http://www.atvinderby.co.uk/2.4Ghz_SWR_Meter.htm

Regarding the dimensions of the wire loops:

ant: 28 mm square
reflector: 30 mm square
separation: 10mm
wi 14ga solid (1.6mm dia)

Are these dimensions center-to-center of the conductors? Or are these
dimensions the space between the conductors? With a conductor diam of 1.6mm,
this is more than 10 percent of one dimension (separation), so I think it's
important to get it right.

It's probably not actually all that important. Quads are not highly
in their dimensioning. Altering the positioning between the driven
(active) element, and the reflector, will change the antenna's pattern
(gain and front-to-back radio) slightly, and will alter the feedpoint
impedance slightly. Neither of these changes is going to be
particularly significant (and quite possibly not easy to detect) in
this situation. The fact that there's a lot of additional metal in
the neighborhood of the antenna (the BNC connector, etc.) is probably
going to alter the pattern more than the spacing different will
(compared to a theoretically perfect quad in free space).

The pattern of a two-element quad is fairly broad, and the
front-to-back ratio isn't terribly high. It's not the best design for
pinpoint aiming / location of a microwave source... but that isn't
really what this device was intended for. As a field-strength meter,
it'll work fine (and probably just about equally well) with a 10mm
separation edge-to-edge or center-to-center.

Also, is the reflector to be grounded or floating?

In a quad, it is usual for all of the elements other than the driven
one to be floating. Since the article says that the reflector is
fastened with a dollop of glue, I think it's safe to assume that it's
intended to be floating.

That detector is a slick little design - thanks for posting the link!

--
Dave Platt AE6EO
Friends of Jade Warrior home page: http://www.radagast.org/jade-warrior
I do _not_ wish to receive unsolicited commercial email, and I will
boycott any company which has the gall to send me such ads!