[HBR] HR-10 to HBR project

Thu Oct 20 14:21:10 EDT 2011

Excelent plan, Walt. my hat is off to you.
73 Kees K5BCQ

---------- Original Message ----------
From: Walt Hutchens <waltah at earthlink.net>
To: HBR <hbr at mailman.qth.net>
Subject: [HBR] HR-10 to HBR project
Date: Thu, 20 Oct 2011 14:12:23 -0400

Yep, I did it -- got a Heath HR-10B with the intent to make an HBR out of
it.   

This project seems to make sense because:

1. The HR-10 was a 7-tube entry level ham band only superhet receiver kit
that was never a great performer.   Available through the 1960s it is
sometimes fondly remembered as a new ham's first receiver, but never as "a
great way to start" or "a good value for the money."   (One reviewer thought
the S-38D a better performer ...)  However,

2. The set has nearly all the hard to find parts for a decent HBR-type
design and many of the other parts, as well.   At average selling prices of
$35 at hamfests, $50-75 on eBay (plus shipping!) starting an HBR project
here rather than by casting a net for the individual parts will save a bunch
of time and money.  And because it IS a complete kit of parts, it's possible
to come up with a design that can be duplicated.

The HR-10 is a single conversion receiver having an IF of 1681 kcs and a
half lattice crystal filter -- the approach of the 'Long, SLOW ...' project
and probably the most straightforward way to do HBR-type construction today.
Among the foundation parts in the set are the power supply, IFTs, dial and
string-type drive (five bands, 80 through 10), S-meter, tuning capacitor,
crystal filter and maybe some tubes.   But much other stuff rides along,
from an output transformer to most of the tube sockets.

Perhaps the most important compromise is that the dial isn't the equal of an
Eddystone 898: It's strictly string drive, with a tuning rate of ~50
kcs/knob revolution.  (The CAL control can be used for fine tuning).  But
then it won't cost you $100 (plus shipping) and five hours of painstaking
work to lay out and install, either.  It is already calibrated and looks
good.

There is a review of the HR-10 in QST, July 1963:  I would not call it
highly favorable.   My unit looks decent and shows decent (beginner)
construction skill.  It has the sort of mechanical issues you expect in a
Heathkit of this age -- dragging dial pointer, stiff tuning, bent needle on
the meter (but the movement is okay), and doubtless more to be noticed when
it's turned on.  It has mostly Daystrom tubes and does not appear to have
piled up many service hours.

This kit seems to have been Heath's answer to the popularity of the W6TC
HBR-series.  It is somewhat the same approach to receiver design, with the
exception that in the Heath market bandswitching was essential and costs
were so tightly controlled -- the kit sold for $70-90 -- that performance
was seriously compromised compared to the real thing.

Geezzz ... if Heath had spend another $8 on parts, you wouldn't be able to
touch these for $200 now.   And you know those engineers knew that.

One could ask "Why not just improve the HR-10 as what it is and gain the
convenience of bandswitching?"   That has been attempted by others (a web
search will find details); the problem is that the HR-10 is limited by
fundamental design compromises that are not practical to fix (as a
bandswitching unit) within the boundaries set by the power supply and
physical size.   

1. Coils are low Q -- small, slug tuned, paper forms -- and there's no great
fix for that, short of much larger coils that wouldn't fit in a
bandswitching configuration on this chassis.   This lowers gain on every
band and that's the GOOD news, because with low Q, overload from adjacent
channels is a serious issue.

2. Second harmonic injection is used on 15-10M but with a wimpy oscillator.
Together with the low Q coils and very limited overall gain, this makes the
receiver pretty much deaf on those bands.   Even hams who used the HR-10
with a beam found 10M to be 'quiet.'

3. Overall gain is minimal, with spec'd sensitivity of 1.0 uV.   That works
in this design (if you don't demand sensitivity on 20-15-10M) because of the
...

4. Very poor AGC setup  With only two controlled stages -- the 6BZ6 RF stage
and the 6EA8 2nd IF -- you've got at most some help maintaining a
comfortable listening level on a round table.

These two controlled stages are respectively a semiremote cutoff tube (CO at
about 20 volts) and a sharp cutoff one (CO at around 9 volts.)   There is a
manual RF/IF gain control; by backing it down on strong signals you can
prevent overload of the 2nd IF but AGC effectiveness is further reduced.

As a rule of thumb it takes three controlled stages to get good AGC action,
that is, really strong signals sound stronger than those that barely move
the meter but they don't put your ears or speaker cone at risk.   Simple AGC
systems should reduce gain comparably in all controlled stages to avoid
pushing one to cutoff (distortion!) while others are still amplifying, so
mixing sharp cutoff and remote cutoff types is generally a bad idea.

5. Complete lack of AGC for CW.   Yep -- this design is that old.   AGC is
used on SSB but has only a backup function: The instructions tell you to
turn up the audio and control the volume with the RF gain control so you
know the AGC isn't doing a whole lot.

6. No product detector.  They just shoot the BFO into the diode detector, as
in a 1940s design.  

7. AGC action shifts the LO frequency via the shared plate power supply.
Any triode oscillator -- the HR-10 uses a triode-pentode as oscillator and
mixer -- is going to have that problem unless the supply is exceptionally
stiff or a separate LO supply is used.

Each of these problems can be addressed in an HBR-type approach, working
with the basic HR-10 parts.   We will have to keep looking over our
shoulders at the power transformer: While the set is rated at 50 watts
dissipation, transformer output is probably closer to 35 and filaments take
~25W of that.   Fortunately, savings are available in several places so the
necessary functions can be added without pushing the xfmr.

A big plus:  The ~11" x 13" chassis has lots of room.  With careful layout
at least a couple of tubes could be added.    Plug in coils?  With the
removal of the existing coil/bandswitch assembly, there should be be plenty
of space.  

Another plus: The existing coils can simply be duplicated as plug ins -- a
coil wound to resonate at the same frequency as the one in the set will
track the dial.  In the original set there are no trimmer caps, even on the
oscillator:  The designers chose fixed caps and calibrated the dial
accordingly, leaving only the adjustment of the slugs in the coils and the
front panel calibration control: That would work as well for an HBR-type
design as it did for the HR-10 and it's worth exploring since it saves
finding a hatful of APCs.

My guess: 'Trimmerless' coils will be fine for the mixer and RF but APCs
will be needed for the oscillator.  The HR-10 has temperature compensating
caps for every oscillator coil and these (25-50 mmf, N750!) will be far too
large for coils mounted above the chassis.  Once you have to find precise
fixed cap values, APCs start to look better.

(The Long, SLOW receiver was overcompensated on 80M with 2 mmf, N750 located
near the oscillator socket under the chassis ...)

The oscillator is a Hartley -- a good choice for a set of this type. A new
front panel and cabinet will be required but those are standard issues, no
tougher with this chassis than any other.

The HR-10B schematic can be downloaded here:

http://www.vintage-radio.info/heathkit/

And the entire assembly manual (pdf) can be downloaded here:

http://www.mods.dk/manual.php?brand=heathkit

(Free membership required for this one.)

The obvious first steps are

1. Get the receiver working as it stands in order to avoid 'oops -- bad
part' surprises later, and,

2. Make a plan.

I've hardly a clue about the plan.   The extremes are 'convert to plug in
coils using existing circuitry' -- i.e., keep most of the worst features! --
and 'strip the chassis, make a new top plate, and ...' which is the biggest
job.   Probably the circuits would be somewhat like what I've been working
with but emphasizing ease of duplication.

Just using a transformer-type power supply will eliminate most of the ...
challenges of the 'Long, SLOW ...' project.  There will be other issues: The
QST reviewer noted audio hum that passed through a minimum as the volume
control advanced.  I.e., TWO sources of hum, opposite phase.   The filament
return -- 3+ amps total -- is via the mild steel chassis which is also
signal ground, so ...

On a very quick look, rearranging things on the existing chassis will be a
challenge.   For example, the BFO now shares an envelope with the 2nd IF.
You can't do that, use a product detector,  and have AGC with the BFO on
because there's way too much coupling into the IF -- the AGC will be driven
by the BFO.  One logical place to move the BFO is the triode half of the
audio output tube, but that's half a chassis away from the BFO coil and
tuning cap.   

That suggests a  partial top plate.   But the IFTs and BFO coil are the type
that mount with thorough-chassis clips and that's a nasty hole  to make with
hand tools.   There were adapter plates (for a round or square hole) that
came with these transformers when they were sold for replacement use but
those are hens teeth now.

The front end needs to add three holes for coil sockets plus one for a
separate local oscillator tube.   I think there's room, but rules like don't
heat up the tuning cap, keep stuff away from the oscillator coil, and don't
let the antenna coil see the mixer coil will complicate matters.   And the
'plenty of space' is very evenly chopped up.  It's not going to be easy to
find room for new and relocated stuff.

Things are always awkward at the start.   The 'Long, SLOW ...' set needed a
complete redo on a new chassis even though I had spent days scratching my
head about the layout of the first one.

It'll be a while before this gets very far: While the LS project is winding
down there are still important odds and ends to clean up, four more bands
worth of coils to wind, and both a coil storage and receiver cabinet to
make.   

Lots to think about.  Suggestions welcome!

Walt 
KJ4KV

______________________________________________________________
HBR mailing list
Home: http://mailman.qth.net/mailman/listinfo/hbr
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:HBR at mailman.qth.net

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html