[Scott Dorsey]

Adrian Tuddenham wrote:

The two inputs of the 'scope were connected to the Tascam, one to the
input and one to the output. A sinewave signal generator was connected
to the input and the recorder was set to Record (actually in Pause mode,
but that should not matter). The two traces were accurately centred
(vertically) and the sensitivity controls were adjusted until the two
traces were of similar height.

At a lowish frequency (e.g. 100 c/s) the traces were compared and, if
necessary, one could be inverted to make them the same polarity. The
frequency was then increased until they were exactly out of phase. it
was further increased until they came back into phase. Those two
frequencies ought to be related in the ratio 1:2, if they aren't, it
indicates that the starting frequency was too high.

The frequency at which they come back in phase (in cycles per second) is
the reciprocal of the delay time (in seconds).

See, I don't like this method. What I like to do is to use lissajous
method, with one signal going to the X input of the scope and the other
signal going to the Y input. If they are both in phase, you get a nice
diagonal line... if they are 90' out of phase you get a circle... and you
can make some rough judgements about values in-between.

If you don't have a scope, you can do it with a VU meter, just summing the
two inputs together.... adjust the frequency and the meter will dip lowest
at the frequency where there is best cancellation. I don't really recommend
this method but I have done it before. It's not all that accurate but it
is sometimes accurate enough.

There are always three ways to do anything: the right way, the wrong way,
and the Army way.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."