I posted part of this to the AM bulletin board last week; thought it
was appropiate for Amplifier reflector too:
I just lit off a Johnson Thunderbolt that has been sitting in storage
for at least 20 years, and is probably about 45 years old. As part of
my job, needed about a kW CW at 5 MHz, for a test. Started Friday
with a bunch of oil and lube on all the shafts and gears as they were
solid stiff. Today added a small relay on the back to kick it into
conduction via a GPIB controlled relay. Formed the 'lytics with a
small HV supply. After rewiring it to 240 VAC from the bastardized
120 VAC wiring someone had done years ago (and not documented their
mess), it came on real nice. The 866s and the VR tubes lit up like a
christmas tree, and my physicist coworked was amazed. RETRO!
In the Johnson Thunderbolt amplifier, T103, the bias/screen
transformer, has a tap (red/black wire) on the primary winding. It
goes to a blank terminal. It appears to be a tap on the 120 volt
winding, say around 100 volts. The schematic that I got from BAMA
online doesn't show a tap. Maybe a later serial number amplifier
would have the tap shown on the schematic and explain what it is for?
If I move the primary voltage to the tap, obviously will be jacking
up the filaments for the two LV rectifiers as well. Just wondering
why its there. I assume one could mess with the idling bias but since
there is a VR tube also, this doesn't make
sense either.
By the way, the rig has plenty of SNC transformers and chokes from up
in Oshkosh. They are/were a quality company, I used a bunch of their
iron in the power supplies I designed for the Broadcast Electronics
FM1.5A, 3.5A and 5A FM transmitters almost 20 years ago. A few years
ago I got their just retiring chief engineer to design a saturable
reactor for the AC ramping of a big filament power supply for a 250
kW tetrode. Worked perfectly.
I'd appreciate any info on the T103 primary tap for the T'bolt. Maybe
SNC still has drawings of it......
UPDATE 10/6/'02:
The T'bolt runs 1 kW DC input as stated, but it has a hard time doing
this, goes to AB2 condition. In Class C I can get about 65% efficient
at some tunings, with more grid current than 20 mA.
Saturation comes quickly, so about 600 watts output into a dummy load
is max. Looks like the plate supply is down to 1900 at the current
for this, so it really has a problem getting much more. Looking at
the 4-400A data from Eimac, and other similar vintage amplifiers, 2.5
KV - 3 KV is much better, if it can take the heat. (and the caps and
insulation can hack it).
I found a Drake L4B power supply sitting around, someone had lost the
RF portion of this pair years ago, so I only found the little block
of a PS. It is more typical SSB design, with tranny feeding a C input
filter, 8 'lytics in series, voltage doubler config. It says 2400 VDC
under load (>0.5 Amps). While it is for intermittent duty SSB, it
should work for my 3 second computer controlled blasts that I need,
maybe a little droop but....
Tomorrow the L4B PS gets strapped into the Thunderbolt. What shall I
call this beast ? A Thunderdrake? A Drakebolt?
73
John
K5PRO
--
|