[HBR] pondering an HBR

[Walt Hutchens]

Chris Howard posted:

> About a month ago, at the Jackson, MS hamfest, I picked up a homebrew
> receiver from a table full of interesting stuff.  It was my only
> purchase  that day, $5.  I just bought it because I hate to
> go home empty handed.

Thanks to a swap deal I now have that receiver sitting on my work table.
Chris's pictures are at:

http://w0ep.us/OM/?p=2244

As a whole, this is a truly extraordinary piece of work.   As Chris has
discussed it doesn't work -- at least, not very much -- but the mechanical
design is quite unusual and the electronic design is perfectly competent.

Tube lineup:

6BZ6  -- RF
6CS6  -- 1st mixer
6C4     -- LO
6BE6   -- 2nd mixer
6CB6   -- 1st IF
6CB6   -- 2nd IF
6AU6  -- xtal calibrator
12AU7 -- 1/2 is S-meter driver, other half unused
6T8     -- AM detector/AVC/1st audio
6AQ5  -- 2nd audio 
6AK5   -- some audio function but ???
6C4      -- BFO
5Y3      -- Rectifier

It's a double conversion superhet; the first IF being 1415 kcs with a
crystal controlled conversion to 85 kcs.   With one crystal above the 1st IF
and the other below, you can switch sidebands easily.  War surplus FT-243
holders, one marked 1330 kcs  (1415 kcs - 85 kcs), the other I can't see the
markings but it should be 1500 kcs.

The BC-1306 and GRC-9 used crystals running from 1 mcs to 4.5 mcs in 10 kcs
steps (I think ... little hazy on the high end) and these were widely
available in the 60s and modestly later.  The crystals were used in the
transmitter which doubled and maybe tripled to get the output frequency.
Big chest with a bunch of drawers full of the things.

There's a 100 kcs crystal calibrator.    That tube is missing.

All but one of the small knobs are BC-342 parts; the one might be Heathkit.
The basement box-o-knobs has supplied a proper (BC-342) replacement.

The tuning cap drive is all zero backlash gears and it looks to me like the
builder may put that mechanism together from parts -- a real achievement, as
anyone who has tried to get proper gear mesh without a machine shop can
attest.   Some of the gears might be from the BC-342 but some clearly are
not and the mounting is different.

~12 turns of the tuning knob to go from one end of the dial to the other.
Not outstanding, but slow enough to allow receiving SSB.  As with many gear
drives there is no flywheel.  This knob is a large fluted style

The dial is a string drive for a spring wire pointer that slides on a rail.
This was competently done and works fine.  Dials like this are "That's about
the frequency" anyhow (scale is only maybe 6" long) and with zero backlash
gearing driving the tuning cap there will be no backlash!

The dial calibration is on a paper scale rolled around a plastic drum maybe
1" in diameter.   This is intended to be driven from a drum on the
bandswitch shaft (below the chassis) by a single cord that winds around a
pulley on the end of the dial drum.   The bandswitch shaft rotation pulls
the cord against a clock spring arrangement on the end of the dial drum, so
it rotates as you switch bands.   This setup is disconnected and may never
have worked properly as the drive drum is about 2-1/2" diameter while the
driven one (drum shaft) is only about 3/4" dia.   So 1/3 or so revolution of
the bandswitch shaft (to cover all bands) would give you  more than a full
revolution of the drum.

It will be something of a reverse engineering project to figure out exactly
how this mechanism should work and I'm thinking some change in the drums may
be needed, to give a step-up ratio of more like 1:2 so the bands will be
separated enough to allow room for the scales for each band without going
clear around and overlapping one or more other bands as now occurs.  I hope
the junque box will supply a drive drum of the right size and if not, I
think I can make one of plywood or something.

The dial set control is a cam operated by a knob to the right side of the
dial drum shaft.   As you turn it, the dial drum is pushed sideways against
a coil spring, thus changing its position relative to the dial pointer.

These arrangements probably are not unique but I've never seen anyone try to
home brew them.   

The LO is tuned by a single very large -- maybe 250 mmf? -- capacitor.
That's way too much for a ham-band-only set and as a result there's serious
non-linearity in the calibration.  It appears that the mixer coils aren't
tuned at all while the antenna coils are tuned with a separate knob.   You
can build a receiver that way but it's not going to have great performance.
There is room for a three gang cap in place of the current oscillator unit
and the separate antenna coil tuning control can become the usual antenna
trimmer.   

What's really needed is a three-gang bushing mount cap of the usual 15-20
mmf/section size but I don't think I've ever seen one of those.   So the
capacitor mounting scheme will have to be changed to accept the more common
FM receiver style cap.

There are five marked bandswitch positions: 80-40-20-15-10.   Only three
oscillator coils are installed; perhaps some are used for more than one
band.   There are no holes for others.  There are four mixer coils and five
antenna coils.  

The coils will have to be redone from scratch when the tuning cap is
replaced.   However they're pretty decent quality 3/8" or so slug tuned
phenolic and I think they'll be okay.

Mechanical construction is good throughout: Special credit for a very nice
job cutting the long slot (with beveled edges, yet!) for the tuning dial.
This may have been a salvaged panel but I've no idea from what, and it has
neither marks clearly indicating that the design was adapted to the panel
nor extra holes: I'm thinking the builder did it from blank.  Electronic
work -- soldering, choice of parts, etc. -- fair generally but poor in
places.   There's a moderate amount of corrosion, mostly under the chassis.
One transformer has been slightly mouse-chewed.   Looks like an output
transformer: One winding is disconnected so maybe it was used as a choke?
There is another output transformer.

I don't think this set ever worked all the way through on all bands.
Although all the bands except 20M are calibrated.

Getting this thing really working is going to be a big job; it'll need a
complete re-kitting after diagramming everything.   It's worth doing though
-- the plan was both remarkably creative and competent and the execution
fundamentals are mostly good to excellent.  The result should be a receiver
that's the equal of most mid-priced ham sets of the 60's:  Since the tuning
is done at the channel frequency (rather than in a 1st tunable IF) it'll
never be an R-390 and the small coils will keep it somewhat below the ranks
of a good W6TC project, but it should be plenty usable.

I don't see a product detector but there's plenty of room since two tubes of
the existing complement are non-essential.

The project will have to wait until the current two HBR jobs are done,
probably in the fall.

Does anyone recognize this design?   I know it isn't in any of the usual
handbooks but maybe written up in QST?  Or ???   I have seen no identifying
markings -- no call letters, no nothing.

Does anyone remember what  years the El-Menco paper caps in ceramic tubes
with hard wax filled ends were on the market?  This thing has a bunch of
them.  

Walt Hutchens
KJ4KV