diy solar

diy solar

New Solar - Seeking Advice

MrSam

New Member
Joined
Jan 6, 2021
Messages
8
Location
NC, USA
Hi everyone, lurker for a while but now it's time to post my thread and glean recommendations and advice from you that are much more experienced and knowledgeable at this than I am.

About Me
I'm a former software & database developer / "ladder climber" of the healthcare Information Technology realm turned avid DIYer within the home renovation biosphere with experience in all areas of residential systems (foundations to furnishings). I live in a grid-connected home in NC and want to offset or negate my energy bill while also building a battery backup system to run my home if needed during a grid outage. I've dreamed of having a solar system for years and the time has come to make that a reality as best I can. I will not be going into debt for or contracting out any of this project to anyone so I feel this is the right forum to take part in.

Project Specific Information
A friend of mine convinced me to "go in together" and purchase a pallet of off-lease commercial solar panels. I have a good start with the 5Kw PVs but know I will eventually need more. I want to build my system with the plans of scalability as best I can while remaining budget friendly.
I want to run as much of my home from PV + batteries as possible with the ability to run critical loads during grid outages.
I will not feed power to the grid.

My System To Date
(20) 250w polycrystalline solar panels

My Home's Energy Usage
45 Kwh per day average (stay-at-home family, homeschoolers, and wife and I run two home based businesses -- so we're always home using the juice). Estimated PV array needed to fully run the home: 15 Kw.

My Starting Question(s)
1. The all-in-one inverter/charger systems are appealing for their simplicity, but are they a good match for my starting setup and eventual goals vs. buying individual components?
2. Can the all-in-one inverter/charger systems simply connect into my home's panel or do I need to break out my home circuits to a panel that does not exceed the inverter/charger's capacities? For example, my home's dryer, well pump, tankless water heater all exceed the capabilities of the inverter/charger. Do I need to separate them to prevent killing the device or are they "smart enough" to just wire into and feed what they can into my home's power?
 
Last edited:
Welcome to the forum.

I personally don't trust MPP Solar/Growatt class systems for primary or backup use. I wouldn't hesitate to use one for a project or solar shed or something I see as non-critical. I'm in the process of incrementally expanding my off-grid system that will eventually be the sole power for a dwelling. I went with all Victron hardware. Victron, Outback, Midnite Solar, Xantrax, Schneider, etc. are the names you trust because those products have a 20+ year history of reliability. You pay for that reliability instead of spending less on products that you hope last.

Most of the AiO products are high frequency inverters meaning they have negligible surge capability, so if you're running anything with motors (air compressor, table saw, HVAC, etc.), you need to consider those limitations.

AiO is appealing until one component fails and you have to replace the whole thing.

If you're comfortable with spending the money for an AiO as a low cost to entry option and then relegate it to "spare" or "project" status once it's time to upgrade to your final configuration, it's probably fine.

The best way to minimize cost is to fully plan your final system and then find a way to break it up into pieces.

45kWh of DIY LFP batteries will cost you about $8K. Since you know your usage, I recommend you use link #6 in my signature to simulate your PV system to establish what it takes to meet your energy needs. Hopefully, your power company breaks your usage out into months, and you can compare them that way rather than just a 45kWh/day average.
 
Is grid-tie net metering available?
So long as the rates are good, that is the most cost effective. Your "battery" is accounting for the bill.

How many effective hours of sun do you get according to web sites, for your location and weather?
For me in San Jose, 5.5 hours/day average throughout the year.

Your 5 kW (STC) of PV can probably deliver 4.0 to 4.5 kW actual power.
4.0 x 5.5 = 22 kWh/day average, so doubling that to 10 kW (STC) might supply your 45 kWh/day on average, with net metering.
To get by off-grid, summer and winter production must both meet actual consumption.

What I have and am happy with is an excessive amount of PV panels on Sunny Boy grid-tie inverters, with net metering.
I have Sunny Island battery inverters which function as a UPS. They have an undersized AGM battery, just enough to make it through the night.
This provides backup when grid fails, and there's enough production to run A/C and everything else direct off PV during the day. Battery bounces along fully charged.

With grid-tie, the excess production kept as accounting on the books for winter electric heating.
If disconnected from grid, not enough PV for heating in the winter so I use the gas furnace, consume less electricity.

Starting with grid-tie would make your system affordable up front and scalable.
Select equipment compatible with future battery system.
There are a bunch of rules regarding how the inverter interacts with the grid, how panels are mounted on the roof, shutdown system for fireman safety. That means not every new in the box old model product you see for sale can be used for grid-tie. (For off-grid some could be used, but safety rules would still apply.)

Yes, separating "excessive", "normal", and "critical" loads is a good approach.
Some can be with manual interlocked breaker, some with load-shed relay based on battery state of charge or other signals.

There might be some you only operate with grid or generator.
Laundry maybe on-grid or when the sun shines.
TV when battery SoC is OK.
Phone and alarm all the time.

I have 200A service, 100A pass-through the UPS, GT inverters are only critical load at this time, 100A load-shed relay for the entire house.
Plan to have additional load-shed for things like laundry.
I can manually flip interlocked breakers to feed the 200A panel and garage from the UPS.
 
My Starting Question(s)
1. The all-in-one inverter/charger systems are appealing for their simplicity, but are they a good match for my starting setup and eventual goals vs. buying individual components?
2. Can the all-in-one inverter/charger systems simply connect into my home's panel or do I need to break out my home circuits to a panel that does not exceed the inverter/charger's capacities? For example, my home's dryer, well pump, tankless water heater all exceed the capabilities of the inverter/charger. Do I need to separate them to prevent killing the device or are they "smart enough" to just wire into and feed what they can into my home's power?
I'm running dual 3kw Growatts to power a critcial loads circuit that I installed. Runs about 90% of the 120V in the house. No operational issues. Been online with cycleing between solar and grid twice a day for 3-4 months now.
 
Welcome to the forum.

I personally don't trust MPP Solar/Growatt class systems for primary or backup use. I wouldn't hesitate to use one for a project or solar shed or something I see as non-critical. I'm in the process of incrementally expanding my off-grid system that will eventually be the sole power for a dwelling. I went with all Victron hardware. Victron, Outback, Midnite Solar, Xantrax, Schneider, etc. are the names you trust because those products have a 20+ year history of reliability. You pay for that reliability instead of spending less on products that you hope last.

Most of the AiO products are high frequency inverters meaning they have negligible surge capability, so if you're running anything with motors (air compressor, table saw, HVAC, etc.), you need to consider those limitations.

AiO is appealing until one component fails and you have to replace the whole thing.

If you're comfortable with spending the money for an AiO as a low cost to entry option and then relegate it to "spare" or "project" status once it's time to upgrade to your final configuration, it's probably fine.

The best way to minimize cost is to fully plan your final system and then find a way to break it up into pieces.

45kWh of DIY LFP batteries will cost you about $8K. Since you know your usage, I recommend you use link #6 in my signature to simulate your PV system to establish what it takes to meet your energy needs. Hopefully, your power company breaks your usage out into months, and you can compare them that way rather than just a 45kWh/day average.
Thank you for the information and the input, I greatly appreciate it!
 
Is grid-tie net metering available?
So long as the rates are good, that is the most cost effective. Your "battery" is accounting for the bill.

How many effective hours of sun do you get according to web sites, for your location and weather?
For me in San Jose, 5.5 hours/day average throughout the year.

Your 5 kW (STC) of PV can probably deliver 4.0 to 4.5 kW actual power.
4.0 x 5.5 = 22 kWh/day average, so doubling that to 10 kW (STC) might supply your 45 kWh/day on average, with net metering.
To get by off-grid, summer and winter production must both meet actual consumption.

What I have and am happy with is an excessive amount of PV panels on Sunny Boy grid-tie inverters, with net metering.
I have Sunny Island battery inverters which function as a UPS. They have an undersized AGM battery, just enough to make it through the night.
This provides backup when grid fails, and there's enough production to run A/C and everything else direct off PV during the day. Battery bounces along fully charged.

With grid-tie, the excess production kept as accounting on the books for winter electric heating.
If disconnected from grid, not enough PV for heating in the winter so I use the gas furnace, consume less electricity.

Starting with grid-tie would make your system affordable up front and scalable.
Select equipment compatible with future battery system.
There are a bunch of rules regarding how the inverter interacts with the grid, how panels are mounted on the roof, shutdown system for fireman safety. That means not every new in the box old model product you see for sale can be used for grid-tie. (For off-grid some could be used, but safety rules would still apply.)

Yes, separating "excessive", "normal", and "critical" loads is a good approach.
Some can be with manual interlocked breaker, some with load-shed relay based on battery state of charge or other signals.

There might be some you only operate with grid or generator.
Laundry maybe on-grid or when the sun shines.
TV when battery SoC is OK.
Phone and alarm all the time.

I have 200A service, 100A pass-through the UPS, GT inverters are only critical load at this time, 100A load-shed relay for the entire house.
Plan to have additional load-shed for things like laundry.
I can manually flip interlocked breakers to feed the 200A panel and garage from the UPS.
Technically grid tie is available with our power co-op, but notoriously difficult and most all in our area opt to not grid tie to avoid the hassles involved. The local power co-op does not play well with others... so I'll have my own playground. :)
 
I'm running dual 3kw Growatts to power a critcial loads circuit that I installed. Runs about 90% of the 120V in the house. No operational issues. Been online with cycleing between solar and grid twice a day for 3-4 months now.
This sounds and appears (I read through your build thread in your signature) like what I would like to do.
 
Technically grid tie is available with our power co-op, but notoriously difficult and most all in our area opt to not grid tie to avoid the hassles involved. The local power co-op does not play well with others... so I'll have my own playground. :)
Zero-export then.
This is possible with some grid-tie inverters and some all-in-one.

What is fairly trivial is for an all-in-one to have a transfer switch, either feed loads from grid or with inverter from battery/PV.
Some would run off battery until low, then reconnect to grid. Others might just serve as a UPS.

What is more difficult is to synchronize with grid and source power up to the point where power draw from grid drops to zero.
Ideally battery never cycles, just sits at float while your loads use up to 100% of available PV plus any extra needed from grid.
If you can make this work you use most or all of your PV production, assuming a smaller system.

Batteryless grid-tie can also curtail production to avoid export.

Some all-in-one units would support downstream loads and avoid export into their grid connection.
Ideally, you have current transformers at the utility connection by the meter, and a system backfeeds into your house but makes sure it doesn't backfeed the grid. Several product support this including Enphase and Sunny Boy Storage.

Price out AC coupled PV (or batteries optional all-in-one), batteries, and the grid in terms of $/kWh.
I put PV + inverter at $0.05/kWh (over 10 years), batteries $0.05 to $1.00/kWh (over their life, depending on technology and source), and my utility rates are $0.15 to $0.45/kWh (depending on time of day.)
Based on this, if net metering isn't available, "use it or lose it" and wasting PV capacity is sometimes preferable to storing power.
 
I think this may be helpful in understanding my unique electrical setup and needs. We live in a singlewide mobile home and in our area that means the meter gets pole-mounted externally from the home and the home is fed by its own service entrance from the meter base. This makes it complicated to set up like I see many others do where their service entrance is in a garage, basement, or other utility space in their home and they easily put a second panel beside or move circuits. Our home's panel is in our kitchen/laundry hallway and there is no room for equipment in the home.

Here is a fancy-pants drawing of our property and power runs. I ran "mobile home feeder cable" from the meter base to my wife's home office shed, then again down to my workshop. I cannot recall the feeder cable size off the top of my head, but remember putting the 60A breaker at the meter base was "playing it safe" on the amps for the outbuildings.

In a perfect world I could somehow have my inverters and batteries in my workshop building and feed the home's grid. But it's a long distance from my workshop to the meter base for any sensing clamps to know when to not push power to the grid or supplement the home's usage. Probably close to 300 feet once buried and routed as needed.

Nearby goal is to offset the power consumption of my home and to learn more about this world of solar and home generation. I've even wondered about taking the two outbuildings (wife's office & my workshop) "off grid" and letting the feed from the meter base be the grid the inverter senses (if this is possible) to get part of the way there... What would you do?


Our Grid.png
Hi-res image link from Amazon Photos: https://www.amazon.com/photos/shared/mTDOoQHfT5uqxnedmqltww.qcb4QpSLC0tT4JfnIFw8Er
 
Last edited:
Attachment:

"Oops! We ran into some problems.
The requested page could not be found."

Typically grid-tied inverters are rainproof but most other equipment is not.
Mobile home is likely high enough electronics and non-combustible batteries could fit there.
Or a shed by the pole.

Any excessive loads like electric furnace or range you would probably keep on grid side of the inverter.
During power failures it is convenient to be able to power the home. I do that with a main breaker and interlocked branch breaker for backfeeding. Some brands you can have two interlocked branch breakers for two sources.

My SMA Sunny Islands have 56A relay to pass grid through and are 120V. They can be stacked series and/or parallel for 120/240V split phase or more current. They are expensive ($5000 msrp) but have been seen for $2500 and occasionally $1250.

I think some other brand battery inverters are as low as $500. Wide range of price, performance, quality.

Long sense wires probably work OK. I'm just not very familiar with the zero-export systems. I have some current transformers from a watt meter that I just use for testing loads.
If you do trench, put in conduit so you can pull wire for loads you may want to make battery backed-up during power failures.

Zero export ought to be just a grid-tie inverter with a signal. I understand Enphase (microinverters per panel) with Envoy can do that. I think SMA can do it with a similar management interface.

There are also battery inverters systems (Tesla Power Wall, Sunny Boy Storage) that capture power from GT inverters and can prevent export. But those come at a high price, and it isn't clear to me which configurations smoothly curtail production rather than banging the PV inverters off-line and cycling batteries.

See if you can identify a product that provides continuous adjustment of power output and supports optional battery backup for the future.

Since I like SMA, I would build what I've got but prevent it from backfeeding by telling it the grid was a generator. When production is less than consumption it would be connected to grid and Sunny Boy's full output plus current from grid feeds the house. When production exceeds consumption it disconnects from grid and raises frequency above 61 Hz to request reduced power. At night it would connect to grid and keep battery charged. At least I think it would do that. (but no remote current transformer for sensing).
 
Back
Top