On Wed, 4 May 2005 11:23:39 -0700, "John Smith"
wrote:

The resistance in the anode circuits of the LEDS only serves as an attempt
at limiting the current through the FWD LED to full brightness

It, in fact, serves to reverse bias the other LED, turning it OFF
(until the first LED depletes the capacitor which then reverses the
situation - the original RDL flip-flop running asynchronously). Which
one lights up is a matter of an indeterminate logic race, not some
proportionality of powers.

(and a constant current source would best replace this)--

If you think about this statement, you would recognize that
either/both LEDs would always be ON and equally bright. [the word
"constant" should be a dead give-away]

and to keep it from "burning out."

Then you put a current limiter (not constant current device) in EACH
LED lead (not sharing). To select a 1KOhm limiter presumes a 20V drop
across it for a 20mA LED excitation current (a typical current
specification). You may achieve the 20mA limit, but it will come at
the cost of more than 40V across 100 Ohms backing through the current
transformer (it is NOT a directional coupler) which can only be
supplied by a transmitter excitation of several KW to a matched load.
If you did nothing other than simply connect an LED across the current
transformer's terminals and adjust the transmitter for 20mA through
the LED, you would need 8W into a matched load.

To put it simply, you are fixated on a problem so remote from your
goal, that solving it basically accomplishes nothing towards measuring
(or even indicating) SWR. As several suggestions have been offered to
research a real topology, you might want to pursue that first.

73's
Richard Clark, KB7QHC