[HBR] Another HBR Project -- Chapter 5 -- Pictures

[hbr@mailman.qth.net]

Kees wrote:

>  Since you are using the 85Khz cans in pairs how is the sound on SSB
> ? AM ? with such a narrow passband or are they stagger tuned ? 

Can't say about the sound until it can hear signals -- probably at 
least a couple of days, the way things have been going.   But the 85 
kcs cans (4 total, in 2 pairs) have adjustable coupling.   With the 
coupling minimum -- basically that works out to critically coupled, I 
think -- the 6db passband looks like around 1 kcs.   Or maybe a trifle 
less.   The ARC-5 receivers were to be operated with the coupling at 
maximum and for that condition, the manual gives a passband at 6db 
of 2.2 kcs.   Going from three transformers (as in original sets) to four 
should have little effect on that number but should steepen the slope 
further out.   So if aligned on the nose it should be about ideal for 
'congested conditions' SSB reception but a bit narrow for AM.   
Stagger-tuning by 1 kcs or so ought to take care of that.   The builder 
desiring a CW receiver could leave the coupling at minimum.

The coupling adjustment consists of a plastic rod that sticks up 
through the center of the can top under those plastic caps.   Pull up 
for minimum, push down for max.   The mechanism isn't very sturdy --
 it probably has a useful life of around 50 down-up cycles -- so it's not 
suited for operational use.   The intent was (I believe) just to simplify 
alignment:  pull rods up for minimum coupling, peak all trimmers, 
push rods down.   Loading resistors (to 'kill' the other winding during 
alignment) were not practical with these sets due to very poor 
access to the underside of the transformers.

> I like the command set plug-in coils from underneath used on the
> front panel, good shielding, small, great plugs. I never saw an
> article suggesting that in QST/CQ/HR. 

I don't remember seeing the idea anywhere.   However it has a few 
'issues' and disadvantages:  (1) Not all receivers have enough pins on 
the plugs to do the job right.  In particular you need to be able to plug 
in the trimmer and padder caps as required for each band.   That 
takes (for example) 5 pins on the oscillator coil -- 2 for each winding 
and one extra for a padder cap.   It is possible that only the *rare* 
ARA receivers have the full set of pins; I haven't studied this.   (2) 
With the full compliment of pins, the insert/remove force is 
considerable.   That means the plug bracket in the set must be rigid 
and very sturdy and some sort of handhold is required near the coils --
 that's the handle below the coils in the front picture.   (3) One must 
be wary of shock hazards should it be possible to tilt the coils so the 
can touches a B+ pin when inserting/removing.   (4) Shields must be 
firmly grounded -- that's the reason for the brass thumbnuts on each 
side of the coils.  (5) I can't think of a clean way to actually put 
trimmers inside those cans.   Accordingly I plan to optimize them by 
selecting fixed capacitors -- this can be done with the caps outside 
the coil can, then installing them inside the can so they get changed 
with the coil.   A panel trimmer on the oscillator will allow calibration --
 that's the 'SET' knob, to the right of the tuning knob.  

I think it's a novel and somewhat appealing idea but I wouldn't be 
ready to recommend it as a *good* idea!   

Ah -- I do have an idea for trimmers!   When using a toroid, there'd be 
plenty of room to make a small trimmer adjustable through a hole in 
the bottom of the can.   But I'll probably try the set and forget 
approach first.   

I'm going to be very interested in seeing how the warm up drift (and 
operational stability) turns out.  On the one hand, things are pretty 
spread around, some of the oscillator parts are above the chassis, 
some below, some leads are pretty long, the coil mounting and plug 
structure can expand/contract -- all unfavorable.  On the other, 
though, the main heat sources are pretty well away from critical parts 
and the airflow through the set should tend to keep the oscillator coil 
can -- containing the coil and two of the most important fixed 
capacitors -- near room temperature.   

More detail:  The audio output tube mounting will be on a plate 
slightly below the chassis with the tubes projecting through 
clearance holes in the chassis.   So the audio stage (most likely 2 
6AQ5's) will suck air out of the chassis.   The oscillator and mixer 
tubes are mounted on 1/4" spacers above the chassis so the pins 
project down through -- they too will draw air up and out, cooling their 
wiring (the main source of heat reaching stuff connected to the 
oscillator tube socket) at the same time.   With the bottom covered, 
the only significant inlet is around the coils -- about a 3/32" gap on all 
sides.   The tuning cap frame is heat sunk to the front panel and I'd 
bet that's a pretty stable device in any case, what with glass 
insulators, spring mounting, and fairly wide spaces to the frame.   

The worst thermal design glitch I ever saw was National's NCX-5 
transceiver, with a wonderfully intricate and expensive tuning cap 
(bronze gears yet!) and an elaborate compensation scheme -- but the 
oscillator mounted in a big hole at the front of the chassis, the 
chassis was typically tilted up by longer feet at the front of the set, 
and there were three sizable power resistors *under the chassis*.   
They put maybe $100 on the cost of the set with that oscillator 
design and then used it for a chimney for under-chassis heat.   It's 
not hard to see why National fell out of the ranks of ham gear 
manufacturers.

Walt 
KJ4KV