Off Grid PV Micro inverter connected directly to immersion heater?

[solarbot]

Hi,

Here is a very simply idea but not sure what the implications might be so wondering what the community think?

I have a 250L hot water tank which is heated by a gas boiler but it also has and 230V immersion heater.  My hope would be to connect say 2 PV panels (say 0.5kWp) via an inexpensive micro inverter such as the WVC-700 directly to the the immersion heater which would be disconnected from the 230V mains supply.  I would then have most of my hot water needs met by this simple system with the tank protected from overheating by the thermostat on the immersion heater.  Two questions which come to mind are:  How can I persuade the micro inverter to provide an output to the immersion without a connection to mains 230V?  What happens when the immersion heater is switched off by the thermostat when the tank is full, say when I'm away on holiday?  The advantage of doing this is that it avoids having to negotiate with the network operator as it is off grid.

Has anyone done anything like this before? Thanks for any thoughts.

Here is a link to the micro inverter: https://en.kaidengdg.com/product/21.html

[fourtytwo42]

It will not work! The device is a GTI and contains an anti-islanding function that all compliant GTI's are required to have by law in every country I have ever heard of.

I hope you know what islanding is ?

[tszaboo]

It will not work! The device is a GTI and contains an anti-islanding function that all compliant GTI's are required to have by law in every country I have ever heard of.

I hope you know what islanding is ?

I guess he des, because:

How can I persuade the micro inverter to provide an output to the immersion without a connection to mains 230V? 

So if we put islanding aside for a moment:
The inverters provide you with a constant voltage, variable power.
The load, the heater expects constant voltage, and constant power, with two different power levels.

So this simple system clearly wouldn't work, because if your load is 1KW @ 230V, and your solar panels only output 700W, then the voltage cannot be 230.  You either need to store energy and periodically turn on or off the heater, or make sure the heater works with variable voltage (good chance it is, because it's likely just a resistor) and vary the voltage in the system, that is not possible with standard inverters. It could be possible but there is a bunch of software protections against it.

If I were you, I would monitor the outputs of the inverters, and turn the heater on with an SSR or similar based on the output. Depending on your electricity meter, this might be basically invisible to the power company.

[Psi]

Simplest plan would be use more panels, perhaps smaller ones, just to get the total panel string voltage up to around 250 at the same power as the heater needs.
like 5x 50v panels.  Then, assuming your immersion heater is just a resistive load and thermostat, you can connect them together and whatever solar power is available will turn to heat.  It won't be perfect because there is no max power point tracking. but it is simple, no inverter needed. Just make sure you replace the thermostat with something solid state so it doesn't cause huge DC arcs and blow the switch.

And make sure you have all the needed safety systems to prevent over-pressuring the boiler.

[NiHaoMike]

Consider a zero export inverter instead, then it can power any load in the house.

[DenzilPenberthy]

Use a grid connected inverter then a device like an Immersun or iBoost to divert power to the immersion heater such that your export is zero.

[solarbot]

Hi All,

Many thanks for your thoughts and help, much appreciated.

- Re: Islanding, I guess my hope was that there was a trick that could be used to bypass the anti-islanding function as it would not be required if the output was not connected to the grid.  I did wonder about having a small inverter connected to a small PV panel to provide 230V that the micro inverters need to start outputting power?

- I've not come across a zero export inverter before, sounds like this might be a good option and I will take a look - how does it prevent excess power being exported?

- I also like the SSR option. I guess a CT clamp round the immersion side 230V, perhaps using a microcontroller, could be used to turn the micro inverter output off when the immersion stops drawing power?  I'm guessing the micro inverters can handle not exporting all the available power from the PV panels if its not required?

- And I also like the solar diverter option :-)  Does anyone know if any of these devices can be setup to simply divert excess power to the immersion and then not export anything to the grid when the immersion stops drawing power?  I have been looking at the Eddi by Myenergi which looks good - I think they did the Immersun device before selling the company?

- And finally the direct connection option is interesting too as there would be no conversion losses and less kit.  I guess the AC immersion thermostat would still work as this is likely to be a bi-metalic switch?  If the AC immersion doesn't mind running on DC then I guess it would be a case of ensuring the DC output stays below 230V DC?  (about 4-5 panels).  Although the thought of having a high voltage DC line in the house unsettles me for some reason!

[NiHaoMike]

- I've not come across a zero export inverter before, sounds like this might be a good option and I will take a look - how does it prevent excess power being exported?

There's a current sensor that's installed on the grid connection into the main panel. The inverter tries to get the (in phase part of) current to near zero.

[SeanB]

Would not recommend the DC link with the existing thermostat, but instead set the thermostat to maximum, and instead have a separate temperature controller operating a DC rated contactor instead, as the thermostat will not last long switching DC, as it will absolutely arc and burn out. Separate small panel charging a 12V battery, probably something you will need to have lighting that is not grid dependant, and this runs a 16A 3 pole contactor with sensitive 12VDC coil, and all 3 16A contacts wired in series, so as to guarantee opening with DC use. then add in a voltage sense relay on the DC input, so that the voltage sense relay disables the contactor when solar panel voltage is below around 100V, as that will mean it is night, and probably no longer going to heat the tank anyway, but will leave the contactor on all night, wasting battery power. Temperature controller should have a LCD display, and run off 12VDC, so as to use minimal power during the night.

Contactor, temperature controller and voltage relay are all common industrial logic parts, and available world wide. Temperature sensor will need to be in contact with the inner tank, preferably stuck under the insulation at the electrical panel opening, where the foam fill will be visible and easy to dig a hole in to hold it in contact with the tank.  Temperature sensor can be any type, thermocouple, PT100 or thermistor, and turn on failed probe detection in the controller.  If using a thermocouple you will need to insulate it electrically from the tank, using some polyamide tape covering the bead tip. the others normally come with insulation anyway.

[Geoff-AU]

Then, assuming your immersion heater is just a resistive load and thermostat, you can connect them together

As SeanB says, do NOT do this.  DC is more challenging to interrupt than AC and an AC thermostat will simply arc continuously and catch fire.

You can use the AC thermostat to control a DC contactor, which switches the PV to the heating element.  You will need a source of low voltage DC (12V is plenty).

[solarbot]

Thanks - I forgot about the arcing problem so nice to be remended :-)  I've just spoken with a solar diverter company and they have said that it would only divert surplus from the PV to the immersion so if I want to just feed the immesion I'm going to have to do something bespoke I think.

[Psi]

Thanks - I forgot about the arcing problem so nice to be remended :-)

yeah, i forgot about that as well.

[max_torque]

if you sit down and "do the math" it quickly becomes clear that it's really hard to ever break even on this sort of low total energy transfer project. Electronic stuff is expensive, especially when not chinese tat from alliexpress etc, and grid 'lecy is not actually that expensive.  First move that is actually the best return is to see if you can get a dual traffif that allows you to use cheap rate night time lecy, and simply turn your immersion heater on with a timer.

One other option is to fit a dual element immersion heater, that has a 3kW 240Vac element AND a 12v (normally around 200w) element. That low voltage element can therefore be driven directly with most solar controllers (PWM / MPPT).   The problem (again) is that these immersions are of course popular for off-grid and RV useage, so cost a huge amount more than any normal immersion, making the payback impossible.

A trick you might be able to do is to reduce the power of a std 3kW 240Vac immersion by 4, by simply buying a cheap 240-120V isolation transformer, that can often be found cheap on ebay second hand, this will mean you can drive the system with around 600 to 700W, making driving it a lot easier!

[Geoff-AU]

if you sit down and "do the math" it quickly becomes clear that it's really hard to ever break even on this sort of low total energy transfer project.

Usually, yes.  I worked out that our PV diverter saved around $100 a year, which made it a 7 year payback.  The economic landscape is always changing (usually, economic advantages being whittled away). But it feels good to have solar-powered showers and making a conscious choice not to use offpeak, because offpeak is a coal-supporting technology.

If you're doing a "from scratch" hot water system for home hot water, solar thermal collectors on the roof makes WAY more sense.  The technology is mature and reliable, and system thermal efficiency is at least triple the electrical method for a given collection area (around 75% efficient, compared to PV panels at 20%).

Some PV panels and a DC contactor feeding a conventional immersion element is easy though.  If your hot water demands are low, it's a viable method and can be very cheap (lots of people basically throwing out their old panels)