diy solar

diy solar

What Can I personally do to help against climate change?

The leaking roof in both pictures is enough to make a grown man cry.
They are old stains on the timber, not sure from what. There is no leak up there. Having pieced together a little of the history of the building, it would not surprise me if they were like that from before a roof replacement occurred sometime in the 1990s.

The ceiling cavity is vented.
The current vent is in the gable. I was think of adding a vent (either a whirlybird or a solar powered vent) to the ceiling itself, near the ridge line. Seems to be a lot of debate about the merits of such vents here.

Aside from the the roof itself, the NE side facing the sun (southern hemisphere) is also covered with 20 solar panels, so they take a bit of the sting out of the heating from the sun.

IMG_2723.jpeg
IMG_2176.jpeg
 
Each fall we put up clear plastic inside some windows to keep ALOT of cold from flowing into the house. 3M calls it a Window Insulator Kit.
There are DIY kits for double "glazing" but the company here which makes them does not supply to my region. They are very effective and considerably cheaper than replacement windows, although it is not cheap. This is a typical room window:

IMG_3046.jpeg

And another, also showing the wide eaves which provide good protection from Summer sun.

IMG_3045.jpeg

And the aluminium sliding glass doors (there are eight of these, a few will get replaced with a renovation). Terrible thermally but it was the 1970s and that's what was done (and is still being done here because cheap).

IMG_3047.jpeg

Then there is the 2m² of single pane skylight (with no blinds or shade) in the vaulted ceiling. On a hot Summer day those things are killers. I plan to replace with Velux double glazed skylights with inbuilt blinds:
IMG_2287.jpeg

That part of the ceiling is also pretty much impossible to do much about without lifting all the metal roofing (some of which has solar PV on it).
 
I had my hot water solar diverter installed today. I wont know if it's working properly until a day or three has passed.

View attachment 89847
That’s an interesting device.
I looked them up, and read all the sales stuff. Cool idea.
I’d think someone with solar grid tied would have this on a water heater tied into a primary water heater, with another water heater downstream on the regular service. That way, here is always cold water in the tank to dump excess in, and the secondary won’t have a need as often.
 
I’d think someone with solar grid tied would have this on a water heater tied into a primary water heater, with another water heater downstream on the regular service. That way, here is always cold water in the tank to dump excess in, and the secondary won’t have a need as often.
We export excess production for some credit. That's normal here. Feed in tariffs are much lower than import tariffs but the energy isn't wasted.

Plus our 315 litre tank is already more than ample - there's just the two of us.
 
How is the catch power working the boiler? Is it a relay type thing, or is it one if those things that control the power exactly?

Also, it says it doesnt need internet, but I don't see how it is wired. Is it measuring the grid connection current with a clamp?
 
How is the catch power working the boiler? Is it a relay type thing, or is it one if those things that control the power exactly?
It controls/varies the power delivered to the resistive heating element.

Assuming the hot water tank's thermostat controlled switch is closed, it will send available excess PV output using a CT clamp around the grid main feed to monitor the available PV output capacity (normally excess production is exported to the grid). I believe it uses a burst fire method to manage that power control.

My tank's heating element is 3.6kW (we are a 240V supply here) so of course that sets the upper limit of power it can divert. It's rated to work with up to a 4.8kW heating element.

It will preferentially use PV output to heat the tank during the day. If the tank's thermostat switch has not opened before the end of the solar day is done, then the unit will use late night off-peak to top up the tank if required.

It has quite a bit of smarts built in to assess the amount of overall heating typically required, adjusting the time window to best operate, learning if it has also an off-peak power supply also available (mine does) and not necessarily defaulting to off-peak top up every night as that's often not required.

There are manual over rides if you need and a range of settings which more or less aggressively rely on solar PV - but it will always ensure water ultimately gets heated with off-peak supply if there is a prolonged period of poor PV output.

Hopefully today it will kick into action using solar PV to heat the water, provided the water needs heating of course. Water was heated last night - the unit on its first days goes through a learning cycle.

It will perform an initial late night heating cycle, partly to ensure there is hot water, secondly it goes though a process of learning if a dedicated off-peak supply exists (as distinct from off-peak time of use tariff), and if there is, it learns when that supply normally becomes available. In my case my dedicated off-peak supply is a "controlled load", which is a separately metered supply circuit which uses a ripple controlled relay to switch the supply on and off each night. Usually the switch comes on a bit before midnight.

I was able to verify it did exactly this by monitoring the power and voltage data. My tank was heated for 1 hour and 12 minutes last night. The Catch unit started the heating at 10:46pm, and at ~11:50pm the controlled load circuit became available, so the Catch unit switched over to use that supply instead, but only for another 8-minutes until the thermostat switch opened.

Today is a sunny day so we'll see!
 
In unrelated news, while the electrician was here I had him install an extra general power outlet in the office supplied by my off-grid system. The off-grid system is right outside the wall of the office, so it was a pretty simple job - but it now means the office equipment is powered by the off-grid system - which were the only computing devices I did not have covered by some form of UPS.

Just takes another little bit of load away from my grid tied system and uses energy from the off-grid system which would otherwise go to waste.

I also did a phase assignment swap for my two aircon systems - this was to help solve an another unrelated issue of particularly high grid voltages on one of my phases which were getting bad enough to not only throttle my grid tied inverter's output but also shut it down entirely at times.

I went through an extensive process of investigation and analysis to arrive at suitable solutions for this. The two most effective have been invoking the Volt-Var setting on my inverter and this phase assignment swap.

Since doing the former the inverter shut downs have been eliminated, and I'm pretty sure today will be a good example of the latter significantly eliminating the need for my inverter to throttle production due to grid over voltage.

What this all means is fewer losses or curtailment of solar production, and that's a good thing.
 
In unrelated news, while the electrician was here I had him install an extra general power outlet in the office supplied by my off-grid system. The off-grid system is right outside the wall of the office, so it was a pretty simple job - but it now means the office equipment is powered by the off-grid system - which were the only computing devices I did not have covered by some form of UPS.

Just takes another little bit of load away from my grid tied system and uses energy from the off-grid system which would otherwise go to waste.

I also did a phase assignment swap for my two aircon systems - this was to help solve an another unrelated issue of particularly high grid voltages on one of my phases which were getting bad enough to not only throttle my grid tied inverter's output but also shut it down entirely at times.

I went through an extensive process of investigation and analysis to arrive at suitable solutions for this. The two most effective have been invoking the Volt-Var setting on my inverter and this phase assignment swap.

Since doing the former the inverter shut downs have been eliminated, and I'm pretty sure today will be a good example of the latter significantly eliminating the need for my inverter to throttle production due to grid over voltage.

What this all means is fewer losses or curtailment of solar production, and that's a good thing.
Again, good for you. Nice going.
I think it's pretty cool participating in a worldwide classroom. I'm getting smarter by the minute.
 
@wattmatters Thanks for explanation.
I am working on similar product ideas, but in a more simple way.
Apart from a boiler which has this all integrated (and is in two stages) I would like to control boilers on/off.

Here in (mainland) Western-Europe things typically go a bit differently.
We typically don't do the large direct electric boilers. They are typically heatpump boilers. (or existing stuff like gas boilers and so on)
But anything under 200 liters and with strict constraints for not being able to use heat pump boilers, a lot of times an electric boiler is used. Typically smaller households or not so much water demand.

What this means is we rarely have these high power boilers like 3.6kW or higher.
Typically max 2.2kW (10amp at 230v) at a normal socket
In my opinion, that should just be switched on or off at the right times, just to keep cost down and make installation much simpler.
Because there will be enough moments where you will have power like that and no consumption.
For smaller solar installations a boiler with a less powerful boiler (less than 2kW are available) can be used.
 
What this means is we rarely have these high power boilers like 3.6kW or higher.
Typically max 2.2kW (10amp at 230v) at a normal socket
In my opinion, that should just be switched on or off at the right times, just to keep cost down and make installation much simpler.
Definitely.

One of the options I had was this exact approach, and it is a good one.

I analysed it extensively (more on that below) however I also had some special considerations which meant this option carried some extra risks for me and the diverter was a better option in my case (also more on that below).

So to carry on with your thoughts - I had the following possibility:
1. Replace the heating element in my hot water tank with a lower power element. 2.4kW or 1.8kW are the commonly available options.
2. Install a timer switch and relay controlled contactor to turn on the power to the hot water circuit at either set times of day, or based on some basic home automation.

As you say, this is a fairly simple on/off scenario which if the timing of the switch is aligned with peak solar PV production hours it will result in a high proportion of self-consumption of PV for hot water and not be overly reliant on the grid for supplemental energy. Obviously how much energy comes from solar PV and how much from the grid depends on the individual home's hot water energy consumption and available solar PV output, as well as total hot water storage capacity and daily heat loss rates (our losses are ~2kWh/day).

I had already done a LOT of analysis on this approach last year as I have years of detailed data on our excess solar PV availability and hot water energy consumption to work with. I know my exact HW energy demand for every day and also have my excess solar PV data in 5-min intervals, so I was able to run a model to know how such a system would perform as well as run an optimisation to determine what was the best time of day to activate a (dumb timer) relay.

In the charts below the height of the columns shows our hot water energy consumption each day for a full year, and the yellow part is that which could have been supplied by our solar PV, while the grey is what supplemental energy would be required from the grid on that day.

e.g. this is the daily chart of a 12 month period had I used this approach with a 1.8kW heating element.

qH4JHGD.png


and with a 2.4kW heating element:

GTdtbd4.png


and compared with leaving the 3.6kW heating element in situ:

9xhTJkF.png


I also did consolidated monthly assessments, here's the 2.4kW element scenario to show the (minor) seasonality (Jun-Aug is Winter here) in the self consumption rates.

VR5dfTi.png


Late Winter and Spring here are very mild and usually reasonably dry conditions so our home consumption overall is lowest and solar PV is abundant while, Summer (highest overall loads + rain) and our short Winter see lower availability of PV.

What we can see from the above charts is that the 1.8kW element wouldn't provide all that much more benefit over a 2.4kW element but they were both quite a bit better than the existing 3.6kW element.

This approach, combined with a programmable relay controller using data on excess solar production, could see better than 80% of our hot water energy being supplied by our solar PV.

So yes, that was most definitely an option I was considering and there are already a number of product options for doing just this, some even interface directly with the solar PV inverter's data and/or can be controlled by the inverter directly.

So why did I choose the smart diverter instead? A few reasons:

i. It will perform even better
However this was not the primary factor.
With the hot water diverter I am expecting better than 90% of hot water energy demand to be self-consumption of solar PV. So it's a step up from basic timer/relay control the smart diverter, although not massively better as the timer/relay for us would already be pretty good.

ii. We have a strict per phase grid supply limit
By moving the hot water heating load to daytime it may well find itself being on at the same time as high power draw appliances which share the same supply phase. When hot water is heated overnight it is far less likely to be operating at a time when other (typically daytime) high power draw appliances on that phase are being used.

As a result I also looked for a solution which on the days when solar PV was poor would not risk exceeding my supply limit. That was not going to be possible with a basic timer/relay controller - it would need to move up to the next step of being programmable based on available solar PV / loads.

So why don't I just rebalance our loads across phases? Well that's already been looked at carefully and done - I am covering for two occupied dwellings / three buildings with a total of five aircon systems, two electric ovens and two electric cooktops, so balancing such loads is not quite so simple.

The smart diverter will use only the available solar during the day and not draw from the grid if the weather is poor (or we are using a lot of power elsewhere) - if those conditions persist it will then revert to drawing from the off-grid circuit overnight. This means that if during a day with modest solar availability we happen to be turn on the aircon system which shares the same phase and the wife wants to boil the kettle to make a cup of tea, then the smart diverter will ensure a rapid reduction in hot water power consumption and grid imports will not exceed our supply limits. It will be fine with aircon + kettle + other general loads on the phase, but adding a 3.6kW load would tip us over the load supply limit. When those appliances turn off then it can revert to drawing whatever available solar PV is there.

iii. We also have daytime grid over-voltage issues
This is a more involved and complex problem I've been grappling with for a long time (resulting in regular PV output throttling and even inverter shut downs). I have recently been able to resolve this with changes to my inverter's reactive power control settings, phase load changes and some other stuff.

It's partly related to ii. above. Fundamentally, adding daytime load is a good thing as that helps to drop our supply voltages - however I also need to do it in a way which is more responsive to the actual solar conditions on any given day. The smart diverter complements that need as it only draws daytime power when excess solar is there and this just happens to coincide with when grid voltages are high and need load to be lowered.
 
Definitely.

One of the options I had was this exact approach, and it is a good one.

I analysed it extensively

So yes, that was most definitely an option I was considering and there are already a number of product options for doing just this, some even interface directly with the solar PV inverter's data and/or can be controlled by the inverter directly.
Which products on the market this ability? I might not be aware of Aussie specific products.
Of course you can build this functionality with all kind of home automation stuff (open source or commercially available) but that is already to difficult for most people.
Connectivity of PV inverters can be nice, but typically I find it not so useful because you only want to know how much you are exporting or not. So you only really know the situation when you know grid connection status.

Great analysis. I want to display self-consumption (for the whole house, and for the boiler) in a similar fashion as the last graph. So users get an overview of what they have achieved. How much grid powered / Solar powered but also battery powered for those who have that (which is me)
Because when the hot water demand is no longer delay-able you can still have situations where you can run it on stored solar energy from your battery. This number needs to be as low as possible, but therefore very useful to show it.

So why did I choose the smart diverter instead? A few reasons:

i. It will perform even better
However this was not the primary factor.
With the hot water diverter I am expecting better than 90% of hot water energy demand to be self-consumption of solar PV. So it's a step up from basic timer/relay control the smart diverter, although not massively better as the timer/relay for us would already be pretty good.
Yes agreed, a power diverter will always perform better, the question is if it will outweigh the costs.
A big part is the installation. Which does not only cost money, there is also a huge shortage in qualified personel.
With my products I would like to focus on low cost and easy achievability for consumers. Lets say you buy something that costs 100 up to 200 Euro and you plug it, connect it and go.
ii. We have a strict per phase grid supply limit
By moving the hot water heating load to daytime it may well find itself being on at the same time as high power draw appliances which share the same supply phase. When hot water is heated overnight it is far less likely to be operating at a time when other (typically daytime) high power draw appliances on that phase are being used.

As a result I also looked for a solution which on the days when solar PV was poor would not risk exceeding my supply limit. That was not going to be possible with a basic timer/relay controller - it would need to move up to the next step of being programmable based on available solar PV / loads.

So why don't I just rebalance our loads across phases? Well that's already been looked at carefully and done - I am covering for two occupied dwellings / three buildings with a total of five aircon systems, two electric ovens and two electric cooktops, so balancing such loads is not quite so simple.

The smart diverter will use only the available solar during the day and not draw from the grid if the weather is poor (or we are using a lot of power elsewhere) - if those conditions persist it will then revert to drawing from the off-grid circuit overnight. This means that if during a day with modest solar availability we happen to be turn on the aircon system which shares the same phase and the wife wants to boil the kettle to make a cup of tea, then the smart diverter will ensure a rapid reduction in hot water power consumption and grid imports will not exceed our supply limits. It will be fine with aircon + kettle + other general loads on the phase, but adding a 3.6kW load would tip us over the load supply limit. When those appliances turn off then it can revert to drawing whatever available solar PV is there.
Typically I would say any heater that knows of its surrounding (meaning local grid and so on) should already prevent breaching the supply limit.
But yeah, 3.6kW is pretty strong on a grid connection when you got other stuff to power too.
However, running an electric boiler on battery power over night is the scenario I would like to avoid. Sometimes it is necessary, but it feels a bit like charging one 'battery' with another battery, and causes unneeded wear. It is also in scenarios where there is little power anyway (cloudy days) so the energy is useful for other things.
iii. We also have daytime grid over-voltage issues
This is a more involved and complex problem I've been grappling with for a long time (resulting in regular PV output throttling and even inverter shut downs). I have recently been able to resolve this with changes to my inverter's reactive power control settings, phase load changes and some other stuff.

It's partly related to ii. above. Fundamentally, adding daytime load is a good thing as that helps to drop our supply voltages - however I also need to do it in a way which is more responsive to the actual solar conditions on any given day. The smart diverter complements that need as it only draws daytime power when excess solar is there and this just happens to coincide with when grid voltages are high and need load to be lowered.
The big difference you can make in case of battery charging or boiler heating, is that you make it a tiny bit smarter, so that you consume power not when it is there, but at it's peak.
Say you can charge/heat with 2kW and you have a 4kW solar installation on a summers day. You want all the power consumed between 2kW and 4kW solar output. Flattening the curve.
 
Yes agreed, a power diverter will always perform better, the question is if it will outweigh the costs.
In general, no. At least not with the tariff structures we have in place.

I had already done that analysis and it's the advice I give to most here - the simple timer+contactor solution is best value option for achieving effective money and emissions savings.

As I pointed out though, there were technical factors at play in my decision which meant variable power diversion was a less risky option for me than simple on/off switching. Plus I was prepared to spend a bit more to achieve a greater level of CO2 emissions reduction.

A big part is the installation. Which does not only cost money, there is also a huge shortage in qualified personel.
With my products I would like to focus on low cost and easy achievability for consumers. Lets say you buy something that costs 100 up to 200 Euro and you plug it, connect it and go.
No such device would be legal to be plug and play in Australia.

The power supplies for off-peak water heating are hardwired (licensed electrician only) and any alternative supply to a 3.6kW unit requires its own dedicated circuit and hardwiring (again all requiring licensed electrician). In fact the element size isn't relevant, they all require a licensed electrician to install.

So even if I had chosen to use a simple timer + contactor to power my hot water at set times of day, it would still have required an electrician to do the installation. The labour cost for installation between each option is not all that much different. Mine was more complicated and took longer but for a "normal" circuit board here, it's roughly a 45-min job. Try getting an electrician out and get billed for under an hour of labour. Not going to happen. There will be a minimal call out charge either way.

So the cost differential is mostly the hardware. Timer + contactor or the diverter.

We are not permitted to do any work in our circuit boards, install our own circuit boards, nor install wiring, general power outlets or light switching. All such things are required by law to be done by licensed electricians.

The only water heaters which can plug in to a regular GPO are continual flow gas units which just require an electrical energy source for ignition, and heat pump water storage heaters which only draw ~1kW or less and so can be legally "plugged in" to a suitable GPO (often an outdoor GPO which meets relevant IP standards). Most are hardwired anyway.

The big difference you can make in case of battery charging or boiler heating, is that you make it a tiny bit smarter, so that you consume power not when it is there, but at it's peak.
The diverter is supposed to be smart enough to identify the optimal times of day to operate, IOW not to start diverting energy too early or too late.

But yep, I completely agree - especially in Australia where grid-tied PV systems often operate with an export power limit, e.g. a typical home with single phase supply in Australia will have an export power limit of 5kW. So if your system is capable of generating more than that, generation will be curtailed to remain within export limits and so adding load (such as thermal or electrical energy storage) at such times is particularly advantageous as you cannot get export credits for energy you can't export. It increases the PV system's capacity utilisation. My phase export limit is lower at 3kW.

I will be monitoring the performance of the diverter in the coming days and weeks as there are many aspects of my home's setup which are atypical.
 
With the "no installation" I was referring to the lower power boilers (up to 2.2kW typically) going on a regular socket, in which case you can use an on/off plug. Which anybody is allowed to do of course.
Yes with power diverters and 3.6kW, this is very different. I think you are still allowed to do it yourself, but you still need to comply with NEN codes.

- Catch Green I have installed
- Fronius (Fronius OhmPilot) and
- Solar Edge (SolarEdge Smart Energy) have solar diverters for hot water which integrate with their grid tied inverters.
- AWS Sunmate
- AC Thor
- Powerdiverter
- myenergy Eddi
- ImmerSun

And for those DIY inclined there is a self install kit available for the MK2 PV Router

No doubt there are others.

Yes I was aware of almost all of them.
They are all power diverters which typically cost at least 500 Euro/GPB/$ and then need installation cost and people too.

My question was meant for the simple on/off solution. This is something a lot of people can build through all kinds of home automation stuff, but for most people, it is not so easy. And I have not seen any plug and play product on the market which does exactly that. Using PV/Grid data to utilize a boiler in on/off style.
Again this is for 2.2kW or lower boilers on a regular socket output.
 
Again this is for 2.2kW or lower boilers on a regular socket output.
For frequent long term reliable switching of even a 2.2kW load I think I would want to be using a relay/contactor, rather than a plug-in smart switch.

For sure a plug in smart device can control a contactor but I'd hate to think of one of those plug in units ending up with a contact welded closed due to arcing.

With something like that I could program it with Home Assistant - I already have all my Fronius inverter data integrated into HA - so it knows how much excess PV capacity there is. I've just recently automated the aircon system in my mancave with a Sensibo device integrated with Home Assistant to operate the dehumidifier mode - it takes advantage of when excess solar PV is available. I have high humidity in there and so want to help ensure no mould problems take hold. Took a while to get the logic right but it's working a treat.

But how reliable would it be? The more devices/systems in the way of getting your water heated, the more things can go wrong. That's why I ruled this option out.
 
My question was meant for the simple on/off solution. This is something a lot of people can build through all kinds of home automation stuff, but for most people, it is not so easy. And I have not seen any plug and play product on the market which does exactly that. Using PV/Grid data to utilize a boiler in on/off style.
I'm less familiar with Euro stuff. "Boilers" here are simply not of the plug in variety, they are hardwired. Exceptions are very small water tanks - 25 litre and some 50 litre models and are only suitable for indoor installation. Majority of water tanks here are 180-400 litres, are hardwired (if electric - a lot are gas) and they are installed outdoors. Mine happens to be indoors in the laundry.
 
For frequent long term reliable switching of even a 2.2kW load I think I would want to be using a relay/contactor, rather than a plug-in smart switch.

For sure a plug in smart device can control a contactor but I'd hate to think of one of those plug in units ending up with a contact welded closed due to arcing.

With something like that I could program it with Home Assistant - I already have all my Fronius inverter data integrated into HA - so it knows how much excess PV capacity there is. I've just recently automated the aircon system in my mancave with a Sensibo device integrated with Home Assistant to operate the dehumidifier mode - it takes advantage of when excess solar PV is available. I have high humidity in there and so want to help ensure no mould problems take hold. Took a while to get the logic right but it's working a treat.

But how reliable would it be? The more devices/systems in the way of getting your water heated, the more things can go wrong. That's why I ruled this option out.
It is just a smart plug. Even if it would weld shut, it would just become a regular socket agian, the boiler still works on its own with an analog or digital thermostat.

I already did a project over here with 10 houses which were connected to a very limited construction grid power, and for months on end, kept all the boilers in check with my software, not to overdemand the grid connection until they got a proper one. No issue at all.

The risk is more in the software or in network connections dropping. Therefore, if I make a product for a customer, and it loses connection for hours on end, at some point it will just start to heat again.
I'm less familiar with Euro stuff. "Boilers" here are simply not of the plug in variety, they are hardwired. Exceptions are very small water tanks - 25 litre and some 50 litre models and are only suitable for indoor installation. Majority of water tanks here are 180-400 litres, are hardwired (if electric - a lot are gas) and they are installed outdoors. Mine happens to be indoors in the laundry.
Haha so very different over here. Because of the local climate nobody will put their boilers outside. And electric boilers are done with regular sockets up to 200 liters sometimes (although typically more like 120 liter or smaller)
The larger boilers are typically not needed (instant gas boilers) or are heated by gas indirectly or nowadays more and more by heatpump.
People find it way to expensive to run the huge boilers directly on electricity only.
 
Back
Top