wrote:

I have noticed that the rooms with skylights are MUCH cooler than other
rooms, making me conclude that heat is being lost through the skylights.
How do I minimize this heat loss? The skylights are flat (not the bubble
type)... I live in Chicago.

December is the worst-case month for solar house heating in Chicago, when
460 Btu/ft^2 falls on the ground and 740 falls on a south wall on an average
26.6 F day with a 34.0 daily max, so 1 ft^2 of single of R1 horizontal glazing
with 90% transmission would gain 414 Btu and lose 24h(65-26.6)1ft^2/R1 = 922
for a net loss of 508 Btu per day.

I have thought about getting plexiglass installed to the inside frame

Good idea. That might reduce the gain to 373 and reduce the loss to 461
for a net loss of 88 vs 508.

but now concerned that it may not be a good idea since during summer,
the 'pocket' between the plexiglass and the skylight might get too hot.

In full sun (say 250 Btu/h-ft^2) in Chicago, 225 might pass through the
first glazing and 203 might pass through the second, a difference of 22.
If it's 83.7 F indoors and outdoors (the average daily max in July), R1
resistors to that indoor and outdoor temp might make the temp between
the glazings 83.7+22xR0.5 = 94.7 F, which seems OK.

I once made a hinged Styrofoam shutter for a west-facing skylight with a 1:1
slope and several layers of glazing and painted the skylight side black to
make it warmer in wintertime, if the shutter were closed most of the time.
It was, but after the dead of winter, the Styrofoam melted into a black
mountain range with pink crevasses and peaks around the screws that held
the foam to the plywood backing. It got nice reviews in an art show :-)

An R2 skylight might collect 0.81x1940 = 1571 Btu/ft^2 of unwelcome solar
heat on an average July day in Chicago. A reflector hinged at the north edge
of the glazing might eliminate most of that and enhance winter collection.
Sun elevations in Chicago (N lat 41.8) are 90-41.8+/-23.5 = 71.7 and 24.8
degrees above the horizon at noon on 12/21 and 6/21, so the upper edge of
a reflector can block all the direct noon summer sun if it's on a 71.7
degree elevation line up from the south edge of the glazing, like this,
viewed in a fixed font:
. - - -


. . |


-- north 1.5' . . 1.12'


. . |


. .

48.2 degrees 71.7 degrees
............----------------------..................
1'

If the glazing is 1' wide in the north-south direction, we can bounce all
the noon winter sun into it if the reflector edge has a 90-41.8 = 41.2 degree
elevation angle from the north edge of the glazing, which makes the upper
edge 1tan(48.2) = 1.12' above the south glazing edge, with a sqrt(1+1.12^2)
= 1.5' slant height, so 1 ft^2 of skylight would collect about 0.9x0.81x740
= 539 Btu and lose 461, for a net gain of 79 Btu/ft^2 per day.

If we replace the skylight glazing with hinged R10 foil-foamboard and use
1.12' of R1 vertical glazing and raise the board up to 48.2 degrees during
the day, we might collect 0.9x0.9x740 = 599 Btu and lose 6h(70-30)1.12ft^2/R1
= 269 during the day and 18h(65-26.7)1ft^2/R10 = 69 at night, for a net gain
of 261Btu/ft^2 per day.

If we leave the R2 glazing in place and use 1.12' of R1 vertical glazing,
we might collect 482 Btu and lose about 6h(70-30)1ft^2/R3 = 80 during the
day and 57 at night, for a net gain of 344.

Nick