tank on a north-facing roof. The booster ran on a time-controlled circuit that cut power during the evening peak. We used this with a time switch that turned on at 3 pm so that if the tank was under-temperature after the main period of sunlight, it would heat for a couple of hours until the control relay turned the power off (usually sometime around 5 pm). The time switch ensured that we maximised the solar contribution before using the booster. We could bypass the switch if necessary or switch off booster power completely. With this setup we could usually run on full solar for seven months of the year, from late September to late April. The only maintenance the system needed was to occasionally clean dust off the flat plate collector panels.

Evacuated tube solar

When we moved into our current place in the western suburbs of Brisbane in 2008, we installed solar PV power (initially 1.4 kW of panels feeding a 2.5 kW inverter) and a Hills Esteem Solar EB-E22 solar hot water system with a 250-litre ground-located tank linked to a 22-tube evacuated tube collector. The tubes are installed at a 25-degree tilt. The tank and collectors cost $5217 minus rebates for RECs of $1032 and Brisbane City Council Rebate of $400, for a total cost to us of $4307.

The only maintenance issues we have experienced with the HWS is replacement under warranty of the pump controller six months after the system was installed, and more recently we replaced the pump when it failed due to a power surge during a thunderstorm. Replacement by a tradesperson cost around $350.

Initially we ran the booster on the existing HWS off-peak circuit. We had the relay removed and switched the circuit to regular power because our energy supplier had a minimum off-peak consumption charge, yet for most of the year we drew no booster power. With this system we can manage more or less full solar for ten months a year, from mid-July to mid-May. During this period we may have to switch on the booster after three or more overcast days. For the other two months we leave the booster switched on.

I think the superior performance of our current system compared with our old one is partially explained by the greater solar energy available in Brisbane compared to the Hunter region. The evacuated tubes with their insulation and ability to collect solar energy from shallow angles may also be a factor.

On hot summer days water in the return line can get as hot as 100 °C, but this hasn’t caused any problems. The water in the tank would be pretty hot, but the hot water outlet has a tempering valve installed to limit the temperature to the taps by mixing in cold water. At the other end of the heat scale, to avoid the possibility of Legionella we have a thermometer probe installed on the return line from the collector. If the temperature drops below 55 °C we switch on the booster, but it rarely happens.

The solar hot water system does a great job supplying more than sufficient hot water for a household of three most of the time, and more if we have guests.

Don’t dismiss resistive element hot water

For Dave Southgate, converting to an all-electric house did not involve using a heat pump for hot water. Here’s what he did instead.

In 2015 I decided we had to do an energy makeover of our house in Canberra. Among other things I wanted to get rid of gas and move seriously to solar PV. At that point we used gas for heating, hot water and cooking. Making this transition is of course now pretty commonplace. While I started out thinking that we would go down the heat pump route for both hot water and space/personal heating for our new energy configuration, in the end we did not use heat pumps (read more about it in my 2015 report, listed at the end).

Opting for resistive element hot water

My initial reason for not choosing a heat pump for hot water was because of noise concerns—the location of our hot water tank meant that the heat pump would be just a few metres from our neighbours’ main living area. But I was also very interested in energy diversion devices: devices which divert solar-generated power that would otherwise be exported to the grid, instead using it for various uses on site, including to heat water via a resistive heating element. It seemed to me that these would be a great way to optimise the use of solar PV power at home and I was keen to see how well they work. At that time there were a few available on the Australian market but in the end I chose a UK-made device, the Immersun, because it appeared to be the most well developed of the available options and provided an excellent energy monitoring system. Energy diverters typically cost around $1000.

We replaced our gas hot water system (about 200 L) with a standard 250 L electric element system. We had to replace the 3.6 kW element with a 3 kW element because of power constraints in the Immersun. I set the thermostat to 60 °C.

When opting for the resistive element approach I was aware that this would involve using more energy than if I chose a heat pump, but my prime aim was not to reduce total energy use; I was looking to minimise the use of grid electricity.

How well has it worked?

We are a family of four, living in Canberra, and the system works perfectly for us. We generally use our hot water for showers in the evening and have never run out of hot water even when we have had house guests. We normally use cold water for all our washing.

Deciding to go down the energy diversion route was also very good for us since, as well as taking care of most of our hot water needs, it opened up the potential for direct solar charging of our electric vehicle; see p. 56.

We have 6.5 kW of solar PV available to the Immersun and I have set it up so that if we have a bad solar day the hot water reverts to grid electricity at 3 pm to top up the heat. On nearly all days the water is hot well before 3 pm (in summer it is commonly hot before 10 am even though I have set up the system to give EV charging priority over water heating).

In 2016 about 95% of our hot water came from our solar PV, a much higher proportion than would normally be reached with a thermal solar system (usually about 80%) and most likely more than I would have achieved with a heat pump.

The proportion of grid electricity we use for hot water is of course not uniform across the year—the graph shows the extent to which our performance dropped over winter 2016. On average, we used 4.7 kWh per day for hot water during 2016. Maybe we would have used 2 to 3 kWh per day less if we had used a heat pump?

Next steps

I am planning to install an additional 4 kW of solar PV and a Tesla Powerwall 2 over the next few months. I’ll be very interested to see to what extent this will improve our level of grid independence for our hot water.

Maybe this is a bit of a purist’s point, but being a retired engineer I love ‘non-moving parts’ solutions. We now have a household energy/transport system based on solar PV,  resistive heating hot water, far infrared (FIR) space/personal heating and an EV. Not too many moving parts there! I’ve yet to work out an equivalent solution for cooling (we use fans) but maybe it’s worth thinking about. S

Dave Southgate is the author of Our Household Energy Transition: Becoming a Fossil Fuel Free Family.

How to save money with a hot water heat pump

Jonathan Prendergast shares his quest to reduce his hot water bills by switching to a heat pump.

After 15 years of inner city living, we moved to the quiet leafy suburbs of the Sutherland Shire, along the south coast of NSW, in 2015. We were used to low electricity bills in our apartments, of around $150 per quarter with usage around 6 to 7 kWh per day. We were shocked when I checked our meter a couple of weeks in to see we were now using 30 kWh per day.

Granted, we used to get our hot water and heating with gas, so our previous low figure was just for lights, refrigeration and appliances. But it shocked us into action, starting with disconnecting half the downlights. Coming out of winter into spring, we saw our consumption drop down to around 16 kWh per day, which was better.

We have our hot water supplied on a different meter which made it easy to track energy use. It is on an off-peak service and only turns on at night to ensure household hot water does not exacerbate network peak demand. We measured that of our 16 kWh per day, our electric element hot water tank system used 6 to 8 kWh per day. Almost half!

I had heard of hot water heat pumps and started researching them. Our action was brought forward when the existing hot water system failed. I ordered a heat pump and new tank from a supplier I know, and contacted the local plumber to arrange installation.

The (approximate) costs were:

  • new tank $1000
  • heat pump $2200
  • installation $800
  • government rebate $1000
  • The net additional cost of choosing a heat pump over an element hot water system was approximately $2500, after the rebate.

The $1000 to $1200 rebate is from the federal government as part of the Renewable Energy Target. Why do they qualify? Heat pumps are just a pump. They don’t heat the water directly. They absorb heat energy from the ambient air and use pressure and refrigerant to transfer the heat into the water. This is very efficient, using around 75% less electricity than an electric element hot water system. The majority of the energy comes from ambient air heat, which is provided by the sun and so is renewable.

The results?

Our hot water heating now only uses around 2 to 2.5 kWh per day, so about a 75% energy saving as expected. In the graph below, you can see our hot water electricity usage and general energy usage over time. You can see the few days we went without hot water when our previous system failed and we organised the new tank and heat pump, but more importantly, you can see the reduction in electricity use.

We buy GreenPower using Powershop, which costs us around 30 cents per kWh. So the heat pump’s daily energy saving of 5 to 6 kWh saves us $1.50 to $1.80 per day, or around $550 to $650 per year.

Other considerations

The heat pump fan does make some noise. Ours currently runs at night on the off-peak circuit and I don’t think it is noisy enough to bother anyone. In any case we are getting solar soon so will switch it to the general circuit and program the heat pump to run from 10 am each day; a little noise certainly won’t be a problem in the middle of the day.

We got a Siddons Bolt-on heat pump as it is what the company I knew supplied. It has a good efficiency rating with a COP of 3.6. There are other heat pumps available that are even more efficient, but while efficiency is important, to me it doesn’t matter too much if it is using 2.1 kWh per day rather than 2 kWh per day (a 5% or $10/year difference).

Eighteen months later, we are happy with our heat pump and cheaper electricity bills.

Here is what our system looks like, with the heat pump in the foreground and the tank behind it.

Troubleshooting issues with solar hot water

Ewan Regazzo’s electrical engineering background was put to good use troubleshooting a faulty solar hot water installation. It’s now working well, but there were several issues along the way.

We have a near-new solar hot water system (SHWS) on our recently built house in East Maitland, near Newcastle in NSW. The system works well now, but it’s been a long road.

Our original intention was to follow our builder’s recommendation and install instantaneous gas hot water. However, during the house design phase, I found a Bosch induction cooktop at a ridiculously low price. This made going all-electric, including solar hot water, more financially attractive—despite the SHWS costing $2000 more than a gas system (after a $1000 rebate).

We chose a Thermann evacuated tube system (TE-250-GL-BOT-22) with a roof-mounted array and a ground-mounted 250 L storage tank. The boost is electric (3.6 kW element at the bottom of the tank).

Getting the location right

Unfortunately, there were several installation issues. Firstly, the builder sited the SHWS storage tank where the original gas hot water heater would have gone—at the gas outlet. To keep the water piping between the array and storage tank as short as possible, the array ended up on the eastern roof. Not an ideal situation! The builder agreed to move the array to our north-facing roof, but this got delayed until cooler weather and a suitable time for the plumber. The unconnected array remained on the roof, soaking up the sun’s heat, but not able to transfer that heat to any good use.

But still not working

Finally, the array was moved to the correct location and it was now time to move in. But at this point, close examination of the wiring next to the storage tank showed the roof-mounted temperature sensor was not actually connected to the SHWS controller. There was confusion about whether it was a job for the electrician or plumber, but eventually the sensor was connected and the system was declared operational—array on the northern roof and the controller happily turning the pump on and off.

Unfortunately, the performance was woeful. Over the next fortnight, despite sunny winter weather, I had to boost every two days. Regular monitoring showed that despite the array manifold heating up, the tank temperature at best remained constant throughout the day and usually dropped.

The plumber came and looked at the system, but couldn’t find anything wrong. Out of frustration I contacted the manufacturer (Apricus, who look after the Thermann brand). A technician came out and together we started fault-finding. The first issue we found was with the adjustable flow valve, which sets the rate at which the water flows through the array manifold. It had been set to zero flow, so no water had been flowing through the manifold—ever! The next issue was a build-up of air in the system, so we went through the purge procedure.

It was with a great deal of excitement that I went to check the temperature after the next day of decent winter sun. My hopes were dashed though, as it seemed there was still little heating occurring.

A problem for an electrical engineer!

To make fault-finding easier, I decided to make a wireless controller to read the controller’s data bus and extract the water tank upper and lower temperatures and array manifold temperature. I could then log this data to help determine what was (or was not) happening with my system—oh, the joy of being an electrical engineer!

With the data logging capability up and running, I began comparing my system to what other people were experiencing with their SHWSs. For most people, once the pump came on in the morning, it would remain on all day until the sun went below a certain angle (or clouds obscured the sun). In my case, the on/off cycling occurred throughout the day, which indicated that insufficient heat was entering the system at the array manifold or excessive heat was being lost in the system through the pipes between the array manifold and storage tank.

I discussed this issue at length with the builder, pointing out that the SHWS had not been installed in accordance with the manufacturer’s instructions. Happily, the builder agreed to remove it and reinstall it. I was also pleased that the local Apricus technician was willing to attend and provide some expert guidance to the plumber.

Reinstall becomes reassembly

The day of the reinstall eventually arrived. With the evacuated tubes down from the roof, it became clear what had happened to the system. The heat transfer pipe in each evacuated tube—a small, copper tube filled with a transfer fluid—had completely over-heated. The shiny copper surface was almost black and the fluid had managed to leak out due to extreme pressure caused by the high temperature inside the tube. Although the SHWS is designed to withstand full sun and no water flow, it still needs water in the array manifold. The months my array had sat on the roof in the summer sun with no water in it convincingly destroyed those small pipes and rendered the system totally ineffective.

The technician and I replaced the heat transfer pipe in every evacuated tube and reassembled the array. We purged the system and checked that everything was working properly. After months and months, was now the time it would finally work? I sat back in the late afternoon after the tradespeople left and looked at the temperature in the tank. Slowly, it began to rise and finally, six months after moving in to my new house, I had my first sun-powered shower!

The next day I saw the SHWS exposed to a full day of sun. I was reassured to see the manifold temperature stay well above the tank temperature instead of cycling above and below it and the tank temperature solidly rise. When the tank reached 75 °C the manifold temperature began to rise rapidly as the pump switched off; the controller had reached the set temperature for the storage tank and water was no longer flowing through the manifold. When enough hot water was used during the day to measurably lower the temperature in the storage tank, the pump turned on and the manifold temperature dropped as water flowed through the manifold again. Finally, it was all working as it should.

It has now been more than two months since the system was reinstalled and we haven’t had to use the booster once. With fairly low hot water demand for showers, dishwasher and occasional white clothes wash, we’re yet to run out of hot water.

As you can see, it’s been a frustrating, but illuminating, journey: I’ve learnt more than I expected about the controller in my solar hot water system and how solar hot water systems work in general. I also gained insight into the workings between different trades and how poorly they sometimes understand each other.

Resistive versus gas

Linda and Mike Dahm were surprised when the energy costs for their dual occupancy homes, one with solar PV and an electric resistive hot water and one with gas hot water, worked out about the same. Here’s what happened.

We built a dual occupancy in northern NSW: two similar houses with two bedrooms and an office each. The first house was set up with instantaneous gas hot water and a gas cooktop, with an option to add solar panels later. The second house was set up with 3 kW of solar panels and a Fronius Primo 3 inverter so that batteries could be added later (when the price falls). We decided to go all-electric with this house. But when we compared the total energy costs for the two houses, they were very similar.

The company who sold us the solar PV setup suggested an 80 L electric resistive element hot water system would be enough to meet our needs (2 to 3 adults in the household, with regular visitors since we lived in a holiday area). We decided to get a Rheem 160 L but even this size turned out to be too small.

We had the electrician fit a timer so that the hot water system only came on between 10 am and 3 pm so that on most days it only used electricity from the solar panels. The problem with this was that if we all had showers in the evening, on days when we used the washing machine and/or dishwasher the hot water would run out, usually during the last person’s shower! So we ended up having to manually switch the hot water booster on at times, and then we had to remember to turn it off again, which of course we didn’t always do. So our electricity bills were somewhat higher than we had hoped for.

We used 11 to 14 kWh per day from the grid in the all-electric house with solar PV, with higher usage in winter. The cost of electricity was $3.18 per day in summer to $3.29 per day in winter, plus the supply cost of $1.50 per day. In the house with gas hot water and gas cooktop and no solar panels we used 4 to 8 kWh per day (higher usage in winter). The electricity cost was $1.60 per day in summer to $2.00 per day in winter plus $1.50 per day for supply. LPG use came to approx $1.50 per day.

That adds up to $4.96 per day for energy use in House 1 (gas and electric) and $4.76 in House 2 (all-electric with solar). But in House 2 we still fed back into the grid around 2000 kWh over one year, so if feed- in tariffs improve or we could work out ways to use that excess power produced (installing batteries, not leaving the hot water switched on, etc) we could reduce the energy cost and use of the grid for House 2.

It’s a little difficult to write definitively about our electric hot water setup, as we didn’t stay in the house long enough to sort out the issues. But it is still interesting to look at what we did and what we will change next time. The biggest lesson we learnt for our next house is to buy a heat pump (greater than 160 L) for heating our water.

ATA’s energy policy expert, Andrew Reddaway, has this to say:

It looks like your main problem is a too-small tank. The best option when using solar electricity with a resistive element tank is to get a big hot water tank with two elements. You connect the bottom element on a day-time timer or smart switch like Immersun, or one included in the solar inverter (investigate whether your inverter has this function). It heats up the whole tank during the day, mostly from excess solar. The element would need to be sized to the solar system, e.g. a 1.8 kW element rather than the usual 3.6 kW.

The top element can be on a night-time timer, using a cheap off-peak or controlled-load tariff if possible. This will switch on only if the top of the tank is below temperature, which should be rare. And it will only heat up the top part of the tank. This setup should avoid running out of hot water, while minimising use of mains power. For further savings you can add extra insulation around the tank (making sure the safety valves still work and can be inspected). You could also consider adding more solar panels to cover more of your hot water use.


Thank you
Several of these case studies were found via My Efficient Electric Home’s Facebook page, a good source of stories and information on all-electric homes and appliances.
Totally Renewable Yackandandah

Totally Renewable Yackandandah

Totally Renewable Yackandandah (TRY) is at the forefront of regional towns making the shift to renewable energy, as Matthew Charles-Jones described at a recent Renew branch meeting in Melbourne.

Read more
All-electric and hydronic: A model of efficiency

All-electric and hydronic: A model of efficiency

There's a lot to learn from this highly insulated and well-sealed renovation in Melbourne, not least how a heat pump is providing both hydronic heating and hot water. Cameron Munro explains the house's modelling-led upgrades and the tweaks made along the way.

Read more
Planning ahead: Three steps to all-electric

Planning ahead: Three steps to all-electric

Thinking about going all-electric, but unsure what’s involved? Here we present an overview of the steps to going all-electric and where to find more information.

Read more