[time-nuts] Lady Heather numbers

Arthur Dent golgarfrincham at yahoo.com
Wed Feb 27 14:09:01 EST 2013

There has been a lot of speculation about the innards of the OXCO in the
Thunderbolt so here is what I found inside a dead one that I opened. I
didn’t trace the circuit out so most of the description that follows is just
from a cursory look and may not be totally correct. Below are links to a
couple of photos of the oscillator board.

The OXCO in the Thunderbolt (Trimble 37265) is a single oven unit. The
circuit board is mounted on the connection pins and has fiberglass mats
on top and bottom for thermal isolation. There are two +12VDC input
pins, one for the oven and the other for the oscillator power. The +12V
pin for the oscillator goes through a 1 ohm resistor to a 8L08A 8V three-
terminal linear regulator which powers the oscillator and most of the other
electronics. The ‘heater’ that is powered by a separate +12V pin is a
2N4921 transistor mounted on the copper block with the crystal and the
thermistor. There is a diode between the input pin and the transistor for
reverse polarity protection but not on the oscillator input pin. From reading
some of the past post about people connecting the power to the Tbolt
incorrectly, this would explain why the heater still works but there is no
output. The 1 ohm and/or the 8L08A are destroyed while the heater circuit
was protected. Other stuff on the Tbolt board probably gets destroyed as

The crystal is marked OFCG-P, Div of Oak Ind and this one is marked 86.3C
for the temperature. The oven controller is an LM358 dual OP-AMP and there
is a 10-turn pot that I believe sets the oven temp. It looks like the 10Mhz
output goes through a couple of transistors, a transformer, and some L-C stuff
before going to the output pin. There is a SOD-323 device marked VD1 which
may be a BB639 varactor for the EFC adjustment.



More information about the time-nuts mailing list