18 month DIY project finally done: 39kW inverter power off grid system

[fmeili1]

Been trying to follow your thread since you finished, but it got lengthy. I plan to cancel my utility service as they are robbing me on a rural "delivery" charge with ~zero consumption through a 3rd party co-op (which never existed until last year).

Really bad how easy they can take our money

But in the next year or 2 I will probably get an EV, hoping to not have shortages, but if that turns out the case I plan to go the same route as you with the double isolation/charger as they also got picky about a backfeed spike on an analog meter no less, or go small generator+rectifier, as my PoCo is just plain unreliable. Wife works 1/2 days, most likely could charge EV mostly via PV every day (except rainy), I currently generate ~30kw excess PV per (sunny) day with room for more.

I like these online double conversion rebuild a lot and I'm really happy that I've done that (now I know I should have done this right from the first day... but as always, learning by doing...)

All the issues I had before were related to AC-in problems (bypass relay, switching timing, find the correct points to switch to grid and back, F60 fault when connecting to the grid, etc.). These problems are all gone now and the certainty that a feedback by accident could NEVER happen. Also I now can use a mix of grid, battery and PV at the same time if required.

If you really want to control the rectifier, you need to DIY them (Huawei R4875G1) to be fully flexible. I've read in other threads, that the EG4 Chargeverter is very limited with it's programming capabilities. Let me know when you need the ESPHome application which I would be happy to share with you.

 

[fmeili1]

Ooops... Maybe I've made a mistake by changing my previous POCO basic energy plan too fast to the Demand TOU plan

I've chosen the "Demand TOU" plan instead of the usual "TOU" plan because I was tempted from the lowest kW prices during off-peak hours (I thought, no problem to prevent on-peak hours via the DIY chargeverters) - I thought, okay, if it costs a few cents more when I need more than 7kW at a time (the "Demand" part of the plan), it's not so bad... unfortunately, I somehow mistook the dollar sign in front of the price on the Demand section as a cent sign...
Also I had a wrong understanding about how this "Demand" part works!

That's my current price plan for "Demand TOU":

They ask from $6.82 (if <7kW peak) up to $9.61 (it >7kW peak) per kWh if using energy at on-peak hours.

The four DIY chargeverters consume about 0.06 kWh per day while on-peak (1.8kWh per month) when not used (hibernate). This will result in only $0.12 energy charge but puts $12.28 just for the Demand charge - 100 times more for the "Demand" part!!!

Ok, this $12 are not really a big deal BUT now it's absolute crucial, that I never use grid support while an on-peak time range is active - otherwise the grid support costs will be exploding!

Unfortunately this will likely not work in summer with our high A/C cooling requirements!

This is our house power consumption chart from last summer (mid of June to mid of August) and it's very likely that I will not be able to make it over a day without using some on-peak grid energy (with my existing PV size). E.g in last July 2,700kWh grid support was required to cover the house load of 5,300kWh. Also I need to stay below 7kW with the chargeverters in the on-peak times to not get the highest demand price of $9.61 per kWh.


It looks like I should change my plan to the standard "TOU" before next summer which has the following price structure:

The energy charges are higher but the "Demand" part will not be longer a problem!

I have to watch the usage of the DIY chargeverters very closely over the next weeks to be sure that my ESP32 controlling program is able to prevent on-peak usage at any circumstances. Maybe it's necessary to extend the current pre-charging time (now it's 2h) before the next on-peak time frame begins to gain enough SOC to make it over on-peak without grid usage.

 

[Brucey]

Usually the demand charge is based on the highest demand period during the billing cycle.

I.e. in February your highest demand was on the 2nd where you hit 5kW, then you'd have a demand charge of 5×$6.82, which would cover whole month. Its not assessed daily.

 

[Brucey]

They ask from $6.82 (if <7kW peak) up to $9.61 (it >7kW peak) per kWh if using energy at on-peak hours.

It's per kW of demand, not kWh of consumption.

Still, if you have demand peaks of 18kW thats going to get expensive, additional fee of 7kW×$6.82+11kW×$9.71, $47+
106.81~$154.

You definitely need to get off that plan you'll never make up the "savings" on the kWh rate.

 

[fmeili1]

Usually the demand charge is based on the highest demand period during the billing cycle.

I.e. in February your highest demand was on the 2nd where you hit 5kW, then you'd have a demand charge of 5×$6.82, which would cover whole month. Its not assessed daily.

Thanks for clarification! The explanations which my POCO added to this plan were confusing me (maybe it's because my English is still not perfect). Good to know that's not on a daily base.

 

[fmeili1]

It's per kW of demand, not kWh of consumption.

got it

Still, if you have demand peaks of 18kW thats going to get expensive, additional fee of 7kW×$6.82+11kW×$9.71, $47+
106.81~$154.

You definitely need to get off that plan you'll never make up the "savings" on the kWh rate.

I'll change the plan to the "standard" TOU as soon as possible.

 

[vietxtlife]

After the 30% federal incentives and the $1000 Arizona solar incentives, my net invest was about $34,000.

With different AIO's, which may not require equipment cooling, I could have saved the ducted mini split, the ductwork and an the other mini split with a typical indoor unit to cool down the ambient in the garage. Also the controller with the emergency circuit and the blackstart UPS was not a must.

I think, it's realistic that one can build a DIY system with this performance data for about <=$30,000 (after incentives). Maybe cheaper, because I bought the components 18 month ago and now all major components are cheaper.

Does federal incentives still going? Or they cut it already?

 

[fmeili1]

Does federal incentives still going? Or they cut it already?

From my knowledge the 30% incentives are still in use and will be valid until 2032 and reduced to 26% in 2033, 22% in 2034 and retired in 2035.
Currently, I have no idea if this will last until 2035..

 

[ShirBlackspots]

I wonder, can I apply for the credit in my taxes in 2026 for multiple years (2022, 2024 and 2025)? I kept track of what I spent.

If just 2025, then that would be almost enough to pay for what I plan on buying for 2026.

 

[vietxtlife]

I wonder, can I apply for the credit in my taxes in 2026 for multiple years (2022, 2024 and 2025)? I kept track of what I spent.

If just 2025, then that would be almost enough to pay for what I plan on buying for 2026.

I have to use the tax credits through multiple years taxes to use them all from 2023-2025, not that im complaining lol