Solar panels?

[Ljohnsaw]

This is a really nice video / build.  Looks very doable.  My question is, why?  With the availability of LEDs for automotive (12v dc), why bother converting 12v to 120v?  Yes, you can use smaller wire to carry the amps but the LEDs draw so little, current draw should not be a problem.  I replaced 5 incandescent lights in my trailer with LEDs - Brighter and use less than one of the old bulbs.

Also, I wonder what the power curve looks like.  Is is a sawtooth 120v pulse at some crazy frequency?  Probably useless for any other purpose than lighting a light bulb.  But still, really cool idea.

[Thehardway]

Lot's of good advice here.  I would chime in on doing as much with straight 12V as possible.  It is much more efficient. You can run all of your lights, TV, a DVD/blu-ray and an array of other appliances made specifically for 12V.  A good source for these is an RV supplier. 

12V LED lights can be found for low voltage outdoor landscape lighting in an array of styles.

Size your panels, charge controller and battery bank for your average load, not your peak load.  Use a small generator for your large incidental peak loads.  For example, trying to size a solar system to run my welder, saws or my planer would be silly.  It is much more cost effective to have a small generator available and a few AC circuits to power large, short term loads and stuff that does not run continuously.

There are some youtube videos out there which show how to convert some old UPS units to work as an inverters.  This is a cheap way to get useable AC power from a battery bank if you are not grid tied.

There are some tricks to getting grid power run to your house more affordably.  One is installing 4" conduit on your own.  If it buried to the appropriate depth and has pull points at correct distances, they will usually run the wire at cost.  This can offer significant savings and also keeps them from having to maintain large power easements trimming trees, etc.  Some basis of the cost calculation is done on the potential for electric sales.  This means when you do your load calculation, don't skimp.

[mad murdock]

This is a really nice video / build.  Looks very doable.  My question is, why?  With the availability of LEDs for automotive (12v dc), why bother converting 12v to 120v?  Yes, you can use smaller wire to carry the amps but the LEDs draw so little, current draw should not be a problem.  I replaced 5 incandescent lights in my trailer with LEDs - Brighter and use less than one of the old bulbs.

Also, I wonder what the power curve looks like.  Is is a sawtooth 120v pulse at some crazy frequency?  Probably useless for any other purpose than lighting a light bulb.  But still, really cool idea.

They are stricly to power lights, the reason you would want one is that they will light the same number of bulbs on about 1/3 the power. 

[OlJarhead]

I run 615watts of solar with 880ah of battery bank and a 2500watt modified sine wave inverter -- I expect to get 2-3 days without sun, perhaps 4 if I'm not too greedy.

Just a quick read of your system and I'm thinking:  no.  It's not nearly enough to run a TV for a few hours in the evening.

There are calculators out there which can help with sizing but I think you are asking a lot of what you're putting together.

Just my 2c

[Ianab]

I just looked at our little TV in the bedroom, new Akai 22", with a DVD player built in. Low and behold, it runs on 12V (3 amps) and has a little laptop type brick to power it.  So it would work fine of a 12V system, no inverters, no power loss etc.

This is just a standard LCD TV that you can buy from any appliance store, nothing fancy or expensive.

Same with lighting, go 12v LED there too. LEDs are inherently a low voltage device, each individual LED needs about 3 volts applied across it. So a 120V LED will have some fancy electronic in there to step the voltage down. Extra cost and wastes some power. A 12V led lamp is more likely to just use 4 x individual LEDs in series, and some very simple circuitry to prevent an over-current meltdown.

Now whether your panels will supply enough power to keep the batteries topped up is debatable, but I'd think that a week of charging would have them pretty much topped up. Then if your TV and lights are only drawing about 5 amps (?) that's plenty of run time for the weekend.

You can always upgrade to a ~300w panel, and would then have plenty of power for a light electrical load.

Ian

[maple flats]

Panels are cheap now. My first panels were $3.00/watt (5-6 yrs ago), non UL panels, last year I added 4830 watts @ $1.75/watt, UL panels. I now have 6.32 KW solar dual system. Check the prices carefully. Don't buy from the so called cut rate companies, their prices are much higher than many other sources. PM me and I'll give you a link to compare prices across the board, hundreds of panels and dozens of vendors.
Adding more panels is by far the cheapest way and keeping away from an inverter if possible helps. A top notch MPPT charge controller is between 95-98% efficient, a PWM controller is at best 50-60% efficient. A cheap inverter is less than 50% efficient, a good pure sine wave inverter is about 97% efficient. These efficiencies however don't come cheap. My first good inverter (an Xantrex 6048, 6000 watts constant, 12,000 watts surge) now on 1480 watts with a battery bank and set to sell excess power by net metering ran $3400+ frt 4 yrs ago. My second inverter is a Fronius 5KW , no battery bank, inverts to 240V was about $2800 + frt. Both of these inverters are at or above 95% efficient, the higher efficiencies are when the inverter is at higher wattages in it's rated size range. Before I got the XW 6048, I got educated in solar, (school of hard knocks + the internet) I bought a cheapo 2500, 5000 surge watt inverter for $279+ shipping. Efficiency, ~40%!
Anytime you can stay with all DC you lose none of these inefficiencies in the inverters. To make the most of your array you do need a MPPT (maximum power point tracking) charge controller, without one you will throw away about 1/2 of the energy the panels generate.
Next, make sure you mount the panels in full sun. Any shade radically drops the power generated. You might think if 25% is shaded you will get 75% of the watts, not so, even a 10% shading drops the wattage produced by over 80%. Full sun is needed. Try to face the panels towards solar south and mount them at the degrees of latitude for your location for tilt. I.E. 43 degrees latitude is mounted at 43 degrees tilt angle for year round, if you use this mostly in the summer, lower the angle about 10 degrees and if mostly in the winter raise it about 10 degrees. When I first started my solar system I was off grid, last year when I added 4840 watts of sloar, I also went the whole shot and brought grid power in, at a cost of about $5000 to the grid and another $2500 in wiring, panel, meters and misc. I now generate excess power and net meter it to the grid. I only have my sugarhouse (for maple syrup production) there, but we will build a new home there someday. The solar system is designed to power the home to be built.

[John Mc]

Maple Flats gives some good advice.  It's good to think about your intended uses - the seasons of use, as well as you electrical loads.  If the site will get year round use, some off grid folks will mount their panels on a rack with adjustable tilt to maximize the output.  This may be too much hassle for some, or may not be possible for some types of mounting systems, but may be something to consider.

This is off the topic for the OP, who is off grid, but I'd also like to point out that for grid-tied systems that are doing net metering, what you typically want to do is maximize ANNUAL production.  (Most states net metering laws let you carry your surplus generation as a credit on your bill month-to-month for up to a year.)  If you are mounting a fixed tilt array, this can often mean setting the tilt at something other than your degree of latitude.

For example here in VT I'm at about 45˚ latitude, but the ideal tilt for a fixed, grid-tied array is closer to 35˚.  This is because the summers have longer days, and historically there are more sunny days than in the winter.  If you want to maximize total production, you tilt to favor the summer.  However, this may be exactly what you DON'T want for an off-grid array used year round: you may need to make the most of what little sun there is in the winter to meet your wintertime needs (shorter days often mean more need for artificial lighting, for example).

On a roof mount system, the question may be moot.  Generally, roof mounts in this area just match the angle of the roof.  Fortunately, most roof pitches in the area from 6 to 12 pitch (26 to 45˚), and 7 to 10 pitch roofs are very common.  These make decent mounting platforms, though you do giv eup a bit from the "ideal" tilt. (It's possible to mount at a different tilt than the roof slope on some roofs, but that can dramatically increase the wind loading if you are in a high wind area, not to mention complicating the mounting system.)

John Mc

[OlJarhead]

I adjust my panels for summer (15 degrees up) spring and fall (48 degrees) and winter (15 degrees down) to maximize year round production.  If I move to my cabin (which I may well do soon) then I'll tweak that a little more.  I can also rotate the panels into the sun if I need a little more production (no tracker) but for the most part I leave them facing true south.


As I haven't needed much power this summer I've left the panels at 48 degrees but again, I'm not living there right now or spending as much time as I could there.  Raising them would increase production but again, no need for me at the moment.

[doctorb]

What time of day was that picture taken?

[OlJarhead]

What time of day was that picture taken?

Early morning before the sun popped up over the trees.  The panels are usually in full sun by 10am and stay there until after 3pm.  At the time of the pick the MPPT controller was actually charging with about 59watts despite the shade.

[Thehardway]

Maple Flats brings up some good points on efficiency reduction and decreased production from shading.  There are some ways around this which can be helpful especially if you want to build out your system panel by panel and not take the hit of buying a big, expensive inverter all at once. 

Using microinverters such as the Enphase product

Enphase uses an individual inverter for up to two panels.  The DC power is converted locally at the panel to AC and allows each panel in the array to operate independent of the rest, as such, if one panel in the array is shaded, it does not effect the output of the remaining panels as it would when you have the panels in a string. Other pros include the ability to use a cheaper, smaller gauge wire from the inverter directly to AC breaker panel, no single point of failure (one inverter can fail and the rest continue to operate supplying critical power), ability to stage panels for time of day specific needs, and the ability to add panels and inverters as budget and needs increase and allow without major changes to wiring etc.

Note that these are only beneficial when you are setup as a grid-tie system.  The biggest drawback to Enphase microinverters IMHO is that they require input from the grid to power on.  This is a safety measure.  It disconnects the inverter from the grid in the event of power loss from the grid to prevent back feeding current during a power outage and electrocution of line workers.  The down side of this is that in the event of a power outage, you can't use the power you are generating on site with your PV panels!  What a bummer.  I understand there are several homebrew and creative workarounds for this but approval is a bit sketchy.  They need to make an autoswitching circuit which transfers to DC mode and charges a battery bank if the AC feed sense is lost or there is no AC load present. 

[Thehardway]

Maple Flats brings up some good points on efficiency reduction and decreased production from shading.  There are some ways around this which can be helpful especially if you want to build out your system panel by panel and not take the hit of buying a big, expensive inverter all at once. 

Using microinverters such as the Enphase product

Enphase uses an individual inverter for each panel. The DC power is converted locally at the panel to AC and allows each panel in the array to operate independent of the rest, as such, if one panel in the array is shaded, it does not effect the output of the remaining panels as it would when you have the panels in a string. Other pros include the ability to use a cheaper, smaller gauge wire from the inverter directly to AC breaker panel, no single point of failure (one inverter can fail and the rest continue to operate supplying critical power), ability to stage panels for time of day specific needs, and the ability to add panels and inverters as budget and needs increase and allow without major changes to wiring etc.

Note that these are only beneficial when you are setup as a grid-tie system.  The biggest drawback to Enphase microinverters IMHO is that they require input from the grid to power on.  This is a safety measure.  It disconnects the inverter from the grid in the event of power loss from the grid to prevent back feeding current during a power outage and electrocution of line workers.  The down side of this is that in the event of a power outage, you can't use the power you are generating on site with your PV panels!  What a bummer.  I understand there are several homebrew and creative workarounds for this but approval is a bit sketchy.  They need to make an autoswitching circuit which transfers to DC mode and charges a battery bank if the AC feed sense is lost or there is no AC load present.