[time-nuts] Lucent KS-24361 15 Mhz and 10 Mhz

Arthur Dent golgarfrincham at gmail.com
Tue Dec 2 00:31:54 EST 2014

I had Planned to eventually convert the 5 Mhz output I added from
J8 on the Lucent RFTG-u REF1 (that I described previously) to 10 Mhz.
Since I made that 5Mhz modification 4 years ago I have been using the 5
Mhz sinewave output for some of the equipment I have around the bench
that can use it directly plus I have a modified Spectracom 8140
distribution amp that take a 5Mhz input and will output a clean 10Mhz
sine wave.

The 5-10Mhz doubler circuit described by John Roos seems like a good
way to accomplish what I wanted to do. I started looking around for
parts I might already have that I could use to construct the circuit
and I found a lot of what I needed. Having dismantled about 200 of the
wireless locator units that held the Thunderbolts that I sold on Ebay,
I had a large pile of the machined R.F. subassemblies left over and
they had a large number of MCL (Micro-Circuits) and other really neat
R.F. stuff. I found a MCL RMS-2 (5-1000 Mhz) mixer that would work
well for the mixer and then found a 10Mhz 2-stage amp with filters on
one of the circuit boards that should give me the clean 10Mhz sine wave
that I want. I have tried the amp and it seems to work well. When fed
with either a sine or a square wave the output is a clean sinewave that
can drive 50 ohms. The power required is +/-7VDC at low current and
that voltage goes to 2 on-board regulators to provide +/-5VDC that supply
the 2 amps. Feeding the +/-15VDC from the REF 1 power supply through two
200 ohm resistors provides the +_7VDC for the amps' on-board regulators.

The inductors I need for John Roos's 90 degree phase shifter circuit should
be here in a couple of days and then I can permanently mount the amp board
right behind the front panel connectors near the middle of the REF 1. On
the outside chance that the unit actually needs the 15MHZ signal for some
purpose this added board will not affect that signal in any way. At least
for me this is a good solution. Here is a link to a photo of 2 different
revisions of the board, one with a discrete lowpass filter and one with
the MCL SCLF-10.7 package. After stripping off the other stuff I don't need
there is plenty of room for the rest of the parts from John Roos's circuit.



More information about the time-nuts mailing list