1000kWh are good for 1 month

If this were true, the biggest monthly bill for electric heat would be $90-$160. I'd review where your estimate came from (Florida?).

I use natural gas and it's cheaper than PV + heat pump.

This morning I have -20C here somewhere in rural Saskatchewan. Is very unlikely that you live in a colder climate.
The house is new this will be the first winter in this house.
House is relatively small 700sqft (65sqm) but well insulated with R30 wall's (It will be R36 in the end but was not able to finalize for this winter) and R70 roof also the concrete slab on grade is well insulated R32 I still need to install the heat recovery ventilation.
There are very few windows all facing south so they in average will not lose energy.
Search for my name on google and you will find my google+ page Dacian Todea there are quite a few photos of the house.

Posted By electrodacus on 11 Nov 2013 11:12 AM

> estimated monthly heat requirement in the coldest month is under 1000kWh

That would last me about 4 days of cold weather, not a month.

If I were generating heat from PV, I'd use a heat pump and then store heated water in a tank.

That must be a huge house and, or low level of insulation but no matter if 1000kWh are good for 1 month or 4 days of heating you still need to heat the house and PV in the long term will probably be the least expensive option.
What sort of fuel do you use currently to heat the house and what is the cost of that fuel in your location?

1000kwh/month is an average load 1.39 kilowatts, or about 4750 BTU/hr.  According to utility surveys from about  average home in the US state of Massachusetts has a design heat load 14 kilowatts at the 99% outside design temp, an average January heat load of about 9 kilowatts. The average size of the house is a bit less than 2000 square feet of conditioned space.

To hit 1.39kw average load in this climate requires PassiveHouse levels of air tightness and insulation, and this is by no means the coolest climate in the lower 48 states.   To go through 1000kwh in 4 days is an average heat load of 35,500 BTU/hr, which would be pretty common 4-day situation for many code-min insulated mid-sized houses in my area during cold snaps (or just normal mid-winter conditions in the colder US climate zone 7 regions.)

1000kwh/month is an average load 1.39 kilowatts, or about 4750 BTU/hr. According to utility surveys from about average home in the US state of Massachusetts has a design heat load 14 kilowatts at the 99% outside design temp, an average January heat load of about 9 kilowatts. The average size of the house is a bit less than 2000 square feet of conditioned space.

To hit 1.39kw average load in this climate requires PassiveHouse levels of air tightness and insulation, and this is by no means the coolest climate in the lower 48 states. To go through 1000kwh in 4 days is an average heat load of 35,500 BTU/hr, which would be pretty common 4-day situation for many code-min insulated mid-sized houses in my area during cold snaps (or just normal mid-winter conditions in the colder US climate zone 7 regions.)

I'm an electrical engineer is unlikely that I did huge mistakes in my calculation and also the calculation seems to be supported by real data for now.
Currently I use a temporary propane heater to heat the house there are freezing temperatures in the last 3 or 4 weeks.
I use a 20lbs standard barbecue propane tank 8kg of propane that stores about 100kWh and it last for about 7 days even if this last one is close to empty after just 5.5 days but It was extremely cold for this time here -20C (-4F) last night.
So according to this data I use about 400 to 600kWh/month at this moment for heat so I expect not to exceed 1000kWh in January.
Hose is small and quite close to a passive house standard but not really there.
The 1.39kWh heat load are my calculated average load for January walls are by far the higher loss since they are just R30 the roof is R70 and slab on grade is insulated with R32 on average.
I mentioned in the post above but you can see detailed photos with the house build on my google+ page search Dacian Todea (electrodacus)

Using very rough figures (sloppy, I know), say your house is twice as well insulated, 1/2 the surface area and you have 2x the number of degree days as Dana's data. So a net difference of 1/2. Yet 9 kw/hr = 6500 kwh/mo and 1/2 of that is 3250 kwh/mo. So your 1000kwh/mo seems low. On the other hand, you have actual data and that's much better than estimates. And I don't see any mistakes. A Manual J would be interesting.

Using very rough figures (sloppy, I know), say your house is twice as well insulated, 1/2 the surface area and you have 2x the number of degree days as Dana's data. So a net difference of 1/2. Yet 9 kw/hr = 6500 kwh/mo and 1/2 of that is 3250 kwh/mo. So your 1000kwh/mo seems low. On the other hand, you have actual data and that's much better than estimates. And I don't see any mistakes. A Manual J would be interesting.

Dana provided data for an average 2000sqft house mine is about 3x smaller and not average thermal insulation.
I think that mine has better than 2x thermal insulation compared to average house is not a typical construction if you seen the photos.
Also I had at least one window open when I used the propane heater since my heat recovery unit is not installed yet that is additional loss at the moment.

We cross-posted on my last bit, handn't seen the details yet.

OK, so your conditional "...providing you have the right house and geographic location..." means you're talking about a tiny super-insulated (PassiveHouse style) house on the sun-drenched prairies.

Deep cycle lead-acid batteries have about a 50 cents/kwh lifecycle cost of operation, so that 1000kwh/month DOES have significant cost, primarily associated with it being off-grid. A tiny superinsulated house ON the grid, if net-metered could hit Net Zero Energy at a lower lifecycle cost.

Given that you are your own power company, with a very high capital cost in battery power, if you really intend to heat with PV, a 3/4 ton cold-weather mini-split (Fujitsu AOU 9RLS2-H or Mitsubishi MUZ-FE09-NA) would probably be a worthwhile investment, since it would cut your peak heating power by nearly half, and overall heating season power use by over half. Both have a specified output at -25C. The Mitsubishi would turn itself off at about -28C (to self-protect), and automatically re-start when it warmed up to -26C, the Fujitsu just keeps going (at some un-rated output.) http://smartgreenbuild.com/pdf/Fujitsu-RLS2H.pdf

What makes sense for most off-grid homes is to use bulk bio-heat (wood stove, corn stove, pellet stove, etc) for space heating, saving wear & tear on the batteries for some function that can't be delivered by un-powered or low-power means, where the bulk of the energy use isn't from the batteries. (I would think/hope that there is a straw-pellet fuel source somewhere in Sask.)

We cross-posted on my last bit, handn't seen the details yet.

OK, so your conditional "...providing you have the right house and geographic location..." means you're talking about a tiny super-insulated (PassiveHouse style) house on the sun-drenched prairies.

Deep cycle lead-acid batteries have about a 50 cents/kwh lifecycle cost of operation, so that 1000kwh/month DOES have significant cost, primarily associated with it being off-grid. A tiny superinsulated house ON the grid, if net-metered could hit Net Zero Energy at a lower lifecycle cost.

Given that you are your own power company, with a very high capital cost in battery power, if you really intend to heat with PV, a 3/4 ton cold-weather mini-split (Fujitsu AOU 9RLS2-H or Mitsubishi MUZ-FE09-NA) would probably be a worthwhile investment, since it would cut your peak heating power by nearly half, and overall heating season power use by over half. Both have a specified output at -25C. The Mitsubishi would turn itself off at about -28C (to self-protect), and automatically re-start when it warmed up to -26C, the Fujitsu just keeps going (at some un-rated output.) http://smartgreenbuild.com/pdf/Fujitsu-RLS2H.pdf

What makes sense for most off-grid homes is to use bulk bio-heat (wood stove, corn stove, pellet stove, etc) for space heating, saving wear & tear on the batteries for some function that can't be delivered by un-powered or low-power means, where the bulk of the energy use isn't from the batteries. (I would think/hope that there is a straw-pellet fuel source somewhere in Sask.)

Dana1 it seems you really know something about power generation and the cost involved.
I do not intend to use chemical battery in conjunction with PV for energy storage but store the energy directly as heat.
My 14 cubic meter concrete slab can absorb 10kWh/degree Celsius so I can go as high as 100kWh stored with 10 degrees Celsius difference the concrete slab has pex tubing so I will heat water and circulate that.
My current PV system is much smaller 3x240W panels + 2x85W panels for backup circuit and a small 300W wind turbine that allow me to have a much smaller battery. Main battery is a LiFePO4 100Ah 24V so about 2.5kWh storage capacity and the backup battery is 20Ah 24V LiFePO4 made of A123 Cells and the cost on this is about 20cent/kWh stored during lifetime they have 3000 cycles with 100% discharge after that they will still have 80 or 90% of the initial capacity if I remember correctly.
Still 20cent /kWh is huge compared with PV cost so no chemical storage will be involved on the heating part.
There will be a separate PV array from the current one probably about 10x as large and the energy will be stored in the large thermal mas of the house.
Of course this solution works for my house but for an existing house with very little thermal mass a large water tank will still be a good solution.
A 200 liter 55 gallon drum will store 7kWh at a 30 Celsius temperature delta.
This is what I will use now since this winter I will use a diesel heater to heat the house is an Espar Hydronic M and the 200 liter barrel will be used as a buffer before heating the concrete slab and there will also be a small radiator to reheat the air after exiting from the heat recovery ventilation.

As for the use of a air heat pump not sure it makes economic sense I will need to do some calculations. I had one on my other house that was in Europe with milder climate and in freezing temperatures the COP was just above 1 almost did not made sense to use.
I will need to know the parameters COP vs temperature and also life cycle and cost but I do not think the increase in complexity and cost will be able to make up for the investment.
My solar panel array for heating will probably consist of 30 x 240W panels at around 8000$ + some wiring and support so not sure a heat pump can add any value.
Wood, corn or pellet may be competitive with some battery technologies but they are not with direct Solar PV and heat storage in thermal mass.

COP on a pretty good mini-split runs about 1.8 at -25C when running full speed, a bit higher at part load. At -15C it's about 2.0 at full speed, 2.2-2.5 at part load. See Figure 5 on page 10, (p18 in PDF pagination) and Figure 9, p14 of this document:

http://www.nrel.gov/docs/fy11osti/52175.pdf

Of course it means you would need sufficient battery & inverter to run one, and battery power is still pretty expensive. I'll be (pleasantly) surprised if you can get the lifecycle cost of the lithium batteries down to 20 cents/kwh.

If you have sufficient thermal mass in the slab to store the heat at low temp, you get better solar conversion efficiency at lower cost than PV out of windows. (Albeit at an increased peak heating load.) The conversion efficiency of commercial PV panels is still at or under 20%, whereas passive solar can easily break 50% average, even in cooler climates.

Storing heat in water adds complexity to the heat distribution system- whether hydronic storage & distribution would cheap enough to make it preferable to ductless heat pumps with more battery & less PV isn't immediately obvious, but maybe. Rarely does electric heating (of any sort) make sense for off-grid homes, but the crashing price of PV over the past five years has broken some paradigms, to be sure. For on-grid homes with 5-10 kw design heat loads, ductless heat pump solutions can get to net-zero-energy with with a PV array that still fits on the roof.

COP on a pretty good mini-split runs about 1.8 at -25C when running full speed, a bit higher at part load. At -15C it's about 2.0 at full speed, 2.2-2.5 at part load. See Figure 5 on page 10, (p18 in PDF pagination) and Figure 9, p14 of this document:

http://www.nrel.gov/docs/fy11osti/52175.pdf

Of course it means you would need sufficient battery & inverter to run one, and battery power is still pretty expensive. I'll be (pleasantly) surprised if you can get the lifecycle cost of the lithium batteries down to 20 cents/kwh.

If you have sufficient thermal mass in the slab to store the heat at low temp, you get better solar conversion efficiency at lower cost than PV out of windows. (Albeit at an increased peak heating load.) The conversion efficiency of commercial PV panels is still at or under 20%, whereas passive solar can easily break 50% average, even in cooler climates.

Storing heat in water adds complexity to the heat distribution system- whether hydronic storage & distribution would cheap enough to make it preferable to ductless heat pumps with more battery & less PV isn't immediately obvious, but maybe. Rarely does electric heating (of any sort) make sense for off-grid homes, but the crashing price of PV over the past five years has broken some paradigms, to be sure. For on-grid homes with 5-10 kw design heat loads, ductless heat pump solutions can get to net-zero-energy with with a PV array that still fits on the roof.

I'm impressed by the amount of knowledge you seem to have regarding this issues.
I can only say that 20 cent for LiFePO4 is based on manufacturer data hope that is true I will need quite a few years to confirm that by that time I hope they will drop under 10cent/kWh stored.
Is a bit funny but the windows will not be more cost effective for heating than PV.
For example I got the cheapest 2 pane argon filed windows at a cost of 199$ the size of the window is 60"x36" that is about 1.5mx0.9m and this also include the frame so about 1.35sqm but effective glass is probably closer to 1sqm. Solar heat gain for this window is 53% and all windows on my house are facing south so quite optimal but they are vertical so there will be additional reflexion.
On my location I get in the winter an average of 3.5 to 4kWh/sqm of solar radiation.
If I take ideal condition this window will probably only capture less than 2kWh of solar energy not taking in consideration the thermal loss compared to the wall for now.
At the same size I can get a 200W solar PV 1$/W that will capture about 0.8kWh for now it sounds like the window is wining but when you take the thermal loss in 24h for that window compared with the wall (usually 8 to 10x more) in this case 1.7W/m2K so at a temp delta of just 30C (+20C inside and -10C out) 1.7Wx30Cx1.35sqmx24h= 1.652kWh/day
Now the window is still a net producer but of only 2kWh-1.652kWh=0.348kWh
I need also to consider the 1.35sm of wall to subtract from solar if I replace the window with solar in my case 0.20W/m2K 0.2Wx30Cx1.35sqmx24h = 0.194kWh so 0.8kWh-0.194kWh = 0.606kWh
So in conclusion at just 30 degree Celsius temperature difference a solar panel will provide almost 2x more heat compared to a similar priced window 0.6kWh vs 0.35kWh
Also another problem with large windows is that there will be large temperature fluctuation in the house and reduced comfort compared to solar PV and in floor heating hydronic or direct electric but I did not installed electric heat wires in mine.
So PV heating better than windows at least in my climate and with the prices that I got for them. I do not think you can get substanialy less expensive windows with the same characteristics or better.
One way that can make windows perform better will be to add an insulated external shutter but it will need to be at least half the thermal insulation off the wall or else is still not good.
I just realized I left out the price of the 1.35sqm wall out of the equation but the PV will probably still be better.

The maximum heating energies computed by electrodacus do not seem out of line to me. My house is slightly less well insulated, and more than twice as large (1602 versus 700 sq. ft.) My maximum monthly natural gas usage for space heating, hot water heating supplemental to solar hot water, and cooking has averaged 1819 kWh over the last three years. Let me guess that 88% of that is for space heating, so that is 1601 kWh, so at 97.5% efficiency, that would be 1561 kWh delivered to the house. Degree (F) days in Regina, Saskatchewan are 10190, while an average for my area is 7388 (although the past winters have been lower), so that would scale my usage up to 2153 kiWh in that climate. Scaling this for the area ratio of 700/1602) would give 941 kWh, close to the 1000 kWh estimated by electrodacus. Scaling linearly for the living area is not quite right (actual usage would be higher), but his numbers for heating energy look reasonable to me since he is better insulated. Some of my heating needs are met by passive solar, but that will be the same for electrodacus.

I heat with natural gas, which costs about $0.022 per kWh, excluding the connection fee. The largest monthly bill that I have seen is $49, excluding the connection fee which is $11 and change. In my area, natural gas appears to be a very efficient way to heat, and I expect that the natural gas furnace will outlast a heat pump.

I have estimated the costs for my solar PV system amortized over 25 years and including all subsidies and rebates at $0.064 / kWh (http://www.residentialenergylaboratory.com/comparison_of_pv_systems.html). (This would be the cost if I used all the electrical energy generated, anyway.) However, this is a grid-tied system with net-metering, although they only pay $0.03 per kWh for excess electricity, while charging about $0.108 per kWh. The guys at the wind-sun forum that have experience with off-grid systems quote MUCH higher costs, on the order of $1.00 per kWh, with much of that cost due to batteries and their regular replacement. Compared to my natural gas costs of about $0.022 per kWh or $300 per year (including connection fees), and my lack of need for A/C, the off-grid electricity using batteries appears very expensive.

I have estimated the costs for my solar PV system amortized over 25 years and including all subsidies and rebates at $0.064 / kWh (http://www.residentialenergylaboratory.com/comparison_of_pv_systems.html). (This would be the cost if I used all the electrical energy generated, anyway.) However, this is a grid-tied system with net-metering, although they only pay $0.03 per kWh for excess electricity, while charging about $0.108 per kWh. The guys at the wind-sun forum that have experience with off-grid systems quote MUCH higher costs, on the order of $1.00 per kWh, with much of that cost due to batteries and their regular replacement. Compared to my natural gas costs of about $0.022 per kWh or $300 per year (including connection fees), and my lack of need for A/C, the off-grid electricity using batteries appears very expensive.

Thanks for the input I did not realized that natural gas is so inexpensive at 2.2cent/kWh is really hard to compete with.
I actually have a natural gas pipe crossing my property but is on the other side of it about 250m (800ft) from the house the cost to connect will probably be to high but I can be wrong since I did not investigate that.
As I explained in a perversion post you do not need batteries for solar PV heating you just need a large enough thermal storage that can be watter or concrete.
As I already have a massive 14 cubic meter concrete slab that is well insulated I just need the solar panels about 1$/Watt for the large 240W or 250W 60 cells panels as the ones I currently have.
The panels can be used with heating mats embedded in concrete and run directly from DC you can make any configuration 3 panels in series will deliver about 90 to 110V
Or you can use an normal inexpensive electric hot water tank connect it directly to solar PV and circulate that water to heat the slab.
My slab can store up to 100kWh of heat that is enough for up to 3 days of extremely cloudy days.
Today was sunny no clouds and I got about 3kWh from my 3x 240W panels. November and December are here the worst months for solar but I still can get in average 60kWh/month in this two months the rest of the months will be over 80kWh so if I get about 10x the amount of panels I can get close enough to heat the house about 800kWh in January and just the panels will cost around 7000 to 8000$ delivered.
I will also need probably a small backup heat source or get about 2x 600W wind turbine the one 300W small wind turbine delivers about 60kWh in the winter months here usually the cloudy days are the windiest so they complement well the solar PV.
Heating season here is quite long at least 6 months.

Solar PV at 1$/W in my location get to about 3 cent/kWh but that is if you can use all the energy from them and I can do that in the winter months I just need to find a use for that energy in the summer but even If I don't I will be happy even with 9 cent/kWh for heating is still better than any alternative I have and just not use the solar panels in summer.

I think natural gas price will go up based on the current cost I know in Europe is much more expensive that is why I was surprised by the price here and the solar panel prices will only go down even the batteries that are extremely expensive right now (the one that I use probably the best one is about 20cent/kWh stored) but with the introduction of electric vehicle the price will probably go down to even 5cent/kWh in 5 years but that is just speculation.

Solar PV at 1$/W in my location get to about 3 cent/kWh

I'd like to see the math on that.

Solar PV at 1$/W in my location get to about 3 cent/kWh



I'd like to see the math on that.

I think I already did that math somewhere here but I can do that again.
Guessing you agree with the 1$/W price of solar PV.
Also that rate is based on my location Saskatchewan Regina.
My current 3x240W = 720W solar panels produce about 999kWh/year according to PVWATTS version1 online solar calculator and also based on my measurements that match quite well the calculator.
If you go to that calculator select SA Regina and then 0.72kW panels then leave the rest of the data as it is you will get 999kWh/year notice the DC to AC factor is default at 0.77 that is 77% efficiency from the panels DC to the battery and inverter but if you use the panels directly to heat the house that value can be 1 and you will get more power form the panels the calculator only allows 0.96 as max with that you get 1242kWh/year.
I will make the calculation with 999kWh/year divided by 720W I get 1.3875kWh/year from each Watt of solar panel installed.
Then I multiply with 25years x 1.3875/kWh/year and get 34.68kWh in 25 years from 1W of solar panel
So you pay 1$ for 1W of PV and get 34.68kWh in 25years that will be 1/34.68 = 2.88 cent /kWh
There are small details that I left out of the calculation for simplicity but the end result will still be around 3 cent/kWh for the solar PV provided you use all that power when is generated or store it in the form of heat.
We just talk about the solar PV panels as if they where fuel at 3 cent/kWh not to bad even if natural gas seems a bit better at the moment at 2.2cent.

If it were really 2.88 cents/kWh, we would see a lot more PV solar panels. A major factor not included is the interest on the money needed to buy the solar panels (or missed income from investing it).

electrodacus is taking a novel approach to using PV panels. By not using an inverter (isn't that your approach electrodacus?), you cannot use the electricity generated for anything other than resistive heating. This made no sense in the past when PV panels were expensive, but now that they are so cheap, it might make sense to use them in remote locations where natural gas lines are not available. This approach might have problems meeting codes where they are enforced, so it might be limited to very remote locations with no code enforcement, and making house financing and resale very difficult. And I see no way to use the electricity for cooling without the inverter. When used for heating, the problem of storage must be addressed, as discussed.

This is a whole new paradigm, avoiding the costs of inverters, and other licensing issues.

And I see no way to use the electricity for cooling

A heat pump, a water tank and fan coils work for that too. Controlling a heat pump to use exactly the energy being created by the solar panels is an issue to be addressed. It's not clear if an inverter would be needed or not - the variable heat pumps already have one.

electrodacus is taking a novel approach to using PV panels. By not using an inverter (isn't that your approach electrodacus?), you cannot use the electricity generated for anything other than resistive heating. This made no sense in the past when PV panels were expensive, but now that they are so cheap, it might make sense to use them in remote locations where natural gas lines are not available. This approach might have problems meeting codes where they are enforced, so it might be limited to very remote locations with no code enforcement, and making house financing and resale very difficult. And I see no way to use the electricity for cooling without the inverter. When used for heating, the problem of storage must be addressed, as discussed.

This is a whole new paradigm, avoiding the costs of inverters, and other licensing issues.

Yes that is my approach using just the PV panels without inverters or batteries.
You may be able to use for other applications other than resistive heating like some DC motors for irrigation or maybe even a DC heat pump also can be used for refrigeration with peltier elements.
I was investigating solar evacuated tubes or vacuum tubes for heating and it seems even if they are more efficient they are not even close at the moment to be cost competitive with solar PV not even taking in account the complexity difference (no pipes or pumps) It was not the case just a few years back.
I do not see any problems in meeting the code.
For example use 120V AC heat elements (like water tank heaters for energy storage) with 3x 30V max power point panels will give you 90V DC the element will be lasting forever even if just at close to half the spec power. Or use just groups of two panels in series that will reduce the working voltage to around 60V and it will be considered safe on almost any electrical code. I'm referring to 60 cells panels usually the best price/watt. Yes you will pay more for thicker cables at lower voltages but maybe is better than getting a certified installer
I do not actually need cooling since summer are not that hot here and I have good thermal insulation plus the large thermal mass that will keep the temperature inside at about the average between the exterior night and day temperatures.
If I will have needed cooling I will have probably used an underground water loop with a pump to cool the concrete slab.
I'm sure as soon as some solar PV companies realize that this inexpensive packages for heating will be available.

If it were really 2.88 cents/kWh, we would see a lot more PV solar panels. A major factor not included is the interest on the money needed to buy the solar panels (or missed income from investing it).

Yes is true that you pay the "fuel" for the next 25years in advance but it makes sens in many situations.
For my small well insulate house I only need 7000 to 8000$ invested that will probably be just the connexion fee to power or natural gas if is a new house and you have those in front of the house but then you still need to pay for the electricity or natural gas after the initial investment.

I only need 7000 to 8000$ invested that will probably be just the connexion fee to power or natural gas if is a new house

I agree with that. Not so clear is the heat pump. If you can pay $2000 for a HP and then reduce the $8000 in solar panels to less than $4000, that probably makes sense.

Posted By jonr on 12 Nov 2013 05:56 PM

I only need 7000 to 8000$ invested that will probably be just the connexion fee to power or natural gas if is a new house

I agree with that. Not so clear is the heat pump. If you can pay $2000 for a HP and then reduce the $8000 in solar panels to less than $4000, that probably makes sense.

Maybe- or maybe not. Most PV panels are good for 40 years with next to zero maintenance, but if you get more than 25 years of efficient service out of a mini-split with low or zero maintenance you'd be beating the odds.  There is a certain elegance to the crude-but effective resistance heating of a large thermal mass, now that the panel cost of PV has been decimated.

Of course the cost of money factors in too.  A net present value calculation using realistic lifecycles and reasonable discount rates would be necessary to really sort it out.