On Fri, 14 Nov 1997 07:20:06 +0000 w8jitom@postoffice.worldnet.att.net
writes:
>> To: amps@contesting.com
>
>I wrote:
>> >That's right Carl, using a series resistance and a shunt diode for
>> >bias is just plain dumb.
>
>Carl sez:
>> But the manufacturers keep using it even when Motorola and TI at
>leas
>> have published good bias circuits.
>
>They will continue to use it until more people start to care about
>signal purity on VHF, and let them know they care.
>
>> I suspect the original "pattern" started out as AM CB for the HF
>bricks
>> and FM for VHF.
>> The "engineering" in either case did not have a clue about linear
>> operation of those biplolar devices. When SSB started to become
>popular
>> on VHF it was "one size fits all" .
>
>Actually the diode works very well if the stage has very little RF
>drive induced current. It is the rectified RF signal that tends to
>drive the bias voltage negative that ruins the linearity, if the
>quiescent current through the devices shunting the base is ten or
>twenty times higher than the RF induced base current, the cheapo
>system works great.
>
>Look at the series resistance most people use. It is about 50 ohms.
>
>13.8(source)-.7(bias)=13.1 volts drop. 13/50 is .26 amperes bias
>current. If the RF induced base current is over .026 amperes the bias
>will usually waddle around too much. If the collector current is 15
>amperes peak, and the beta at the operating frequency is 10, the base
>current will be 1.5 amperes.
>
>1.5 amperes will force the base bias far negative, and the
>device into class C.
>
>IMD will actually become much "less bad" if bias is removed and
>the base is grounded through a low dc resistance choke!
>
>73, Tom W8JI
I certainly agree Tom,
Motorola lists such a wide range of Beta for a specific device, MRF-247
as an example, that it takes a good bias circuit to handle the worst case
condx of maximum base current.
What special parameters do those SRF House numbers have?
In the amps I have worked on here recently (TE, Mirage, RFC) the ~ 0.7V
of bias rapidly decreases much beyond about 10% of rated drive. At 30-50%
of rated drive the bias is zero. Measured base current in the 150-180W
bricks that I have data on (limited, granted) is in the 1.1 to 1.7A range
I use a pair of 7A T0-220 devices (2N6121 or equiv) in a Darlington
config with a 78L08 providing the reference from the 13.8V keyed line.
The temperature compensating diodes are the good old 1N5408. Basically a
junk box copy of a 1988 TI manual circuit that was described in a VHF
Conference book in 1995. Probably a bit of overkill, but parts are cheap
and the circuit rugged.
73 Carl KM1H
--
FAQ on WWW: http://www.contesting.com/ampfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|