Water heating DC Element with PV vs Inverter

Good day everyone

Anyone faintly familiar with Solar Power would recognize the situation where the goal posts are always shifting… You get a system and your battery(ies) are the limiting factor. Then you upgrade the batteries. Suddenly you realize you need more power and hence get another MPPT and a few solar panels. Suddenly you have too much power (DC) and now the inverter is the limiting factor.

After having installed a second MPPT and additional panels, I downgraded by geyser element to a 2 KW element (200 liter) in order for the inverter to power the geyser instead of the grid. The problem is that the inverter is now running for the whole day, basically running the geyser with very few watts to spare. To make things worse, I realized that even now, at the beginning of winter, I have a bit of solar power going to waste as the batteries are full and the inverter is running flat out (midday-eish) and the MPPTs start to throttle back. Now, for most this is not an issue, but as one of the members of the “other forum” put it a long time ago… he considers wasting solar power a mortal sin due to the capital lay out for it in the first place. I tend to agree with this.

Getting to my question… I have a few options:

  1. More batteries (not gonna happen at this point in time).
  2. Grid-tied inverter (should have gone this route in the first place due to the benefit of having a second inverter but chose not to due to my affinity for the blue stuff… :smiling_face:) - so not gonna happen.
  3. Convert geyser to DC geyser via geyserwise or some such device (more about this below).
  4. Replace Victron Multiplus II 3000VA with bigger baby brother 5000VA.

Basically I am left with only option 3 and 4?

With regards to option 3. I hoped that I could use the old 4 x 370Watt panels and 150/35 MPPT to power the geycer DC element. This would free up space on the inverter to power other stuff. Having spoken to Geyserwise it seems that I will not by able to use the Victron MPPT and will only be able to use three of the 4 panels with their Geyserwise Eco MPPT. I still want to investigate this myself since I am sure the Geyserwise controller would be able to communicate just as well with the Victron MPPT as with the Geyserwise ECO MPPT. Should this option be viable this would be the most cost effective. R5000 odd for the DC element and Geyserwise Controller as long as I can use the Victron MPPT.

Alternatively, I need to go for option 4 with the added benefit of having additional capacity?

What would the collective knowledge of this forum recommend? Explore Option 3 further, option 4? Other?

PS. The obvious answers in terms of heat pump, gas powered geyser, flat panel collector, vacuum tubes etc is not an option that I want to explore for various reasons.

Bonus Question: Is it ok for a Victron Multiplus II 3000 to run continuously at maximum or self-limited derated maximum?

Thank in advance

So if I understand you correctly you have 4 x 370W panels that you are willing to dedicate solely for heating the geyser? And you are looking for a value proposition on how to do it?
If so, can you tell me the Vmpp and Impp of those panels please?

1 Like

How about a cheap Chinese modified sine wave inverter just to run the geyser element?


What I have done is connect my 4 panels in 2S2P.
This works fine…

1 Like

I would get a PV inverter to add to the output of the 3kVA MK II. Just make sure you don’t add more than 2.5kW PV (4x 370W will be fine).
You can get a second hand 1.6Kw Sunny Boy PV inverter for less than R4000

1 Like

Specs on the rest of your system, other MPPT and panels on it, battery size?

Option 5: Parallel up another Multiplus II 3000VA, IF you can find one with a compatible model number. Best to send a photo of your data plate to your Victron supplier to help you confirm this. It’s seamless with Victron, and both will present effectively as one larger inverter.

Then heat your geysers off the combined power, but do some smart programming to determine when the geyser(s) should get power (battery DoC/level, time of day, PV output, other loads, etc). Go mad with NodeRed if you’re a tinkerer (sounds like you’re up to the challenge).

You’ll need a GX and some comms cables, but I “assume” you have a GX in there already.

Then add more panels and let the cycle begin again… :wink:

SA is in this for the long-haul, so “goedkoop koop is duurkoop” - even if the US and EU governments slapped a trillion buck contract in front of the government today to fund 10GW of free power stations and have Siemens start building on Monday, I doubt that anything would be producing power in under ten years (there’s precedent and the former is pie in the sky).

1 Like

Hi Phil.g00, yes, that is the idea. I may regret that route once it is summer again with panels that are then effectively out of play to provide - for example - air conditioners…

Multiplus II 3KVA, Pyltontech 3 x 2.4 kw, MPPT 150/35 (2s2p 4 x 370W), MPPT 150/60 (2s3p 6 x 470W), Venus GX

OK, in a nutshell, the cheapest solution:
You don’t need an MPPT.
The idea is you select an element with a resistance that makes the panels run at their maximum power point. It is just a resistance, not a voltage-sensitive battery.

What isn’t mentioned in the clip below is that AC is needed through your thermostat to a DC contactor ( with an AC coil, obviously) to break the DC from the PV. The thermostat probably won’t break the DC current. You will easily be able to determine the DC rating after watching the clip.

Now, rather than re-explaining this, watch this clip:

Things should be pretty self-explanatory.
There have been other suggestions offered that may suit, but this is by far the least expensive solution to the way you framed the question.

Edit: I’d work with the NOCT Mpp values, not with the STC values, and the 25% wattage safety factor David uses is overkill.

1 Like

Is a 2KW element not a bit weenie for a 200L geyser? I thought they come out standard with a 4KW. If the geyser is on all day, and there is constant hot water usage, that geyser will run all day. I have a 150L where I have installed a 2KW element. I have it on a timer to run twice a day for 2hr slots and it pretty much takes 1.25 to 1.75 hrs to heat up on most of those periods. Thermostat is set to 55deg. A 200L set at 60 or 70deg will run all day.

1 Like

You are correct, in this instance, the situation is actually worse as the 4 panels in question will not make 2kW.

Thank you for the information and the video. Wish I spend more time paying attention in High School during physics class… :smiling_face:

The geyser indeed had a 4kw element which I removed and swapped out for the 2 kw one. Idea was to work with the capacity of the 3kva inverter. Heat water for longer at a lower wattage in order to use more solar and less grid.

Just remember, just the 2kW rating of the element doesn’t mean you’ll get 2kW. It means you can power it safely with 2 kW without burning it out.
Your four panels are not going to make 2kW, probably closer to a 1kW max.
I think a 200l geyser will lose about 2kWh/day to maintain the temperature.
Working with a (10 am-2 PM) 4-hour sunshine window doesn’t leave a whole lot of power left over to do significant heating.

1 Like

This being a 220V element means it will only draw 2Kw approx.

Standing loss. It is usually printed on the “nameplate” on the side of the tank, and I think it is even legally mandated to be within a certain limit. 2kWh is close enough as an estimate, though it should be noted that this number is measured at 65°C with an ambient of 25°C (if I recall). On a cold day your losses will be more, but since most of us only heat to 55°C, the losses will be less. Somewhere in the middle it will meet. I still use 2kWh as a rule of thumb though.

These losses are significant, so much so that one cannot depend on a geyser retaining it’s heat overnight for showers in the morning. :frowning:

My 150L geyser (no blanket, but the first 2m of all pipes are insulated) loses around 1°C per hour at 80°C, down to 0.5°C per hour at 50°C. Need to heat it to around 55°C in the afternoon to have an acceptable shower the next morning. Not perfect, but certainly not a non-starter.

Or get a class B geyser, most old geysers are D or E