[Greenbuilding] A simpler PV ground mount, rev 2
Nick Pine
nick_pine at verizon.net
Fri Nov 18 13:34:06 CST 2016
How about bending a 20’ pipe into a flattish 15’ section with 2 legs that slip over an 8’ south ground sleeve and a 10’ north sleeve, both pounded 2’ into the ground to make a storage/parking structure? Like this, viewed in a fixed font:
| 15’ |
........ ---
/ ........
| ........ 2’
| ........ ---
| \
| |
| 10’ | south >
| | 8’
| |
| 14.9 |
|-----------------------------------|------
| |
| | 2’
| |
Most ground mounts have lots of pipe and concrete and posts in the middle so you can’t park a vehicle lengthwise, and the PV roofs are not waterproof...
In this one, 2 groups of 6 $25 164” Everest or Iron Ridge PV rails would run east and west on top of the 15’ length to support 3 5’ portrait length PVs in 15’, totaling 24 PVs in a 10 meter east-west length. Krannich sells 245 W Trina modules for about 50 cents/watt.
http://www.engineersedge.com/fluid_flow/steel-pipe-schedule-40.htm says 2” schedule 40 pipe has modulus Z = 0.5607 in^3, so with a 40 psf load (35 for snow + 5 for PVs, rails, and the pipe itself) and no wind and a d’ rafter spacing, M = Wl/8 = 15d40x15’x12”/8 = 13500d in-lb, and 48K psi = M/Z = 13500d/0.5607 makes d = 2’. That’s a lot of big pipe.
A shallow pond along the north side with an EPDM liner and Styrofoam on top could provide warmish groundwater to melt up to 4”/hour of snow, with a layer of greenhouse polyethylene film over the PVs and a 12 volt battery-powered pump moving water up to an east-west pipe with holes at the peak. (What keeps the water from running down the grooves between modules instead of over the modules?) Without the foamboard, this could also provide summertime cooling for more PV output with a Trina PMAX = 0.41%/°C tempco.
| 15’ |
........ ---
/ ........
| ........ 2’
| ........ ---
| \
| | gutter
| 10’ | south >
| | 8’
| |
| 8’ | 14.9’ |
--~~~~~~~~~~~~~~~ |-----------------------------------|------
1’ deep pond | |
--------------- | | 2’
| |
But melting 4”/hour of snow is like melting 0.4”/h (2.08 lb) of ice at 144 Btu/lb or 2.08x144 = 300 Btu-h/ft^2, ie 300x15x30 = 135K Btu/h (40 kW) for the whole roof, which would require 13.5K lb/h of 42 F water, ie 27 gpm. That’s a lot of water. So it seems better to increase the PV tilt from 7.6 to 35 degrees so the snow is more likely to slide off, which would also increase the yearly output from 7183 to 7856 kWh in Philadelphia, according to PVWATTS.
.....
/ .....
| ..... 15’
| .....
| 8.6’ .....
| .....
| 35 degrees .....
| ................................. \
| | gutter
| 8’ | south >
| | 8’
| |
| 8’ | 12.3’ |
--~~~~~~~~~~~~~~~ |-----------------------------------|------
1’ deep pond | |
--------------- | | 2’
| |
Some 55 gallon water drums inside the south wall could help melt snow. The PV roof conductance is about 15’x30’/R0.3 = 1500 Btu/h-F. The north wall adds 8.6’x30’/R1 = 258, and the lower south wall adds 8’x30’/R1 = 240, totaling about 2000. PVWATTS says 3.18 kWh/m^2 (1008 Btu/ft^2) of sun falls on a south wall on an average 30 F January day. If the drums are insulated on the north side and the south wall has 90% solar transmission, 0.9x8’x30’x1008 = 218K Btu = 24h(T-30)1500 makes the average drumwater temp T = 36 F. If the drums are also insulated on top with a 2’x30’ bench and a fan with up and down ducts, 218K Btu = 24h(T-30)240 makes the average drumwater temp T = 67.8 F... 15 drums could store about 15x450(67.8-32) = 242K Btu of snow melting energy, ie 1008 Btu/ft^2 of roof, enough to melt 1008/144 = 7 psf of snow, ie about 1.1”. Melting a little snow above the PVs might start the rest sliding. Blizzards might require Mr. Longarm http://solarchargeddriving.com/2012/01/04/a-dozen-tips-for-getting-snow-off-solar-panels/ with a $7 angle adapter http://mrlongarm.com/product/angle-adaptor/ or electrical heating of the PVs.
Summertime cooling can also increase the electrical output. Lowering the 44 C Nominal Operating Cell Temp http://www.trinasolar.com/HtmlData/downloads/us/US_Datasheet_PD14.pdf to the 25 C Standard Test Condition (or less, eg the 18 C average July dew point or the 13 C ground temp), would increase the power output by (44-18)0.0041 = 10.7%, ie 0.107x5880 = 629 watts (about 2 PV modules), about 10X more power than a 64 W $99 3 gpm pump (or 2, for greater reliability. We don’t want to melt the film.) https://www.altestore.com/store/solar-water-pumps/surface-solar-pumps/shurflo-surface-pumps/shurflo-2088-443-144-12v-std-surface-pump-p1088/ On an average July day with an 86.1 F max outdoor temp, 15%-efficient PVs under greenhouse film with 94% solar transmission could absorb 0.85x0.94x15’x30’x317Btu/ft^2 = 114K Btu/h of sun... With 6 gpm of 55 F water, 114K = (T-55)3000 + (T-86.1)2x15’x30’/R1 makes the PV temp T = 91 F (33 C), ignoring evaporation and radiation losses, which could be high to a low sky temperature on a clear day.
Perhaps guy wires to 2500 pound ground anchors https://smile.amazon.com/Ground-Anchor-System-50016-Piece/dp/B000AMQC46/ref=sr_1_7?ie=UTF8&qid=1479493764&sr=8-7&keywords=ground+anchors ? can resist wind loading. A 90 mph wind would make 0.00256x90^2 = 20.7 psf, eg 8.6’x30’x20.7 = 4977 pounds on the north wall, with guy wires to 2 anchors.
With a 5 psf load and a d’ rafter spacing, M = Wl/8 = 15d5x15’x12”/8 = 1688d in-lb, and 48K psi = M/Z makes Z = 0.035d = 0.239 in^3 for d = 82”/12 = 6.8’, half of a 164” rail, so we could use 1.25” pipe with Z = 0.2346 in 5 rafters for the entire structure, which would be about 10 meters long with 6 rows of 2 rails end-to-end, containing 3x8 = 24 245W PV modules (5880 Watts.)
We could use 1” pipe ground sleeves, since 1.25” pipe has a 1.380” ID and 1” pipe has a 1.315” OD. A 10’ pipe clamped in the ground and pinned at the top would have a critical buckling load Pcr = Pi^2EI/(0.7L)^2 = Pi^2x29Mpsix0.0873in^4/(0.7x10x12)^2 = 3541 pounds and a dead load of 15’x6.8’x5psf/2 = 255 pounds.
Schwartz & Sons in Wilmington DE sell 1” schedule 40 pipe in 20’ lengths for $1.30/foot and 1.25” pipe in 21’ lengths for $1.25/foot. Mike Schwartz also sells 15 ga 1” swaged pipe for purlins in $15 21’ lengths and $4 crossover clamps to attach them to 1.25” pipe. http://www.harborfreight.com/16-ton-hydraulic-pipe-bender-35336.html might bend the pipe.
Drew suggests attaching a 4’ downward spreader pipe (“king post strut”) perpendicular to each 15’ pipe with a Kee clamp barrel T over the 15’ section and a north-south cable connecting the eave and the ridge under each spreader and an east-west horizontal purlin cable that goes under each spreader. It looks like that isn’t needed here, but a north wall purlin is.
https://www.revisionenergy.com/at-home/solar-trackers/ also looks nice, but expensive, at $35K+ for 24 modules, about $6/watt, with about 40% more energy production than a fixed array.
Nick
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.bioenergylists.org/pipermail/greenbuilding_lists.bioenergylists.org/attachments/20161118/55417132/attachment.html>
More information about the Greenbuilding
mailing list