This is how I am doing it this year. I only have enough room on my McSuburb parcel to have small-scale structures, so this year for experimentation one of the things I have is a 4 x 6 coldframe (walls insulated at ~R-5) with a below-ground insulated structure, a water-tubing subsoil heating system and ~30 liters of water for thermal mass. The glazing is 8mm twinwall polycarb at a 30º angle and at night I cover the glazing with an old sleeping bag. Further, I take some of the heated water and put it in the soil, which is insulated @ R-6 - down to 10" in a raised bed next to the house. Despite numerous days with highs in the single digits F (-15ºC ) and several lows well below 0ºF (-25ºC), last time I checked my soil was 64ºF in the sun and 58ºF in the shade (~18ºC and ~14ºC). Air temps during the day always need to be vented on sunny days and at night I have never dropped below 35ºF (2ºC), even on the night we hit -14F (-26C). So my dark-colored water and soil are my thermal masses. So, to reinforce anecdotally what is above, IMHO the original link had inadequate mass to keep from freezing for that volume. The glazing insulation is important. Any text on solar houses or solar greenhouses spend considerable time on solar access and solar mass, as well as insulating the glazing at night and opening the glazing to the sky as soon as the sun is up. For example, right now the coldframe has heated up ~4ºF in 3/4 of an hour's exposure despite cloud cover allowing only ~40% of the expected cloud-free solar radiation through. The sun is free energy. We must do more than just collect it to make it work for us. We must store it, distribute it, and insulate our glazing. Dan

Fuzzy, You want to cover your GH with a heavy solar pool cover. Okay. The heavier the better and the heavier the guage, the better the insulating quality and longer it will last. But in a single piece large enough to cover the GH; too heavy for two people to handle. So the rational solution is to lay it on in sections and that's where the problem comes in: How to hold them on? That's why I asked if you could install bolts through the aluminum: As fasteners for battens. If you cut the cover into 4 sections, with each of the 3 wall pieces cut to allow just a foot of overlap on the top and around the corners, you could then cut a piece for the top that came just to the edges and with holes punched into it around the edges to fit over the bolts, lay on the battens, also with holes, drilled to fit the bolts. Than, just secure them in place with wing nuts. Secured down in that manner, the top piece would also secure the side pieces, that would also have holes, in the overlapping parts. Then use a long rope or better, a long cargo tie-down all around the bottom connected to eye bolts installed at the bottom of the GH base at each end and the cover would never budge in the strongest wind and be as close to a custom fit as you can get. For battens, 1 x 4's would do, but for bolts and nuts, stainless steel only, with a bit of silicone sealant. Slip them into the holes from inside and use a standard nut outside, to hold them in place. Set about 2 ft. apart, once installed they could just stay there and serve as extras to secure a sunshade, but why they also need to be SS only. Ditto eye bolts. I don't know if this scheme appeals to you or not, but I believe it would work. If it did prove to be the solution for you, you could also use solid vinyl 1 x 4 trim board instead of wood, and the battens would be literally forever and you could use them to hold the covers in summer. Roll the cover pieces onto them, tie them up and they would not make too large or heavy a bundle to just lay onto some wall brackets in the garage. Again: 4 pieces and 4 battens, so none too heavy to handle with relative ease. Also, you previously wrote that you had a means of securing a cover to the top, where it's connected to the house, but another batten run along the top would do the same job as on the edges, even if you could only reach in far enough to secure it to about 4 ft. from the end walls. Anyway- for what it's worth- hopefully, a solution.

nathanhurst I'm not going to get into an electronics lesson with you, but you (with all due respect) are complete wrong on how amperage, wattage and voltage work. Real quick. Using your idea of connecting the panels to get 120V is not correct. Two problems with your scenario 1) the heaters we are talking about are 120V AC. Your method only provided DC current. While the elements can handle DC, the fan cannot. There may also be other electronics that are running off of a power supply thatÂs designed for AC. You need an inverter to change DC to AC. While there does exist inverters that will convert, say, 120DC to 120AC, this is not widely used in the Americas. 2) Parallel or SeriesÂ. That is the question. What you first need to understand is that in order to increase AMPERAGE, you need to connect the panels in PARALLEL. To make things simple, letÂs say that a single panel is 15 watts (thatÂs approx. 1 amp at 12V)  (which each panel is in the kit aboveÂ. 3 panels make 45 watts, which is approx. 3 amps at 12Volts). And also, letÂs use small numbers just to get the point across here. 1 panel producing 12V = 15 watts = 1 amp To product 10 amps at 12V, you connect 10 panels together in a parallel manner (positive to positive, negative to negative). Now, if you take those 10 panels and connect them in series (positive from 1 panel to negative on the other panel and repeat for each panel you end up with 1 positive and 1 negative when done). That would make 120V DC, BUT, it would be 120V at 1 amp only. To be clear: - Connecting panels in parallel increases AMPERAGE (watts) but the VOLTAGE stays the same. - Connecting panels in series increases VOLTAGE but the AMPS (watts) stay the same. So using your idea, in order to product 1,500 watts (which is approx. 12.5AMPS) you would need: 10 panels in series = 120V at 1 AMP (letÂs call this 1 set). To make the MINIMUM 12.5 amps (we are not including startup amperage or other necessary things to take into account when configuring a system)Â. you need to connect 13 "sets" (rounded off) of the above connected in parallel. ThatÂs 130 panels (individual panels). AND THEN, that would ONLY product the required amperage ONLY in direct, full sun (we havenÂt calculated for dimmer light levels). You still need an inverter here, which of course, also costs in amperage loss. Hopefully, you now understand how this works. Ok ok, so IÂve been a technician for over 30 years! LOL I just hope that I explained it well enough. I admit, itÂs difficult for me to put my knowledge into short words. LOL But like IÂve said before, itÂs better to use a water heating system for a greenhouse rather than solar electric panels.

I covered just the north wall and a little down on each side wall. That was the largest cover I could get without paying freight charges. I put a layer of regular plastic inside on one sidewall that rans NorthEast. And I had already put a layer of plastic inside on the back North wall before I bought another solar cover for it. Yes, it did reduce the condensation. I don't remember having any at all on the ceiling. Just the front wall that was not covered. Look at 'My Page' to see how far the cover came down on the sides. The wind was a problem esp. in March. So it's a good idea to seal it tight so air doesn't get under it. I finally just took some very heavy rope and threw it over the ends and tied it down so wind couldn't get under it.