I have had decent success using a very different approach
than those listed here.
My "gobo" looks very similar to the illustration in the second
post, at the top of the thread. However, the contents of
the panels are neither cellulose, nor fiberglass.
First, using fiberglass can be a health hazard that doesn't
seem remotely worth the risk, especially for a recording
location likely frequented by singers. EVIDENCE GROWS ON POSSIBLE LINK OF FIBERGLASS AND LUNG ILLNESSES - New York Times
Second, many posts on this topic focus on "mass" as an
essential component for sound dampening, especially at
the low end ("bass traps").
I find this quite odd since one highly effective sound
dampening material (aerogel), is next to massless,
and is particularly effective at the low end. Kalwall
Third, another neglected topic is the concept of using
viscous materials which actually convert audible energy
(sound waves) into low grade heat. Sound Damping Products for Automotive, Marine, and Industrial
My panels are made up of layers, using the inexpensive
foams that provide 1/2 the high frequency dampening as
the designer foams, at 1/10th the cost, on the outer
surfaces. Next is to an irregular viscous material in a
very thin layer, with a center layer of the closest thing
to aerogel that I could find in an inexpensive form.
Four 2' X 4' panels, configured as stated above, similar
to the illustration in the second post, with one more
across the top, to block ceiling reflections, make up my
"gobo". All are supported on a simple frame made of PVC
The gobo is my back drop, and my mic is surrounded by
a homemade version of the "Harlan box". Build a $21 Portable Vocal Booth - O'Reilly Digital Media Blog
My tiny studio apartment is transformed into a recording
studio, with my gobo as 1/2, and the "Harlan box' as the
other. Believe it or not, I can record, sitting inside the gobo,
with a through the wall air conditioner running 6 feet behind
me, and a computer, and refrigerator also running in the same
room, with no significant noise evident in the recording.