Sizing your backup: how to judge risk vs cost tradeoffs

danielh · July 2012

Let's assume that one is building a PV system which is meant to be grid-tied; so you size it's panels to maximize efficient energy production.
While this power maximization is the primary goal, being able to use your fancy system during outages, and as a most-of-house UPS, has appeal. Enough appeal to warrant a few thousand extra bucks, but not more than few thousand. One also learns that a third potential usage -- demand shifting and time of use sale -- is not cost effective.

So one considers "grid-interactive" setups. Wiser heads advice you to start with your energy needs, and then work backwards -- determine the size of your battery bank, then the size of your inverters/chargers/etc, and the number of panels to use. These decisions are somewhat independent of the decisions you made for the primary purpose -- one can always split one's panels into "grid tied" and a "grid-interactive" (or off-grid) sets.

Thus, one should start with an assessment of power needs. But that is not so simple. There is the relatively easy technical details of determining what your loads demand, especially the startup oads (anyone have an idea of what the surge load of a natural gas boiler is?). Then the lifestyle question of just how much do you really need. Or how much are you willing to pay to NOT be without. Then the behavioral questions of household discipline: how much can one police oneself (and your goofy kids) into carefully spacing out usage, say to not run high demanding loads (microwave, toaster oven, hair dryer) at the same time.

Depending on worst case scenarios -- from complacency or bad luck -- the peak demand could be quite high -- perhaps several times ones average demand. In other words, the probability distribution of instantaneous power demand could have skinny tails -- large demands happen, but rarely.

Which leads to my question: how much does one need to worry about an infrequent large demands that stresses the capacities of your system; including transient (< a few seconds) demands?

In particular: when sizing a battery array, what are the dangers of these infrequent large demands. If they shorten the battery life a bit, that may not be a problem worth spending a lot of $$ to cure; since if the number of days the grid is down is low, even a once-a-day-while-on-battery-backup event will not happen a lot over the 5-10 year life expectancy of a battery. But if the consequences are large, say ones household equipment (computers, etc) are damaged, that's a real cost. Alternatively, if the costs are largely annoyance -- system shut downs leading to loss of power and a need to restart equipment -- that is annoying, but may not be worth a lot to prevent. After all, your neighbors and friends may not have power at all, so putting up with a few minutes of their all day suffering really isn't much to complain about. But if the costs are safety -- a fire could happen -- then thats a real issue.

Basically, battreries are expensive, so I am trying to figure out how to minimize the battery pack size. If some inconvenience is a price of minimization, that is acceptable. But jeopardizing safety is not.

Hence this query: how much is enough (in the sense of choosing where you should be on the probablilty distribution of demand), if you don't want the great to be the enemy of the good?

BB. · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

A bag of AA batteries and/or CR123A lithium batteries, a handful of LED flashlights, headlamps, and small radio... Can run for a year or more at night pretty easily.

These lights are great (at least I think they are, I have the two year old version). From very dim (don't trip on stuff in home, read a bit) to very bright (like a small spot light). From about 30 days of run time for 1 set of batteries on very low to ~1-2 hours on very high...

https://www.foursevens.com/product_info.php?products_id=2903 (2x AA batteries)
https://www.foursevens.com/product_info.php?products_id=2911 (2x CR123A batteries)

Got a 7 watt solar panel to recharge the smart phone--If there is cell phone network then will have Internet (but cell phone networks usually don't last very long before their batteries/local gensets go dead).

Do you need a refrigerator/freezer for long term outages?

So, rule of thumb for death:

3 minutes without air
3 hours without shelter (if bad weather/etc.)
3 days without water
3 weeks without food

I live in an urban area--City water/sewer/utility power/etc...

How long would a city allow me to live in my home without power/city services before they get around to red tag (if ever)?

My "end of the world" event would probably be an earthquake (rather than weather). Interestingly, destruction patterns for earthquakes (and explosions) tend to skip around quite a bit. Most likely my home will still be standing. However, if it is not--Having a permanently attached solar system+battery bank does not sound as useful as a small 50 lb genset and 20 gallons of fuel (plus stove that can burn gasoline/other fuels) to bug out.

-Bill

niel · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

unless one takes bill's advice for alternatives, the battery costs are one of our harsher realities to face, but until they come up with something else for a large on demand rechargeable power source, there isn't much choice.

to stop the need for these large demands being placed upon the batteries are the conservation efforts that one would take and it is the only way to reduce the battery size until the minimal required amount is reached for a said inverter. there is a limit though that one is willing to go through with conservation efforts before biting the bullet and just go with what you need. what you truly need is up to you and educating yourself on power consumption is important. for instance, you mention the gas boiler's surge load and if it has a pump on it then it would be the surge created by that pump for burning natural gas does not consume electricity. also note that some boilers do not consume electricity at all like my gas water heater. education is part of the battle and to win it, all in the family must get this education. it's either that or you become the power police, which is more often the case, because most people don't want to learn about electricity, only use it.

danielh · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

Hi Bill.

BB. wrote: »

A bag of AA batteries and/or CR123A lithium batteries, a handful of LED flashlights, headlamps, and small radio... Can run for a year or more at night pretty easily.

These lights are great (at least I think they are, I have the two year old version). From very dim (don't trip on stuff in home, read a bit) to very bright (like a small spot light). From about 30 days of run time for 1 set of batteries on very low to ~1-2 hours on very high...

https://www.foursevens.com/product_info.php?products_id=2903 (2x AA batteries)
https://www.foursevens.com/product_info.php?products_id=2911 (2x CR123A batteries)

Got a 7 watt solar panel to recharge the smart phone--If there is cell phone network then will have Internet (but cell phone networks usually don't last very long before their batteries/local gensets go dead).

Do you need a refrigerator/freezer for long term outages?

So, rule of thumb for death:

3 minutes without air
3 hours without shelter (if bad weather/etc.)
3 days without water
3 weeks without food

I live in an urban area--City water/sewer/utility power/etc...

How long would a city allow me to live in my home without power/city services before they get around to red tag (if ever)?

My "end of the world" event would probably be an earthquake (rather than weather). Interestingly, destruction patterns for earthquakes (and explosions) tend to skip around quite a bit. Most likely my home will still be standing. However, if it is not--Having a permanently attached solar system+battery bank does not sound as useful as a small 50 lb genset and 20 gallons of fuel (plus stove that can burn gasoline/other fuels) to bug out.

-Bill

After going through 3 or 4 multi day outages over the last decade, the basic survival stuff is in hand. So the goal is convenience. For example, in the last ice storm event the wife just happen to have the flu. Misery compounded with 40 degree house temps led us to a hotel suite for 3 nights. Not so cheap (around here), and a pain (especially doing what one can to keep the house above freezing). It's that kind of thing that motivates the consideration of battery backup/ grid-interactivity.

The goal is to balance cost, versus quantity, of more convenience-during-outages (I am ignoring for the moment the most-of-house-UPS non-outage benefits). The bigger the system the more convenient: more stuff can be run, one doesn't have to worry about balancing loads over time, the genset doesn't have to be run so often, etc.
But the bigger the system the greater the cost.

So if a reduction in system size only causes some occasional inconvenience -- such as once a day having to wait a minute or so while an inverter resets itself due to over demand (and please do tell it that is what happens or is more operator intervention required) , or having to replace batteries in 9.5 years rather than 10 years -- and could yield significant cost savings, then it is worthy of serious consideration.

That is the issue I am hoping to get some guidance on.

BTW: interesting that you say 50lb genset; something you can take with you (rather then a 200+ beast that will run the house for a while)

Eric L · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

The goal is to balance cost, versus quantity, of more convenience-during-outages (I am ignoring for the moment the most-of-house-UPS non-outage benefits). The bigger the system the more convenient: more stuff can be run, one doesn't have to worry about balancing loads over time, the genset doesn't have to be run so often, etc.
But the bigger the system the greater the cost.

This may be an obvious point, but when I was planning my latest system, I went to all my local battery distributors. Each time, I would bring quotes from the other distributors (or the web), and ask them if they could beat the price for similar batteries. In the end I got a very low price on a largish (17.5 KWH) battery bank. My point is that you're not really going to know what the cost-benefit sweet spot is for you until you're out there haggling with the local battery distributors, or perhaps e-tailers (although in my case, battery e-tailers weren't even close after shipping costs).

I'm guessing from your original post that you're interested in a hybrid grid-tie system, and that you already know that a back-up generator is much more cost-effective for emergency purposes, but this isn't deterring you. Remember that if you are looking at a grid-tie hybrid, you're going to need a minimum-sized battery bank, determined by the hybrid grid-tie inverter arrangement, which would be your starting point for haggling purposes on batteries.

FWIW, I have yet to have my 4.4 KW inverter reset due to overloads, despite having some fairly heavy surge loads (like a large well pump). Of course I don't do things that would almost certainly shut it down (like having it start the well pump while the air conditioners are running).

danielh · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

HI Eric

Eric L wrote: »

This may be an obvious point, but when I was planning my latest system, I went to all my local battery distributors. Each time, I would bring quotes from the other distributors (or the web), and ask them if they could beat the price for similar batteries. In the end I got a very low price on a largish (17.5 KWH) battery bank. My point is that you're not really going to know what the cost-benefit sweet spot is for you until you're out there haggling with the local battery distributors, or perhaps e-tailers (although in my case, battery e-tailers weren't even close after shipping costs).

I'm guessing from your original post that you're interested in a hybrid grid-tie system, and that you already know that a back-up generator is much more cost-effective for emergency purposes, but this isn't deterring you. Remember that if you are looking at a grid-tie hybrid, you're going to need a minimum-sized battery bank, determined by the hybrid grid-tie inverter arrangement, which would be your starting point for haggling purposes on batteries.

FWIW, I have yet to have my 4.4 KW inverter reset due to overloads, despite having some fairly heavy surge loads (like a large well pump). Of course I don't do things that would almost certainly shut it down (like having it start the well pump while the air conditioners are running).

The plan is to have the bulk of the panels on a separate grid tie inverter (or perhaps microinverters).
A smaller set of panels would be on the hybrid system, feeding critical loads (the largest of which would be a microwave and toaster oven). In addition to emergency purposes, the hybrid inverter would provide most-of-house UPS.

For example, a possibility is 2kw of panels with 200AH of 48v batteries connected to an Outback or Xantrex. The battery backup could run the house for 1/2 day; and on a good sunny day I might be able to get through 24 hours (being frugal, and with batteries providing minimal power at night). For longer outages, or less than ideal conditions, the genset will be run. I have generac 6500 now, but I might want to replace it with a Honda eu2000 (the generac's noise/stink/fuelhoggishness are tolerable, but not enjoyable).

After all this writing, I realize that a question I need help with is:

What are the limits an AGM battery can provide (in terms of percent of AH capacity)?

For example is drawing 3.5kw from a 200AH battery, say for 10 minutes twice a day during outages, acceptable?
Or 5000kw for 5 seconds once a month to handle grid glitches?

BTW: shopping for batteries like shopping for a car .. that could be fun

)

BB. · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

AGM batteries cost around 2-4x the cost of flooded cell batteries. But they can supply huge current surges (upwards of C*4 of capacity--you can drain some AGM batteries flat in 15 minutes--such as for UPS systems--Although, the AGM batteries may only last a few years in such an application).

A typical emergency backup system may run the home at night (quiet time), and use a genset during the day for water pumping/HVAC (heating, ventilation, and air conditioning). Then the batteries can be charged by solar/genset/etc. during the day.

An issue with designing for an off grid capable system is to balance the load against the battery bank against the charging system.

While an AGM battery can supply lots of current, they may have problems with a very large solar array (once the battery bank is fully charged, a "huge" array may over voltage the battery bank with a MPPT charge controller--so the minimum AH rating for AGM vs Flooded Cell lead acid batteries is the same when looking at the charging side).

Remember that batteries usually last somewhere around 5-8 years or so (unless you get forklift or other high end battery). And pure float service (i.e., UPS type operation) is hard on lead acid batteries too (they actually need to be cycled every so often).

And the inverters/charge controllers/etc. only have about a 10+ year life too... So, there can be a lot of maintenance costs with off grid power battery/inverter systems.

And why a genset can be a good alternative for emergency backup power. And if you have natural gas, then you don't have any fuel storage issues.

So, lets look at what a "balanced" design for a 200 AH @ 48 volt system would look like:

8x Trojan T105-RE 6 Volt, 225 AH Deep Cycle Battery $166 each ($1,328 plus shipping). Life of 6-10 years?

To recharge such a bank, we suggest 5% to 13% rate of charge (Trojan recommends 10% minimum):

225 AH * 59 volts charging * 1/0.77 panel+controller derating * 0.05 rate of charge = 862 Watt array minimum
225 AH * 59 volts charging * 1/0.77 panel+controller derating * 0.10 rate of charge = 1,724 Watt array nominal
225 AH * 59 volts charging * 1/0.77 panel+controller derating * 0.13 rate of charge = 2,241 Watt array maximum cost effective

Next, the inverter... Minimum "nicely supported" inverter would be C/8 and the maximum inverter surge output would be around C/2.5:

225 AH * 48 VDC * 1/8 max continuous * 0.85 inverter eff = 1,148 Watts max continuous load
225 AH * 48 VDC * 1/2.5 max surge * 0.85 inverter eff = 3,672 Watts max recommended AC surge

So, somewhere between 1,200 watt and 1,800 watt (assuming inverter supports 2x rated power for surge) would be a nice fit.

To recharge the battery bank with a genset... The rough generator/battery charger requirements would be:

225 AH * 5% = 11 amp @ 48 volts "minimum"
225 AH * 10% = 23 amp @ 48 volts "nominal"
225 AH * 25% = 56 amp @ 48 volts "maximum recommended"

The size genset required (poor PF AC battery charger):

11 Amps * 59 VDC * 1/0.67 PF * 1/0.80 charger eff = 1,211 watt genset at 5% rate of charge
23 Amps * 59 VDC * 1/0.67 PF * 1/0.80 charger eff = 2,532 watt genset at 10% rate of charge
56 Amps * 59 VDC * 1/0.67 PF * 1/0.80 charger eff = 6,164 watt genset at 25% rate of charge

If you had a "good" PFC (power factor corrected) power supply (or inverter/charger), you could reduce the required VA ratings above by *0.67 (assuming near 1.0 PF). To look at picking the "optimum" AC battery charger for a small genset, here is a good technical thread:

Question about battery charger selection with EU2000 generator.

If you used AGM batteries instead, say 4x PVX-2580L 12 volt @ 252 AH for $692 ( $2,768 ), you could use a much larger inverter--But I would suggest following the charging rules as above. (5-8 year life?).

Or, fork lift batteries at 2x Crown Industrial Battery - 24 Volts, 530 Amp-hours $2,850 ($5,700)... But these are 2x the capacity battery bank--so they will require 2x the charging resources and supply 2x the amont of power (and last upwards of 20 years).

Note, I am just guessing on battery life--You will need to confirm with somebody in the solar/battery business their estimate of battery life. Note that all it takes is one "aw heck" moment to full discharge a battery bank and you are left with a lot of scrap lead.

-Bill

icarus · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

I just have to go back to this again! Since th OP is asking about balancing the cost and the risk for what is admittedly "luxury" during outages here is my take. $1000 of Eu 2000 genny, 10 gallons of gas (plus what you might have in your car!) would give you pleanty of run time for 24/7 luxury for most items, including at least furnace heat, single room air, radio/tv etc.

Show me the price for a battery bank with enough capacity to run room air, or furnace and fridge for the duration, and I'm betting it is close to the cost of the genny.

I have to chuckle (and I am not without sympathy to those in this situation) at some of the commentary a out the recent heqt wave in the States. People are commenting that you simply can't live without air conditioning, when the fact is people have been doing so forever. Arabs have lived on the desert for yers, and have learned how to cope. People in the US have only had regular A\C for the last generation or so, and field workers till routinely work outside of A/C. It is all a matter of what we are used to and how we cope.

I for one couldn't live in Florida with temps in the 90s regularly and RH high, but folks mow lawns, build houses, work on roofs ect because they adapt. Just like many of you wouldn't like to live in my environment where -40 is not uncommon. We work outside routinely, but we have adapted.

Sorry for he digression. I just think investing in a bunch of life cycle batteries for a POSSIBLE multiple day outages every NW and again is a good bet. I would buy a genny. (I know the OP already has one, but in general)

Tony

Cariboocoot · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

Expanding on Tony's first paragraph:

I bought new batteries this year. $800+ for 232 Amp hours @ 24 Volts (yes, cheaper in the U.S.: everything is!). That's up to 2500 Watt hours AC and does not include the price of the inverter & other equipment to keep them going.
Cost of Honda EU2000i (which I also have one of): $1,200 (last time I checked). To get 2500 Watt hours out of it you need to run it full speed for a bit more than 1.5 hours. That will use up almost 2 litres of gasoline @ $1.27 per (price here) or $2.54.

Once the batteries are depleted, they're done without some way of recharging them. The generator can be "recharged" for $3 and produce another 2.5kW hours. And again and again. A 20 litre container can run the thing for 15 hours and give you 24kW hours of power. Total cost: $1,225.

1500 Watt inverter: $500. Batteries for 24kW hours: 4000 Amp hours @ 12 Volts about $7,000. Or a smaller basic bank for $300 (about 220 Amp hours, 1200 Watt hours worth, 1/20th the size) and enough panel to recharge it for $600: $1,400. If the sun shines every day you can have 700 Watt hours AC every day for that cost. If the sun shines. The batteries will age regardless of use and need replacing.

You can play numbers many ways. Some things aren't tangible costs, though.

On the whole, we keep using fossil fuels because they offer a lot of concentrated energy in a fairly easy to manage "package" at quite a cheap price.

danielh · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

Thank You Bill... this the kind of info that really helps.

BB. wrote: »

AGM batteries cost around 2-4x the cost of flooded cell batteries. But they can supply huge current surges (upwards of C*4 of capacity--you can drain some AGM batteries flat in 15 minutes--such as for UPS systems--Although, the AGM batteries may only last a few years in such an application).

A typical emergency backup system may run the home at night (quiet time), and use a genset during the day for water pumping/HVAC (heating, ventilation, and air conditioning). Then the batteries can be charged by solar/genset/etc. during the day.

An issue with designing for an off grid capable system is to balance the load against the battery bank against the charging system.

While an AGM battery can supply lots of current, they may have problems with a very large solar array (once the battery bank is fully charged, a "huge" array may over voltage the battery bank with a MPPT charge controller--so the minimum AH rating for AGM vs Flooded Cell lead acid batteries is the same when looking at the charging side).

Remember that batteries usually last somewhere around 5-8 years or so (unless you get forklift or other high end battery). And pure float service (i.e., UPS type operation) is hard on lead acid batteries too (they actually need to be cycled every so often).

And the inverters/charge controllers/etc. only have about a 10+ year life too... So, there can be a lot of maintenance costs with off grid power battery/inverter systems.

And why a genset can be a good alternative for emergency backup power. And if you have natural gas, then you don't have any fuel storage issues.

So, lets look at what a "balanced" design for a 200 AH @ 48 volt system would look like:
8x Trojan T105-RE 6 Volt, 225 AH Deep Cycle Battery $166 each ($1,328 plus shipping). Life of 6-10 years?

To recharge such a bank, we suggest 5% to 13% rate of charge (Trojan recommends 10% minimum):
225 AH * 59 volts charging * 1/0.77 panel+controller derating * 0.05 rate of charge = 862 Watt array minimum

225 AH * 59 volts charging * 1/0.77 panel+controller derating * 0.10 rate of charge = 1,724 Watt array nominal

225 AH * 59 volts charging * 1/0.77 panel+controller derating * 0.13 rate of charge = 2,241 Watt array maximum cost effective

Next, the inverter... Minimum "nicely supported" inverter would be C/8 and the maximum inverter surge output would be around C/2.5:
225 AH * 48 VDC * 1/8 max continuous * 0.85 inverter eff = 1,148 Watts max continuous load

225 AH * 48 VDC * 1/2.5 max surge * 0.85 inverter eff = 3,672 Watts max recommended AC surge

So, somewhere between 1,200 watt and 1,800 watt (assuming inverter supports 2x rated power for surge) would be a nice fit.

To recharge the battery bank with a genset... The rough generator/battery charger requirements would be:
225 AH * 5% = 11 amp @ 48 volts "minimum"

225 AH * 10% = 23 amp @ 48 volts "nominal"

225 AH * 25% = 56 amp @ 48 volts "maximum recommended"

The size genset required (poor PF AC battery charger):
11 Amps * 59 VDC * 1/0.67 PF * 1/0.80 charger eff = 1,211 watt genset at 5% rate of charge

23 Amps * 59 VDC * 1/0.67 PF * 1/0.80 charger eff = 2,532 watt genset at 10% rate of charge

56 Amps * 59 VDC * 1/0.67 PF * 1/0.80 charger eff = 6,164 watt genset at 25% rate of charge

If you had a "good" PFC (power factor corrected) power supply (or inverter/charger), you could reduce the required VA ratings above by *0.67 (assuming near 1.0 PF). To look at picking the "optimum" AC battery charger for a small genset, here is a good technical thread:

Question about battery charger selection with EU2000 generator.

If you used AGM batteries instead, say 4x PVX-2580L 12 volt @ 252 AH for $692 ( $2,768 ), you could use a much larger inverter--But I would suggest following the charging rules as above. (5-8 year life?).

Or, fork lift batteries at 2x Crown Industrial Battery - 24 Volts, 530 Amp-hours $2,850 ($5,700)... But these are 2x the capacity battery bank--so they will require 2x the charging resources and supply 2x the amont of power (and last upwards of 20 years).

Note, I am just guessing on battery life--You will need to confirm with somebody in the solar/battery business their estimate of battery life. Note that all it takes is one "aw heck" moment to full discharge a battery bank and you are left with a lot of scrap lead.

-Bill

This makes sense. Perhaps I am learning (couldn't of done it without this forum).

A couple of questions stated as comments.

First, I will keep in mind that one should aim for a ratio of 100AH to 1kw of PV input, even with AGM batteries.

As for AGM vs FLA -- the better surge capability of the AGMS may be a real benefit -- one could buy a smaller size set but still ride out occasional large draws (say, a +5 kw draw if the boiler and fridge start up while the microwave and toaster oven are going, and I am on the dual monitor computer). Also, I will probably be keeping these batteries in a basement utility area, that has ng hotwater heater and boiler 6 or so feet away. The area is opened along one end to the rest of the basement, so it has air circulation; but for safety sake it seems that AGMs would be better (or should I plan on building some kind of vented-to-the-outdoors compartment for them).

There is also "non-outage" sizing concerns. In normal times, the setup will provide UPS backup to both critical loads (that I want available during outage) AND "electronics to protect". That is, stuff that during an outage I will just turn off and do without. Since I don't want to think about power during non-outages, the system has to be able to provide more than enough power to both kinds of loads for extended periods: say, 6kw for over an hour, perhaps more for short periods. Just to clarify: this is non-outage demand supplied from the grid, as supplemented by my full PV system (which includes panels not connected to the hybrid inverter).

Thus, do I need to oversize the inverter so that its pass through capability is a lot larger than its outage requirements?
More precisely, to size it to UPS protect against transient outages that coincide with periods of high demand (say 6kw)/
Note that if a transient event lasts longer than a few minutes, we will reduce loads -- we will switch to "outage" mode.

Or am I misunderstanding the power rating of these inverters. Perhaps under normal cases, when they are NOT inverting, these ratings are irrelevant -- they can pass through as much power as it is being fed to it from the main cb panel (say, 12kw). Hmm, given that these inverters are also feeding PV power to the "protected" loads (or back to the main panel), somehow I think these ratings ARE relevant.

IOW: I really don't know!

BTW: an(other) ignorant questions: I assume that inverters will NOT overdraw a battery, that it will stop pulling power and shut down when it detects the batteries approaching a target DOD (say, 50%). If so, under what circumstances would you end up with "a lot of scrap lead"?

BTW2: the above "normal times provision" makes me think the Xantrex 6048, with its 6k continuous and 12kw surge, is a better choice than the smaller Outback 3648.

danielh · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

icarus wrote: »

I just have to go back to this again! Since th OP is asking about balancing the cost and the risk for what is admittedly "luxury" during outages here is my take. $1000 of Eu 2000 genny, 10 gallons of gas (plus what you might have in your car!) would give you pleanty of run time for 24/7 luxury for most items, including at least furnace heat, single room air, radio/tv etc.

Show me the price for a battery bank with enough capacity to run room air, or furnace and fridge for the duration, and I'm betting it is close to the cost of the genny.

I have to chuckle (and I am not without sympathy to those in this situation) at some of the commentary a out the recent heqt wave in the States. People are commenting that you simply can't live without air conditioning, when the fact is people have been doing so forever. Arabs have lived on the desert for yers, and have learned how to cope. People in the US have only had regular A\C for the last generation or so, and field workers till routinely work outside of A/C. It is all a matter of what we are used to and how we cope.

I for one couldn't live in Florida with temps in the 90s regularly and RH high, but folks mow lawns, build houses, work on roofs ect because they adapt. Just like many of you wouldn't like to live in my environment where -40 is not uncommon. We work outside routinely, but we have adapted.

Sorry for he digression. I just think investing in a bunch of life cycle batteries for a POSSIBLE multiple day outages every NW and again is a good bet. I would buy a genny. (I know the OP already has one, but in general)

Tony

OP?

Well, although one should not be overly swayed by recent events; the recent outage did help clarify the benefits.

I do wonder how well a 1600kw genset would cover a family's needs; such as cooking with the wave, while the fridge andni tv is one. That isn't an extravagant luxury!

More importantly, the UPS protection is just as big a benefit (if not bigger).

And if one thinks of this as an appendix to a much larger project, and this project may or may not be be a solid investment, then spending <10% on batteryization doesn't seem too frivolous.

At least that is what I am telling myself these days. Perhaps I have become irrationally enamored of the idea.

BB. · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

danielh wrote: »

As for AGM vs FLA -- the better surge capability of the AGMS may be a real benefit -- one could buy a smaller size set but still ride out occasional large draws (say, a +5 kw draw if the boiler and fridge start up while the microwave and toaster oven are going, and I am on the dual monitor computer). Also, I will probably be keeping these batteries in a basement utility area, that has ng hotwater heater and boiler 6 or so feet away. The area is opened along one end to the rest of the basement, so it has air circulation; but for safety sake it seems that AGMs would be better (or should I plan on building some kind of vented-to-the-outdoors compartment for them).

AGMs are nice and "clean". And they don't vent unless they are over charged and/or failing from age... So, I would still suggest making sure they are well vented.

I am not sure what code says about open flames and battery banks--But it is something to consider. If you build a system that can take either AGM or flooded cell--then you will error on the "safe" side.

There is also "non-outage" sizing concerns. In normal times, the setup will provide UPS backup to both critical loads (that I want available during outage) AND "electronics to protect". That is, stuff that during an outage I will just turn off and do without. Since I don't want to think about power during non-outages, the system has to be able to provide more than enough power to both kinds of loads for extended periods: say, 6kw for over an hour, perhaps more for short periods. Just to clarify: this is non-outage demand supplied from the grid, as supplemented by my full PV system (which includes panels not connected to the hybrid inverter).

Now to be clear... Are you talking about 6kW of peak load, or 6kWH of outage protection (say 1kW * 6 hours = 6kWH)?

6kW load is A LOT OF POWER. That is 50 amps of 120 VAC current or ~4x 1,500 watt electric heaters...

The "cheap guy" in me says "...do you really need that kind of power?"

Thus, do I need to oversize the inverter so that its pass through capability is a lot larger than its outage requirements?
More precisely, to size it to UPS protect against transient outages that coincide with periods of high demand (say 6kw)/
Note that if a transient event lasts longer than a few minutes, we will reduce loads -- we will switch to "outage" mode.

It is not unusual to install a relay to prevent two loads from cycling on at the same time. It will cut the maximum power requirement and can save you a lot of money.

Remember a 6kW inverter depending brand/model can source upward of 12kW of surge.

Or am I misunderstanding the power rating of these inverters. Perhaps under normal cases, when they are NOT inverting, these ratings are irrelevant -- they can pass through as much power as it is being fed to it from the main cb panel (say, 12kw). Hmm, given that these inverters are also feeding PV power to the "protected" loads (or back to the main panel), somehow I think these ratings ARE relevant.

Yes, many inverters can pass higher peak current than they can invert.

BTW: an(other) ignorant questions: I assume that inverters will NOT overdraw a battery, that it will stop pulling power and shut down when it detects the batteries approaching a target DOD (say, 50%). If so, under what circumstances would you end up with "a lot of scrap lead"?

There are lots of ways to kill batteries (run plates exposed, over/under charge, let set partially discharged for days/weeks/months will sulftate). Running occasionally below 50% state of charge will not kill a battery, but it can reduce its cycle life (from several thousand cycles to less than a thousand if always cycled below 50%).

The killer is to cycle a battery bank below ~20% state of charge--Below that point is is possible for a "weak" cell to actually go to zero volts and begin to reverse charge. Reverse charging will pretty much kill a cell/battery.

Cutoff voltage of ~11.5 volts (~46 vdc for a 48 volt bank) will help reduce the chance of a below 20% state of charge damage--But a Battery Monitor is really the only accurate way to "estimate" the SOC of an operational lead acid battery bank. And with sealed/AGM batteries, you cannot even use a hydrometer to measure the specific gravity.

BTW2: the above "normal times provision" makes me think the Xantrex 6048, with its 6k continuous and 12kw surge, is a better choice than the smaller Outback 3648.

Look at the battery bank as the "heart" of your system. Oversizing inverters/loads and/or undersizing charging is not doing it any favors. Large inverters are (relatively) cheap--Large battery banks, not so much:

Deep Cycle Battery FAQ
www.batteryfaq.org

-Bill

mikeo · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

I have to chuckle (and I am not without sympathy to those in this situation) at some of the commentary a out the recent heqt wave in the States. People are commenting that you simply can't live without air conditioning, when the fact is people have been doing so forever. Arabs have lived on the desert for yers, and have learned how to cope. People in the US have only had regular A\C for the last generation or so, and field workers till routinely work outside of A/C. It is all a matter of what we are used to and how we cope.

I would like to point out that in the last 2 months in Arkansas we have had 13 heat related deaths due to some cities having 15 days of temps exceeding 100F in a row with the highest temps at 115F. This has happened before but it's uncommon in our climate. Older people just can not adapt in time for an emergency like this. Fortunately our power has held up due mostly to our two nuclear plants that generate about 30% of our energy locally. If there had been a power outage, the death toll would have been much higher. Unless you design your house with survival in mind like I have, it would be very hard to live in these climate extremes with out power for a few weeks.

danielh · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

BB. wrote: »

I am not sure what code says about open flames and battery banks-

That one can't be ignored!

Now to be clear... Are you talking about 6kW of peak load, or 6kWH of outage protection (say 1kW * 6 hours = 6kWH)?
6kW load is A LOT OF POWER. That is 50 amps of 120 VAC current or ~4x 1,500 watt electric heaters...
The "cheap guy" in me says "...do you really need that kind of power?"

Peak load -- for those moments where a lot of the "critical loads" happen to be on (or turning on).
I figure that during outages an average of 1kw an hour is more than enough, and if we want to avoid running the genset (or if it is cloudy), that could be halfed.

That said: in normal times there may be some periods where demand from these "critical loads" is high -- say 5 computers, the microwave, the big screen tv.
Well, that's probably about 3-4kw, but the point is that during normal times I just don't want to think about balancing and offsetting loads.

It is not unusual to install a relay to prevent two loads from cycling on at the same time. It will cut the maximum power requirement and can save you a lot of money.
Remember a 6kW inverter depending brand/model can source upward of 12kW of surge.

Yes, many inverters can pass higher peak current than they can invert.

If the oversized portion of the inverter would be used almost always during normal times (when power is from the grids or non-battery-connected panels), is that a problem?

To repeat, my concern is that the cb subpanel that contains the critical loads (the loads that the inveter powers during outages and are UPS protected) gets all the power it needs the 99% of the time the grid is working fine.

And that is why the idea of a Xantrex 6400 is appealing. Now, if the Outback 3648 could "passthrough" +6 kw of grid power for several minutes at a time, it might be more appealing (its a little cheaper, etc.). I currently assume that is not the case -- that the rated output (continuous and surge) always apply.

There are lots of ways to kill batteries (run plates exposed, over/under charge, let set partially discharged for days/weeks/months will sulftate). Running occasionally below 50% state of charge will not kill a battery, but it can reduce its cycle life (from several thousand cycles to less than a thousand if always cycled below 50%).

The killer is to cycle a battery bank below ~20% state of charge--Below that point is is possible for a "weak" cell to actually go to zero volts and begin to reverse charge. Reverse charging will pretty much kill a cell/battery.

Cutoff voltage of ~11.5 volts (~46 vdc for a 48 volt bank) will help reduce the chance of a below 20% state of charge damage--But a Battery Monitor is really the only accurate way to "estimate" the SOC of an operational lead acid battery bank. And with sealed/AGM batteries, you cannot even use a hydrometer to measure the specific gravity.

Look at the battery bank as the "heart" of your system. Oversizing inverters/loads and/or undersizing charging is not doing it any favors. Large inverters are (relatively) cheap--Large battery banks, not so much:

Deep Cycle Battery FAQ
www.batteryfaq.org

I read that people like their battery monitors. But that leaves the question of what to do if you leave the house unattended, say everyone goes to work for 8 hours. If the batteries get too low,
I would much rather the house be a little cold when I get home (or the fridge a little warm), then to ruin my batteries. Would these battery monitors help with that -- can they inform the inverter to stop drawing power. Or do they just provide accurate information to the human eye.

BTW: That's one advantage of a battery backup over just genset: they aren't as likely to walk away. Or perhaps you guys don't worry about loud gensets attracting unwanted attention?

Cariboocoot · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

danielh wrote: »

OP?

Original Poster. A forum acronym.

I do wonder how well a 1600kw genset would cover a family's needs; such as cooking with the wave, while the fridge andni tv is one. That isn't an extravagant luxury!

I think you mean 1600 Watt genset. 1600 kilowatts would power a small town.

It might surprise you to know that a fairly large segment of the world's population lives without refrigerators, televisions, and microwave ovens. More likely it would surprise your children to know this.

The cabin here is far out in the woods. Before the solar install there was no electricity save running a big generator. Refrigeration was by propane, as were lights. No microwave, no television (still none), none of the "must haves" so many people think they can't live without. Before we bought it a family lived here year-'round, accessed by lake only (frozen in Winter). The improvements I've made certainly make life easier and more enjoyable, but they are hardly necessary.

There is just a small chance that the recent wide-spread blackouts will clue a few people in to just how much they take for granted. Maybe some of them will prepare for it. I don't mean the "doomsday scenario" which is highly unlikely to happen, but the inevitable next time the lights go out that absolutely will happen.

But too many people wander through the day blissfully unaware or purposefully ignoring the actuality that things do go wrong from time to time.

One thing we off-gridders have learned: when you have this much power, you tend to use that much more.

icarus · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

Wow,

Critical loads of 5 computers, wide screen and Micro?

Cooking can be accomplished in a variety of ways in an emergency situation. Barbeque, camp stove (I think ever gone should have a camp stove and a bunch of fuel stockpiled for just such emergencies!) the big issue is to be able to store water for a long term if city water goes out, or well power is not available. That said most houses have a pretty good stock pile of water if you are clever about getting it. hot water tank for example holds 40-60 gallons, toilet tanks, pressure tanks etc all hold water even if there is no pressure to get it out.

I guess it comes down to a matter of what one considers a "critical" load.

As a side note to mikeo, I am not m asking light of the situation in the efctd parts of the continent. What I am saying is that A/C has allowed explosive growth in climates where people generally didn't live in previous generations in large numbers. Those that did live there had specific adaptations to the climate, like building design, clothing design, and indeed work specific designs for labor etc.

Nowadays, we have generic buildings and people who leave in a hermetically sealed environment,, and when the lights go out,, they are not inured to the heat and humidity.

Tony

ggunn · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

icarus wrote: »

Wow,

Critical loads of 5 computers, wide screen and Micro?

Cooking can be accomplished in a variety of ways in an emergency situation. Barbeque, camp stove (I think ever gone should have a camp stove and a bunch of fuel stockpiled for just such emergencies!) the big issue is to be able to store water for a long term if city water goes out, or well power is not available. That said most houses have a pretty good stock pile of water if you are clever about getting it. hot water tank for example holds 40-60 gallons, toilet tanks, pressure tanks etc all hold water even if there is no pressure to get it out.

I am from south Louisiana; a standard part of preparing for a hurricane strike is to fill all the bathtubs in the house with water. Of course, hurricanes often spawn tornadoes, and they say that if a tornado comes, a bathtub is a good place to take refuge. It would be a bummer to drown in your bathtub.

BB. · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

For me, 5x computers would be laptop which use around 20-30 watts--So a 150 watts of computer, another few tens of watts for cable modem and router. Only turn on the laser printer when printing (even do that when we have power).

It is a tough set of questions to answer. Yes the XW6048 is a really nice unit. And you need 600+ AH worth of 48 volt battery bank to feed it (assuming 6kW of load and 6kW of solar panels).

Around here, we have had entire multi-kWatt arrays walk away from the schools/buildings they were mounted on. Solar arrays are not exactly "stealth".

Two things to "protect the batteries". Say the low voltage limit to 46 volts (plus a couple of minutes delay to allow for starting surges) and there are battery monitors (Xantrex/Schneider and Victron) that do have a programmable external alarm contact which you may be able to connect to your inverter/load relay if things go south (battery monitors are typically more accurate than a voltage trip alarm).

I wonder if the Outback Flexmate battery monitor subsystem can be programmed to do similar?

-Bill

danielh · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

icarus wrote: »

Wow,

Critical loads of 5 computers, wide screen and Micro?

Cooking can be accomplished in a variety of ways in an emergency situation. Barbeque, camp stove (I think ever gone should have a camp stove and a bunch of fuel stockpiled for just such emergencies!) the big issue is to be able to store water for a long term if city water goes out, or well power is not available. That said most houses have a pretty good stock pile of water if you are clever about getting it. hot water tank for example holds 40-60 gallons, toilet tanks, pressure tanks etc all hold water even if there is no pressure to get it out.

I guess it comes down to a matter of what one considers a "critical" load.

I am defining "critical loads as "loads that I want to have available in an outage"
and "UPS loads" as "loads I want to have protected against transient events"
In particular, the 5 computers and wide screen are in the "UPS loads" category. During any kind of outage (that lasts more than 10 minutes), they would be turned off.
The microwave is in between; it is a convenient cooking option (a lot less power needed than the electric range). We also have the grille, and cook stoves, if we want to cook
something more elaborate.

I want the inverter to provide both functions. So it should be able to deliver enough power to deal with transient events, but it doesn't have to deliver that kind of power
over an extended period of time.

That's why I was hoping the Xantrex would do the job. BUT, Dusty warns me that 200 AH and 2kw of panels and 6kw of Xantrex inverter is likely to stress your batteries,
or make you choose to disable selling excess power (from these panels) to the grid.

Ugh, this is complicated

danielh · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

BB. wrote: »

For me, 5x computers would be laptop which use around 20-30 watts--So a 150 watts of computer, another few tens of watts for cable modem and router. Only turn on the laser printer when printing (even do that when we have power).

It is a tough set of questions to answer. Yes the XW6048 is a really nice unit. And you need 600+ AH worth of 48 volt battery bank to feed it (assuming 6kW of load and 6kW of solar panels).

Around here, we have had entire multi-kWatt arrays walk away from the schools/buildings they were mounted on. Solar arrays are not exactly "stealth".

Two things to "protect the batteries". Say the low voltage limit to 46 volts (plus a couple of minutes delay to allow for starting surges) and there are battery monitors (Xantrex/Schneider and Victron) that do have a programmable external alarm contact which you may be able to connect to your inverter/load relay if things go south (battery monitors are typically more accurate than a voltage trip alarm).

I wonder if the Outback Flexmate battery monitor subsystem can be programmed to do similar?

-Bill

Sigh, is it true that matching a 6kw xantrex with 200AH of battery and 2kw of panels is not so good -- that it will lead to unnecessary cycling of the battery.

That it won't be straightforward to achieve the goal of keep the batteries ready for use as a UPS and for use during an outage backup (but to otherwise not use them), to
use available PV and the grid for local loads (when the grid is up), and sell excess PV to the grid?

Is this a setup that the Outback does a better job at providing?

Cariboocoot · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

Even in a purely off grid situation there is no sense in having a 6kW inverter with only 200 Amp hours of battery. If you ever came close to the inverter's total output capacity the battery bank would fail very quickly: 6000 Watts @ 48 Volts is 125 Amps. That would be like trying to suck half the battery capacity in an instant (the higher the discharge rate, the lower the total actual battery capacity).

And yes, it is complicated.

westbranch · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

danielh wrote: »

choose to disable selling excess power (from these panels) to the grid.

Ugh, this is complicated

I don't comprehend how you would be 'selling' during an outage as GT inverters need incoming power before they can sell...???

Oh, I see you corrected that in the second post...

Cariboocoot · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

westbranch wrote: »

I don't comprehend how you would be 'selling' during an outage as GT inverters need incoming power before they can sell...???

Not during an outage: when the grid is up. The XW with a tiny battery bank has fits trying to Sell to Grid because there's not enough buffering.

What I'm confused about is needing UPS for laptop computers. :roll:

mtdoc · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

danielh wrote: »

I am defining "critical loads as "loads that I want to have available in an outage"
and "UPS loads" as "loads I want to have protected against transient events"
In particular, the 5 computers and wide screen are in the "UPS loads" category. During any kind of outage (that lasts more than 10 minutes), they would be turned off.
The microwave is in between; it is a convenient cooking option (a lot less power needed than the electric range). We also have the grille, and cook stoves, if we want to cook
something more elaborate.

I want the inverter to provide both functions. So it should be able to deliver enough power to deal with transient events, but it doesn't have to deliver that kind of power
over an extended period of time.

That's why I was hoping the Xantrex would do the job. BUT, Dusty warns me that 200 AH and 2kw of panels and 6kw of Xantrex inverter is likely to stress your batteries,
or make you choose to disable selling excess power (from these panels) to the grid.

Ugh, this is complicated

Based on my experience, the Outack GVFX 3648 would handle this job fine. Unless you're running supercomputers and microwaving at 5KW at the same time, I would think it's 6000 watt surge capacity would handle this. It handles my baseline loads for a 3000 sf modern home (200-300 watts) and occasional very high surges -e.g. toaster oven on when 1 hp septic pump - (with 30 amp inrush current) kicks in and runs for 15 seconds.

Worst case I suspect is if you happened to have almost all of your "critical and UPS loads" on while simultaneously running microwave on high when the grid power goes out- you might trip a 60 amp breaker or fault the inverter and lose power briefly - but really what are the odds that the grid power goes down at exactly that time.

No doubt the XW6048 would handle even that scenario.

You really need to measure your "critical and UPS loads" and have a well defined power need to know for sure which inverter would cover you.

The battery needs are another issue. The AH capacity will be defined by the "non UPS" loads you need to run during an extended outage. Once you know how much battery you need for that purpose - you can limit the number of panels you devote to this purpose to avoid the AC ripple issues (using the 100AH per KW of PV rule of thumb). i.e. it's not really the inverter size that determines the size of battery bank you need - but the size of battery bank you need to maintain your critical loads will determine the number of panels you devote to this back-up function. Your situation is a bit different than most in that you have plenty of panels to start with and are working backwards from there. 8)

BB. · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

There alternative configurations.

You can do a XW6048 inverter + 600 AH battery bank and no solar panels. Let the AC mains/generator keep the batteries up (and add solar panels later as funds/needs permit).

In theory, you can have your 6kW XW + 600 AH battery bank and up to ~6kW of solar panels and GT inverter.

Put the GT inverter on the protected side of the XW. Normally, XW operates as a battery charger and passes power from the grid to the protected panel, and the GT inverter will also feed the local loads and push any excess power out to the grid (and turn your meter backwards).

During a power failure, the XW will switch to off grid mode (and switch off from grid supply). At that point the XW will supply power or absorb power form the GT inverters. When the battery bank is full, the XW will shift frequency from 60 Hz to 62 Hz which will "knock" the GT inverter off line (or micro GT inverters) for 5 minutes, drop frequency back to 60Hz for GT inverter(s) to restart... Repeat cycle as battery state of charge dictates.

And still works with a backup genset (more than likely, the GT inverter will not turn on when AC generator is operating--Standard generators cannot hold 60Hz +/- 0.5 Hz (inverter generators may sync--which raises a whole more questions about maximum AC voltage, back feeding inverter/generator, etc.--much of it potentially not good).

Anyway, when I did some sample cost of power before (capital costs+solar array+electronics with 10 year life, battery bank with 8 year life, 20 year life on panels+wiring), I got a cost of around $0.45 per kWH, even with Net Metering and grid power--So you are not saving money (but, as always, I suggest you do the design/math yourself for your needs--my numbers are just "ball" park to see what your threshold of pain is).

If you have a few computers and only need short term (15 minutes of power) backup--You may be better off with standard computer UPS systems... Overall, less cost and less energy losses. UPS systems do have their own maintenance issues (small UPS for desktop systems may last 2 years before replacement--usually not even worth replacing the battery).

-Bill

danielh · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

mtdoc wrote: »

Based on my experience, the Outack GVFX 3648 would handle this job fine. Unless you're running supercomputers and microwaving at 5KW at the same time, I would think it's 6000 watt surge capacity would handle this. It handles my baseline loads for a 3000 sf modern home (200-300 watts) and occasional very high surges -e.g. toaster oven on when 1 hp septic pump - (with 30 amp inrush current) kicks in and runs for 15 seconds.

Worst case I suspect is if you happened to have almost all of your "critical and UPS loads" on while simultaneously running microwave on high when the grid power goes out- you might trip a 60 amp breaker or fault the inverter and lose power briefly - but really what are the odds that the grid power goes down at exactly that time.

No doubt the XW6048 would handle even that scenario.

You really need to measure your "critical and UPS loads" and have a well defined power need to know for sure which inverter would cover you.

The battery needs are another issue. The AH capacity will be defined by the "non UPS" loads you need to run during an extended outage. Once you know how much battery you need for that purpose - you can limit the number of panels you devote to this purpose to avoid the AC ripple issues (using the 100AH per KW of PV rule of thumb). i.e. it's not really the inverter size that determines the size of battery bank you need - but the size of battery bank you need to maintain your critical loads will determine the number of panels you devote to this back-up function. Your situation is a bit different than most in that you have plenty of panels to start with and are working backwards from there. 8)

I think I have a handle on the "non ups" loads; it is similar to what Bill sketched out earlier. Basically, 200AH of battery fed by a 2kw of pv panel, with 6500w generator as a supplement. Or, if I am careful and/or willing to be more frugal, a honda 2000eu as the generator.

The question is then how to provide UPS to a larger set of loads, perhaps (on occassion) a lot larger. As you succinctly point out, should the UPS be needed (a transient grid event) during one of these much larger load occasions, a breaker might be tripped. Oh well, that's a rare event with minor consequences, and can be easily lived with.

However, there is a concernt: will this strategy, where battery capacity is undersized relative to inverter size, lead to rippling and other battery stress.
I might be misunderstanding, but it seems that this is likely to be the case with the Xantrex, at least if you want to sell excess PV production to the grid.
And to your knowledge, the Outback does not suffer from these same dilemnas

danielh · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

BB. wrote: »

There alternative configurations.

Put the GT inverter on the protected side of the XW. Normally, XW operates as a battery charger and passes power from the grid to the protected panel, and the GT inverter will also feed the local loads and push any excess power out to the grid (and turn your meter backwards).

During a power failure, the XW will switch to off grid mode (and switch off from grid supply). At that point the XW will supply power or absorb power form the GT inverters. When the battery bank is full, the XW will shift frequency from 60 Hz to 62 Hz which will "knock" the GT inverter off line (or micro GT inverters) for 5 minutes, drop frequency back to 60Hz for GT inverter(s) to restart... Repeat cycle as battery state of charge dictates.

And still works with a backup genset (more than likely, the GT inverter will not turn on when AC generator is operating--Standard generators cannot hold 60Hz +/- 0.5 Hz (inverter generators may sync--which raises a whole more questions about maximum AC voltage, back feeding inverter/generator, etc.--much of it potentially not good).

This sounds like AC coupling? Is the xantrex frequency switching dependable enough (or do you still need load diverters).

Anyway, when I did some sample cost of power before (capital costs+solar array+electronics with 10 year life, battery bank with 8 year life, 20 year life on panels+wiring), I got a cost of around $0.45 per kWH, even with Net Metering and grid power--So you are not saving money (but, as always, I suggest you do the design/math yourself for your needs--my numbers are just "ball" park to see what your threshold of pain is).

If you have a few computers and only need short term (15 minutes of power) backup--You may be better off with standard computer UPS systems... Overall, less cost and less energy losses. UPS systems do have their own maintenance issues (small UPS for desktop systems may last 2 years before replacement--usually not even worth replacing the battery).

Thats what I have now... and what I am hoping to improve upon: avoiding replacing the desktop ups, getting a longer run time if needed, and having a broader arrange of equipment covered).

The question in my mind is: consider a battery backed inverter capable of providing more than 1kw to 100AH of battery capacity. Is that an invitation to trouble, even if the times when this excessive output (say, > then 2kw of output with a 200AH batterY) will almost always be when the grid is up (and the batteries aren't needed).

Cariboocoot · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

I don't think you're getting the hang of this.

Assuming that battery bank is 200 Amp hours @ 48 Volts you won't need 2kW of array to recharge it. 1200 Watts is more likely. Putting too large of an array on a battery system is a waste of money.

An inverter will put out what it is capable of and no more, and even then only if there is enough battery to supply the power demands. If you have more than 3500 Watts (actually Volt Amps) of stuff on a 3500 VA inverter it will fault. If it's less than that and there's not enough battery to supply the power the input Voltage will sag and the inverter will shut down.

The trick is to size the inverter and battery bank to accommodate the loads. You can't get away with scrimping on either and expect it to work. It does not matter which inverter you choose.

The same thing goes for hybrid GT: not enough battery, it doesn't work. No matter which inverter you choose.

The same thing goes for AC coupling to a battery-based inverter: too small a battery bank (even with dump loads) and you run the risk of failure.

For all these applications you need to be able to store a certain amount of power. In the case of the hybrid GT it is "short term storage" as the batteries act as filters but must be large enough to do the job. In the case of emergency power supply it is "long term storage" as they must be able to handle the loads until power returns.

Your trying to do it without batteries, so to speak, and that does not work.

mtdoc · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

danielh wrote: »

However, there is a concernt: will this strategy, where battery capacity is undersized relative to inverter size, lead to rippling and other battery stress.
I might be misunderstanding, but it seems that this is likely to be the case with the Xantrex, at least if you want to sell excess PV production to the grid.
And to your knowledge, the Outback does not suffer from these same dilemnas

Ah, I see. BB or Cariboocoot would be better to answer that question, but my admittedly limited understanding of AC ripple concerns is that it's not a rare brief surge of inverter output beyond the 100 AH/1KW panel rule that is a concern it's more the regular "too high" inverting/selling to the grid that would come with too many panels relative to battery bank size that causes problems. I could have that all wrong though and will let the experts answer.

BB. · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

I am going to punt on the XW answer... The hardware appears to do what most people need. The support and firmware updating (if needed) tend to be a bit iffy at the moment (from what I have read here).

And, while peak/surge power is important, most of the time, what drives the design of the system is average power * hours of operation per day. Running a 1,500 watt inverter 20 minutes per day is usually much less of an issue (overall) vs running a 250 Watt desk top system 12 hours per day:

1,500 watts * 20 min/60 min per hour = 500 Watt*Hours = 0.5 kWH per day
250 watts * 12 hours = 3,000 WH = 3 kWH per day

And on and on...

The rules of thumb for battery/load sizing (100 AH @ 48 volts for 1kW of load/charging) is one rule of thumb.

Obviously, there is the important rule for Watt*Hours of load per day have to be replaced the next time the sun rises. So you need to meet both requirements to have a satisfactorily operating system. If you have a low peak power running many hours per day--The size of the array will probably be defined by hours of sun per day to recharge the battery bank.

If you have a low average loads but a lot of heavy power surges (pumping, cooking, etc.), then you may be forced to have a large battery bank to support the surge current--and a larger solar array to properly recharge the battery bank.

For example, a 3.3 kWH per day (~100 kWH per month) system is about right for a full off grid home, lots of conservation, yet still have an "electric home" experience (refrigerator, well pump, washer, lights, TV/computer, etc.).

Using PV Watts for Sterling Virginia, fixed array tilted up 39 degrees from horizontal, we get hours of sun per day by month:

Month    Solar Radiation (kWh/m 2/day)
1      3.59     
2      4.28     
3      4.80     
4      5.34     
5      5.32     
6      5.66     
7      5.46     
8      5.38     
9      5.07     
10      4.72     
11      3.56     
12      3.03     
Year      4.68

Lets say run without a genset for 9 month of the year... 4.28 hours of sun minimum for February:

3,300 WH per day * 1/0.52 * 1/0.77 panel+controller derating * 1/4.28 hours of sun per day = 1,925 Watt Solar Array Minimum

That is one simple array calculation and assumes that you will have to run a genset during bad/cloudy weather in the winter to make up for poor production days.

Next, sizing the panel based on the battery bank... Basically, we suggest 1-3 days of "no sun" and 50% maximum discharge. Use 2 days as a "nominal" setup:

3,300 WH per day * 1/0.85 inverter eff * 2 days no sun * 1/0.50 max discharge * 1/48 volt battery bank = 324 AH battery bank @ 48 volts

Now, assuming a C/8 typical maximum continuous discharge, the AC inverter rating would be:

324 AH * 48 volts * C/8 discharge rate * 0.85 inverter efficiency = 1,652 watt rated inverter "nominal max continuous power"

And assuming a flooded cell battery bank that can support C/2.5 rate of discharge:

324 AH * 48 volts * C/2.5 discharge rate * 0.85 inverter efficiency = 5,288 watt rated inverter (max surge for flooded cell)

Since many "good" inverters support surge at 2x of rated power (Watts or Volt*Amps -- VA):

5,288 Watt surge / 2 surge rating = 2,644 Watt max inverter rating based on battery surge capabilities

Now we need to check the Rate of Charge that we would suggest for such a battery bank--5% to 13% rate of charge:

324 AH * 59 volts charging * 1/0.77 panel+charge controller derating * 0.05 rate of charge = 1,241 Watt Array "minimum"
324 AH * 59 volts charging * 1/0.77 panel+charge controller derating * 0.10 rate of charge = 2,483 Watt Array "healthy nominal"
324 AH * 59 volts charging * 1/0.77 panel+charge controller derating * 0.13 rate of charge = 3,227 Watt Array "max cost effective array" (my humble opinion)

So, combined with the minimum array to supply 3.3kWH for 9 months of the year of 1,925 Watt Solar Array, and the above 1,200 to 3,200 watt array based on battery capacity... The recommend range for "your system" would be around 1,925 watt to 3,227 watt solar array...

Notice the minimum array of 1,925 Watt and the "healthy" nominal array based on battery size is 2,483 watts--That is a "balanced" system... If you go to 1 day of "no sun" and can live with less surge current--Then the array size based on battery charging goes down... However, if you need 3 days of "no sun" and/or higher surge current then you need a larger array to properly recharge the battery bank...

Anyway--I hope the above makes some sense... You really need to be "iron fisted" on your loads--both in what you will want to power and how many hours per day (vs available sunlight, generator fuel, etc.). Solar PV power (true off grid capable) is not a "fire and forget" type system. You will need to monitor it closely when first installed. And you will need to train others (spouse, kids, guests) on the basics (and how not to leave things on when nobody is around) to keep from "killing/murdering" the battery bank.

-Bill

PS: The above is just my opinion and based on feedback I have read here from others who actually live off grid... I am on grid with GT power--So my daily power is pretty simple. The utility is my main power, and the GT inverter helps keep the bill low ($6 per month, with excess balance at the end of the year--to allow room for growth later).

Also I carry out to 3/4 places for math so you can reproduce (and catch any errors). In reality, if you are within 10-20% of the numbers, that is "good enough" for solar calculations (10-20% variation in solar energy averages is normal--weather patterns play a large role in day to day operation).

And, this is a starting point... Suggestion/questions/changes are always good for another few posts.

Cariboocoot · July 2012

Re: Sizing your backup: how to judge risk vs cost tradeoffs

Amen to that!
Why just last night at 11:00 PM I was explaining to my wife that staying up all night running the laptop and satellite set-up was like running a second refrigerator and the system was only designed for one.
Maybe I'll crank the LVD up to 24.5. :roll:

Sizing your backup: how to judge risk vs cost tradeoffs

Comments

Categories