A couple of tools on the internet claim ~15,000 BTUs of cooling for your building assuming standard insulation. Better insulation and shading should be able to cut that significantly. Plant lots of trees/bushes and consider a tracking mount for your panels.

This site: http://www.dcairco.com/ has a 24VDC 4400BTU/h truck unit that pulls 1300W. So for every hour of operation, you'd need ~2.5 hours of sun (~80% system efficiency)

Considering your budget and location, have you looked at wind power? Not a consistent, but may be a better bang/buck.

Sean

So what are you using to take your DC and make AC ?

Your AC will need to be 240 VAC ... all the of grid inverters are 120 VAC , so two would have to be stacked ...

You really want to be grid tied and have the solar suppplement the AC , not replace it our you'll be in the dark , with dead batteries allot of the time

I have at FERRUPS UPS that is 3.2KW and uses 48v DC. I bought it at an auction for $35. It provides either 120v or 240v. It was in operation when removed. It has a unique feature in that it can be manually controlled via RS232 port. My garage has a 100Amp 240vAC panel. I'm only attempting solar AC, not solar garage.

I bought a Hawkeye current detector and placed it around the A/C power input line. When the A/C switches on, the Hawkeye sends me a 12vDc signal. I programmed a PIC micro controller to wake on input of the signal line and when it senses the +12, it sends a cntrlI (switch to inverter) command to the UPS. When the +12 goes away, it sends a cntrlR (reset) to the UPS and it goes back to monitor mode. There is a lot more to the program, like check for power failure and check batt voltage (don't take the batteries below this voltage, record times, etc., etc.) but that is the basic process. I also programmed in 3s delay to let the current spike from the motor startup go away. In the purest since I guess its a bit of a cheat, but again, it ain't the space station.

None of the system is installed. I have been collection pieces and I'm about to begin with the A/C installation.

Grid Tie is not an option. In TX some utility Co.s are a regulated monopoly and mine is one of them.

As for the A/C size, when I asked around I get a pretty consistent 1 ton per 750sqft residential and 900sqft for metal buildings (which mine is). Since I have a tall ceiling, I went for the 1.5 ton sizing. Plus the fact that there is a 16'x8' garage door and everytime it gets opened there will be some pretty good recovery going on. Fortunately, it only happens about once a week.

Thanks for the tree suggestion. I'm a little worried about the shading of the panels but on the West side for sure that will a huge help.

As for wind, I just don't see it. First I'm not in a good location for it. There are to many trees. Secondly, I have to get everything approved by a board (in TX it is actually against the law to not allow solar panels on a roof I'm told) who I don't think would ever go for a tower. And thirdly, it just doesn't seem to be as developed a power source, though I could be wrong there. I also think the solar will be much more maint. free.

Thanks guys for suggestions and scrutiny of the project. I've spent a lot of time thinking about this and it's always good to have the design ideas punched around. Builds a better mouse trap in the end.

Glenn

Your ups is not designed to handle motor startup surges ... when the AC kicks in your looking at a 4-5x load ... the UPS will just shutdown

Also , I don't know what the efficiecny of the UPS is , but usally , UPS's aren't that great , more PV watts turned into heat

Seem like you have a great can do attitude and you have lots of talents

keeps us informed of your progress !

Hi Glenn,

There are several things that you can do to reduce your A/C load.

1. Insulate with celulose not fiberglass. Cellulose is opaque to radiant heat so will not transfer as much radiant energy to the inside of the garage.

2. Put awnings over east, south and west windows.

3. Keep the structure as tight as possable so you don't lose conditioned air and pull in hot air.

Don't forget that if the sun is not shining, you won't need as much A/C as on a sunny day, so you can size your array for a good summer day instead of an average yearly day.

Put your electrical loads on a sub panel with a transfer switch so you can use as much solar electricity as possable when you have little or no need for A/C but have a surplus of electricity.

By the way, about 15 or 20 years ago, I built a ground sourced heat pump that had a DC motor that I ran directly off of a 650W PV array.

Thanks for the encouragement Ron. I wish I would have gone the cellulose route. That's a definate on the next one I build. That fiberglass nearly killed me building the building. I used the 4" 'stiff mat' kind (no chicken wire) and what a pain. Hard to keep tight and harder to put a 2.75" tek screw through without oil caning the metal. There is a real interesting material that is simular to celluose that is made just a few miles from here. It is compressed, epoxied wheat straw. 4'wide x 2.5" thick and any length you want. I may add later. It is real easy to install. You just shoot down baseangle at the bottom and screw into the 7'4" header perling at the top. No studs required. I went with the owner and toured an installation where they textured it and it looks just like plaster on block. R18 and R36 available if I recall. Where this not just a Garage/Workshop I'd probably do it. At the end of the day it was higher than the mat, but not much and 1.5x the R value; increase noise damping, and really attractive.

There are no windows, just an overhead 8x16 3x insulated door and a walk-in door. All the rest is sealed and all penetrations are through the slab except the feed from the solar panels on the roof to the inverter. It is a 2" EMT with weatherhead.

As for tight I think it is pretty tight for a metal building. The one exception, and this doesn't work out in my brain, but I'm taking a leap of faith from a guy who lives near me with 35 years experience, is the 10' commercial vent with baffels and damper. He claims leave it open 1/2 notch in the winter and 4 in the summer and all will be better than if it were sealed up tight. I can see in the summer that the hot air will go out. But it seems like it can also come in when the wind blows. Except I understand that if it is sealed tight there will be no pressure differential so maybe it won't want to come in. I'm as afraid of heights as anyone you have ever met. I bet I have spent 5 hours on a scaffel 13' in the air testing this out over the summer. Maybe it's correct. Not sure and not willing to spend the time to instrument to find out.

The subpanel idea sounds good, but I'm going to keep it in reserve for now since I am only trying to make the CHA system solar at the moment. I was thinking of a golf cart charging station for the spring/fall time when I don't need to run the A/C very often. The only other thing inside that uses power is a 1000 bottle wine tomb I built and I'm entrusting that to Greenville P&L for now. Switching that one doesn't pass the risk/reward test.

So congratulations on the GS system. I would like to no more. Was it cost effective? What kind of pipe did you use? How would you summarize the performance. What size convential system do you think it would have taken to do the same job? Did you make you're buddies buy you a beer after they told you it couldn't be done?

Again, thanks for the input. And to SG, thanks for hosting this web forum. If we get too far off of PV let me know. (come to think of it I'm pretty far of Solar Panels for sure)

Glenn

Hi Glenn,

It was a hybrid horizontal and vertical loop of Polyethlene irrigation pipe. The first 250 feet of pipe were about 3 feet under ground and the last 150 feet went into a 75 foot hole with a welded u-bend. The compressor was from a 1980 Honda Civic. The compressor was driven by a 90 VDC permanent magnet motor thru a Gates toothed belt. The PV array was in strings of 3 panels so the normal operating voltage was about 50 volts. I had 12 panels so I could configure the array for 50 volts or 100 volts. In bright sunlight, the heat pump could run continiously at 50 VDC. I also had an unregulated 50/100 volt DC supply that I could put in parallel with the array and "blend" grid power with the solar. I had to blend grid power when I was operating the system on 100V since the array was about 1/2 the power that was needed on high.

I got an E.E.R. in the low 20's in cooling and a C.O.P. of just over 4 in the heating mode.

The heat pump was water to water so I had to circulate the heated or cooled water thru a fan coil/air handler attached to the ductwork. The ground loop circulating pump had a 90VDC PM motor in parallel with the compressor motor so the ground water flow somewhat tracked the compressor load. I also controlled the speed of the air handler fan to control the load so the PV stayed at about 50 volts (crude MPPT).

The nice thing is that you get paid to do this. The rest of us slobs, like poor Glen on a shoestring budget, has to tow the line himself. At least we get good data from the labs (one good usage of our tax dollars)

Now, let's get back to solar, shall we ? Unless da Gup wants to open a groundhog section of his site up (aka Solar Moles)..

Hi Larry,

No, I didn't get paid to do it. It was on my own home. I was too cheap to by central air. As an interesting note, I did not use Freon as a refrigerant. I used Propane as the refrigerant. It's temperature/pressure characteristics are similar to R-12 but it is slightly more efficient.

It really was a good example of the load matching the available solar energy.

Glenn,

Have you considered the cost/benefit of mounting your panels on a tracking mount instead of the roof? You could probably find an old C-band dish mount/motor and set up a PIC to nudge it every five minutes or so.

Here's a decent article on shade trees and the amount of cooling reduction they can generate
http://hem.dis.anl.gov/eehem/95/950307.html

Sean