Most class C bipolar transistors have no bias at all. Especially VHF FM
and CB "linears." CB linears for AM depended on the carrier for forward
biasing but were dirty besides being outside FCC regulations. Often the
base circuit includes a lower resistance RF choke or coil from base to
emitter or ground. Stage gain for low level signals is zilch until the
rf peaks get up past the silicon junction barrier voltage. While that's
a decent operating point for efficiency, it increases harmonics and key
clicks. It takes away from envelope shaping done at lower levels.
That note in the G3VTT manual version is too neatly done, it doesn't
stand out. I had to read the manual twice to notice it. I can't argue
with the suggested values, have to remember those two stages are drawing
current on receive also according to the schematic. And are the only
stages on the PA board running from the +12 connection to that board. So
the current in the 12 volt connection consists of both collector
circuits plus the bias circuits current.
Typically the bias current for any device trying to be linear works out
to be the current that dissipates rated power with no signal because any
smaller current results in more distortion. Since its CW only,
distortion is less of a problem except when it creates key clicks. Given
the G3VTT information, I'd turn both bias pots to zero voltage, and
disconnect the +12 and reconnect it through my meter. I'd notice the
bias circuits idling current, and I'd increase the driver pot setting to
increase the 12 volt current by 30 ma. Then I'd increase the PA pot
setting to increase the 12 volt current another 140 ma and with my meter
shorted, I'd check for key clicks. The PA will draw 5 amps and a meter
that can read 30ma won't like 5 amps for long.
Or I could measure the receive current since the PA and driver
transistors take current all the time there's 12 volts applied. Turn the
audio gain down to minimize receiver current, turn the two bias pots to
zero volts, note the supply current, turn up the driver pot to increase
that 30 ma (presuming the half amp spec is correct I'd look for a little
less with no receiver audio), then turn up the PA bias to increase it
another 140 ma. And with the meter protected, I'd check for key clicks.
On the other hand, if the RF transistors in the two stages were not
overheating, the power output is close to rating, and there are no
complaints about harmonics or key clicks, if I hadn't replace
transistors in one stage or the other, I'd let them set where they are
and worry about something else.
73, Jerry, K0CQ
On 1/29/2011 10:24 PM, Bwana Bob wrote:
It's easy enough to unsolder the wire. Otherwise I'd have to borrow a
Tektronix current probe from the work QTH or bring the rig to work on a
Saturday.
I measured the bias voltages as they are set now and I see 0.64 volts.
This is the normal base-to-emitter forward voltage drop for a Silicon
bipolar transistor. Trying to push it to 0.7 volts would not increase
the voltage but would dramatically increase the current. That is why
bipolar transistor biases are specified in terms of current not voltage.
It would be a different story if it were a vacuum tube or MOSFET final.
The 30/140 mA spec came as a result of an internet search. I downloaded
a copy of the manual which had a typed notation on it, from G3VTT, who
apparently got it from Ten-Tec. I don't know how old this information
is, and I would prefer to use a "primary source", so I e-mailed Ten-Tec
myself and got the 0.07 volt recommendation. The only way that I can
explain the difference is that the bias was originally set up for class
AB linear service and that Ten-Tec at some later time decided to bias it
for class C. I have seen a few CW transmitter circuits in W1FB's Design
Notebook that use zero d.c. bias. The drive from the previous stage
turns on the transistor. Biasing for class C is fine for CW and FM
transmitters, but for SSB or AM, the driver and final would have to be
biased for linear operation, of course.
73,
Bob WB2VUF
On 1/29/2011 9:39 PM, Dr. Gerald N. Johnson wrote:
"Cheap" and "clamp on ammeter" don't come in the same package for DC.
.07 volts on the base of a silicon bipolar transistor won't turn it on,
.7 may turn it on too much. I find nothing in the manual to help.
73, Jerry, K0CQ
On 1/29/2011 8:05 PM, Bill Miner wrote:
Bob,
I believe that the driver and PA bias voltage should be set to about .70
volts not .07 volts. That could just be a typo from Paul at Ten Tec.
Not sure about what the total current to the PA board should be but that
seems like a better way of setting bias as the voltage would just be a ball
park figure.
Are you sure about the 30 mA and 140 mA values???? If you have access to
a clamp-on mA meter you would not even need to unsolder the power lead to
the PA board.
Thanks and 73, Bill - K6WLM
----- Original Message -----
From: "Bwana Bob"<wb2vuf@verizon.net>
To: "Discussion of Ten-Tec Equipment"<tentec@contesting.com>
Cc: "carlos popelier"<carlos.on6cn@gmail.com>
Sent: Saturday, January 29, 2011 10:51 AM
Subject: Re: [TenTec] century 22 setup driver and pa
Carlos:
To read the current, you can unsolder the red +12 V lead to the driver
and final board and insert a milliammeter in series with it. This will
give you total current into the board. I don't know if the 30 mA and
140 mA settings refer to total current or collector current, so this
would affect the procedure, which I don't have. If it's total current,
you set the current into the board to 30 mA, then go key down with no
drive and set the current to 140 mA. If the values represent collector
current, you will have to set the driver bias to min, read the current
into the board and adjust the driver bias for an increase of 30 mA, then
go key down and adjust the final bias to increase the current by 140
mA. I'm looking at the manual for my Corsair, which used the same
driver transistors (MRF-475) as the Century 22. The Corsair manual says
to set the driver bias to 4 to 8 mA, so the 30 mA must refer to total
current into the board. After I replaced my C-22 finals, the total
current into the board was 159 mA, which I reset to 140 mA TOTAL.
I wasn't sure about this procedure, so last week I contacted the factory
by e-mail, asking how to set the biases and Paul Clinton replied and
said that I should set the base voltage on the drivers to 0.07 volts,
then with the drive at zero, lock the radio into transmit and set the
final bias voltage to 0.07 volts. There is a test point pin next to the
base of each transistor. Here is the text of his e-mail:
"With the 579 in receive mode set the base voltage of the driver
transistors 0.07 volts DC using a digital voltmeter. Place the 579 in
LOCK with no drive and adjust the base voltage of the final transistors
to 0.07 volts DC."
This is a very much different procedure. My guess is that the C-22
driver and final were originally set up for class AB linear operation
and Ten-Tec later changed it to class C, since CW transmitters can run
OK in class C. Maybe someone else has some insight into this. I haven't
tried the 0.07 volt settings yet. My C22 is presently on the bench with
a possible intermittent short that causes excessive current drain in the
final. I'm still working on it, but will let you know when I get it
back on test.
By the way, Ten-Tec replaced the original drivers and finals with type
2SC1969.
73,
Bob WB2VUF
On 1/28/2011 4:08 PM, carlos popelier wrote:
hello,
current settings for driver and pa by tentec are 30 mA and 140 mA,
this is done by setup driver and pa potentiometers,
question: how can i measure this current values and the best way to do
it,
or the entire setup procedure.
thanks carlos
|