Linear DC load for water heating from solar?

[max_torque]

I'm looking at various options for adding some renewable energy into our houses energy mix, and as the price of solar panels falls these start to look like a sensible option. However, currently there is no sensible feed-in tarrif in the UK, and electricity still isn't actually very expensive, so reducing fossil fuel consumption without spending more money is actually really difficult.

However it struck me, that if you start with solar panels for just water heating, then a DC load, acting in input voltage control mode, is basically an MPP controller for solar panels, and as what you want is heat anyway, if you made that load water cooled, then your solar electric energy is pretty much driven straight to heat, and there is no need for an expensive DC -> AC inverter.

A 1kW nominal system would be reasonably easy and relatively cheap, perhaps using 20 parallel pass elements (transistors will be fine because you don't care about how the current gets to ground, so the high base current is ok). That would be 50W per device, just about  inside the SOA for a typical TO-220 package. You could also wire your solar panels in parallel and run at a low voltage to avoid any HV issues, obvs your resistive losses will be a bit higher than running at HV, but for a 1kW system at say 30Vdc we are only talking about 33 amps, so nothing too wild.


One question i did wonder about is how is the life of a BJT transistor affected by the temperature at which it runs?  Given your water outlet temp is going to need to be lets say 80 degC (aim for 60 to 65degC max in the hot water tank), running at max junction temp (150 degC) gives an overal deltaT for the heater of around 70 degC, so to pull 50watts out of it, we'll need a thermal impedance to the junction of lower than 1.4 degC/watt, and that is getting tricky (typical TO-220 junction to case impedances seem to sit around 1.2 to 1.5 degC/watt)

So if we run our trannies at maximum junction temp is that going to cause long term reliability issues??

[Berni]

Heating the water using the sun directly using solar collectors sounds like a more sensible idea to me, its cheaper and more efficient, just use a small solar panel to run the water pump instead.

But if you do want to dissipate >1kW of power id suggest dissipating it mostly trough high power resistors or heating elements. They are cheaper for the same power rating and will happily work at even higher internal temperatures for long periods. You can then use MOSFETs to switch more of them in parallel and if you want continuously variable current draw a small array of power transistors can be used to only cover the range of one resistor.

You don't want to run them close to junction temp all day long. Besides if the design is close to its limits to begin with, its going to slowly get worse and exceed limits as crap build up on the heater walls and thermal paste ages. When you want to dissipate a lot of power its usually a good idea to go for even bigger packages than TO-220 and even then you might only want to do about 25W per package at these temperatures. On the upside tho waterblocks do a really good job of cooling, so good thermal contact with water might not be so bad if you can afford having a small pump actively move water trough a block.

[max_torque]

Installing solar water panels is a PITA though and significantly more expensive thanks to the plumbing required, and the panels themseleves that can manage a high temperature output aren't cheap either!

You can disipate 1kW with a 15 degC upheat into about 1 l/min so not a high flow rate at all, might be able to do that with thermo-syphon (unpumped)  if the system plumbing is kept short and large diameter

Big wattage power resistors aren't actually that cheap, about the same cost as a mosfet / transistor, and you loose the ability too fully dynamically control impedance.

Might even be able to design the active elements to dump their heat straight into the wall of the hot water tank itself, after all there is plenty of surface area. Might need a copper tank rather than a stainless steel one perhaps?

[David Hess]

The systems I have seen used a separate hot water tank before the main hot water tank which preheated the water using solar but there is no reason photovoltaic could not be used to do that same thing except for much lower efficiency.

[Marco]

Big wattage power resistors aren't actually that cheap

I don't think transistors will win against resistance wire though.

[themadhippy]

Cut out the middle man and use  the water pipe as the the resistor.

[NiHaoMike]

The main problem with solar hot water is that they produce the most when it is needed the least, unless you have some large summertime use for hot water. See if you can get a "zero export" inverter for cheap, which will offset any load connected to it. (Most only zero export to loads connected to the output. A few support external current sensors to allow offsetting loads too big to connect to the output, but those are really expensive.)

[Berni]

Installing solar water panels is a PITA though and significantly more expensive thanks to the plumbing required, and the panels themseleves that can manage a high temperature output aren't cheap either!

You can disipate 1kW with a 15 degC upheat into about 1 l/min so not a high flow rate at all, might be able to do that with thermo-syphon (unpumped)  if the system plumbing is kept short and large diameter

Big wattage power resistors aren't actually that cheap, about the same cost as a mosfet / transistor, and you loose the ability too fully dynamically control impedance.

Might even be able to design the active elements to dump their heat straight into the wall of the hot water tank itself, after all there is plenty of surface area. Might need a copper tank rather than a stainless steel one perhaps?

Well you might be able to attach transistors straight to the tank if you can make a heat spreader plate first. So attach the transistors to a 2mm thick aluminum or copper plate and then thermal glue that plate to the side of the tank. Once the heat is spread out over an area the low thermal conductance of steel should not be a problem anymore, perhaps a local convective current might develop on the inside in the water.

And don't look at heatsinkable power resistors on Farnell or Digikey. They indeed cost a lot. You can get cheap resistors from china that work fine. Alternatively you can use resistance wire as suggested, you can also buy carbon resistive heating cable where you just cut it to length and connect power to the ends, wrap a bunch of that around the tank. Chinese also make plenty of flat heating pads, some of them adhesive so that you can simply stick a bunch of them to the side of the tank(even if its curved).

When it comes to solar power we have a better deal with the local power distributor, they just subtract the power we generated from the power we take from the grid, so if i make 10kWh during the day and then use up 10kWh during the night the bill is zero (Its also summed up at the end of the year, so power produced in the summer can be taken back from the grid in winter).

[james_s]

The main problem with solar hot water is that they produce the most when it is needed the least, unless you have some large summertime use for hot water. See if you can get a "zero export" inverter for cheap, which will offset any load connected to it. (Most only zero export to loads connected to the output. A few support external current sensors to allow offsetting loads too big to connect to the output, but those are really expensive.)

I don't think my consumption of hot water varies all that much with the weather. Summer or winter I still take hot showers, wash the dishes, wash the laundry, etc. It's not really determined by the outside temperature.

[fourtytwo42]

If you are lucky enough to have a hot water cylinder with an immersion heater like some UK homes including mine then all the plumbing and mechanics are taken care of already and all you have to do is to add an MPPT controlled buck/boost converter between your PV panels and the immersion heater (standard 2.4Kw/240V style no need for a new low voltage heater).

[NiHaoMike]

I don't think my consumption of hot water varies all that much with the weather. Summer or winter I still take hot showers, wash the dishes, wash the laundry, etc. It's not really determined by the outside temperature.

Solar produces more in the summer and the water is warmer as well, so poor utilization really is a big deal.

A cheap way to make a zero export inverter is to get an inverter module that accepts high voltage DC and have a string of panels connected to it (through a blocking diode) such that the open circuit voltage is well within the ratings of the inverter and the maximum power voltage about the peak of the AC voltage. Then add a capacitor to handle peak currents and a bridge rectifier to the mains in order to supplement when solar production is too low to power the load by itself.

[SolarMan]

I looked into the topic very deep a few months ago.

I basically concluded that it makes more sense to stick with a standard DC-AC inverter plus a power diverter such as the MyEnergi Eddi.  With the standard AC solar PV system, firefighters can make everything safe by disconnecting your house from the electrical grid (at least if the system is installed right).  With an independent DC system, there's no easy way to isolate that DC power in an emergency (correct me if I'm wrong).  And with low volts you have higher amps.  6 solar panels in parallel can give you 32V 60A on a good day - think ark welder.  Yes you can convert the DC to square wave AC to reduce the risk of arcing but at this stage things are starting to get more complex.  Plus, with a grid-tied system you at least get some benefit for running other base loads like your fridge, etc.

If you want to do it for the sake of a project and don't expect to save much money then cool.  Otherwise, the research I've done suggests this isn't a great option financially  .  But I would only be delighted for you to prove me wrong 

Relevant links:
http://energyd.ie/d-lumen-a-system-for-cost-effective-water-heating-from-solar-pv/
http://electrodacus.com/
https://www.eevblog.com/forum/renewable-energy/device-for-water-heating-using-solar-pv-energy-feedback-wanted/msg2849452/#msg2849452

[Gregg]

A friend of mine has an off grid cabin powered by a 24 volt DC system consisting of a Pelton wheel that produces 18 amps at 27VDC continuously plus a 500Wsolar grid that helps charge the large flooded telco type batteries during the day.  A 4500W true sine inverter provides 120VAC.  There is a Xantrex charge controller that dumps the excess power to large resistors in the power house to keep it from freezing during the winter as well as one under the cabin to help keep it dry.  The controller could be connected to a water tank to help heat water, but there is a propane water heater; adding electric water heating wouldn't make sense here, but it could be done.
Here is a link to the charge controller: http://www.xantrex.com/documents/Accessories/C%20Series/c-series-manual-975-0004-01-02-rev-d_eng.pdf

[f4eru]

Use resistors designed to heat water, for example :
https://alexnld.com/product/48v-1500w-submersible-immersion-water-heater-1npt-flange-solar-energy-heating-element/

you can stack and switch those to provide variability, or use PWM control on them.

[Seekonk]

I had to chuckle a little bit about SOA. I've built a number of PV water heating systems.   I just blew out two 110A FET in parallel in just a second with only a couple amps when the drive went linear. In fact, the most dangerous time for control circuits is early morning when they start up and the array can only produce 10W. 

These PVHW systems are quite easy to build.

[dohhomer]

Hello, i’m new to posting on the blog, so please feel free to straighten me out if I don’t hit the right tone. I also don’t have a background in electricity/electronics/science. I’m just trying to learn some stuff and it’s making my old head ache.

Anyway I thought I’d add my two cents for a change rather than just read what others have written, sort of test the temperature of the water and see if it’s suitable.

I found a website that uses the specific heat of water to make calculations about time/energy required to lift the temperature of water. I will attempt to attach a screenshot, if someone can confirm that the image can be viewed that would be good.

The calculation relies on 100% of the power (1kw) being used to heat the water with no losses. I just picked some parameters that seemed reasonable and plugged them into the calculator; 100kg of water being lifted in temperature from 10degC to 60degC using 1kw of power and this will take 5hours and 49mins.

In order to use the power efficiently it would seem a good idea to immerse the heating element completely in the water.

If the energy is “free” then any lift is better than nothing.

Just trying to further the conversation 😀

[Seekonk]

My quicky calculation for those non metric people is  60F rise, 6 gallon, 1KWH which does factor in loss.