[ARC5] PS Hum in BC-453-B (solved!)

[Leslie Smith]

           Power Supply Hum - problem solved

I would like to thank all the readers who helped me find (and remove)
hum in a BC-453.
In particular a single posting by Elden Meyer was most helpful.

Detail.
The problem turned out to be obvious, once the proper observation was
made.
The common earth line carried BOTH 24V AC at 450mA (to filaments) & AF
(to the headphones).  Once I separated the current flow into two
separate conductors the hum dropped, (and even tolerable).  There is
still a small amount - but it's now coming via the 12A6; it appears
about 15 sec after I turn the set on.


False leads.
(1) First, I run my set from a home-brew PS, where I double the 24V
filament line to generate +60V for the B+ line.

(2) I understood the possibility of getting induced hum 'down' the
common earth line, and even provided a second common for the lower level
AF signal.  Hum wasn't a problem with a loudspeaker.  When I connected
modern 'phones, I continued to put the heater current & AF thru the same
common cable (external to the set, via J3 on the rear panel).  -
Stoopid.

(3) I measured the B+ line with a 'garden variety' DMM.  B+ voltage was
59V - marginally below the +60 I expected.  I believed hum was coming
down the B+ line - in other words the unfiltered supply fed to the zener
didn't have enough head-room.  Wrong.  When I used a Fluke meter I
discovered the B+ line was +60V, and clean.

(4) I noticed the hum appeared the moment I turned on the power - before
the 12A6 heater could even warm a little.  Elden wrote:

  "Since you notice hum before the tubes have warmed up,
    the 12A6 should not yet be drawing plate current. In that
    case there are other possible sources  [apart from 
    a common conductor] for AC to get into the audio."

And a little further on:
   "Also there can be voltage drop in a wire or ground 
    connection common to power and audio circuits."

It's obvious, now, isn't it?
Thank you EM!

Les

Now I must track down the fault in L.O. 
It doesn't start easily.
Some have suggested 'tin wishkers' in the tuning gang.
I doubt this, 
  (1) because I see the signal (on the plate) diminish from approx 20V
  to zip
      as I tune the set from 200 to 280kHz.
  (2) If the set doesn't oscillate (even at 200kHz) I can 'shock' it
  into starting by flicking the P.S. B+ line.
  (3) I replaced the 12K8.  Three times. (But that still doesn't prove
  anything).
  (4) The voltages around the 12K8 socket appear "about right".
        Pin #       Voltage/comment
         1          Metal shield - looks good.
         2          Heater - just under 12V, looks good.
         3          Plate - mixer.  +60V (approx) seems ok
         4          +60V.  G2+G4.  seems ok
         5          Oscillator grid.  Cant remember the voltage.
                    The grid lead has gone high - from 51k to 60k.
         6          Oscillator plate.  1/2 B+ (about right)
                      CRO shows LO signal at times.
                      Amplitude 16 to 20V AC, drops with incr.
                      frequency.
                      Above 300kHz, the oscillator starts for a moment.
                      So - get heaters warm, toggle B+, watch signal
                      appear.
                      Signal decr. over about 6 seconds - then zip, it's
                      gone.
         7          Filament voltage, 24V AC.
         8          Cathode, cant remember the voltage.  

Dave Stinson's hint:  Check for oxidation somewhere - esp in the tuning
"C".
My hint:  Use CRO carefully, remember Mr. Ohm, think carefully.


Les

-- 
http://www.fastmail.com - Does exactly what it says on the tin