Adding storage to my Enphase system

[pvdude]

Looks like you have a good plan to help the thermostat sense the actual cooling requirements!

When I lived in upstate NY back in the 1970's, I put an incandescent lamp near the thermostat to reduce the
heater operation cycles below what the old "mercury tube" thermostat could sense.
Heating used #2 Diesel - -uggg...
I lived alone in the 1850's farmhouse, did not bother me that the house was 40-50°

I'm going to try an experiment next month with the HVAC here.
Wife is going to be traveling first 2 weeks of August, so I can interrupt house power at will!
Temps here have been in the high 90's day, high 70's night.

There are only a few loads connected to the Schneider inverters now.
I moved things w/ electric motors off grid and on to the inverters, but most of the house is still on grid.
Normally the PV powers the loads shown in the screenshot.


The XW Pro's are configured for "grid support", and use the grid when the PV is not producing enough to support the loads.
The batteries are always in "standby", and only used for the brief transition between PV output "sag", and grid taking over the loads.

When I have the house to myself in August, I'm going to try to run everything off batteries overnight.
Disconnecting the house from the grid.
With the upgraded batteries, it will be possible to source current in the 8-10 hour rate,
until the Sun is up again.

As the HVAC compressor only cycles a few times per hour @ night, I am optimistic.

If this experiment works, next project is to add the HVAC air handler, so that the entire (compressor & air handler) Heat Pump is powered by the inverters.

The problem with this scheme is Winter.
There are usually 10-20 Winter nights here per year that get cold enough that the HVAC activates the 15KW resistive heat coils for a few minutes.
No way I want to subject the inverters to that.

So the plan would be to put the switch in "grid" when the HVAC thermostat is set to " heat", and switch to "solar" when the thermostat is set to "cool".

It might work!!!!

 

[GXMnow]

We are into day 5 of this heat wave. Here is this billing period of daily power as seen by So Cal Edison.

Starting on July 12, the solar production was not quite keeping up with production. But I had some reserve power still in the battery, but that depleted as the hot days with the A/C running took it's toll. The battery has been hitting the low voltage shut off for the last 3 days now. Each day, it has had to switch to using grid power a little earlier. There is only one more full day left on this billing cycle, and even with it being another crazy hot day, I expect it to look like the 17th again. It should not really be any worse. For the whole month, I am still about $60 worth towards energy credit, averaging 8 KWHs of export each day. Yesterday, my son decided to cook in the oven, and I think he left it on a while after his food was done. It is gas, but it heats the whole main floor, so the A/C started vey early trying to keep the house cool. Here is what the XW did trying to keep up.

The A/C was cycling even before noon. Between cycles, the extra Enphase power was still going into the battery, and the DC system again produced about 10 KWHs. The battery had hit cut off before 4 am, and it ended the day a little lower than it began. This morning, it hit cut off again at 4:26 am, so it actually did a little better. And my son is at work, so the energy use should be a bit lower today. Hopefully it will recover a bit and use a little less grid power on this final day of the billing period. Here is the hourly SCE usage for this same day.

Due to their auto scaling, it looks bad, but that tallest bar still never hit 1.2 KW, or just 5 amps on the 240 volts into my house. Considering the A/C compressor alone is pulling 14 amps at 240, I would say the solar panels are doing a great job. During that 3 pm hour, the A/C was on about 1/3 of the time. And the XW was still charging at 5% (7 amps x 55 volts = about 385 watts) even with the A/C running. The Enphase system was cranking out about 3,000 watts. Here is the Enphase production for this day.

We did have a few clouds go by early in the day, but beyond that, production did well even in the over 100F heat.

All in all, I think my system handled this quite well. My electric bill is still going to be a credit, not a charge. But I did need the grid to make it through this heat wave. If I was off grid, I would be miserable with the power going out. But these days only happen a few times a year. Can I justify adding even more solar panels to cover the few times this happens? I am still thinking about what it would take to add 4 more panels to my Enphase system. That would make another 6 KWHs on a day like that. Still not enough to match the usage on that day, but with that extra production every day, the battery would not have depleted in just 3 days into the heat wave. And with the newer panel, I could put up 4 400 watt ones instead of my 300 watt panels. That puts me over 8 KWHs more per day.

But if I expand the Enphase system, I have to deal with my 120% rule limit on my main panel. I would have to either drop my main breaker to a 90 amp, install it as a line side tap, or change out the old 100 amp panel for a larger one. The only one that is cost effective is the 90 amp main breaker. But will that be an issue when the A/C starts with no solar coming in?

At this point, I am very happy with the system performance. This month last year, I had to pay a $100 electric bill. This year, it is still going to be a $50 credit. August and September are going to be the real tests. I am sure I will start to eat into my existing $214 in credit as I am sure we will have a few more 100+ days in the coming weeks.

 

[GXMnow]

On another note, I have not been able to swap in the new Envoy board just yet. I still can't log into it because of the token issue. Enphase sent me a token to try, and I still get the "Bad Token" error. So I powered up the new Envoy board again, and I have it open out to the internet, and I get the info the the Enphase support guy I have been working with. I hope they will be able to remote into the board and get it configured, or at least get me a valid token so I can get it on WiFi and install it in the Combiner soon. The old Envoy board is losing connection with the panels for a few hours every day. It seems to drop out when it gets hot outside. I know, big shocker there.

 

[Hedges]

But I did need the grid to make it through this heat wave. If I was off grid, I would be miserable with the power going out. But these days only happen a few times a year. Can I justify adding even more solar panels to cover the few times this happens? I am still thinking about what it would take to add 4 more panels to my Enphase system. T

Used panels can be cheap enough that easy to justify. Even retail priced panels GT PV hardware costs around $1.00 or so per W, makes power for $0.025 or so per kWh. Added PV is easier with string inverter than (already overpaneled) microinverter. Since grid costs you $0.25 per kWh or more, overbuilding anything less than 10x average usage still saves you money.

If you can get a signal before SoC drops so low as to shut system off, use that to interrupt A/C thermostat signal. "Load shed". System continues to operate, but house heats up for a while until there is enough stored power to reenable. (higher thermostat setting sufficient to avoid this might be preferable. PID loop bumping thermostat to achieve some SoC or voltage and bring net battery current to zero?)

I think my wall thermostat uses one wire to enable furnace, which turns on its fan when heat exchanger is hot. For the add-on A/C unit, one wire enables compressor and another wire enables the "fan only" function of the furnace. I discovered that if I externally enable "fan only" (via "fan" output of an electric duct heater), that is connected in parallel with A/C signal, so duct heater and A/C played tug-of-war. I had to add a relay to isolate those signals.

If your system can't keep whole house comfortable on such hot days, how about running a window A/C for one or two rooms instead?
Add a mister at air inlet of condenser? (use demineralized water, probably not softened either.)

But if I expand the Enphase system, I have to deal with my 120% rule limit on my main panel. I would have to either drop my main breaker to a 90 amp, install it as a line side tap, or change out the old 100 amp panel for a larger one. The only one that is cost effective is the 90 amp main breaker. But will that be an issue when the A/C starts with no solar coming in?

I consider 120% rule a guideline and safety net.

First point is to have PV breaker at far end of busbar from main. That way, busbar never exceeds 100% of main breaker current. I read that PV breaker and main breaker could have each been 100% of busbar for a 200% rule, but concern was it might be relocated later and overload. Only 100% rule would be completely safe, but doesn't allow adding PV if main breaker equals busbar. With 120%, they threw us a bone.

Second point, which I figured out but haven't seen elsewhere, is that if you have sufficient (excessive) 120V loads on a single phase, it is possible for neutral wire to carry the 120% or 200% while L1 and L2 do not exceed 100% (which main breaker would limit.) Reason is that split-phase secondary of utility transformer acts as an auto-transformer. If you have 2400W of PV at 240V, 10A supplied to L1 and 10A to L2, consider 2400W of load on L1 only. 2400W / 120V = 20A. PV supplies 10A on L1, consumed by load. Main breaker supplies the other 10A needed by load. PV puts 10A on L2, which goes out through main breaker to transformer. N supplies all 20A to load (even though main breaker carries only 10A per phase, just in opposite directions.) Same would happen with 24,000W, 100A through main breaker, 200A through neutral. For a 100A main panel, that would be way over spec for neutral busbar and utility drop.

I think you can bend the rules that would break if you keep your eyes open.

 

[Shimmy]

But if I expand the Enphase system, I have to deal with my 120% rule limit on my main panel. I would have to either drop my main breaker to a 90 amp, install it as a line side tap, or change out the old 100 amp panel for a larger one. The only one that is cost effective is the 90 amp main breaker.

Is replacing the panel itself with a 200A bus unit in-place (and re-using the existing breakers) not viable? Some of the panels can just have the internal bus assembly replaced, although you often have to buy the whole box.

 

[zanydroid]

Not sure where the main panel discussion started but in case it hasn’t come up yet you can also address this by moving out all load breakers to a new adjacent subpanel and have just a 100A breaker on main feeding it. I currently have 32A of AC-out backfeeding through a 100A main with only 100A breaker and am going up to 72A shortly. All permitted.

 

[zanydroid]

Is replacing the panel itself with a 200A bus unit in-place (and re-using the existing breakers) not viable? Some of the panels can just have the internal bus assembly replaced, although you often have to buy the whole box.

It may not be that easy to do a panel board replacement in a main. Depends on mechanical size details and how the power gets fed in. Also replacement panel boards in 200A might be hard to fit in a 100A chassis.

 

[Hedges]

Not sure where the main panel discussion started but in case it hasn’t come up yet you can also address this by moving out all load breakers to a new adjacent subpanel and have just a 100A breaker on main feeding it. I currently have 32A of AC-out backfeeding through a 100A main with only 100A breaker and am going up to 72A shortly. All permitted.

Sounds the same as a line-side tap. Except doing that for a 100A main. Easier I think to get a tap for 40A.

Are there magnetic/hydraulic breakers that fit common AC panel boards? They can run 100%, some maybe 95% continuous rather than having to be derated to 80%.

 

[zanydroid]

It might be sort of similar to a line-side tap, but it's load side from the perspective of the POCO so they don't have to approve it, which is a huge bonus in my book. I think of it as just making the MSP a straight "wire" between the main breaker and the 100A subpanel feeder. Then the backfeed calculation is based entirely on the busbar rating of the subpanel. If you get a 200A subpanel feeding from that 100A breaker you can put solar breakers wherever you want in that subpanel and it won't matter.

Note that the panelboard rating also comes into play, so even if you find a 100% rated breaker you have to figure out how to make an adjustment on the panel side.

 

[Hedges]

Oh, no tap.

I've heard of PV aggregator panel, PV breakers only no loads, not subject to 120% rule.
And of course you can put a 100A breaker for loads anywhere on a 100A panel.
You're describing putting a 100A breaker on end of busbar for bidirectional power flow. And no breakers anywhere else. Fine if that complies with NEC and flies with AHJ; is it considered compliant?

I'm doing similar by buying a Square-D box with just a 200A main breaker and lugs, no busbar.
You could pretty much do that if you removed busbar from box and installed lugs. (may not fly)

A Midnight/CBI breaker is guaranteed no trip under 105% load. So a 20A breaker could carry 20A continuous, instead of derating to 16A. If such was available for QO or Homeline panel, would still comply with 120% rule but carry 25% more power.

Install a tap (switched or unswitched) in last position of busbar (instead of a breaker), then wire to a 100% rated DIN breaker? This is just to get around the 20% derating.

 

[zanydroid]

Oh, no tap.

I've heard of PV aggregator panel, PV breakers only no loads, not subject to 120% rule.
And of course you can put a 100A breaker for loads anywhere on a 100A panel.
You're describing putting a 100A breaker on end of busbar for bidirectional power flow. And no breakers anywhere else. Fine if that complies with NEC and flies with AHJ; is it considered compliant?

It is compliant and is common. You can put that 100A load side breaker anywhere, it doesn't matter as long as it's the only breaker on the 100A bus (besides the main breaker)

The Community Forum

Solar and energy storage Discussion forum, one of the most interactive and well organized forum.

www.solarsean.com

"The sum of the ampere ratings of all overcurrent devices on panelboards, both load and SUPPLY devices, excluding the rating of the overcurrent device protecting the busbar shall not exceed the ampacity of the busbar:"

So you don't have to count the main breaker. For a 100A main busbar the only sensible thing (IMO) is to feed a subpanel from that main panel. For a 200A you have a few more options I guess.

It might even have a nickname:

Hawaiian Tie In

Has anybody heard of the term ‘Hawaiian tie in’? We are a solar company in California. We have been told that 'Hawaiian tie in' is common practice out east. With a Hawaiian tie in, we remove all load breakers and put a lug kit to feed a sub where we relocate all house loads. This makes the...

forums.mikeholt.com

 

[Hedges]

As commenters there observe, Hawaiian tie-in has two breakers feeding the panel. 100A main and 100A backfeed. If a non-breaker tap, it's still 100A or whatever from downstream panel. Technically violates 120% rule.

As you note, if all load breakers removed, no way to overload the L1/L2 busbars.
(I proposed you could as well physically remove the busbar, except you might not then have a UL listed way to connect cables to output of main breaker.)

Not sure where the main panel discussion started but in case it hasn’t come up yet you can also address this by moving out all load breakers to a new adjacent subpanel and have just a 100A breaker on main feeding it. I currently have 32A of AC-out backfeeding through a 100A main with only 100A breaker and am going up to 72A shortly. All permitted.

But I say you can end up with 100A + 72A = 172A going through neutral busbar and utility drop neutral wire.

Put 172A loads on single-phase breakers all on L1 of your new sub panel. Connect 72A @ 240V of PV to (200A?) sub-panel. You can now supply 100A + 72A = 172A on L1 to your loads - no problem there, it's a 200A panel!

But the loads draw 172A on neutral from grid, through the poor little 100A main panel and utility drop. Smoke and fire!

 

[zanydroid]

As commenters there observe, Hawaiian tie-in has two breakers feeding the panel. 100A main and 100A backfeed. If a non-breaker tap, it's still 100A or whatever from downstream panel. Technically violates 120% rule.

120% rule doesn't apply. Look at 705.12 (B)(2)(3)

Load-Side Source Connections | UpCodes

Explore a searchable database of US construction and building code. Code regulations are consolidated by state and city for easier navigation.

up.codes

"
(3) Busbars
One of the following methods shall be used to determine the ratings of busbars:

"
120% rule is option (2) (end feed) and (4) (center feed)
Sum of breakers rule is option (3)

Since it passes option (3), you can ignore (2) and (4).

Actually in my ongoing installation, I cited 100% rule (1) when 120% rule (2) was possible, because I had a big enough busbar and I didn't want to get into a food fight with the AHJ over whether tandem breakers at the end of the busbar are really at the end of the bus bar, and whether I've gone over the stab limit by putting 60A of tandem breakers across from a 40A combiner panel feeder.

Re: neutral overload. Yes, we've discussed that neutral overload before. NEC doesn't care, and neither do I, since I don't have anywhere near 100A of 120V loads at my house. I hope insurance company doesn't care if NEC doesn't care, lol.

 

[Hedges]

"
3. The sum of the ampere ratings of all overcurrentdevices on panelboards, both load and supply devices, excluding the rating of the overcurrentdevice protecting the busbar, shall not exceed the ampacity of the busbar. The rating of the overcurrentdevice protecting the busbar shall not exceed the rating of the busbar. Permanent warning labels shall be applied to distribution equipment displaying the following or equivalent wording:
WARNING:
THIS EQUIPMENT FED BY MULTIPLE SOURCES. TOTAL RATING OF ALL OVERCURRENT DEVICES EXCLUDING MAIN SUPPLY OVERCURRENT DEVICE SHALL NOT EXCEED AMPACITY OF BUSBAR.
The warning sign(s) or label(s) shall comply with 110.21(B).
"

I think this means 100A main breaker on 100A busbar, OK to have 100A PV breaker backfeeding and no loads, or 80A PV breaker backfeeding it and 20A load breakers, or 60A PV breaker backfeeding it and 40A load breakers, etc.

Ordinarily you could have loads totaling any amount, say 200A of loads. Of course, if you exceed 100A it trips the main breaker. The moment you have 10A PV breaker backfeeding the panel, to use (3) the loads can't exceed 90A of breaker.

This one is safe, can't overload either busbars or neutral. And it allows repurposing main panel as pass-through, 100A main breaker to 100A PV or 200A main to 200A PV.

Yours under 100A single-phase loads won't overload it either (got any split-phase loads that are imbalanced? A clothes dryer for instance is 240V, but it's motor may be 120V.)

Do you have a Square-D plug-on neutral? Until you've taken care of the screw torque recall, it could overheat within spec.

 

[zanydroid]

"
3. The sum of the ampere ratings of all overcurrentdevices on panelboards, both load and supply devices, excluding the rating of the overcurrentdevice protecting the busbar, shall not exceed the ampacity of the busbar. The rating of the overcurrentdevice protecting the busbar shall not exceed the rating of the busbar. Permanent warning labels shall be applied to distribution equipment displaying the following or equivalent wording:
WARNING:
THIS EQUIPMENT FED BY MULTIPLE SOURCES. TOTAL RATING OF ALL OVERCURRENT DEVICES EXCLUDING MAIN SUPPLY OVERCURRENT DEVICE SHALL NOT EXCEED AMPACITY OF BUSBAR.
The warning sign(s) or label(s) shall comply with 110.21(B).
"

I think this means 100A main breaker on 100A busbar, OK to have 100A PV breaker backfeeding and no loads, or 80A PV breaker backfeeding it and 20A load breakers, or 60A PV breaker backfeeding it and 40A load breakers, etc.

Ordinarily you could have loads totaling any amount, say 200A of loads. Of course, if you exceed 100A it trips the main breaker. The moment you have 10A PV breaker backfeeding the panel, to use (3) the loads can't exceed 90A of breaker.

This one is safe, can't overload either busbars or neutral. And it allows repurposing main panel as pass-through, 100A main breaker to 100A PV or 200A main to 200A PV.

Yes this clause is extremely conservative if you think about it. It’s rather painful to employ when there are any load breakers. Also note that most of the other clauses refer to 125% of power source output ampacity while this one specifically refers to OCPD ampacity. So you lose on rounding error…

Also for completeness this is actually the clause used for stuffing combiner panels with a lot of solar breakers. With 120% rule you would need a much more massive subpanel. I think the Envoy combiner which has a load breaker to power the brain has to derate backfeed below the busbar rating of the panel board component they sourced from one of the panel makers (Eaton? It’s in their specs, because you have to buy the breaker to populate it).

Yours under 100A single-phase loads won't overload it either (got any split-phase loads that are imbalanced? A clothes dryer for instance is 240V, but it's motor may be 120V.)

My dryer motor is indeed 120v but I can’t imagine it’s more than 1HP.

Do you have a Square-D plug-on neutral? Until you've taken care of the screw torque recall, it could overheat within spec.

I don’t have any square D panels, but thanks for the heads up.

 

[GXMnow]

My main panel is a GE. There is a slim chance I could find a 150 or 200 amp GE bus bar assembly that might fit. But I don't think that would be a good route for me. There are 2 reasons I don't want to change out the panel. It is a single chassis which also contains the power meter, and it is flush mounted in the exterior stucco wall of my garage. Changing out the meter bas is a huge pain, and having to chisel out the wall to change the panel is no small task either.

Removing ALL load breakers from the main panel, and only having a single breaker feed a new 200 amp panel is legal and works well. I did look into that. I could locate the new panel inside the garage mounted back to back to the existing panel. Most of the circuit leads would be long enough to reach to the new breakers that way. If (when) I get an EV and need a 50 amp Level 2 EVSE, that may be the way to go. Until then, my only serious load is the A/C compressor. It's a 40 amp breaker to handle the 105 amp inrush current. The running current is only 14 amps.

Dropping down to a 90 amp main breaker looks like it would work just fine with all of my existing loads. That would allow a 30 amp back feed breaker and I won't be adding that much more grid tied solar. If I added the 4 more 400 watt panels on iQ7a inverters, that would bring my total back feed current up to 22 amps with all inverters at clipping. That is safe with a little room to spare on a 30 amp breaker with 80% rule.

But now charging an EV at 40 amps, and then having the 105 amp start surge from the A/C while the sun is down, might be an issue with only a 90 amp main breaker.

In a full year, I really have so few days that need the extra energy, it makes it tough to justify the time and money to do this. The big issue will be a plug in car. My brother is doing more mileage now in his Chevy Bolt and even with his Sun Power grid tied system, and the EV rate plan, he is complaining about his electric bills again. I had to remind him how much I still spend on gasoline in my hybrid. I drive up to 300 miles a week just when working in Los Angeles. In a typical EV, that is an additional 86 KWHs a week. The car alone would need another 3 KW of solar panels to cover it. I filled up my hybrid today. It took 12.1 gallons at $4.97 (Just over $60) to fill up after going 528 miles. That is a little over 11 cents a mile out of my pocket. At my lowest electric rate of 28 cents per KWH, assuming 3.5 miles per KWH, that makes the cost per mile drop to 9.33 cents per mile. Not a huge savings, but it is less, even if I buy all of the EV power from the grid. Any extra solar production I can funnel into the EV would reduce the cost even further.

It is certainly a thought game to see what works best. Looking at the new cars available now, I am still leaning towards another hybrid, but the next one will likely be a plug in. My 216,000 mile, 10 year old C-Max is still running great, so I am not in a big rush yet. Would love to see an EV with LFP cells that can exceed 350 miles on a charge. But that may be a while still.

 

[zanydroid]

Removing ALL load breakers from the main panel, and only having a single breaker feed a new 200 amp panel is legal and works well. I did look into that. I could locate the new panel inside the garage mounted back to back to the existing panel. Most of the circuit leads would be long enough to reach to the new breakers that way. If (when) I get an EV and need a 50 amp Level 2 EVSE, that may be the way to go. Until then, my only serious load is the A/C compressor. It's a 40 amp breaker to handle the 105 amp inrush current. The running current is only 14 amps.

It is legal to use main panel as junction box and extend with THHN etc. If the conduit between boxes is shorter than 24” you do not need to do CCC derate.

You might be able to move everything with 1-2 1” or 1.25” conduit.

 

[Shimmy]

As commenters there observe, Hawaiian tie-in has two breakers feeding the panel. 100A main and 100A backfeed. If a non-breaker tap, it's still 100A or whatever from downstream panel. Technically violates 120% rule.

If you protect the bus on both ends with nothing connected in the middle then the main panel is not part of the equation; there is no way to overload it. Likewise, if you have a downstream breaker subfed from lugs off the main that matches the rating for the main then that whole path is protected. The downstream panel can have 200, 225, or 400A bus and the 120% rule only comes into play for that panel.

...unless I am missing something big or completely misunderstanding what they are talking about on Mike Holt. Can't find my old login details so I can't see the single line they are arguing about.

 

[Hedges]

Agreed. Does not overload and complies with Texas Electrical Code (hopefully taken straight from NEC?)

A clause in the rules Zany linked (Texas Electrical Code 2020) is what allows it. Sum of breakers (excluding main breaker protecting panels) does not exceed panel busbar rating.

"
3. The sum of the ampere ratings of all overcurrentdevices on panelboards, both load and supply devices, excluding the rating of the overcurrentdevice protecting the busbar, shall not exceed the ampacity of the busbar. The rating of the overcurrentdevice protecting the busbar shall not exceed the rating of the busbar. Permanent warning labels shall be applied to distribution equipment displaying the following or equivalent wording:
WARNING:
THIS EQUIPMENT FED BY MULTIPLE SOURCES. TOTAL RATING OF ALL OVERCURRENT DEVICES EXCLUDING MAIN SUPPLY OVERCURRENT DEVICE SHALL NOT EXCEED AMPACITY OF BUSBAR.
The warning sign(s) or label(s) shall comply with 110.21(B).
"

I think this means 100A main breaker on 100A busbar, OK to have 100A PV breaker backfeeding and no loads, or 80A PV breaker backfeeding it and 20A load breakers, or 60A PV breaker backfeeding it and 40A load breakers, etc.

Ordinarily you could have loads totaling any amount, say 200A of loads. Of course, if you exceed 100A it trips the main breaker. The moment you have 10A PV breaker backfeeding the panel, to use (3) the loads can't exceed 90A of breaker.

This one is safe, can't overload either busbars or neutral. And it allows repurposing main panel as pass-through, 100A main breaker to 100A PV or 200A main to 200A PV.

Yours under 100A single-phase loads won't overload it either (got any split-phase loads that are imbalanced? A clothes dryer for instance is 240V, but it's motor may be 120V.)

Do you have a Square-D plug-on neutral? Until you've taken care of the screw torque recall, it could overheat within spec.

 

[GXMnow]

My preliminary month end So Cal Edison bill is posted. It may change a little before they issue the actual bill, but it has always been very close. My minimum monthly charges and the Non Bypassable Charges is only about $11 again, so it still has not used up my California Climate Credit. That balance is still a credit of about $16. And my new energy charges total up to another $70+ in credit. At the end of month 10 of my billing cycle, I now have a credit of over $310 from exported energy. This was my first July billing period that resulted in a credit. The DC panels made all the difference.