diy solar

diy solar

How do I cheaply but safely heat domestic water with PV?

szilveszterandras

New Member
Joined
Oct 17, 2022
Messages
3
as the title says, i'm trying to heat water for domestic use via PV. From my research so far, the most popular suggestion seems to be solar thermal. That isn't the right option for me.

Pros of solar thermal:
smaller surface area

Pros of solar PV:
cheaper (depending on the solution)
no/fewer moving parts
ease of installation: channels for PV cabling already installed, that's not the case for solar thermal
i already have an AC immersion heater in my hot water tank
in the future i plan to install a proper 5kw PV grid-tie system so i could reuse the panels
the big one: i don't have to worry about overheating issues in the summer when i don't use hot water for a couple of days

Tank's 200 liters, and my hot water energy demand is roughly 8 kWh/day, so a 700-1000 watt system would suffice. Ideally i'd like to use an off-the-shelf solution as much as possible, not the custom microcontroller route, that said, hooking up a temperature probe to a relay through arduino etc isn't a problem. So far i entertained 3 options:

simplest/cheapest is to get a DC immersion heater and match it to PV panels to be close to MPP. I'd need a temp probe and a relay/mosfet to cut power when water temperature reaches upper limit. I'm worried about the safety of switching relatively high voltage DC current, also about potential electrolysis happening in the tank, not sure what kind of damage could that do. I know AC immersion heaters have a protective outer shell and ground protection, which wouldn't be the case for DC

use a solar charge controller like this one (https://www.epever.com/product/tracer-an-10-40a-mppt-charge-controller/) and hook only a properly rated DC immersion heater up to it. This gives me more efficiency because MPPT, but the questions of cutoff and electrolysis are still there (unless these controllers have safety mechanisms in them). I also don't know if these controllers can work without batteries connected, but i'd like to avoid batteries in the system.

use an off-grid solar inverter that already outputs 240v, and hook it up to the existing immersion heater (AC 240v, 2500W). This is the priciest option, but easiest to install since the immersion heater is already in place, it has internal thermostatic control for cutoff and is grounded, although i don't know how an off-grid inverter handles ground protection. The question is, can off-grid inverters work without batteries hooked up? Every one i saw has battery terminals in addition to the AC plugs.
Any input is appreciated.
 
A couple years ago I found a couple places that made a PV powered solar water heater. It was all in one system. MPPT controller, heater and tank in one unit.
Here's one that uses your tank:

As I recall there were others people doing similar things to make hot water with PV panels.
 
The question is, can off-grid inverters work without batteries hooked up? Every one i saw has battery terminals in addition to the AC plugs.
Any input is appreciated.
I am not sure I think your question is even categorically correct: what is the inverter inverting if not even connected to a battery? Where is the energy actually coming from, in your scenario?
 
If you don't need a lot of hot water, replace the heater with a hybrid. On heat-pump-only mode it draws about 600 watts. Big expense will be a 240VAC (assuming US) inverter.
 
Missour wind and solar has some great emersion water heating elements including one with a built in thermostat. They also have a stand alone dc thermostat. Great. Now I have another project to try.
 
There are a dizzying array of options to do this. I'll throw out one, may not be ideal since you wanted to avoid batteries. There is a guy called Electrodacus who makes a custom designed solar charge controller/battery bms combo. The one he sells now is hard to get due to supply chain issues, but his SBMS0 is at least worth looking at.

I have the older SBMS120 that has a built in solar charger, but the SBMS0 allows you to use your own charger or get a few of his units. His units have a diversion function that goes into effect when the battery is full, where it shunts energy into whatever you designed for it. I've seen videos of people putting a flexible silicone 24v heater on the side of a small water tank in a RV setup.

Advantages, seriously, look at the cost. It is like $120-150 for the unit, plus maybe $20-40 for the diversion units. I think two panels go to each one, but he is supposed to have newer 50a units (60 cell, 24v panels).

I think you could just get a small, less expensive battery which could be full by 10am and the rest of the day it is just dumping energy into your water tank. I know the SMBS0 has some temp inputs and has some control over it all, and has decent software.

I don't have any financial interest. I'm shocked to see the prices of other options that seem to do something similar. I've been using my sbms120 around a year now and it just works. If you used 24v heating units, you wouldn't even need an inverter at all. His website has a forum where people talk about this stuff. Probably some examples there. electrodacus.com
 
Missour wind and solar has some great emersion water heating elements including one with a built in thermostat. They also have a stand alone dc thermostat. Great. Now I have another project to try.
Google reviews on Missouri wind and solar first.
 
Your cheapest and about only option is to build a power point controller. Europe is sold out of all PV water heating controllers during the current crisis. Here is one I built and variations of this are working around the world. I've had people with no experience in electronics build them. One person uses one in parallel with his grid tie system so no need to abandon it later.
GWH22828.jpg
 
Apologies for my ignorance, i don't have much of a background in electrical engineering, i just like to tinker with stuff. Surely, the cheapest and simplest way to achieve what i'm looking for is hooking up some solar panels to a roughly impedance matched immersion heating element so it produced decent power on a bright sunny day. I understand it won't be the most efficient solution in the summer, but it doesn't have to be, there's plenty of daylight going around. It would be even worse efficiency-wise during the short cloudy winter days, but that's ok too, this isn't an off-grid system, i'm just looking to save a buck or two when it's practical.
My hangup about it is how to make it safe. If a full fledged mppt controller/inverter is needed to make it safe i'm ok with that, but i'm wary of taking on the complexity of building one of my own.
Sidenote: living in europe, sourcing some of the recommendations looks tricky.
Sidenote 2: end-goal system is a full grid-tie solar array and heat-pump for both heating and DHW, but that's waay too expensive to justify as things currently stand.
 
decent power on a bright sunny day
..... at noon. If no clouds pass overhead.

I share your desire to build the simplest/cheapest system. But which of your scenarios is simpler/cheaper than the Loadmaster-style circuits above?

1. You have to switch in a DC heating element, add something to handle high temp cutoff and solve DC arcing / in-tank electrolysis. And how do you heat your water on cloudy days if you've swapped out your AC element? LoadMaster addresses or eliminates all these issues, and is more efficient to boot.

2. You still have the same issues, just with better efficiency / higher cost. And you might also need to add a battery.

3. Eliminates some issues (at even higher cost), but still need to solve high temp cutoff and probably add a battery plus inverter/charger.

All things considered, a LoadMaster approach looks simpler and cheaper to me even though I don't have the slightest idea how to actually build one. But no matter which approach you choose you'll have to build something you've never built before, so....
 
The Loadmaster is really overly complicated and expensive to make with no real advantage. There are very simple one chip designs that can do this efficiently and include an arc interrupt circuit so existing thermostats can be used without damage. A dozen parts, take two days to install each component and it is still done in no time. You probably spend more on coffee than to risk learning something.
 
If you don't need a lot of hot water, replace the heater with a hybrid. On heat-pump-only mode it draws about 600 watts. Big expense will be a 240VAC (assuming US) inverter.
Yes, hybrid heat pump or even electric point of use might be an option for something like a sink.
 
Sorry if I've misunderstood but I've got a controller that takes the excess power from my panels to heat the immersion. It detects when I'm about to export to the grid and instead heats the immersion. It's a really simple and small box that sits by the consumer unit and cost around £100 (I'm in the UK).

The bloke who made them doesn't any more but I know there are similar ones on the market over here. I've been really impressed with it - first time in the 20 years that I've lived here that I've had (usually) hot water on tap since I only ever switched on the immersion on demand.

 
The Loadmaster is really overly complicated and expensive to make with no real advantage. There are very simple one chip designs that can do this efficiently and include an arc interrupt circuit so existing thermostats can be used without damage. A dozen parts, take two days to install each component and it is still done in no time. You probably spend more on coffee than to risk learning something.
you keep saying that, yet all i have found so far is not cheap, nor readly available
 
That diversion box is a grid tie diverter which diverts when excess AC power is detected going to the grid. It can not work with panels directly and requires a grid connection.

Here is a recent paper that says PV is the way of the future to heat water. Surprising that it is research done in sunny warm Spain where you would think solar thermal would have an advantage. They say the controller should only cost 60 euro. That design does not have arc interrupt. Still, China has been slow to jump on this product. It is now producing complete water heaters with this control built in.

 
Back
Top