Determing solar and backup capacity on a new house

[ericm]

We want to put solar and backup (preferably batteries) in the house we're getting built in Oregon. I can find a ton of info on determining the loads on an existing house you're living in and nothing on how to do it for a house that's not finished yet let alone lived in.

Just in case anyone's interested it'll be all electric. 4.5 tons of mini splits, double oven, induction stove, about 4.5hp of water pumps for the well and water treatment system, plus the usual washer, dryer, fridge and lights. Yes that might take a lot of batteries if we run all of that, that's why I want to spit ball it before I talk to sales people.

 

[BreeStephany]

Are you looking for whole-house backup or just critical load backup (refrigerators, some lighting, furnace, well pump)? You will have to do a general load calc based on the specs / estimated load of your critical loads and then base your system off of that estimate and a basis of estimated general off-grid time as estimated based on your area / utility provider.

It is important to keep in mind that stand-alone solar systems will not run without a grid tie, either through the 'grid' of your batteries or the grid of your utility. Something to keep in mind when talking with sales teams.

 

[TRLKC]

Sent you a DM

 

[ericm]

Are you looking for whole-house backup or just critical load backup (refrigerators, some lighting, furnace, well pump)?

Whole house if it's not too expensive. Some subset if it is.

You will have to do a general load calc based on the specs / estimated load of your critical loads and then base your system off of that estimate and a basis of estimated general off-grid time as estimated based on your area / utility provider.

All I have is the max load, or even worse, recommended breaker, for most of the devices. I can guess that an oven will run kinda close to that, and I can guess how long it might run. But devices like mini splits rarely run at full output and I have no clue what their real draw is under various circumstances.

It is important to keep in mind that stand-alone solar systems will not run without a grid tie, either through the 'grid' of your batteries or the grid of your utility. Something to keep in mind when talking with sales teams.

Yea I know, thanks.

 

[wyliesdiesels]

no natural gas available in your area?

 

[mm08822]

We want to put solar and backup (preferably batteries) in the house we're getting built in Oregon. I can find a ton of info on determining the loads on an existing house you're living in and nothing on how to do it for a house that's not finished yet let alone lived in.

Just in case anyone's interested it'll be all electric. 4.5 tons of mini splits, double oven, induction stove, about 4.5hp of water pumps for the well and water treatment system, plus the usual washer, dryer, fridge and lights. Yes that might take a lot of batteries if we run all of that, that's why I want to spit ball it before I talk to sales people.

Start listing the loads and their namplate values. A diversity factor can be put on each.

Next define the priority for each of your loads as: needed, wanted, don't care, etc.

Not all loads will run at the same time and you could manually control them too.

How long do you expect you need to be on battery back up? What is the history of the area for outages?

Have you considered LP appliances? That could also provide fuel to a gen instead of batts.

What storage space is needed for batts and what is their life and replacement cost?

 

[billconner]

If I understand OP, this is off grid. Correct? My goal also.

 

[ericm]

no natural gas available in your area?

Just propane, and I'd prefer to avoid it. I already wanted an induction cooktop (a gas stove spewing pollutants into the house turns out to be a bad idea) and a heat pump water heater. The builder only does mini splits and I like them anyhow. Electricity there is 1/3 the price than in my part of California while propane is about the same price, making heating with mini splits much cheaper than heating with propane. Oregon's got net metering which makes solar a good deal.

Start listing the loads and their namplate values. A diversity factor can be put on each.

By diversity you mean some estimate of the actual load? My understanding is the nameplate load on a mini split is a lot higher than the actual load. But I suppose I can just make a guess.

Next define the priority for each of your loads as: needed, wanted, don't care, etc.

Not all loads will run at the same time and you could manually control them too.

How long do you expect you need to be on battery back up? What is the history of the area for outages?

Have you considered LP appliances? That could also provide fuel to a gen instead of batts.

What storage space is needed for batts and what is their life and replacement cost?

Yea I can make a stab at that. We have plenty of storage space and costs are going down pretty fast. My neighbor's couple year old battery was $1k/kw and the same batteries are now $0.7/kw.

My backup plan is a propane generator but I'd like to avoid it.

If I understand OP, this is off grid. Correct? My goal also.

No I wasn't clear, it's grid tied.

 

[inphx]

I have been using this vendors batteries in old computer server racks.. but they also have powerwall options.. seems like $230 to $260 per kwh (eg4 signature solar).

 

[pembol]

@ericm we just built a similar house in Colorado - all electric, ground source heat pump, super insulated, triple pain windows, 10kW of solar, etc. Our goal was to be below net zero, including 20K miles of EV driving per year. We have only been in the house 9 months, but we seem to be on track to meeting that goal. We do not have a battery system (yet) as our grid is generally reliable, so for the moment it is much cheaper/easier to use the grid as our battery.

A starting point for estimating your energy is your HERS rating. That should give you an OK estimate of your HVAC energy requirements, which is typically the biggest load. The HERS rating also includes an estimate of appliance and lighting loads, which is also a reasonable starting point. The well pump you will have to figure separately, but if you know the power consumption, and the gallons per hour produced, you should be able to calculate the Watt-hours per gallon and get an estimate of the energy based on an estimate of the water usage.

What is the motivation for the batteries? For power outages, or for load leveling/PoCo arbitrage? How long a power outage are you planning for, and when do they typically occur (summer, winter etc?). How sunny is your area?

In our case it is unusual to have two cloudy days in a row, and almost unheard of to have 3. So a good design criteria for us would be to have enough battery to last 24 hours at 'normal load', albeit without optional things like doing laundry or running the oven. This would allow us to ride out a cloudy day and recharge the next day. We could also add another 48 hours for critical loads (fridge, lights etc) in case we were to get up to 3 cloudy days. For a typical winter day (our highest energy season) we use about 20kWh per day, so a 30kWh bank would probably be about right - this will be different for your specific requirements.

 

[ericm]

The motivation is for backup power. With net metering there's no need for batteries to store power for night time to reduce grid use like people need to do for new installs in California. The power company in Oregon has imported a bad idea from California: EPSS. If there's a problem anywhere in a large segment they take the whole segment down and don't turn it on until they have inspected the lines. They say it's for safety but it also lets the power company reduce maintenance and instead depend on the breakers. This summer I lost count after 10 outages. People there are pissed but the company seems unresponsive. I don't see it getting much better.

We have a lot of outages at our California place. In part that understandable: steep unstable mountains, big trees, lots of rain. We bought a generator when we moved here 27 years ago. But it's gotten way worse since then as PG&E skimps on maintenance. My wife's sick of power outages, generators and cords. And we're getting older. While the new place is getting wired seems like an opportunity to do something better.

One of my neighbors in CA has a solar and battery setup where he can plug in a gas generator and run it for a few hours to top up the batteries during a long outage where there's not enough light. He says that even on overcast rainy days the panels make some power but it's not enough for multiple days.

One of the cool things about Oregon is that I can have the credits for power I return to the grid applied to more than one meter. The property will have three meters- one for the irrigation pump, one for the house and garage and apartment, and one for the shop. So if the panels produced more power than the house used, that can get applied to the other meters. Or I can put panels on the shop and have the power from them credited to the other meters.

 

[dcg9381]

I do some work in this space, mainly for myself, but have done some work as a "solar sub" for a home builder.
IMHO, "by far" the best way to do this is by throwing a monitor on your main. I use Emporia vue. I know your home is not built yet, but I'd even suggest throwing a monitor on your existing main. The power you use in CA is likely very different from what I use in Texas (climate difference). Design your battery storage so it can "easily" be expanded.

I think you're going to run into an "amps" capacity problem before you have a total storage problem, so the point may be moot. 4.5hp is a LOT of surge power... And even the Tesla powerwalls seem to be limited to 5kW or so each.

Some of these solutions are "inverters and batteries" in a single form factor. Others have separate inverters and then storage systems. You can "parallel" both designs, but it's critical that you design for "flexibility" now if you're going to expand later.. It can save you a ton of money.

 

[pembol]

Three meters may make things difficult - won't you need three separate systems to back it all up?

 

[manwithtools]

Three meters may make things difficult - won't you need three separate systems to back it all up?

If I read it right, the meters are not connected, only the billing address is. If he makes more power than he uses at the single meter, he can chose to have the credit for power applied to another of his meters at his address?

 

[ericm]

I do some work in this space, mainly for myself, but have done some work as a "solar sub" for a home builder.
IMHO, "by far" the best way to do this is by throwing a monitor on your main. I use Emporia vue. I know your home is not built yet, but I'd even suggest throwing a monitor on your existing main.

My current house is very different power wise, with propane for heat, water heater and dryer.

The power you use in CA is likely very different from what I use in Texas (climate difference). Design your battery storage so it can "easily" be expanded.

I think you're going to run into an "amps" capacity problem before you have a total storage problem, so the point may be moot. 4.5hp is a LOT of surge power... And even the Tesla powerwalls seem to be limited to 5kW or so each.

It's over three motors. One 3hp well pump. The others are for the RO system and to deliver filtered water from the tanks. They are or will be three phase motors with controllers that match the pump to the flow rate. The pressure tanks on these are small, like 5 gallons or less. So I think there's not the same surge as the same size fixed speed single phase motor would have.

But not having done the estimate yet I suspect you're right that the limitation may be short term power use rather than long term. Though the mini splits will be running a lot especially for outages in the depth of winter or hottest days of summer (which is when the power is likely to go out). At least in winter I can turn the wood stove up. It may be that I set it up now to not run the mini splits with the option (?) of running them later if we get more battery.

The batteries I like (Enphase) have a higher short term output than long term and the system is expandable.

Some of these solutions are "inverters and batteries" in a single form factor. Others have separate inverters and then storage systems. You can "parallel" both designs, but it's critical that you design for "flexibility" now if you're going to expand later.. It can save you a ton of money.

Yes I was planning on that. We could always start with fewer batteries and add as needed (and as the price drops).

 

[dcg9381]

If I read it right, the meters are not connected, only the billing address is. If he makes more power than he uses at the single meter, he can chose to have the credit for power applied to another of his meters at his address?

Probably not.

It's over three motors. One 3hp well pump. The others are for the RO system and to deliver filtered water from the tanks. They are or will be three phase motors with controllers that match the pump to the flow rate. The pressure tanks on these are small, like 5 gallons or less. So I think there's not the same surge as the same size fixed speed single phase motor would have.

This sounds like a bazillion dollar house. No offense intended, I'm just saying it may be above the level of experience "consumer" systems at my payscale.

CA needs a 3hp well pump? Must be on a ridge somewhere.

When we get into 3-phase, battery backup, and solar - that's above my experience level.. At least I've never wired it. 240V 2 phase, no problem.

I don't understand why you'd need another pump for the RO system, unless it's "whole house" RO. Then it's likely a "booster pump" as RO does better at high prsessure. And what I'd say there is that "whole house" RO wastes a ton of water (like all RO). Collect the waste water (it's filtered) and use it for something else.... Irrigation or the pool. It's gonna be a massive amount of water. Like 10k gallons a month for a whole house system.

There are two homes behind me that are "whole house" RO systems. RO is the only way to clean up the high TDS water that comes out of the ground here. I switched to rain water collection and use a single 100 GPD RO system (no separate pump) to provide drinking water to fixtures in our home. All on a 1 gallon pressure tank.

I'm not a huge fan of variable speed deep-water well pumps. They work great until they don't and then they're really expensive and you're tied to a single service company. Give me a Grundfos well pump and a cycle sensor for protection and stop-cycle-valve. (none of this 3-phase, so it won't work for your application)

But not having done the estimate yet I suspect you're right that the limitation may be short term power use rather than long term. Though the mini splits will be running a lot especially for outages in the depth of winter or hottest days of summer (which is when the power is likely to go out). At least in winter I can turn the wood stove up. It may be that I set it up now to not run the mini splits with the option (?) of running them later if we get more battery.

You've got PV too, right? So as long as the sky is clear, you'll have some ability to charge. The best multi-day solution to back-up is a generator. I know CA doesn't like those, but you could put in a 50A generator inlet and have "some" ability to recharge.

When things really **** the bed around here for length of time, we shut down 50% of the house. Our heat pumps don't work worth a darn in the cold, so I dropped propane inside the home and heat from 20-30k BTU portable ductless heaters... They do create humidity, but a 30K unit can basically heat our whole house for a month if the power was out.

The batteries I like (Enphase) have a higher short term output than long term and the system is expandable.
Yes I was planning on that. We could always start with fewer batteries and add as needed (and as the price drops).

I haven't used Enphase, but have heard good things.
Backup gets complicated if you're trying to optimize power bills buy "buying low" and "discharging high" (time of day rates).

I'll be curious to see what you come up with.

 

[manwithtools]

CA needs a 3hp well pump? Must be on a ridge somewhere

Our house in CA had a 900' deep well and a 5HP submersible pump set at 840'. It fed a 10,000 gal storage tank and only turned on when the tank level fell by 500 gallons. This prevented short cycling the pump. The storage tank also fed a hydrant for fire fighting equipment to hook up to. There was a 2HP booster pump that then pressurized the house.

We had 34kwh of Generac battery backup (2 battery system) and 8kw of solar panels. Propane for heat and 4+tons of AC. Our system could not start the 5HP submersible pump or run the AC equipment. We had critical loads such as fridge, freezer, lights, microwave, 2HP booster pump and furnace control wired to two sub panels, each fed by one of the 17kwh batteries. We would have had up to 10k gallons of water available, plenty to last through a power outage. We chose battery backup over a generator to carry us through the proactive power disconnects that SDGE would do during high fire probability conditions. Ideally a generator would have been the next fall back in case of successive cloudy days. Oddly our system was from a generator company and they did not support that third power source.

The Generac system on paper is nice because the inverters are built into the battery panels, so the solar panels feed them directly. Generac bought Pika Energy to acquire the battery technology utilizing Panasonic batteries. Over the the 3 years we had the system, we had at least 4 months of non service from it. I think Generac ended up with a class action law suit over this product. Consequently, I would not recommend that system.

EDIT: I forgot to mention that we were on TOD billing and we used the batteries every day from 4:00 to 9:00 to augment our SDGE incoming supply. When the system worked it was great, when it didn't - not soo much.

 

[wyliesdiesels]

Our house in CA had a 900' deep well and a 5HP submersible pump set at 780'.

holy moly that is a deep well. what city or town was this in?

 

[manwithtools]

holy moly that is a deep well. what city or town was this in?

Ramona, North San Diego County. We were on a bluff 600 feet above the San Pasquale Valley floor, water table was below that. EDIT: I just looked at the invoice again for the well, they had used 840 feet of pipe and 850 feet of cable, so the pump was closer to 840 feet.

EDIT: God, how I miss that house and view

 

[ericm]

I bet the view was spectacular!

The stuff I'm working on is for the house we're getting built in Oregon. It's not a multi bazillion dollar house. The well is over 400' deep. The geology there is really varied. One well might get 25 gpm at 150' deep and next parcel over they might drill 600' and get 2 gpm. Or nothing. Ours ended up deep but at least it flows well.

The 3hp pump is to pump the water up from near the bottom of the well to the house, which is about 100' higher than the well. The higher the lift the more power is needed.

A lot of well pumps are three phase these days. The pump motor costs less and running 3ph means lighter gauge wiring which is a significant savings. In our case it more than pays for the controller. The controller can run the pump at whatever speed is needed to maintain flow, so you don't need a big pressure tank. Instead of a minimum of 40 gallons they're like 4. Also there shouldn't be a big current inrush when the motor starts.

The water treatment system is for excess boron in the water. It's not good to drink and its especially not good for plants. Needing to treat for boron was an unpleasant surprise late in the purchasing process but the story's long and this post is already too long.

There's an irrigation scheme in the area with canals and pipes that bring water down from the mountains and the parcel has access to it (it's a farm). They turn the irrigation off for the season at harvest time.

We'd need an RO system for drinking anyhow but we'll also need to water the garden after the irrigation shuts off, hence the whole house RO system that's big enough to water a garden. Having a lot of farm land to garden/farm on was a draw for my wife, and since I'm getting out of the SF bay area and getting a big shop out of the deal I can't exactly say no. Most of the time it will just be household use as we'll use the water provided by the irrigation scheme. This year there would have been about three weeks of watering some part of the garden.

Since the waste water from the RO system is loaded with boron I can't use it for irrigation. This system's supposed to have a 1:1 or 2:1 rejection rate, I forget which.

The batteries would be for backup only. Unlike California where new solar only pays if you use batteries, Oregon has pretty good net metering and the power co has not forced people into time of day based metering yet.

 

[ericm]

I made a simple model that covers two worse case situations: early fall when it's hot during the day and the irrigation is off so we're pumping and treating water for the garden, and depth of winter.

I got some numbers for mini split power consumption from a web site (https://learnmetrics.com/how-much-electricity-does-a-mini-split-use-watts-kwh-us-dollars/) and used the SEER and HSPF for the units we'll be getting. In this location it cools off at night in summer and warms up during the day in winter so they won't be running full blast 24 hours a day. I arbitrarily took 1/2 of the full out values for a daily kw number for that and because they are slightly oversized for the calculated manual J loads per manual S. I got about 30kw/day for summer and 50 for winter.

For water usage I calculated the power needed to pump water from the well to the house, and the power for the high pressure RO pump. The delivery pump is low flow and low pressure so I can ignore it. It's about 36kw/day to make 200 gallons for household use and three times that to make 600 if we're watering a garden. Numbers are pulled out of my *** for house use (I think we use less as we're both pretty water concious) and based on numbers from the exiting garden in CA where my wife installed a flow meter.

Powering an oven or stove for a couple hours is only 3-5 kw/day each. Fridge, heat pump water heater, TV, lights, computers, etc are about 8 kw/day. The total is 160kw/day for worst case summer (90 when not watering the garden) and 112kw/day for coldest winter.

Oregon lets you assign unused power credits from power returned to the grid to other meters. So I can generate more off the house and give the excess to the shop or irrigation pump meters. Or run solar panels attached to those meters and give the credits to the house, though they won't charge batteries at the house.

 

[cvairwerks]

Ericm: Be safe in your financial calculations and don't include any credits you calculate that you will get. CA got tons of people to put in solar with a very high energy buyback per kWh, and has now since cut it by about 75%. Figure the credits as a bonus, and assume you won't get any.

 

[WisJim]

A very well sealed and insulated house should require minimal heating even in the dead of winter, so if that is done during construction, the heating system is only a minor concern. Based on my experience with a super insulated house in Western Wisconsin, I feel I could build a house that required no central heating system again today a lot easier than the one we build back in the 1970s.
Just because solar has gotten cheaper in recent years doesn't mean you shouldn't design to use the least energy possible.

 

[ericm]

The house is supposed to be well sealed when its done. The builder has warned me that the wood stove will require an outside air kit (which I would do anyhow) and that it will be easy to make the house too hot with the stove. I picked a stove that can be turned down low. They'll do a blower test and the house is supposed to get Earth Advantage certification, which is a sort of LEED light. There will be an HRV system to get fresh air in.

If the house was a different design without cathedral ceilings I could get more insulation than R30 in the ceilings, but it's the design my wife and I picked. And it's got a lot of windows which will not be as efficient as walls but it'd be a shame to not enjoy the views. Some people are happy living in a bunker with no windows (my father was one of those) but I'm a lot happier when I can look out the window at stuff. So it's a compromise for spaciousness and view vs efficiency.

There's a few pics of the view in my shop build thread: https://www.garagejournal.com/forum/threads/i-bought-the-farm.536525/

 

[pembol]

Your usage estimates seem *really* high to me. As I mentioned before, we just built what seems like a similar house, 3600 conditioned sq ft in Colorado at 5500', so probably a bit colder than you. We are all flat roofs/cathedral ceilings (but those are still R66) and lots of windows, the glazing to floor area is around 25%. We do have ground source heat pumps (geothermal) which are a bit more efficient than air source.

Our worst winter usage day (-13F out) for the whole house was about 60 kWh. In the summer we top out around 30 kWh. Typically we are at about 20 - 25kWh. Well designed modern homes are really efficient compared to their predecessors.

Now we don't have our own well pump of whole house RO system, so I don't really know what those draw - but a quick google says RO systems for desalination on boats use around 20 Wh/gallon, so about 4 kWh for 200 gallons.

Are you getting a HERS rating for your house? That will give the answers you need.

 

[manwithtools]

I bet the view was spectacular!

It was!

 

[ericm]

Wow, that's a nice view!


The mini split power usage I came up with may well be high. Keep in mind though that the ground sourced heat pump has a lot less work to do when it's really cold or hot out since the ground temp that it's working with does not vary like the air temp that a mini split sees.

But even so I could be way off. I didn't take the house into account at all, just assumed that in the worst case days the mini splits would run at half of maxium, on average over a full day. That assumption could be way off, or the power consumption data I used could be off.

I have a lot more confidence in the water pumping power use numbers. I calculated the power needed to pump water to a height out of the well or to a pressure for the RO system. And when discussing this with my wife last night she remembered the RO system will have a 2:1 rejection rate so it's even worse than I calculated above. So 3/4 of the power usage for water will be just pumping water from the well.

We're on a well here in CA and it pumps water about 300' from the water level in the well to the tank up the hill. The power usage goes way up when we're watering the lawn and garden in summer.

So far the solar companies I have talked to have not seemed to have any magic way to calculate the use for a brand new house. I'm thinking of putting as many panels on the house as I can (the roof is partially shaded by trees that I want to keep, so that's not a huge number) on the theory that with tariffs and eliminating rebates, solar prices will go way up soon. If I end up with too much for the house the shop and irrigation pump will likely consume the left over credits.

 

[manwithtools]

I'm thinking of putting as many panels on the house as I can (the roof is partially shaded by trees that I want to keep, so that's not a huge number) on the theory that with tariffs and eliminating rebates, solar prices will go way up soon.

With the land you have available, I strongly urge you to go with ground mount. Having been involved in some large scale utility connected solar projects I can tell you that keeping the panels clean has up to a 25% increase in output. Having them on the ground, makes it much easier to keep them clean and eliminates the complexities of roof mounting.

 

[Fav Onefour]

With the land you have available, I strongly urge you to go with ground mount. Having been involved in some large scale utility connected solar projects I can tell you that keeping the panels clean has up to a 25% increase in output. Having them on the ground, makes it much easier to keep them clean and eliminates the complexities of roof mounting.

That is a good point.
Friends live in an area with a high percentage of homes that use solar. The window cleaning ads include info for cleaning the solar panels. It is a part of life if you want them to run efficiently.

I don't have experience with solar at home so I'm going off the experience of friends and family. There are quite a few variations within the group. (I'm adding a little experience with backup systems.)

1. You are lucky to have net metering. Use that to your advantage when planning the size.
2. Prepackaged systems are easy and monitoring setups are slick. Get an idea of what you size want and do some real shopping. There are a lot of high pressure sales teams in our area. We did bids on three different buildings this year. The pricing and setup variables can be overwhelming. It helps to know the variables and lay out parameters for bidding.
3. Dang lucky once again. Doing this from the beginning of build process really allows you to set it up correctly. Dedicate a room or space just for distribution and storage mechanicals. (That stuff might look pretty at first along a garage wall, but you'll get tired of it soon enough). Make the space large enough to move around and work. Storage will get replaced down the road, and smart home switching always seems to need tweaking. Save garage walls for garage stuff.
4. System size. Again, lucky to be on net metering. Look at enough generating capacity to get you through peak time on typical weather day. Wispy clouds change output graphing pretty quick. Big duration events will knock the socks out of your generation and storage. Power failures and shutdowns are not limited to off peak. If you can, do enough capacity so there is no down transition.
5. K.I.S.S. You might know how the system operates. Family and friends probably will not know. If line power goes down, you want the backup to take over without someone getting in there to "switch" the system. I am shocked how much time people spend dinking around online and on the phone with their high tech systems. Transitions should operate seamlessly if you want to enjoy your time away from home. Easy backup testing should be a part of system design.

 

[ericm]

Good points both of you, thanks.

There's a great south facing place for ground mount but it's about 400' from the house. Closer in the slope is east facing.

There's a 20x40 tractor shed that's near the house and garage which both solar guys who have been out to the property spotted. We might get talked into putting the panels there. It's roof is sloped mostly east-west but it's out from the trees that partially shade the house. My wife has been "OMG it costs how much?" on house stuff lately so an extra 25% for ground mount might be a hard sell.

With our current place there's an 11 step process for getting the generator set up. I go over it with my wife periodically but she's never done it. Seamless is the goal for the new one.

One interesting thing is that all three solar/battery companies I have talked to have said that it's normal to run all the loads off the batteries and then you have to go shut stuff off so you don't run the batteries down too fast. I asked if code there requires load sheds if the backup does not have the capacity to run everything, and they said yes but it gets approved anyhow. One guy said the inspectors don't really understand battery backup systems.

Some of the systems proposed have built in shut offs for heavy loads so if we go with one of those we'll have to get that wired correctly after making some decisions about what we want to run during an outage. Like I realized that the well pump and water treatment system don't need to run during an outage, just the pump from the tanks to the faucets. There's 900 gallons of tanks which should last a while for household use.

 

[dcg9381]

Now we don't have our own well pump of whole house RO system, so I don't really know what those draw - but a quick google says RO systems for desalination on boats use around 20 Wh/gallon, so about 4 kWh for 200 gallons.

Whole house RO is just another water pump, typically higher pressure. You can SWAG that if you know what pump they will install.
I'd expect a UV system to be in a home with "whole house" RO also, both would be critical functions.

The RO system will be rated with GPH production, but it's booster pump powered (at least in the OPs case).

One interesting thing is that all three solar/battery companies I have talked to have said that it's normal to run all the loads off the batteries and then you have to go shut stuff off so you don't run the batteries down too fast.

I don't think you're talking to the right person. Modern hybrid inverters can set "power preference" based on a number of things such as battery charge % and time of day. There is no such thing as "run the batteries down too fast" - modern lithium has a ton of capacity and if the inverter can handle the draw, the batteries better darn well be engineered for it.

I asked if code there requires load sheds if the backup does not have the capacity to run everything, and they said yes but it gets approved anyhow. One guy said the inspectors don't really understand battery backup systems.

Load shed modules are for use with generators. They detect low-voltage and under-frequency (low RPM) and will drop out circuits or entire sub-panels.

I've never seen a load shed for an inverter. You only put circuits you can handle on the inverter. If you overload the inverter, it's going to shut down.

Some inverters (again, don't know Tesla) have "grid pass through" capacity with more than one output, which can make things a little simpler.

But the "best" way to design this is to figure out your critical loads, size the right hybrid inverter for them, then add battery capacity to get you the "backup hours" that you want.

I realized that the well pump and water treatment system don't need to run during an outage, just the pump from the tanks to the faucets. There's 900 gallons of tanks which should last a while for household use.

Are those RO water tanks? If not and you've got a whole-house RO system, your water capacity at the house will be about 1/3 of that on RO water due to the way RO works. If it were me, I'd keep the RO waste water, it's highly filtered, just higher TDS. You're going to produce a lot of it.

One of the cool things about Oregon is that I can have the credits for power I return to the grid applied to more than one meter. The property will have three meters- one for the irrigation pump, one for the house and garage and apartment, and one for the shop. So if the panels produced more power than the house used, that can get applied to the other meters. Or I can put panels on the shop and have the power from them credited to the other meters.

That makes things different. If you get 1:1 "net metering" per kWh, there is no need to discharge battery power. You can keep "full reserves" of battery power.

 

[pembol]

One option for load shedding on batteries are span.io panels. You use them like a regular breaker panel, and then you can set it up to drop loads you don't need to back up when the grid is down and your battery is providing power. It would **** to have a the power go out, and then waste half your battery capacity on the irrigation pump because you were out or forgot to turn it off.

 

[ericm]

One option for load shedding on batteries are span.io panels. You use them like a regular breaker panel, and then you can set it up to drop loads you don't need to back up when the grid is down and your battery is providing power. It would **** to have a the power go out, and then waste half your battery capacity on the irrigation pump because you were out or forgot to turn it off.

The irrigation pump is on its own 480v 3 phase service but I get your point. It could be a car charger.

Some of the solar/battery controllers the solar companies I've talked to suggested have a few ports for loads that get switched off when on backup power. They are intended for big loads like car chargers but could possibly be used for entire sub panels. That's part of why I'm trying to nail this down soon, so the electrician can do whatever needs doing to match.

 

[dcg9381]

I'm working on a separate product with a Victron inverter. This is an interesting feature I was unaware of. It allows you to "reserve" battery for household use and still keep reserve capacity for backup:

 

[ericm]

The numbers I'm getting by the methods I described above are whack: something like 45000 kwh/year just for the house. I looked at our current house and the worst month is September when it's hot so we're running A/C, and I'm watering the 1/3 ac lawn a lot since I reseed it then, plus my wife is still watering two gardens. That's like 1300 kwh per month. The current house is slightly larger, built in the '80s and not all that well so the insulation and sealing is much worse than modern houses. We're on a well with no R/O or anything just pumping about 300 ft from the water level to the tank up the hill. While I have not measured it, we use a lot more water than we'd use from the house system at the new place.

Someone probably suggested this a while back but sometimes I'm slow:

Since there's also the ADU and will be a heated/cooled shop at the new place, we'll use more power, but it'll be more efficient buildings and equipment. So I could go with that 1300 kwh/month and just multiply by 12 and get a guess that's a little high but can be used by the other buildings. Thats about 16,000 kwh/year and I can probably hit that with a 12-14kw array depending on how much south facing roof I can use.