I should have posted a note about this earlier, but got real busy.
As some of you may recall, my 87A developed this minor problem after about
two years of service: in standby with key up, the lowest segment of my LED
Ip meter would stay lit all the time and the computer would report 70ma of
plate current. The symptoms were the same whether or not HV was connected to
the tubes, ruling out leakage. I traced the problem to an op amp that
measures the voltage drop across a plate metering resistor in the HV return.
I confirmed with Karl at Alpha that the chip input voltages were correct but
the output voltage was incorrect. The Motorola MC1536 op amp used in that
rev of the HV board is no longer made, but Scott Erhorn at Alpha searched
out a supplier and sent me a replacement within a week.
As for the Ip meter, the lowest LED segment is no longer lit in standby! I
can't tell you what a rush it was to see that. The CPU still reports about
30mA of plate current in standby, but Karl recalls that this is probably the
nominal reading. When he gets his hands on an 87A from my vintage he's going
to verify that. 30 mA is below the level required to light the lowest LED
segment and I don't recall what the CPU reported for standby plate current
when the amp was new. It's quite possible that this is the normal reading
and the result of an anomoly in the measuring circuit design. It's really of
no importance in operation of the amp. Perhaps coincidentally, there is a
small voltage drop of 30 mV across the 10-ohm meter resistor, which would
translate into 3 mA of current (again, that's with or without HV connected
to the tubes, so it's caused by something else -- perhaps the 36V bias
supply.) Maybe there's a x10 non-linearity in the measuring circuit at that
low level. That's well beyond my expertise.
Perhaps at some point I'll get in there again and measure the chip output.
I'm in no hurry to do so before I hear from Karl about the nominal reading.
No need to work around HV, especially when that annoying LED is out and the
amp is operating flawlessly.
BTW, while I had the amp open to replace the chip, I also installed the only
factory mod developed for my rev 87A, a 100uf capacitor across an op amp on
the HV board. That eliminates erroneous power supply voltage faults (fault
#24) by slightly slowing response in the detection circuit. I had only two
of these hardware faults in four years (one when using another amp in the
shack), so it wasn't an urgent fix. But it was very easy to do if you have a
low wattage pencil iron, a steady hand and aren't afraid of ruining a $6K
amp.
The service from Alpha was friendly, courteous, extremely helpful and
prompt. My thanks to Karl and Scott.
73, Dick, WC1M
--
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
|