PV energy less expensive than natural gas.

[electrodacus]

For most people in a cold climate, 6 days is not a long lull in output.

If you refer to your location then that is not a cold climate not compared to my location where difference in temperature between inside and outside is more than 2x higher thus more than 2x more energy is required assuming the exact same house.
And you design the system for worst winter month in your case December where is mostly cloudy the entire month so the average energy production for that month is mostly based on cloudy days and thus you do not need much thermal storage as the solar output is fairly uniform. From that 250kWh in December from a 10kW PV array that will produce in a bad overcast day around 3.5 to 4kWh and your average for the month is 250kWh/30days = 8.33kWh thus that means i mostly cloudy maybe even every day of that months with a few cloudy breaks and likely not many if any fully clear sky days as a single day like that will produce 40 to 50kWh so you can not have that with just 250kWh for the entire month.
In my case where I have more extreme days with quite a few sunny days in December along with many overcast snowy days I need more thermal mass to smooth that out to an average. So yes you need a slightly larger PV array assuming same house but not a larger thermal mass storage. Also sunny days are the coldest with overcast and cloudy days typically much milder.
If I where to build this same house I have here in your part of the UK then I will need more solar panels about 40% more so 4kWp that will cost around $3.2k the same thermal mass so nothing else different thus instead of 15k it will be 18.2k investment that is still $50 bill for electricity and heating combined and while this is for a small energy efficient house it can not be lower with any other energy source including natural gas (keep in mind just about half of this bill is for heating the other half is for electricity).

You can go to my google+ account and scroll down until you will find some videos I took last winter is a time lapse with different type of days snowy overcast and you can see the PV output for those days.

[Marco]

With water storage when water gets to 20C so empty is basically empty especially with a high temperature water storage and there is no extra buffer there you will just freeze

I think you are severely overestimating the thermal resistance of floor heating. Your house seems to need only a couple kW at most of heating to maintain equilibrium, across 65m2 floor heating if not buried deep in a slab of concrete can supply that down to a couple of degrees difference (see page 16). There's the edge case of pump failure, but thermal inertia has nasty edge cases as well ... if you arrive at the house when heating failed it will take much longer to warm up and worst comes to worst and you have to switch to a wood fire all that thermal inertia really becomes your enemy.

[Red Squirrel]

I always pondered on the feasibility of using PV for electric, vs thermal solar panels.    PV is simpler, because you just have one big system, that also powers everything else.  With thermal you would need some sort of liquid like glycol or something and there's just more moving parts and risk of leaks and environmental issues etc.  Also don't know if you'd get ANY heat when it's overcast.  With PV you still get some power output.

It's my dream to one day buy a big property so I can do stuff like this and live mostly off grid. I'd do a large ground mount array, like 10-20kw maybe even more.   WAY cheaper in the long run, as utility bills and costs of living keep going up.   My current property is just too small to do a decent size PV array though.  I did setup 400w of solar panels on my shed to experiment and get around 60w or so on a good day.  It's facing west as that is all that was really viable due to shadows.   But it's cool to know that even when it's overcast I still get SOME output.   I have not yet my system during a sunny day.   We don't really get much sun here (northern Ontario).  In summer we do get more though. With big enough inverter could easily run AC.

I still have work to do on my system though such as some monitoring/automation (like low voltage cut off), heated/insulated battery box, etc.    For hydrogen mitigation I'll probably build a small HRV system.  Eventually it will power outside lights.


Oh and another thing, in an off grid setting you'd probably want wood heat too, nothing like sitting in front of a wood stove on a -40 day watching the storm outside.  So even if the PV is not keeping up every day you could still start a fire in the wood stove.  Me personally I'd probably also want a force air system just for circulation.  In summer when the sun is out and the days are long there would be enough excess power to run a full size central AC unit.  The inverter would be expensive though...

[electrodacus]

I think you are severely overestimating the thermal resistance of floor heating. Your house seems to need only a couple kW at most of heating to maintain equilibrium, across 65m2 floor heating if not buried deep in a slab of concrete can supply that down to a couple of degrees difference (see page 16). There's the edge case of pump failure, but thermal inertia has nasty edge cases as well ... if you arrive at the house when heating failed it will take much longer to warm up and worst comes to worst and you have to switch to a wood fire all that thermal inertia really becomes your enemy.

Not sure you understood the point I was trying to make with the floor thermal storage vs water thermal storage.
The usable range for floor thermal storage is much smaller and the extreme limits that it can be used is 30C max (that is the spec for any type of floor heating) and minimum is 18C for comfort thus the 12C delta range for max usable capacity.
For water storage if you say have a steel storage tank you can go from 20C up to 80C or closer to boiling point.
Say your house has no floor thermal storage or very little like just minimum concrete thickness needed for floor heating and you have all the thermal storage calculated for say worst case 6 days but then you get a year where you get 8 bad days.
Say both storage are 180kWh and house needs 30kWh/day (say this is for -10C out + 20C inside)  (just round numbers for this example).
Now if all your storage if the concrete floor and you have 180kWh at 12C delta (18C to 30C) typical ambient may be 2C lower than floor temperature
And water storage 180kW at 60C delta (20C to 80C).

Now both will deal with those 6 days without any solar (is just for the example there is no such thing as no solar output unless at north pole).

While both have 180kWh usable after that is spend the remaining energy before house will freeze inside (0C) is 60kWh about 2 days theoretical for water storage but about 270kWh before concrete floor gets to 0C freezing.
There are many situations here in Canada and maybe other cold places where the heating fails and since the house has no thermal storage at all (wood frame houses) the pipes inside the house for water and or heating will just freeze in a few hours and that will be very costly and inconvenient.

A house with proper thermal mass to work with solar will only lose maybe 2C in 24h if heating fails thus there are quite a few days even a week for you to fix the situation.
There is nothing to fail on my heating as I have electric heating wires embedded in the floor (solid state nothing to fail) and if say the DMPPT450 fails for some reason you just remove that and connect the PV directly to the heating wires a bit less efficient but it will still work indefinitely until the DMPPT450 gets fixed or replaced.
Wires can not fail as current is limited both by the PV panels that are constant current limited and the wire resistance in case of heating wire so there is nothing more reliable than direct PV heating.
Sorry for getting over complicated with the answer.

The short answer is that concrete floor has larger margins and best is to combine floor thermal storage with water thermal storage as I will have as then you can control the temperature much better and be able to keep the house within 1 or 2C delta at all times.

I always pondered on the feasibility of using PV for electric, vs thermal solar panels.    PV is simpler, because you just have one big system, that also powers everything else.  With thermal you would need some sort of liquid like glycol or something and there's just more moving parts and risk of leaks and environmental issues etc.  Also don't know if you'd get ANY heat when it's overcast.  With PV you still get some power output.

It's my dream to one day buy a big property so I can do stuff like this and live mostly off grid. I'd do a large ground mount array, like 10-20kw maybe even more.   WAY cheaper in the long run, as utility bills and costs of living keep going up.   My current property is just too small to do a decent size PV array though.  I did setup 400w of solar panels on my shed to experiment and get around 60w or so on a good day.  It's facing west as that is all that was really viable due to shadows.   But it's cool to know that even when it's overcast I still get SOME output.   I have not yet my system during a sunny day.   We don't really get much sun here (northern Ontario).  In summer we do get more though. With big enough inverter could easily run AC.

I still have work to do on my system though such as some monitoring/automation (like low voltage cut off), heated/insulated battery box, etc.    For hydrogen mitigation I'll probably build a small HRV system.  Eventually it will power outside lights.

Oh and another thing, in an off grid setting you'd probably want wood heat too, nothing like sitting in front of a wood stove on a -40 day watching the storm outside.  So even if the PV is not keeping up every day you could still start a fire in the wood stove.  Me personally I'd probably also want a force air system just for circulation.  In summer when the sun is out and the days are long there would be enough excess power to run a full size central AC unit.  The inverter would be expensive though...

Thermal solar is not only less reliable but also more expensive with only advantage that it takes less space about 2.5x less space for same output. You will get some output from thermal solar even when overcast but if is very cold it may only cover the losses so there may not be high enough temperature to get anything usable.
If I will have designed the house I have now I could have integrated all the necessary PV array in to the house structure (probably an A frame type structure to have the right angles for the panels).

Using Lead Acid battery for storage is just a waste of money. There is a replay I made earlier about cost amortization and Lead Acid is just not suitable for energy storage.

I have no backup heating as is not needed and adding a backup system even for 10% of the time is not cost effective. As long as you size the system for worst case PV array size and thermal storage size it will just work in any conditions and will be the most cost effective solution (Wood is not even close in therms of cost that is why I always mention natural gas as that is the one that gets close).
I do not need air conditioning/cooling in summer as the house is well designed and my low energy use means I do not produce much heat inside the house. Most houses are just badly designed with large windows bad roof thermal insulation and use a lot of electricity that all ends up as heat inside the house and needs to be removed.

[Marco]

Not sure you understood the point I was trying to make with the floor thermal storage vs water thermal storage.

I did not ... but it didn't help that you used hyperbole. A quarter of the energy isn't exactly empty.

The short answer is that concrete floor has larger margins and best is to combine floor thermal storage with water thermal storage as I will have as then you can control the temperature much better and be able to keep the house within 1 or 2C delta at all times.

Question remains though, do you wear shorts? 25 degrees is a little more than comfortable.

[Red Squirrel]

The thing with lead acid is they can be floated and don't need to be balance charged so it makes the system and electronics much simpler to setup.   Currently they are also cheaper per kwh, where weight/size does not matter.   I think this will change soon though as prices of lithium ion and LiFePO4 etc come down due to more popularity.    TBH I still trust lead acid more from a safety point of view though.  Less likely to explode if you mess up the electronics, logic etc for charging.  Keep them at float voltage and they are happy and will never overcharge.  Run an equalize once every couple weeks, and good to go. 

[electrodacus]

I did not ... but it didn't help that you used hyperbole. A quarter of the energy isn't exactly empty.

Usable energy as at 20C storage you can not longer maintain ambient above 18C and in less than 24h will be less than +10C very uncomfortable then in less than another 24C permanent damage to pipes will occur
While with with concrete floor you have much larger margins outside the usable region.

Question remains though, do you wear shorts? 25 degrees is a little more than comfortable.

Not sure why you think 25C is not comfortable ? I had a max of 25.5C a few days ago and it was no issue extremely comfortable. Is not the same as 25C outside temperature that is measured in shade and then the sun heats you up way more than that.
An average temperature of 23C with max of 26 and min of 20 is ideal for me. Keep in mind that the fluctuation is no more than 2C per day (24h). It is not like I have 26C during the day and then 20C in the morning.
Also since is floor heating is warmer closer to floor and colder closer to ceiling.

The thing with lead acid is they can be floated and don't need to be balance charged so it makes the system and electronics much simpler to setup.   Currently they are also cheaper per kwh, where weight/size does not matter.   I think this will change soon though as prices of lithium ion and LiFePO4 etc come down due to more popularity.    TBH I still trust lead acid more from a safety point of view though.  Less likely to explode if you mess up the electronics, logic etc for charging.  Keep them at float voltage and they are happy and will never overcharge.  Run an equalize once every couple weeks, and good to go. 

Lead Acid charging is actually more complex than Lithium and Lead Acid are completely inadequate for solar as if they are not fully charged they will suffer (degrade much faster) and adding a second energy source like a generator just to keep them charged in cloudy days adds even more to the cost.
Initial cost can be the same for LiFePO4 as for a good Lead Acid as you need a lower capacity LiFePO4 to perform the same as a Lead acid because of charge discharge efficiency difference and because Lead Acid should not be deep discharged to much.
As for cost is about cost amortization and not investment cost and Lead Acid is significantly more expensive in therms of cost amortization than LiFePO4 and the cost of the BMS is very insignificant typically 5% or less.
LiFePO4 is at least as safe if not safer than Lead Acid.   Lead Acid are only good if you never use them like in a battery for starting the car or UPS where is discharged maybe once a year and the rest of the time they spend fully charged.
I will not have selected LiFePO4 for my own house if there was a lower cost solution (lower cost as in amortization cost).

[Red Squirrel]

Lithium is way more complex because it has to be balance charged and charged at very specific currents etc.  So in a renewable energy application you might have like 20+ cells, you need 22 leads from a balance charger that can then charge each cell individually (by cell I mean a group of cells in parallel) AND it needs to somehow also be able to track how much current is going into the load so it can make sure the right amount goes to the battery.  I'm actually curious how this is done TBH, I've researched it and never found much info.  I imagine it also requires a lot more wire too. 

With lead acid you just put all the cells in series, float them at 2.25v per cell and bob's your uncle.  Of course some chargers will do things that are more fancy, if you want them charged faster, but you can get away with constant voltage 100% of the time and they will never degrade. Though equalization is a good idea once in a while.   There is a reason the telcos still use strings of large lead acid batteries.  They are much simpler.   Price out 18650's and you cannot really beat lead acid for price either, especially if you consider your time to spot weld them together etc.  Now at some point maybe they'll make lithium battery packs that you can buy but I imagine that is quite a niche thing right now so probably hard to find packs in a specific/consistent size.

For something like a RC car it's not as bad because you have the discharge lead, where all the cells are in series, and then you have the balance charge lead.  Both are never used at same time.  So the charger only has to worry about the current going into each cell, not what is coming out. 

So from a design standpoint designing a system to use lead acid is much much simpler than lithium or other cell types that are more picky about current.  Unless there's something I'm missing here.  But based on what I've read lithium is much more complex due to not being able to just string them all in series and charge them at a constant voltage.   I'm not sure about LiFePO4, those are even more niche and harder to find.  Though I punched that into Amazon and there is results, but I honestly have not read up too much on that tech so not sure how complex it is to charge compared to lithium.  I hear they are safer, so there is that. 

[tautech]

Actually not so much in practice and while it is true each lithium cell needs be monitored, the rate of charge/discharge can be monitored for a 'pack'.
I do regret not taking pics of a 22KW lithium setup I saw recently as it was just series cells but each one had a sense pair of wires.

As for charge/discharge rates, well have a look at some of the specs on what lithium can 'really' achieve:
http://en.thundersky-winston.com/product/tqkqsgbdcfy.html

Have a squiz at some of the big cells on P2. 

[electrodacus]

Lithium is way more complex because it has to be balance charged and charged at very specific currents etc.  So in a renewable energy application you might have like 20+ cells, you need 22 leads from a balance charger that can then charge each cell individually (by cell I mean a group of cells in parallel) AND it needs to somehow also be able to track how much current is going into the load so it can make sure the right amount goes to the battery.  I'm actually curious how this is done TBH, I've researched it and never found much info.  I imagine it also requires a lot more wire too. 

With lead acid you just put all the cells in series, float them at 2.25v per cell and bob's your uncle.  Of course some chargers will do things that are more fancy, if you want them charged faster, but you can get away with constant voltage 100% of the time and they will never degrade. Though equalization is a good idea once in a while.   There is a reason the telcos still use strings of large lead acid batteries.  They are much simpler.   Price out 18650's and you cannot really beat lead acid for price either, especially if you consider your time to spot weld them together etc.  Now at some point maybe they'll make lithium battery packs that you can buy but I imagine that is quite a niche thing right now so probably hard to find packs in a specific/consistent size.

For something like a RC car it's not as bad because you have the discharge lead, where all the cells are in series, and then you have the balance charge lead.  Both are never used at same time.  So the charger only has to worry about the current going into each cell, not what is coming out. 

So from a design standpoint designing a system to use lead acid is much much simpler than lithium or other cell types that are more picky about current.  Unless there's something I'm missing here.  But based on what I've read lithium is much more complex due to not being able to just string them all in series and charge them at a constant voltage.   I'm not sure about LiFePO4, those are even more niche and harder to find.  Though I punched that into Amazon and there is results, but I honestly have not read up too much on that tech so not sure how complex it is to charge compared to lithium.  I hear they are safer, so there is that.

I'm sort of an expert in Lithium battery charging as I designed the Solar BMS  here is the link to user manual this is an all in one solar charger + BMS + advanced energy monitor and is fairly inexpensive likely better than a Lead Acid charger + energy monitor of similar quality. Sorry if this looks like advertising it is open source so you can maybe build your own.

Charging a single lithium cell of any type LiCoO2 (laptops or cellphones), NMC (mostly EV's) or LiFePO4 (best for stationary energy storage) is exactly the same as charging a capacitor in the sense that you just push current from a constant current source (PV panel as an example is a constant current limited source) and then just monitor the voltage and when it gets to the voltage limit say 3.55V for LiFePO4 just disconnect the source and done you cell is now fully charged and while discharging you just check that voltage is above 2.8V for LiFePO4 and if is below just disconnect the load. That is all that it is to LiFePO4 charging or any other type of rechargeable lithium battery much simpler than minimum 3 stage charging on Lead Acid.
Any number of lithium cells in parallel are no different from a single larger capacity cell.
As for cell in series that is when you need the BMS (battery management system) that will monitor each cell voltage (only the cells in series) so say 8 LiFePO4 in series will give you a nominal 24V battery pack so BMS will monitor the 8 cells in series (each cell can be made of as many parallel cells as you want but large format cells exist so you do not need to do any soldering).
Then all you do is connect the most common and cost effective PV panel that is made with 60 PV cells in series and you have a very efficient charging system with minimal components.

So LiFePO4 is as simple to charge as a capacitor single stage constant current (also known as bulk) charging is all that is needed.
Charge current can be any value between zero and 0.3C (0.3C for long cycle life but in applications where you do not care about that it can be higher). My A123 20Ah LiFePO4 cells are spec at 5400 cycles 100% DOD (Depth of Discharge) with 1C charge and 1C discharge lab test. The large format Winston LiFePO4 cells are also rated at 5000 cycles at 80% DOD 1C charge and discharge rate so almost as good as A123 cells and for sure better than any other type of battery and much more than you will be able to use in a typical offgrid installation.
My battery after 1 year of full time use has degraded a bit less than 1% thus at least 15 to 20 years of heavy duty offgrid use is expected before battery will be at 80% of original capacity.
The result from recent battery test can be found here .

A Lead Acid battery used in same condition's as my battery will needed to be 2x to 3x the capacity cost about the same or slightly more and not last more than 4 to 5 years best case scenario thus in therms of cost amortization Lead Acid batteries as as bad as it gets and I talk about the best available deep cycle Lead Acid batteries specifically designed for solar.
Telcos use this batteries as UPS / bakup in case of grid failure and that use case is very different than offgrid solar as battery will see almost no use an be just in standby and almost never discharged.

There is no Hydrogen gas generation as with Lead Acid so my LiFePO4 is inside the house just about 1 meter from me on the side where is nice and warm. A Lead Acid can not be inside because of the hydrogen gas generation during charging process.

[Red Squirrel]

Yeah helps to be an expert haha.  I'll have to read your paper when I get the chance.  This is probably the future of energy storage and will slowly phase out lead acid, just trying to wrap my head around how it works due to the more complex charging, having to balance charge etc while it is being discharged at same time.     Don't lithiumion also have very specific charge phases? Constant current for some stages and constant voltage during different stages as well.  If you can treat like a capacitor then yeah it would make it quite simpler.     Though don't you still have to balance charge each one?    You'd have around 13 or so cells (groups of cells in parallel) for a PV setup I imagine so you can feed a 48v inverter.  If you need to constant current charge them you need to somehow be able to take the load into account too as you need to differentiate how much current going into the battery vs the load.   

[electrodacus]

Yeah helps to be an expert haha.  I'll have to read your paper when I get the chance.  This is probably the future of energy storage and will slowly phase out lead acid, just trying to wrap my head around how it works due to the more complex charging, having to balance charge etc while it is being discharged at same time.     Don't lithiumion also have very specific charge phases? Constant current for some stages and constant voltage during different stages as well.  If you can treat like a capacitor then yeah it would make it quite simpler.     Though don't you still have to balance charge each one?    You'd have around 13 or so cells (groups of cells in parallel) for a PV setup I imagine so you can feed a 48v inverter.  If you need to constant current charge them you need to somehow be able to take the load into account too as you need to differentiate how much current going into the battery vs the load.

LiFePO4 will be fully charged with just  the constant current stage as for the high energy density lithium cells you will also need constant voltage if you want to fully charge them but that will have a negative impact on cycle life. In some application like mobile electronics and even EV's the trade off of reduced cycle life for the extra capacity is worth but not in solar energy storage. Charged with just constant current to 4.2V the high energy density cells will be at around 80 to 90% of the real state of charge depending on charge rate but by doing so you double the cycle life compared to cc and cv charging to 100% and this is an advantage for stationary energy storage where energy density is not relevant. Still even so the high energy density cells are not as cost effective as LiFePO4 that are by far the winners in this category.
Cell balancing is done automatically by the SBMS (based on the ISL94203) and the amount of energy needed for cell balancing is extremely low with new high quality cells.
You want to say cells in series not parallel and for LiFePO4 that will be 16 cells in series for a 48V system. My SBMS only supports max 8s so max 24V battery and while two SBMS0 can be used for a 48V setup is not ideal.
Constant current charging does not mean the current needs to have the same value at all times it just means current is limited and will not exceed a max value. Solar PV panels are constant current sources and assuming you have constant amount of light you will have a constant amount of current if light drops to half the current will also drop to half but it will still be constant current irrespective of the current level.
All you need between a solar PV panel and a lithium cell is a switch to turn the charging off when cell got to the max charge voltage level (so it is that simple ).

[Red Squirrel]

Oh interesting so you literally dump as much current as you can into it as long as voltage stays below threshold?  I figured you had to follow a very specific curve and not exceed a certain current during certain stages.    So even if they are discharging at same time (house loads using power) the voltage is just never going to hit that threshold so you can just keep dumping current?  Unless you're using less than what is going in then it will eventually reach equilibrium.

And yeah was unclear when I said parallel I meant a single cell would consist of multiple cells in parallel then those groups would be in series, like a bunch of 18650's or what not.

So is balance charging not all that important then?    Quick look at your doc shows you are monitoring the cells but not actually balance charging? 

[electrodacus]

Oh interesting so you literally dump as much current as you can into it as long as voltage stays below threshold?  I figured you had to follow a very specific curve and not exceed a certain current during certain stages.    So even if they are discharging at same time (house loads using power) the voltage is just never going to hit that threshold so you can just keep dumping current?  Unless you're using less than what is going in then it will eventually reach equilibrium.

And yeah was unclear when I said parallel I meant a single cell would consist of multiple cells in parallel then those groups would be in series, like a bunch of 18650's or what not.

So is balance charging not all that important then?    Quick look at your doc shows you are monitoring the cells but not actually balance charging?

You can only dump as much current as it is available and that depends on amount of sunlight. The max current will not exceed about 0.3C as that is both the ideal max charge current for long life on LiFePO4 and about max of what you will want to have as ratio typically for a offgrid setup.
So say for a 5kWh battery you will want to have a 1 to 1.5kW PV array ideally.  If you talk about the DMPPT450 where I have a 10kW PV array that device will be able to select how many panels are transferred to battery so that charge current always stays below 0.3C

I do cell balancing (The SBMS will do that) and the cell balancing is done only during charging by default is the same type of cell balancing that most commercial EV's use like Nissan Leaf or Tesla and is extremely effective as cell balancing will start as soon as there is more than 10mV delta between any two cells and then higher cells will receive slightly lower charge current than the lower cells until it is balanced within 10mV (all this and over 30 more parameters are user programmable I just mention the default behavior).
SBMS will take care about all aspects with no user intervention for the life of the system.

[f4eru]

Yes, this resistive system makes economic sense for this special off-grid application
Clever!
You could build an extremely simple and very efficient MPPT by switching ON/OFF single or groups of your paralell resistors, and thus modulating the load to the optimum directly.



I personally would have buried some tubes in the concrete slab for an eventual later upgrade to a water based system if ever the grid reaches your house sometime in the future.
Also, one thing lacking in your SBMS hardware is overvoltage protection over the ideal diode mosfets at the input.

[Red Squirrel]

If you can live off grid why even connect?  One of the best things about being off grid is less cost.  I would love to buy a bigger property that is away from the city eventually and do the same.  Bills cost too much and keep going up each year.  The more self sufficient you can be, the less your cost of living is. 

[electrodacus]

Yes, this resistive system makes economic sense for this special off-grid application
Clever!
You could build an extremely simple and very efficient MPPT by switching ON/OFF single or groups of your paralell resistors, and thus modulating the load to the optimum directly.



I personally would have buried some tubes in the concrete slab for an eventual later upgrade to a water based system if ever the grid reaches your house sometime in the future.
Also, one thing lacking in your SBMS hardware is overvoltage protection over the ideal diode mosfets at the input.

That is how DMPPT is very efficient over 93% by selecting the number of outputs effectively selecting the number of parallel resisitve heat elements.

Why Will I ever want to connect to grid as that is many times more expensive than solar PV heating ?

Ideal diode mosfet is perfectly protected not sure what you want to say.  That is a diode ideal or not and the pulse of energy that comes when a load is switched off will be clamped by the TVS that is connected between GND and output of the ideal diode.

[HackedFridgeMagnet]

Yeah I am really keen to try this at my place. Maybe even going off grid. My main trouble is roof space, my roof is angled every which way.
Using water as my thermal mass.
... one day.

[Red Squirrel]

Yeah my roof is my main issue too, split level house, cottage style roof.  Lots of trees and other obstructions around too.   There's only like one section that is viable, the rest get shadows.  I measured once and I have room for around 3kw on my house so not enough to go off grid especially with the lack of daylight for most of the year, but it would be something.  I think the money I would spend on that would be better off saving up for a bigger property though, then I can do a proper off grid setup there.  Decent size acreage properties come up in my general area at times.  There was a 16k one that I was very tempted to buy but I think I will wait till something waterfront comes up. Ideally I should wait till my mortgage is paid off before I buy a property, since that's only half the battle, still need to actually build something.

[electrodacus]

Yeah I am really keen to try this at my place. Maybe even going off grid. My main trouble is roof space, my roof is angled every which way.
Using water as my thermal mass.
... one day.

You have the advantage of a much warmer place and if the house is well insulated or can be insulated better then maybe the roof shape is not that big of a problem. Panels do not necessarily need to face North in your case but east or west facing can generate fairly similar results. Shadow from other buildings or trees will be more problematic.

Yeah my roof is my main issue too, split level house, cottage style roof.  Lots of trees and other obstructions around too.   There's only like one section that is viable, the rest get shadows.  I measured once and I have room for around 3kw on my house so not enough to go off grid especially with the lack of daylight for most of the year, but it would be something.  I think the money I would spend on that would be better off saving up for a bigger property though, then I can do a proper off grid setup there.  Decent size acreage properties come up in my general area at times.  There was a 16k one that I was very tempted to buy but I think I will wait till something waterfront comes up. Ideally I should wait till my mortgage is paid off before I buy a property, since that's only half the battle, still need to actually build something.

You do not need that much land and if properly designed the house can fully integrate the PV array.  I have 20 acres (8 hectares) that is absolute overkill tho this was a cheaper option than getting a much smaller 1 to 2 acres lot.
Even 1/2 acres (2000m^2) is plenty for an offgird self sufficient house.
If I where to design my house now I will probably try to do a sort of A frame so that all 10kW PV array can fit on the south side at a steep 60 degree angle for max output in winter.
But is fine with a ground PV array as I have plenty of space and no shadow's

Here is an older photo from March 3  2018 time 13:36 (I know that as it was in the photo name) and also know that this array currently as seen made of 33x 260W panels 8580W produced that day 5.1kWh

[fourtytwo42]

Yeah my roof is my main issue too, split level house, cottage style roof.  Lots of trees and other obstructions around too.   There's only like one section that is viable, the rest get shadows.  I measured once and I have room for around 3kw on my house so not enough to go off grid especially with the lack of daylight for most of the year, but it would be something.

Better to get some experience on a small scale while you can, I only have 1Kw on a SSW face at 52 degrees elevation and latitude (handy coincidence) but that's enough to heat all my water in the summer and save oil fuel and boiler wear. We had a supermarket slogan in the UK I am fond of, "every little counts"

Great picture electro, I wish .............

[f4eru]

Ideal diode mosfet is perfectly protected not sure what you want to say.  That is a diode ideal or not and the pulse of energy that comes when a load is switched off will be clamped by the TVS that is connected between GND and output of the ideal diode.

The destroying pulses don't happen switching loads, but when lightning(surge) or EMC (Burst) couple onto the long wires. Then those mosfets are unprotected, and will probably be destroyed

[f4eru]

That is how DMPPT is very efficient over 93% by selecting the number of outputs effectively selecting the number of parallel resisitve heat elements.

OK

Why Will I ever want to connect to grid as that is many times more expensive than solar PV heating ?

The solar resistive PV heating being economically preferable over PV heat pump is an edge case IMHO. It still has big constraints, like a massive PV array, huge copper cross sections outside a ground level which is a cost and thieves problem. Also the DIY nature can distort the cost comparison, but fair enough.

I don't think it's advantageous at all compared to GTI + heat pump, especially if overproducing energy into the grid and thus not paying much electricity at all and using the grid as storage, removing the expensive battery. With a massive PV array, GTI would probably be economically positive for you in the summer (not considering the cost for laying the grid)

A lot of things could change in the next 10 years, economics of all this, grid laying price, etc.
What about future EV charging ?

[Red Squirrel]

You do not need that much land and if properly designed the house can fully integrate the PV array.  I have 20 acres (8 hectares) that is absolute overkill tho this was a cheaper option than getting a much smaller 1 to 2 acres lot.
Even 1/2 acres (2000m^2) is plenty for an offgird self sufficient house.
If I where to design my house now I will probably try to do a sort of A frame so that all 10kW PV array can fit on the south side at a steep 60 degree angle for max output in winter.
But is fine with a ground PV array as I have plenty of space and no shadow's

Still need room with no shadows though. Looks like you got that covered.  My property is measured in feet not acres and surrounded by trees and houses.  Just don't have the room.  My dream to buy acreage property eventually though.   I'd like at least 1/2 but probably do 10+.  Would be nice to get into drones and the rules are super strict here but if I can do it on my own property I'd probably be safe.

[electrodacus]

but that's enough to heat all my water in the summer and save oil fuel and boiler wear. We had a supermarket slogan in the UK I am fond of, "every little counts"

Yes any little counts but a combination of multiple energy sources will almost always be more expensive than a single source especially since that source is electricity and easily converted in any other type.

The destroying pulses don't happen switching loads, but when lightning(surge) or EMC (Burst) couple onto the long wires. Then those mosfets are unprotected, and will probably be destroyed

The DMPPT450 has an array of 16x 3kW TVS so a max theoretical capability to absorb a 48kW 1ms pulse in reality that is lower but sufficient to deal with a 450A (DMPPT450) that is all switched off at once and keep the peak voltage clamped at below 60V that is the mosfet ratings.
When current is interrupted the energy stored in the long PV cables (basically an inductor) will need a place to drain and so voltage on the PV inputs will rise until something will clamp the energy that will be proportional with the total cable inductance and current that was interrupted. This increased voltage will pass trough the diodes (ideal diodes in this case ) and on the other side of the diode the voltage will be clamped even in worst case to below 60V by the array of TVS mentioned before.
Lightning will induce a similar pule depending on how close it is and it will be absorbed in a similar way.
I get quite a few lightning strikes in here as it is an open field with not many tall things around and while I never had any severe ones since DMPPT450 was installed I did had a serious one while SBMS120 was installed and that has a similar protection for the ideal diodes and it was no problem over many years except for a strong thunderstorm I think this summer when a close strike damaged the PV current sense amplifier but nothing else inside the SBMS was damaged. and externally an improper LED light driver failed as short triggering the short circuit protection on SBMS.
Any more serious lightning protection should be done outside the DMPPT450 or SBMS on the PV input side.

The short answer is that those input mosfets are protected if you look closer at the schematic (I know is not the most readable schematic).

Why Will I ever want to connect to grid as that is many times more expensive than solar PV heating ?

The solar resistive PV heating being economically preferable over PV heat pump is an edge case IMHO. It still has big constraints, like a massive PV array, huge copper cross sections outside a ground level which is a cost and thieves problem. Also the DIY nature can distort the cost comparison, but fair enough.

Not an edge case an yes as mentioned you can reduce the PV array size but it will be significantly more costly so if you have the space for the needed PV array size it is the most cost effective heating solution. If you do not have the space and you already have a natural gas connection then that is the next best option from a cost perspective.
Yes DIY nature is a factor bot not as large as you may think as any sort of heating will need Labor and when making the comparison I did not included the labor on any of the other solutions.

I don't think it's advantageous at all compared to GTI + heat pump, especially if overproducing energy into the grid and thus not paying much electricity at all and using the grid as storage, removing the expensive battery. With a massive PV array, GTI would probably be economically positive for you in the summer (not considering the cost for laying the grid)

Using grid as storage will be way more expensive than using thermal mass storage. I think I mentioned that thermal mass storage can have a cost amortization as low as 1 cent/kWh where grid will be at least an order of magnitude more.
The LiFePO4 battery has nothing to do with PV heating that only requires inexpensive thermal storage.
Personally I do not need cooling in summer as house is designed close to passive standards but DMPPT450 can also be used for cooling with peltier elements (again DIY as there is nothing of the shelf for that but very inexpensive).

A lot of things could change in the next 10 years, economics of all this, grid laying price, etc.
What about future EV charging ?

Yes many things can change in time but my energy cost is paid for the next 30 years.
Currently EV's are more expensive to drive than gasoline cars and if battery price will drop enough (at least about 3x from current price) then they may be cost effective and I'm sure that will happen but likely not before self driving cars that will take personal car out of the equation. I currently drive a old gasoline car and I hope self driving car's will soon become common place so that I can rent one any time I need to.
I only drive in average around 6000km/year and most of that is for grocery shopping about once a week. This wasted time traveling to the city and back can be completely eliminated if I can order the food online and have that delivered by a self driving car.
So in my prediction energy may become decentralized no grid or pipes thanks to very low cost solar and transportation will be the other way around and become a service instead of current wasteful car ownership.