Special Pages

Wednesday, February 22, 2012

A Solar Cell Backup Power System - 2

The purpose behind yesterday's post was to start thinking about how big the problem is and how to solve it.  In engineering and the sciences it's called finding the boundary conditions or "bounding the problem".  We set a high end size and cost for a system by doing some simple calculations.  The next level is to look at the problem in terms of, as the Rolling Stones famously said, "You can't always get what you want, but if try sometimes, you just might find, you get what you need". 

Costs are dropping on these systems, and like computers, the best approach is always to get the least you can get by with now, and add on to it with cheaper upgrades as time goes by.  The prices you actually pay on the market are usually set less than the real costs by tax incentives and rebates that the fed.gov and your local utility give.  These, of course, are real costs - you're just getting all of the taxpayers in the country and the customers of that utility to pick up parts of the bill.  There is a type of system called a "grid tie", where you sell electricity back to the power company - who is required by law to buy it from you.  I'm not going to emphasize that because I'm talking about a SHTF power supply.  There just might not be anyone left to sell to. 

Among the approximations built into yesterday's analysis was that you actually use the full 1800 Watts 24/7 from your 120V 15A outlets, and I don't have to tell you that you don't.  Your electric utility doesn't sell you power (Watts), it sells you energy (Watt-hours, and our bills are expressed in kilowatt hours; thousands of W-hr).  The system that we outlined yesterday was actually a 3600 Watt-hour system, and it's probably best to start thinking in terms of Watt-hours, understanding that 250W solar panels are specified to deliver that all the time, as long as the sun is bright and sky is clear.  Over the course of a month, should you really run it at full output, that would be 3600 Watt-hours, 24 hours a day for 30 days.  That's 2592 kW hrs (kWh). 

One reason to start thinking this way is that it makes choosing batteries easier.  Since a watt is the product of voltage and current, 120V at 15 amps equals 12V at 150A.  Batteries are sold by the amp-hr, and 150 A-hr is not an extremely big battery. 

Enter the world of planning.  My house is an 1980-era house and the electric panel has never been upgraded.  We have about 20 kW service here.  That essentially means 20kW every hour, or  480kWh a day, and 14,400 kWh in a month.  My electric bill just came, and it said we used around 1/10th of that last month.  The important point is: you may need some extra capacity to run other things, and you may need extra capacity for starting some motors, but you don't need to build out the full fuse panel in your house.  That first number I threw around (for a 20 kW house - that 14,400 kWh system at over $100,000 -based on the last time I looked at prices, a few years ago) is not the way to approach this question.  The system we looked at yesterday, instead of being one circuit in my house, would provide more power than I bought last month. 

The way to calculate how much capacity you need is to list the loads you need to run and determine how many watt hours are there.   People that live on small boats or live in trailers are good at this.  If you are going to use electricity to cook, you don't need the thousands of watts the stove uses all the time, but you need to bank those watt hours in batteries somewhere.  Better yet, cook on gas, or wood, or a microwave if you must use electricity.  If you're going to run some nights at light, you need to bank those watt hours into the battery during the day. 

This is probably a good time to introduce the first law, or golden rule or whatever you want to call the most important thing to remember:
The cheapest watt in a solar power system is the one you don't need.  Conservation is cheaper than solar panels.  
Admittedly, you can go a little nuts this way, but if you want to run a lot of lights, for example, LEDs are expensive but extremely low energy consumption compared to incandescent bulbs. 

This is boring work, looking at the power consumption on everything and entering it into a spreadsheet, but probably essential.  An example for the communications backup might look like this:
 
Load       Amps @ 12.5V           hours/day      Needed A-Hr
HF Radio - RX 1.3 6 7.8
TX at 50 W 10 0.2 2
TX 100W 20 0.05 1


This tells me that I only need about 11 A-hrs to run my Emergency Communications station.  And I could get by with less. 

More to come...

10 comments:

  1. I've got two 130 Watt panels, and two 100 AHr Optima yellow tops. I can keep my HF, VHF/UHF and scanners running for a practically unlimited amount of time. And I have a dual-fuel 3kW Honda generator for those times when the sun doesn't shine, or I need to run something BIG.

    ReplyDelete
    Replies
    1. Yeah, that's a great system. As long as you have fuel for that Honda, you're good for the duration.

      Delete
    2. And in response to some of the other posts here, we have a bunch of 12 Volt LED lamps to use.
      And I agree about the inverters. I have one, but it's only rated at about 400 Watts or so. My laptop adapters can run from either 110VAC or 12VDC, so I don't have to use the inverter.
      We figured we can run the generator once or twice a day to cool down the fridge and freezer, which stay pretty cool inside *IF* you're not constantly opening the doors!
      One other thing I've read, is to keep your generator usage to a minimum after dark. The lights you have blazing away, and the sound from the generator, can attract a lot of pests!

      Delete
  2. It makes sense, when planning for SHTF scenarios, to consider how many electric items could be powered from a lower voltage than 120V.

    It is inefficient to have sun-charged nominal 12V batteries use 120V as an intermediate step for items that use lower voltage. When you go from 12V to 120V to plug in a wall-wart to charge your cell phone or flashlight you lose energy in all of the conversion steps.

    Cut out the "middle man" on as many of your items as possible and bypass the 12V to 120V inverter. A direct 12V source for chargeable items is the most efficient, perhaps efficiently stepped down to 5V (as so many items now charge from 5V USB).

    ReplyDelete
    Replies
    1. North - you're getting ahead of me; that's on the agenda, too. Seriously, you're exactly right and it needs to be considered. The old wall warts are a transformer and rectifier; I don't recall an efficiency number, but 25% wouldn't surprise me. I've notice my newer wall warts are switchers, and probably 75% efficient.

      Meanwhile, an excellent switching inverter (to turn DC into AC) can be in the 80% range. But and 80 to go up and an 80 to go down still turns into 64% efficient, and 36% of those expensive watts turning into heat.

      Everything I'm looking to run is a 12V device. Except the refrigerator. So far.

      Delete
    2. I'm probably telling you something you already know, but when the SHTF if you don't have a generator, use an automobile. I have a couple of smaller inverters that can be used with a vehicle. Smarter and more efficient to have a generator, but a car works in a pinch.

      If you have to steal a car for a generator when the SHTF, steal a hybrid! And gas.

      Delete
  3. I've done alot of thinking about this lately. I've been considering war-torn Bosnia and former states of the soviet union, Central America during their wars. There's a really good documentary about a power company installing meters in one of the former soviet countries and the challenges of getting the grid back to a safe state, and billing for power. Their substations were lethal tangle of wires, overfused, noninsulated, and good portions of "liberated" connections. Just walking into one of these you could easily be electrocuted. I digress, Even in these totally destroyed countries, there was power for at least a couple or few hours a day. A reasonable approach might be to simply have deep cycle batteries and an inverter. You can charge your batteries when the power is sporadically on, and have capacity to run essentials during outages. Truth is we have a very well built power infrastructure. If an all out war broke out on American soil, we would certainly have outages, and if you're rural you're more likely to be in a pickle, but those in cities or near cities should have some power. Perhaps, the money spent on solar for this scenario should be scaled down dramatically, to charge a couple batteries, or run a laptop, cellphone and 3 led lights.

    ReplyDelete
    Replies
    1. Anon - your first part about Central America and Bosnia is certainly a good point, but not what I was considering. I was considering after a hurricane or other major natural disaster. This comes from living in hurricane country. My experience with that is that when the power is off, it's absolutely off, and when it comes back, it stays back.

      In your second part, that's a good approach. Unless the grid is devastated, like after an EMP attack, even the most outlying places are back within a month or so. In the meantime, roughing it and living with bare minimum electricity is fine. Figure a few lights, battery operated fans and a handful of other things, while using propane bottles to cook or charcoal or whatever else is available. As above, it would be most efficient to have DC powered stuff instead of running inverters and then AC powered chargers.

      Delete
  4. if stuck with AC appliances the kilawatt meter does a decent job of defining actual energy use over time:

    http://www.amazon.com/P3-International-P4400-Electricity-Monitor/dp/B00009MDBU/ref=sr_1_1?ie=UTF8&qid=1330142802&sr=8-1

    ideally one would first define critical loads and work from there

    the dc-dc plan would be much better, except with a well pump or welder use...

    ReplyDelete
  5. For most people, AC powered stuff is what they have. And when you go to buy DC powered stuff (ask any cruising sailor ;-) you often pay through the nose.

    The cost of inverters has come down a lot in the past few years. Sine wave inverters - needed for some sensitive electronics appliances like computers and such - tend to be more expensive, but those are affordable as well, if you stay with the lower wattage units and run one computer/printer or similar appliances at a time. Modified-sine wave inverters (usually square wave) have become pretty cheap.

    AC powered equipment is so much cheaper than DC powered (when talking household stuff, not necessarily true for ham radios, etc. - SG would know a hell of a lot more than I on that subject) that the cost of an inverter (you're going to need batteries for the DC stuff, too) will usually be much cheaper than buying DC powered TVs, blenders, coffee makers, or whatever.

    Also, the wire costs of pushing power from the panels to your house is much cheaper with AC than with DC, if your wire run is any significant distance. A short DC run to an inverter followed by a much less expensive AC wiring run to your house (if you can't get the panels close due to trees, etc.) can be much, much cheaper. (This is more an issue with things like hydro-electric and wind generation as opposed to solar panels, but can be an issue with them as well.)

    Something a lot of folks forget when they plan for power outages is refrigeration. When you lose a thousand bucks worth of meat and other perishables when your freezer is out for a week (I once lost a couple of big albacore tuna along with half a steer when I had no power for my chest freezer and I wasn't home to run a genny), you'll wish you had some way of keeping it running, although a generator run for an hour or so each day will usually suffice for that. At least, until you can't get gas anymore. (DC powered reefers and freezers ARE usually more efficient, although quite a bit more expensive).

    ReplyDelete