[Amps] Looking for Alpha 87A expertise

[Jeff DePolo]

A friend's 87A showed up on my bench.  The symptom is fault code 1 (receive
PIN back bias voltage not at minimum required level with key down and no RF
drive) as soon as the key line is closed.  This 87A has one of the newer HV
boards for what it's worth, with a zener diode string that totals 900V for
the RBIAS supply.  The receive PIN diodes are/were KS1001 with a 3.9M
resistor across them.
  
Long story short, I found one of the RBIAS gating transistors on the HV
board was bad and replaced it.  I now have 900V on RBIAS unloaded (i.e. with
the "R" terminal disconnected at the HV board), and 730V when the R terminal
is connected.  Those values were measured with the operate/standby switch in
standby which puts reverse bias on the receive PIN's even though the RF path
through the amp is bypassed via the mechanical relays while in standby.
When switching to operate, RBIAS goes low (a fraction of a volt) indicating
that RBIAS is now switching on and off as it should.

However, one of the receive PIN's measured appreciably different than the
others when testing out-of-circuit, so I replaced it, and its paired mate,
with 1N4007's and retained the 3.9M resistors across each.  Receive-path
loss through the amplifier is low whether in standby (relay bypass) or
operate as measured with the VNA (and before anyone screams bloody murder,
the HV was disconnected from the tubes during that measurement).

The RBIASMON line going back to the 68HC11 for monitoring sits at 3.56V when
RBIAS is active.  That's measured right at the micro, after the op-amp
buffer and everything else.

The microfuses on the output T/R board are good, as are all of the passives
on the board that I can test in-circuit.

Still no joy - amp throws fault 1 as soon as PTT line is closed.

Questions:

1.  Does anyone know what is considered a "good" voltage for RBIASMON?  I'm
unsure whether the port used for RBIASMON on the 68HC11 is A/D or logic I/O
(the port can be used for either).  If the former, what is an acceptable
range for RBIASMON to be considered good?  If the latter, 3.56V should be
just barely high enough for a logic high, but not by a comfortable margin
(spec is 0.7 x VDD for logic high, so 3.5V with 5V VDD).

2.  Is a drop from 900V open-circuit to 730V typical when
disconnecting/reconnecting the "R" bias lead at the HV board?  I'm wondering
if I have another leaky PIN diode or cap to ground that is loading it down.
It doesn't take much to load the bias supply given that it inherently has
880k of series resistance.

3.  Is it desirable or necessary to have the 3.9M resistors across the PIN's
when replacing them with 1N4007's?  I am unclear what their purpose is -
does it affect switching speed?

4.  Any other words of wisdom before I get back to the project later today?

Thanks.

					--- Jeff WN3A







-- 
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus