Non penetrating roof mount?

Volvo Farmer

My neighbor has a flat roof in snow country. He recently spent 5 grand or so to have one of those professional rubberized roofs installed to stop the leaks. They sealed up all his solar rack posts, but now he wants to install 4 more panels and doesn't want to fool with puncturing the roof membrane. He has a concrete block house with a 18" parapet wall around the whole perimeter of the roof that could have something attached to it. Anyone know of a way to mount adjustable racks on a flat roof without screwing through it?

solarix

Re: Non penetrating roof mount?

150lb of concrete ballast blocks per module if the roof is strong enough, or cables or aluminum truss brackets to secure the tilt racks to the parapets.

ggunn

Re: Non penetrating roof mount?

Volvo Farmer wrote: »

My neighbor has a flat roof in snow country. He recently spent 5 grand or so to have one of those professional rubberized roofs installed to stop the leaks. They sealed up all his solar rack posts, but now he wants to install 4 more panels and doesn't want to fool with puncturing the roof membrane. He has a concrete block house with a 18" parapet wall around the whole perimeter of the roof that could have something attached to it. Anyone know of a way to mount adjustable racks on a flat roof without screwing through it?

These guys have a pretty cool ballasted system:
http://www.dynoraxx.com/

Volvo Farmer

Re: Non penetrating roof mount?

That is cool! But he needs a tilting system. Those do not appear to do that. I think we may end up building something out of steel that attaches to the parapet on the north side and is ballasted on the south side.

ggunn

Re: Non penetrating roof mount?

Volvo Farmer wrote: »

That is cool! But he needs a tilting system. Those do not appear to do that. I think we may end up building something out of steel that attaches to the parapet on the north side and is ballasted on the south side.

Dynoraxx does tilt the modules on a flat roof. I'm not sure what tilts are available, but a system I recently designed with it had a 10 degree tilt.

Volvo Farmer

Re: Non penetrating roof mount?

I guess I mean he wants to adjust the tilt for the seasons, I think the current adjustment is on the order of 30 degrees summer to winter.

solarix

Re: Non penetrating roof mount?

Been there, done that. When you go thru the calculations to analyze tilt angles, you'll find that tilting for the seasons only improves your production maybe 5% at best. Not worth it. And, to really help much you have to really get them tilted up in the winter which means spreading your panel rows way out all the time, taking a lot of roof space. Modules are cheap, just add some more in a fixed array and relax.

ggunn

Re: Non penetrating roof mount?

Volvo Farmer wrote: »

I guess I mean he wants to adjust the tilt for the seasons, I think the current adjustment is on the order of 30 degrees summer to winter.

FWIW, the difference in the altitude of the sun at solar noon on the solstices is the same anywhere on the planet outside the arctic and antarctic circles (inside the circles the sun is below the horizon at one or the other times) - 2 X 23.4 = 46.8 degrees. If you are changing the tilt twice a year then the ideal difference would be half that, from the middle of the summer range to the middle of the winter range.

That said, before I would build a system with adjustable tilt, I would run it through PVWatts at latitude + 11.7 degrees tilt for the six months surrounding the winter solstice and add it to the output for latitude - 11.7 degrees for the rest of the year and compare the total to a year's output at latitude tilt to see if the difference justifies the expense and trouble.

Cariboocoot

Re: Non penetrating roof mount?

solarix wrote: »

Been there, done that. When you go thru the calculations to analyze tilt angles, you'll find that tilting for the seasons only improves your production maybe 5% at best. Not worth it. And, to really help much you have to really get them tilted up in the winter which means spreading your panel rows way out all the time, taking a lot of roof space. Modules are cheap, just add some more in a fixed array and relax.

Depends on where you're at. In Arizona there's not much difference between Summer and Winter angles. The further North you go, the greater the difference and the more worthwhile to tilt (including being able to shed snow, which isn't so much of a problem in AZ eh?)

ggunn

Re: Non penetrating roof mount?

Cariboocoot wrote: »

Depends on where you're at. In Arizona there's not much difference between Summer and Winter angles. The further North you go, the greater the difference and the more worthwhile to tilt (including being able to shed snow, which isn't so much of a problem in AZ eh?)

The difference between the sun's altitude at solar noon on the winter solstice and on the summer solstice is the same no matter where you are. The center of that range is straight overhead at the equator and on the horizon at the poles, but the amount of swing is the same - 46.8 degrees.

Cariboocoot

Re: Non penetrating roof mount?

ggunn wrote: »

The difference between the sun's altitude at solar noon on the winter solstice and on the summer solstice is the same no matter where you are. The center of that range is straight overhead at the equator and on the horizon at the poles, but the amount of swing is the same - 46.8 degrees.

Only in theory, not in practice.
The angle isn't the only thing that matters; hours of daylight change too. Nearer the Equator not so much so. Up here ... egad! Ten hours difference!
To that end, changing the panel angle nearer the Equator does not gain you as much daily power as it does when further away.
In fact when you get to the Equator it actually reverses; Summer daylight is slightly shorter than Winter.

At least we have the cold temps to raise panel Voltage! Come visit the Great White North some Winter. Be sure to wear insulated everything.

Mac's Lab panel angle calculations (good explanation of the business, but is geared for grid-tie): http://www.macslab.com/optsolar.html
Calculate length of day where you are with this US Navy site: http://www.usno.navy.mil/USNO/astronomical-applications/data-services/duration-world

Volvo Farmer

Re: Non penetrating roof mount?

Hmm, interesting thread! I did not know that the gain was only on the order of 5% annualized. I will present this information to my neighbor and see if he still wants to re-tilt his panels bi-annually. As he is off grid, the winter is the time of greatest need and he is generally flush with power in the summer. Thanks for the links!

ggunn

Re: Non penetrating roof mount?

Cariboocoot wrote: »

Only in theory, not in practice.
The angle isn't the only thing that matters; hours of daylight change too. Nearer the Equator not so much so. Up here ... egad! Ten hours difference!
To that end, changing the panel angle nearer the Equator does not gain you as much daily power as it does when further away.
In fact when you get to the Equator it actually reverses; Summer daylight is slightly shorter than Winter.

At least we have the cold temps to raise panel Voltage! Come visit the Great White North some Winter. Be sure to wear insulated everything.

Mac's Lab panel angle calculations (good explanation of the business, but is geared for grid-tie): http://www.macslab.com/optsolar.html
Calculate length of day where you are with this US Navy site: http://www.usno.navy.mil/USNO/astronomical-applications/data-services/duration-world

Another thing to consider (and I haven't thought enough about which way it would influence the adjustable tilt vs. fixed tilt discussion) is that the higher the latitude the further around the horizon the sun travels in the summer, so it's possible that for a significant part of the time that the sun is above the horizon, its rays are striking the back of the array. For that reason, tracking arrays are more viable in the higher latitudes.

Cariboocoot

Re: Non penetrating roof mount?

ggunn wrote: »

Another thing to consider (and I haven't thought enough about which way it would influence the adjustable tilt vs. fixed tilt discussion) is that the higher the latitude the further around the horizon the sun travels in the summer, so it's possible that for a significant part of the time that the sun is above the horizon, its rays are striking the back of the array. For that reason, tracking arrays are more viable in the higher latitudes.

You are absolutely correct!
In fact, in mid-June when we get our near 17 hours of daylight most of that adds nothing to the harvest potential. The sun actually rises behind the array, and it still takes 'til mid morning before the panels have any appreciable direct sunlight.

So you see it isn't so much the change in angle of the sun (which alters panel angle by about 38 degrees no matter where just as ggunn said) as the effect that has on total daylight hours. In general, with a single fixed array you still will only get about 6 hours "equivalent good sun" out of those 16 hour days because most of the time the panels are pointing the 'wrong way'.

If you want crazy efficiency, you'd get monocrystaline panels on a dual axis tracker connected to a grid-tie system.
But is it worth the extra money?

RVer's often have to spend more because they have limited roof space and less than ideal insolation much of the time. If you've got the 'roof real estate', however, it's cheaper to just add more panels even if they aren't at the most efficient angles.

inetdog

Re: Non penetrating roof mount?

Cariboocoot wrote: »

If you've got the 'roof real estate', however, it's cheaper to just add more panels even if they aren't at the most efficient angles.

So which is the bigger bang for the buck:
Adding more panels all set to the same compromise angle as the original ones or adding additional groups of panels facing more to the east and to the west (and attached to separate charge controller inputs?) to use the early morning and late evening sun? The latter would average out your power over a longer period and would be better for handling daytime loads or more uniform battery charging.

vtmaps

Re: Non penetrating roof mount?

inetdog wrote: »

So which is the bigger bang for the buck:
Adding more panels all set to the same compromise angle as the original ones or adding additional groups of panels facing more to the east and to the west (and attached to separate charge controller inputs?) to use the early morning and late evening sun? The latter would average out your power over a longer period and would be better for handling daytime loads or more uniform battery charging.

Two things: First, if you are using a PWM controller, one controller can handle two arrays. Second, for battery charging the west array can be smaller, assuming the batteries are floating by afternoon.

--vtMaps

Cariboocoot

Re: Non penetrating roof mount?

Funny you should ask.

Due to circumstances beyond my control I was unable to implement my planned solar expansion this year.

However, the plan was to put a large (940 Watt) array facing South-East on the MX60 and then divide the small (existing 700 Watt) array into South and South-West facing segments. The function would be that the large array would supply maximum current first thing in the morning, facilitating Bulk charging and beginning the Absorb stage. By midday the South-facing panels would be producing fully, and could handle the running loads on their own (based on actual measurement of present usage). The South-West array would attempt to maintain Float level until sunlight dwindled to the point of uselessness. In this instance, the two smaller arrays would be run through PWM controllers (because of the current involved two smaller PWM controllers worked out cheaper than one large enough to handle both). The Winter consideration is easily handled in this install as the power demands are minimal. This is the "fixed tracking" which has been discussed from time to time here but so far has not been implemented by many.

It would be possible to use PV Watts to get a general idea of the production from various arrays at different angles and see which combination added up to the best scenario for a given site. A lot of trial and error and math, but can be done.