I tapped into the T-line and keyed a transistor which in turn was used to
key the 405 Amp's relay.
Same here...
When I owned an Omni Six, I decided to modify the external amp relay circuit
to support a super-fast, high-voltage key line for use with an Alpha Seventy
amp. The Alpha uses a +55VDC key line and is beyond the maximum specified
Vceo voltage rating of most small-signal switching transistors.
First, I replaced the Omni's relay driver with a 2N5551 NPN transistor and
brought out the collector to the RCA connector. The 2N5551 is Vceo rated
for +140 VDC. This allowed me to directly key the amp but it created
another problem: the timing of the open-collector was causing the Alpha to
hot-switch. I had modified the Alpha for full QSK by using a
voltage-doubling circuit that initially doubled the keying voltage to the
vacuum relay and thus increased the initial snap time of the relay. But
when using the Omni Six, the problem was not the leading edge, but rather
the trailing edge of the amp's keyed waveform. For many years, I have been
using a scope to monitor my transmitted output. Had I not had the scope, I
probably would have never caught the hot-switching issue before replacing
several expensive vacuum relays.
The next fix involved finding a better sample point for the Omni's keying
transistor. That turned out to be Ten Tec's "T" line. Sampling the T
voltage line resulted in fast amp switching on both the leading and trailing
edges of a the keyed waveform. I further went on to create a quasi
Darlington pair for better isolation from that +55 VDC line. Two 2N5551
transistors were cascaded to create the Darlington. A diode on the base
added further isolation.
If I had to go back and re-visit this exercise today, I would use a
high-voltage MOSFET. When I owned a Ten Tec Pegasus, I looked at the amp
keying transistor, and they designed this part of the circuit correctly: a
TO-220-style transistor is used with a high-voltage Vceo point -- even
though Ten Tec specified the keying line for much less than it was capable
of switching.
At least one mod exists on the Internet to create an amp key line by
utilizing an RCA "Y" adapter connected to the Omni's TX OUT/TX EN line with
the third RCA connector sent to the amp key line through a small steering
diode. That solution works fine for amps that protect against
hot-switching, but it will not work correctly for amps that are full-QSK but
have less sophisticated sensing circuitry.
It would be interesting to see what type of amp keying transistor is used on
the O2 and Omni Seven.
Paul, W9AC
|