diy solar

diy solar

battery grid feedback

That'll be alright. It just happens to be a s16 280Ah pack.
If you already bought the cells, then that’s that, but having 5kW available for 2 hours is not the same thing as consuming 10kWh.

The only reason it matters is for sizing your AC power supply.

If you consume 10kWh AC utilizing 12.5kWh from your battery, you will need at least a 1.5kW AC charger to recharge the battery in 8 hours.

If you only consume 5kWh AC (~50% duty cycle) utilizing 6.25kWh from your battery, you can easily recharge the battery in under 7 hours with a 1kW AC charger.

And if you start off with a 1kW charger and realize it’s not cutting it, you can always add a second in parallel (though a single 1.5kW or 2.0kW charger may be more economical than two 1kW chargers.. ).
 
If you already bought the cells, then that’s that, but having 5kW available for 2 hours is not the same thing as consuming 10kWh.

The only reason it matters is for sizing your AC power supply.

If you consume 10kWh AC utilizing 12.5kWh from your battery, you will need at least a 1.5kW AC charger to recharge the battery in 8 hours.

If you only consume 5kWh AC (~50% duty cycle) utilizing 6.25kWh from your battery, you can easily recharge the battery in under 7 hours with a 1kW AC charger.

And if you start off with a 1kW charger and realize it’s not cutting it, you can always add a second in parallel (though a single 1.5kW or 2.0kW charger may be more economical than two 1kW chargers.. ).
The combination of a nighttime rate and daytime air temperature is a win-win. I will use all stored energy leaving some buffer for battery health. It doesn't have to be exactly 2h and I will run the heater even if it gets the water a bit past the desired temp.
 
The combination of a nighttime rate and daytime air temperature is a win-win. I will use all stored energy leaving some buffer for battery health. It doesn't have to be exactly 2h and I will run the heater even if it gets the water a bit past the desired temp.
Look at Picture 1 on the first page (‘First Connecting Method): https://m.media-amazon.com/images/I/91joUTV92TL.pdf

The user configuration I’d suggest to power your Heat Exchanger from a battery (with the Heat Exchanger replacing the 240V Loads and your 48V battery replacing the solar panel).

As I said earlier, the battery capability is poorly documented, but I just found proof there GTIL2 inverters are manufactured by Deye and they document the battery-powered capability: https://www.solartex.co/tienda/wp-content/uploads/2019/02/FICHA-TECNICA-INVERSOR-CON-LIMITADOR.pdf

‘-Battery discharge power mode enabled, Can regulate depth of discharge of the battery bank.’

And if you read down at the Amazon link:
you will find this:

  • Only use the 36V/48V battery to power the inverter
  • Use a battery to power the inverter, please use a circuit breaker.
Finally, if you look at Picture 5 on page 3 of the User Manual I linked to earlier, you’ll see the Limiter Mode Configuration page includes: ‘Bat or solar limited Current Mode’ as well as ‘Bat or solar limited Power Mode.’
 
Last edited:
Look at Picture 1 on the first page (‘First Connecting Method): https://m.media-amazon.com/images/I/91joUTV92TL.pdf

The user configuration I’d suggest to power your Heat Exchanger from a battery (with the Heat Exchanger replacing the 240V Loads and your 48V battery replacing the solar panel).

As I said earlier, the battery capability is poorly documented, but I just found proof there GTIL2 inverters are manufactured by Deye and they document the battery-powered capability: https://www.solartex.co/tienda/wp-content/uploads/2019/02/FICHA-TECNICA-INVERSOR-CON-LIMITADOR.pdf

‘-Battery discharge power mode enabled, Can regulate depth of discharge of the battery bank.’

And if you read down at the Amazon link:
you will find this:

  • Only use the 36V/48V battery to power the inverter
  • Use a battery to power the inverter, please use a circuit breaker.
Finally, if you look at Picture 5 on page 3 of the User Manual I linked to earlier, you’ll see the Limiter Mode Configuration page includes: ‘Bat or solar limited Current Mode’ as well as ‘Bat or solar limited Power Mode.’
I get it. I would need 3 of those though + the charger.
 
I get it. I would need 3 of those though + the charger.
Yes, 3 or 4 2000W GTILs plus a 1-2kW AC charger. A little of $1000 if you shop around.

I believe GroWatt may offer a 240V GTIL inverter but every model I’ve seen has a much higher minimum voltage than 48V (so either can’t be powered from a battery or needs to be a much higher-voltage battery than 48V…).

But this market is evolving very quickly and with every year you can wait, there are almost certain to be additional options.
 
Last edited:
There are 2 ways to control a GTIL - disconnect the DC supply (battery) or disconnect the grid.

I started off using lamp timers to control my AC grid signal / output and that worked well, but I’m not sure they make 240V lamp timers and I’m not sure they would be rated for 6kW…
They make dozens of them for Hot Water heaters, you can pick and choose the model with the features you want.
https://www.amazon.com/INTERMATIC-E...=hot+water+heater+timer&qid=1633658312&sr=8-8
 
My original plan was something very similar to this. I looked at those same GTIL inverters, and I was going to use three 600 watt chargers. I was going to have a microcontroller like an Arduino monitor the output from my Enphase solar. When it topped 1,000 watts, turn on one 600 watt charger. When the power exceeds 2,000 watts, turn on a second 600 watt charger, an at 3,000 watts turn on the third charger. 1,800 watts of charging and still 1,200 or more watts feeding my house when the sun is up. I have 16 soo watt panels with 240 watt iQ7 microinverters, so the system peaks out at about 3,800 watts under peak sun. On good days of sun, I figured that setup could push about 6 KWHs into a battery bank. The microcontroller would then turn on the GTIL inverters from 4 pm to 9 pm as long as the batteries were above the cut off voltage. It should all work. But when I figured in the cost of the batteries, I basically chickened out and bought the Schneider. But I got suckered in by their web site. It claims it can do "Time of Use" and "A/C coupling", but what they fail to explain is that to use the built in time of use functions, you need to use DC coupled solar.

I went over my NEM 2.0 contract. I am allowed to export a maximum peak current of just 16 amps into my 100 amp feeder line. To export more, I would need to upgrade to a 200 amp feed and panel, and that would have been a huge cost. My solar was sized o max out a 20 amp back feed breaker, which is 16 amps. If I am using no load at all in my house, my existing solar will just touch my 16 amp export limit. And then there is also the total energy export limit. I am allowed to push 900 KWHs per month back to the grid. I don't think I will ever get even close to that.

I did exceed my 16 amp export for a few minutes by accident. I was looking through the settings in the XW when I first got it, and I did not have the solar routed through it yet, so the Enphase solar and the XW each had their own 20 amp breaker into the panel. I was just having it charge, and then discharge "Sell to Grid" as all of my circuits were still in the main panel. My central air conditioner was running, and that pulls a solid 3,200 watts. So I dialed up the grid sell in the XW as a test, and I got it down to where I was only buying about 300 watts, right as the A/C cycled off. Oops. There was still some sun on the solar panels, so my export power jumped to nearly 5,000 watts with the XW and the iQ's all pushing power to the grid.

Since I moved the Enphase iQ's to the output side of the XW, it works so much better now.
I have sell to grid blocked until 4 pm. At 4, my solar is still typically putting out more power that I need, so the XW waits in standby. As the solar output falls off, the XW just starts pushing enough current to make my total export going out the AC1 terminals match my export limit settings. Any power that the Enphase pushes backwards will cause the WX to reduce it's output.

Today we had horrible overcast so solar production was way down, only produced 8 KWH instead of my normal 22 KWH. So while it as charging, basically from grid, I still saw it reduce charge current to limit the grid consumption to less than my grid support current limit.

My eventual plan is adding about 3,000 watts of additional solar panels, but they will only charge the battery bank. During the hot summer months, that should easily make enough power to run everything in my house from 4 pm to 9 pm peak rate time. And since they are essentially an off grid system, they can't export to grid and mess up my NEM agreement. Ideally, I want to make the XW inverter to grid zero current limit. But like all things Schneider, that is buy another box and configure the software again.
 
Here's another battery inverter that can push power into the grid, and doesn't cost an arm and a leg. About $2500 for 6kW.
Unfortunately, the batteries for it do cost an arm and a leg.



Batteries are expensive, and only the cheapest (DIY LiFePO4, or off-brand/recycled cell batteries) might save money.

I agree with GXMnow's suggestion that additional PV could be cheaper.
If you connected a PV inverter behind the relay controlling heatpump compressor, it will come on after 5 minutes. It will disconnect when relay shuts off. It could be sized, together with a PV array having two orientations, to offset much of the consumption.

With something like 5x difference in cost of power between day and night, What if you just run your heatpump at night? It may be less efficient, but in terms of BTU/$, this may be cheaper than using batteries to store power from nighttime until daytime. May also be cheaper than additional PV.
 
My original plan was something very similar to this. I looked at those same GTIL inverters, and I was going to use three 600 watt chargers. I was going to have a microcontroller like an Arduino monitor the output from my Enphase solar. When it topped 1,000 watts, turn on one 600 watt charger. When the power exceeds 2,000 watts, turn on a second 600 watt charger, an at 3,000 watts turn on the third charger. 1,800 watts of charging and still 1,200 or more watts feeding my house when the sun is up. I have 16 soo watt panels with 240 watt iQ7 microinverters, so the system peaks out at about 3,800 watts under peak sun. On good days of sun, I figured that setup could push about 6 KWHs into a battery bank. The microcontroller would then turn on the GTIL inverters from 4 pm to 9 pm as long as the batteries were above the cut off voltage.
I found lamp timers to be a much easier and straightforward way to do that. Then I discovered my SCC supported a dry relay controlled by an internal timer as well as battery voltage, so I just rigged up a 240V AC relay controller by the SCC relay switching battery voltage.

It should all work. But when I figured in the cost of the batteries, I basically chickened out and bought the Schneider. But I got suckered in by their web site. It claims it can do "Time of Use" and "A/C coupling", but what they fail to explain is that to use the built in time of use functions, you need to use DC coupled solar.
That’s the issue with most of the All-in-One Hybrid inverters I considered - your hands get tied somewhere in terms of them not supporting the precise options you are looking for..
I went over my NEM 2.0 contract. I am allowed to export a maximum peak current of just 16 amps into my 100 amp feeder line. To export more, I would need to upgrade to a 200 amp feed and panel, and that would have been a huge cost. My solar was sized o max out a 20 amp back feed breaker, which is 16 amps. If I am using no load at all in my house, my existing solar will just touch my 16 amp export limit.[/B]

Precisely - exceeding your NEM export limit can be costly to do with permission / authorization and dangerous to do without…
And then there is also the total energy export limit. I am allowed to push 900 KWHs per month back to the grid. I don't think I will ever get even close to that.
I don’t recall seeing a monthly export limit in my NEM agreement but I’m going to go back and check…
I did exceed my 16 amp export for a few minutes by accident. I was looking through the settings in the XW when I first got it, and I did not have the solar routed through it yet, so the Enphase solar and the XW each had their own 20 amp breaker into the panel. I was just having it charge, and then discharge "Sell to Grid" as all of my circuits were still in the main panel. My central air conditioner was running, and that pulls a solid 3,200 watts. So I dialed up the grid sell in the XW as a test, and I got it down to where I was only buying about 300 watts, right as the A/C cycled off. Oops. There was still some sun on the solar panels, so my export power jumped to nearly 5,000 watts with the XW and the iQ's all pushing power to the grid.
When I first played around with my GTILs, J made a point of running a continuous load exceeding their max output capability (like the electric oven on broil) precisely to avoid export while I was figuring things out…
Since I moved the Enphase iQ's to the output side of the XW, it works so much better now.
I have sell to grid blocked until 4 pm. At 4, my solar is still typically putting out more power that I need, so the XW waits in standby. As the solar output falls off, the XW just starts pushing enough current to make my total export going out the AC1 terminals match my export limit settings. Any power that the Enphase pushes backwards will cause the WX to reduce it's output.

Today we had horrible overcast so solar production was way down, only produced 8 KWH instead of my normal 22 KWH. So while it as charging, basically from grid, I still saw it reduce charge current to limit the grid consumption to less than my grid support current limit.
The XW is a nice unit. I plan to keep my experiment with GTILs going but if they flake out on me, will probably fall-back to a higher-quality hybrid solution like the XW or Magnum PAE. My main breaker on the outside of the house is connected to the main panel through a 3” piece of conduit running through the wall, so wiring in a critical loads panel essentially means a full rewiring job. That’s what convinced me to try the GTILs first - zero rewiring…
My eventual plan is adding about 3,000 watts of additional solar panels, but they will only charge the battery bank. During the hot summer months, that should easily make enough power to run everything in my house from 4 pm to 9 pm peak rate time. And since they are essentially an off grid system, they can't export to grid and mess up my NEM agreement. Ideally, I want to make the XW inverter to grid zero current limit. But like all things Schneider, that is buy another box and configure the software again.
That’s exactly the rig I’ve put together - SCC charging battery all day long while GTILs are programmed to offset as much consumption as they can during peak hours.

My SCC actually has a second dry contact to control a generator which I’ve managed to program so it will turn on the GTILs early when battery reaches ~50% SOC (meaning a high production day with energy production to spare). This way I pretty much never fill my battery (SCC has never gotten out of Bulk) which would mean wasted PV production…

My DC array size is essentially limited by my consumption - I get no credit to produce any more so I’ll wait until we are consuming more from a planned EV in our future before adding more panels.

But in theory, the system can expand all the way to supplying 100% of home consumption all year long which would translate to 4-5MWh of credit for charging an EV (meaning 13,000 to 17,000 miles of driving which is more than we’ll ever need).

Expansion just means adding solar panels to the already-established rack and strings and eventually adding another 1 or 2 SCCs…
 
My main breaker on the outside of the house is connected to the main panel through a 3” piece of conduit running through the wall, so wiring in a critical loads panel essentially means a full rewiring job.
Full rewiring sounds overwhelming. The reality is that it may just involve moving some circuits to a new subpanel, I am on my third home doing that and it was not overwhelming. The benefits for me outweighed the extra work,
 
Full rewiring sounds overwhelming. The reality is that it may just involve moving some circuits to a new subpanel, I am on my third home doing that and it was not overwhelming. The benefits for me outweighed the extra work,
I hear ‘ya. That option is always there so if I end up concluding my experiment with these little GTIL is a bust, I’ll sieve some time and money wiring in a critical loads panel and put a good UL listed Hybrid Inverter in between.

But the ease of wiring-in a GTIL has to be seen to be believed. Easier than wiring-in a Microinverter-based grid-tied PV-array. Easier than wiring in an EV charger.

I needed to add a new 10A 240V-with-neutral branch circuit to a new dedicated outlet (with both an L2 side and an L2 side),

That’s it.

Each of the two GTIL inverters needed to be plugged into one leg of the outlet and I needed to be sure that the sensor of each GTIL was clamped to the correct leg in the main panel.

I went for extra credit by plugging my GTILs in through lamp timers so I could control time of use (peak hours).

But the wiring required really was unbelievably easy.

Whether these Deye budget GTILs have the quality needed to last a decade or not is a big question mark (doubtful) but the architecture is so compelling that I’ll be surprised if we don’t see higher-quality UL-listed GTIL offerings from other US-based vendors before the decade is out.
 
Here's another battery inverter that can push power into the grid, and doesn't cost an arm and a leg. About $2500 for 6kW.
Unfortunately, the batteries for it do cost an arm and a leg.



Batteries are expensive, and only the cheapest (DIY LiFePO4, or off-brand/recycled cell batteries) might save money.

I agree with GXMnow's suggestion that additional PV could be cheaper.
If you connected a PV inverter behind the relay controlling heatpump compressor, it will come on after 5 minutes. It will disconnect when relay shuts off. It could be sized, together with a PV array having two orientations, to offset much of the consumption.

With something like 5x difference in cost of power between day and night, What if you just run your heatpump at night? It may be less efficient, but in terms of BTU/$, this may be cheaper than using batteries to store power from nighttime until daytime. May also be cheaper than additional PV.
Basically, this is what I am doing. The heater kicks in at night because the water temp starts dropping. I just figured, with a hybrid inverter, 14kWh cell pack, I can do a few hours during the day and also have a power outage backup. THe heatpump is so effective when it's 90 and humid.

The inverter you are mentioning clearly can do solar to grid but can it do battery-to-grid?
 
Whether these Deye budget GTILs have the quality needed to last a decade or not is a big question mark (doubtful) but the architecture is so compelling that I’ll be surprised if we don’t see higher-quality UL-listed GTIL offerings from other US-based vendors before the decade is out.

Would be nice to see a higher quality GTIL with UL listing. Imagine if a majority of households had a 1 or 2 kW GTIL designed for twin 120v outputs. The energy output would be amazing.
 
Basically, this is what I am doing. The heater kicks in at night because the water temp starts dropping. I just figured, with a hybrid inverter, 14kWh cell pack, I can do a few hours during the day and also have a power outage backup. THe heatpump is so effective when it's 90 and humid.

The inverter you are mentioning clearly can do solar to grid but can it do battery-to-grid?

Sunny Boy Storage is SMA's PowerWall. Looks like a Sunny Boy reprogrammed to talk to a BMS and use 400V batteries rather than PV on each of its three inputs. It doesn't do solar itself, only battery to grid (or house loads.)
It is used for such purposes as zero-export, sucking down PV production to prevent power from flowing into the grid, then producing AC from battery when consumption exceeds PV production. It feeds toward the grid, with your AC loads taking the power, in that case.
Pretty sure it can backfeed into the grid if desired, so surplus PV production is stored for credit when rates are higher.

PG&E has some rules on such usage - they limit battery inverter to 1.5x wattage of PV inverter, and require the batteries be charged from PV not from grid.
I think Tesla programs their PowerWall to recharge exclusively from PV rather than grid, simply so it qualifies as alternative energy equipment for tax credit purposes.

The small (3kWh to 30 kWh) "400V" batteries recommended for use with Sunny Boy Storage are actually 48V batteries with a boost converter. Maybe a BMS like REC that can be cascaded for as many as 128 or more cells in series would work to make a compatible DIY battery.
The commercially available batteries recommended for use with it work out to $0.50/kWh of lifetime usage, or did the last time I ran the numbers.

With additional SMA "Automatic Backup Unit" (120/240V transformer and relay), Sunny Boy Storage can make a grid-backup system. But not needed if just doing grid-support or shifting time of consumption.

Sunny Island doesn't normally do battery backfeed to grid unless SCC drives battery voltage higher than target voltage. What it is supposed to do is limit current draw from grid when excessive by inverting from battery. That is all software - I met a guy from a company which did use them for grid-support development, but I don't know how he got that to happen.

Any GT PV inverter would backfeed the grid if fed from batteries. When first connected, you'd want to limit inrush current to the capacitors. MPPT algorithm would default to PV, but would go to maximum power. Simplest control would just be relay to connect AC.


Regardless of how much more effective heatpump is at 90 degrees and humid, I'd think the decrease in effectiveness running at night is less than the cost increase to run off batteries. The fact you're heating pool to comfortable temperature in the summer means modest temperature delta, not like heating a house during Minnesota winters.

What is pool temperature? What heating effectiveness would you get just circulating water through a radiator and blowing ambient air over it? That would raise water temperature toward air ambient.

With the utility rate delta, I was thinking resistance heating at night could be cheaper than batteries. But maybe not - does "SEER" of 16 mean thermal energy is 16x electrical input?

 
Sunny Boy Storage is SMA's PowerWall. Looks like a Sunny Boy reprogrammed to talk to a BMS and use 400V batteries rather than PV on each of its three inputs. It doesn't do solar itself, only battery to grid (or house loads.)
It is used for such purposes as zero-export, sucking down PV production to prevent power from flowing into the grid, then producing AC from battery when consumption exceeds PV production. It feeds toward the grid, with your AC loads taking the power, in that case.
Pretty sure it can backfeed into the grid if desired, so surplus PV production is stored for credit when rates are higher.

PG&E has some rules on such usage - they limit battery inverter to 1.5x wattage of PV inverter, and require the batteries be charged from PV not from grid.
I think Tesla programs their PowerWall to recharge exclusively from PV rather than grid, simply so it qualifies as alternative energy equipment for tax credit purposes.

The small (3kWh to 30 kWh) "400V" batteries recommended for use with Sunny Boy Storage are actually 48V batteries with a boost converter. Maybe a BMS like REC that can be cascaded for as many as 128 or more cells in series would work to make a compatible DIY battery.
The commercially available batteries recommended for use with it work out to $0.50/kWh of lifetime usage, or did the last time I ran the numbers.

With additional SMA "Automatic Backup Unit" (120/240V transformer and relay), Sunny Boy Storage can make a grid-backup system. But not needed if just doing grid-support or shifting time of consumption.

Sunny Island doesn't normally do battery backfeed to grid unless SCC drives battery voltage higher than target voltage. What it is supposed to do is limit current draw from grid when excessive by inverting from battery. That is all software - I met a guy from a company which did use them for grid-support development, but I don't know how he got that to happen.

Any GT PV inverter would backfeed the grid if fed from batteries. When first connected, you'd want to limit inrush current to the capacitors. MPPT algorithm would default to PV, but would go to maximum power. Simplest control would just be relay to connect AC.


Regardless of how much more effective heatpump is at 90 degrees and humid, I'd think the decrease in effectiveness running at night is less than the cost increase to run off batteries. The fact you're heating pool to comfortable temperature in the summer means modest temperature delta, not like heating a house during Minnesota winters.

What is pool temperature? What heating effectiveness would you get just circulating water through a radiator and blowing ambient air over it? That would raise water temperature toward air ambient.

With the utility rate delta, I was thinking resistance heating at night could be cheaper than batteries. But maybe not - does "SEER" of 16 mean thermal energy is 16x electrical input?

Oh, no. Resistance heating is never more efficient, maybe below freezing temperatures with less efficient units. SEER is not unitless and something like BTU/kW averaged out for the whole season. The unitless ratio of provided heat (or cooling) to energy consumed is COP is roughly in the vicinity of 4. I keep the water at 80-82 into the fall (northeast) when the ambient temp starts dropping below 40 by the morning.
In the summer months, I could save 20 cents per kW from time-shiting. A good battery costs me (DIY) about 15 cents per watt. I would be ahead after 1000 cycles not counting the cost of the inverter. I am willing to chalk the inverter up to the other benefits- adding another 1,5 kw solar, outage backup. And even if I include its cost to the ROI, I need less than 2000 cycles. Doable but you can't do it with a $3,000 inverter and off-the-shell battery storage system.
 
PG&E has some rules on such usage - they limit battery inverter to 1.5x wattage of PV inverter, and require the batteries be charged from PV not from grid.
I think Tesla programs their PowerWall to recharge exclusively from PV rather than grid, simply so it qualifies as alternative energy equipment for tax credit purposes.
I have seen those rules in the case of an installation where SGIP funds are concerned. It is my opinion that for an inverter that is run behind the meter no permission from PG&E is required. A building permit is required. The requirement about only charging from the grid is an IRS requirement for the Investment Tax Credit. PG&E could care less if we charge batteries or add any load because that means more revenue. The only control they have is to restrict how much we can export.
 
Last edited:
In the summer months, I could save 20 cents per kW from time-shiting. A good battery costs me (DIY) about 15 cents per watt. I would be ahead after 1000 cycles not counting the cost of the inverter. I am willing to chalk the inverter up to the other benefits- adding another 1,5 kw solar, outage backup. And even if I include its cost to the ROI, I need less than 2000 cycles. Doable but you can't do it with a $3,000 inverter and off-the-shell battery storage system.

Is that $0.05/kWh at night, $0.25/kWh during the day, so time-shifting save $0.20/kWh?
"15 cents per watt" - do you mean per Wh of capacity? So after 1000 cycles it's cost you $0.15/kWh of use?

I believe in long-term return on PV, less certain about batteries. Maybe they do last cycles according to spec, wait and see.

But how does this investment and anticipated return compare to simply running heatpump at night, with reduced efficiency?
This would mean 40 degree F boost on those frigid mornings.
Not my field, and I'm only finding vague answers with a quick search. I do read reduced efficiency at 40 degrees F, heat pump stops being useful down around 25 to 30 degrees F.

I suggest doing the math on cost to heat by operating at night, vs. using batteries, total per year throughout the seasons.
Other desired application of batteries (backup for critical loads) would be much more modest than heating a pool.

Obviously ground-source for heat would be preferred, but you already have air-source. Laying many feet of underground pipe would cost $$, but I could imagine ground to air heat-exchanger to feed your heatpump in the winter.
 
when the utility approves a NEM agreemebt, that is in the context of reviewing upstream capacity of all transformers - if you push more peak power out to the grid than you agreed you would, it can cause damage and possibly even a safety hazard.
Technically transformers can handle upstream power as easily as downstream power. There is no limitation imposed by physics. The issue that we have seen in places like Hawaii is the control systems at substations are not designed to measure and switch reverse flow. That means the utility cannot switch power from one section of the grid to the other. I suppose there is a safety issue there somewhere. As far as the physics are concerned my exported power flows to my neighbors who are on the same transformer. The transformer doesn't see any of my power unless my power exceeds the sum of my neighbors. My issue is that the utilities do not acknowledge that their switching equipment is not designed to control bidirectional flow but they haven't spent the money to upgrade for the future so they can manage that bidirectional flow. Part of that future is one in which they will see less revenue as more people generate their own power.
 
Last edited:
Is that $0.05/kWh at night, $0.25/kWh during the day, so time-shifting save $0.20/kWh?
"15 cents per watt" - do you mean per Wh of capacity? So after 1000 cycles it's cost you $0.15/kWh of use?

I believe in long-term return on PV, less certain about batteries. Maybe they do last cycles according to spec, wait and see.

But how does this investment and anticipated return compare to simply running heatpump at night, with reduced efficiency?
This would mean 40 degree F boost on those frigid mornings.
Not my field, and I'm only finding vague answers with a quick search. I do read reduced efficiency at 40 degrees F, heat pump stops being useful down around 25 to 30 degrees F.

I suggest doing the math on cost to heat by operating at night, vs. using batteries, total per year throughout the seasons.
Other desired application of batteries (backup for critical loads) would be much more modest than heating a pool.

Obviously ground-source for heat would be preferred, but you already have air-source. Laying many feet of underground pipe would cost $$, but I could imagine ground to air heat-exchanger to feed your heatpump in the winter.
The efficiency drops by about 40% from air temp of 80F to 50F with water temp of 80. And, no my calculations do not compare to running only at night but rather running at rate of $.25 vs $0.05. Then again, I have other benefits of having the batteries and the extra inverter. A ground source heatpump would be a tens-of-thousands project.
 
Back
Top