On Mon, 30 Jul 2012 23:59:10 -0500, flipper wrote:

On Mon, 30 Jul 2012 13:59:54 -0500, legg wrote:

I'm working on a Harman Kardon FM100 "Counterpoint II" that shows
signifigant variations, internally, from the documentation provided by
SAM's HF3 repair and servicing manual, or other references searchable
on the web.

Well, I can really answer why it's broke but some general
observations.

For a start, the valve pins are reversed, by section, in V1 and V2,
from the SAM's schematic. The tube sections are swapped - confusing,
but not a biggy. Second, the tube types differ from SAM's and all
other references.

Can you find a manufacture date on it? The SAMS may be the 'original'
version and they couldn't know what would be done 'later'. Or you may
have a 'pre-production' unit.

The board art itself is copyright 1956.

Anecdotal info suggests that ECC85 (6AQ8) was developed and marketed
after ECC81/12AT7, and that ECC85 was developed specifically for VHF
input/oscillator combinations. An improved inter-device screen
supposedly reduced oscillator stage radiation. This isn't an
input/oscillator schematic application, in any event, it's an
input/mixer combo - same difference? Not sure what 6BK7A does that's
special or not except for the biasing differences.


1) V1 dual triode in the RF amp and mixer stage are the older
6AQ8/ECC81, rather than the later 6BK7A/ECC83 expected in the SAM's
schematic and listed in other web references.

The two are listed as 'substitutes' for each other but I do think the
6BK7 is 'preferred'. Both should work

I got this a little bassackwards in my description. 6AQ8/ECC85 (!!) is
used in the documented 12AT7/ECC81 (!!) and 6BK7 positions, for the
functions noted.

The newer part suggests a later rev. If it's an early one, perhaps
they dumped the 6AQ8 because of higher cost or poor availability.

2) V2 dual triode in the AFC and oscillator sections is 6AQ8/ECC81
rather than the 12AT7 expected in the SAM's schematic and web
references.

3) The printed circuit board has the correct heater connections for
the single 6V heater of 6AQ8, without any rework in evidence (pins 4/5
receive 6V normally - same as the other 6v heaters, pin9 is grounded).

Well, that's obviously different, and not pin compatible, but the two
types are 'similar' and rather close in both mu and gm (6BK7 requires
more heater current). I mean, you could use either when designing, but
not swap once the traces are down. See below.

4) Cathode resistor of the grounded grid 6AQ8 RF input amp is 91R, vs
the 68R expected. No note in SAM's about variations here.

Sounds like a bias tweak, maybe due to the slightly different tube
being used, but it probably doesn't make much difference one way or
the other and might have been done, regardless of the tube, just to
lower current drain.

gm would likely change a bit, perhaps altering gain a bit, but neither
should make it 'work' or 'not work'.

For self-bias, the cathode resistor for 6AQ8 is supposed to be roughly
4X that of the 6BK7, so an increase here would be expected. It's in
the AFC circuit

5) Plate resistor in V2 oscillator is 6800R, vs the 1000R expected in
schematic. No SAM's note on variations here. This seems to be an
extreme circuit change.

The 6AQ8 pulls significantly more current per plate volts than the
12AT7 so the higher plate resistor cuts plate volts to keep current
down. I don't know if that value makes it 'the same' as the other tube
but being a higher value makes sense. It's also not 'operational'. I
mean, it's bypassed and, so, just setting B+ to the plate, which is
probably lower than SAMS indicates because it doesn't need as much
plate volts to get the same current draw. Gain might be a little
different but it doesn't really matter for an OSC, as long as there's
'enough', and they're usually overdriven anyway.

Easy enough to find out. Is it oscillating?

Btw, note that if AFC is screwed then so is the OSC and everything
else.

That seems to be the issue. Oscillator is not oscillating. AFC control
line seems to be functioning, in simple IF signal tests.

The non-functioning mixer self-biases at half the schematic grid
voltage value and 2/3 the plate current - but expect this when input
signal is missing Grid resistor is 20% low with age/dirt. Oscillator
grid resistor seems low in schematic at 22K, but what do I know.

The 22k sounds pretty common. Why I don't know but you see it
everywhere. It's bypass and, so, out of circuit at RF.

I also see that in typical schematics of the time.

As this is the section that seems to be malfunctioning, I'm wondering
if there's any advice related to the earliest versions of this tuner
that might ease in reviving it. Tubes test functional for heater,
emissions and transconductance, if a little slow to warm up.

RL

What is the 'malfunction'?

RF front end went out, on the shelf, 2yrs after last power-on. (It is
mono, after all.)

This is the first time I've actually paid much attention to the
innards - was happy to just leave it alone, as it warnt broke.

I don't have an FM signal generator or a controllable source (like a
wireless audio link). So I've been limited in what I can actually
check.

The SAM's IF and discriminator calibration procedure, using 10.7MHz cw
after the mixer, showed it to be pretty out of tune, but functional in
those sections.

No 6AQ8's in my spares cabinet.

RL