This is only a preview of the April 2023 issue of Silicon Chip. You can view 39 of the 112 pages in the full issue, including the advertisments. For full access, purchase the issue for $10.00 or subscribe for access to the latest issues. Articles in this series:
Items relevant to "500W Class-D Mono Amplifier":
Items relevant to "Wideband Fuel Mixture Display Part 1":
Items relevant to "Automated Test Bench":
Items relevant to "Silicon Chirp – the pet cricket":
Purchase a printed copy of this issue for $11.50. |
Vintage Radio
Browning-Drake Model 6A from 1927
By Dennis Jackson
I’m fascinated by early radio sets, especially
regenerative sets from the 1920s. BrowningDrake is a US company that made
innovative radios in the 1920s, and this set
is one of their later models, which evolved
over the years to become solid performers.
Shown here is the model 6A with a large
Utah horn speaker that came with it.
W
hen listening to the conversations of older relatives who had
made their own wireless receivers
during the 1920s, I remember being
impressed by the names Browning-
Drake (B-D) and Radiokes.
Browning-Drake made tuned radio
frequency (TRF) wireless sets while
Radiokes made tuning coil sets. There
were none better, according to my
father and some of his brothers, who
had lived solid physical lives working
as farmers and builders and judged
things on their merits.
Radiokes were a Sydney-based company that manufactured boxed sets of
tuning coils for various receiver configurations, plus other desirable components that were mainly aimed at the
94
Silicon Chip
amateur constructor.
The story of Browning-Drake is well
documented. In August 1923, Glen
Browning and Fredrick Drake were
students at Harvard University in the
USA. They were asked to explain theoretical losses in the wireless receivers of the times.
Accordingly, after a long detailed
mathematical study, they concluded
that the major losses were due
to unwanted capacitive coupling
between the primary and secondary
windings and within the windings of
the tuning coils between RF stages.
The regenaformer
Their solution was to develop the
“regenaformer” transformer.
Australia's electronics magazine
The secondary consisted of 74 turns
of enamelled copper wire wound on a
75mm Bakelite former that was spiral
threaded so the windings were spaced
one-half of a wire diameter apart.
The primary consisted of 24 turns
of 30 AWG (10thou/0.25mm diameter)
silk-covered fine wire loosely wound
in a slot cut into a ring and placed
firmly inside the tube, level with the
Earthy end of the secondary. The aim
was to reduce capacitive coupling.
The feedback or tickler coil is
wound on a 60mm former placed in
the other end of the secondary former
that is free to be rotated 180°. This
feedback winding used 20 turns of 26
AWG (16 thou/0.4mm diameter) wire
to provide a controllable amount of
siliconchip.com.au
Opening the front of the model 6A’s case reveals five control knobs and primary tuner. From left-to-right, the knobs
control power, variable capacitor C2, station selection, valve filament voltage and feedback coil in the regenaformer.
feedback regeneration from the plate
of the detector to the grid of the first
audio valve.
Browning and Drake’s main contribution to the regenaformer was the
placement of the primary winding
within a slot fitted at the end of the
secondary, with a view to reducing
unwanted capacitive coupling causing RF losses. Hazeltine’s balancing
circuit was used to minimise plateto-grid capacitive effects within the
first RF valve.
Howard Armstrong had developed the concept of regeneration. To
get around his patents, the complete
regenaformer, its associated tuning
capacitor and tuned aerial coil were
initially sold as a boxed kit to people
building their own radio. B-D receivers were popular with amateur constructors as they could wind their own
regenaformer and the other parts were
generally standard items.
Interestingly, the variable tuning
capacitor used in the first RF stage
had a higher capacitance (400pF) than
that used to tune the regenaformer
(300pF). Maybe this was to compensate for aerial loading. I have noticed
a tendency for stations to crowd the
lower end of the tuning range on other
B-D sets.
Complete factory-built B-D receivers were available by the mid-1920s,
and I had the good fortune to acquire
a B-D model 6A from the USA around
2008 after I saw it advertised on eBay.
I rather impulsively placed a bid for
$250, which was knocked back due to
not meeting the undisclosed reserve.
I also had to consider the freight cost
of around $200 at the time.
I was a bit peeved by missing out
on what would have been a once-in-a-
lifetime chance and set about doing the
next best thing, as many an old-time
siliconchip.com.au
amateur would have done, by building my own.
I saved the photos used in the advertisement and whatever other information I could find. I had almost completed the RF section, ready for testing, when an email arrived. The seller
had a rethink, and as I was still the
highest bidder, I was given a second
opportunity.
I lost no time paying up through
eBay, and the model 6A duly arrived
through the back gate to preserve matrimonial bliss. The mid-west USA had
been subjected to severe blizzards,
and the unfortunate seller needed
to buy shakes (wooden shingles) to
repair his roof. I upped my payment
a bit for the goodwill, and he added
his big Utah horn speaker into the
bargain.
The 1927 model 6A
The six-valve model 6A was a complete rethink compared to its basic
five-valve predecessor, the model 5R
from late 1926. It is a table set built of
solid timber, probably poplar, which is
light, soft, workable and stains well for
an attractive finish. The double doors
in front open to display the timber
control panel.
By 1926, dedicated output valves
were becoming available such as the
71A and the CX112A, the latter used
in this set. These gave a modest but
welcome rise in sound output compared to using more general-purpose
valves in the output stage. Still, listeners would have to wait several more
years before the moving coil speaker
(like we use today) provided a broader
range to the audio spectrum.
It was common for these pioneering
wireless receivers to have all the same
types of valves in the line-up. UX201s,
UX199s or the Phillips B405, B409
and A609 were the main types. This
set uses a bit of a mixture; the UV199
and UX201A were from General Electric (GE), the 200A was a generic type
made by several manufacturers, and
These coils and associated tuning capacitor make up an original B-D
regenaformer sold as a boxed kit (not the one used in the 6A). The rotatable coil
which controls feedback regeneration is on top. The secondary is the larger coil
while the primary is wound on a slotted former and slid inside the main tube at
the bottom, Earthy end.
Australia's electronics magazine
April 2023 95
Resistance-capacitive coupling is
used between all stages, except the
first RF amplification stage, which
uses an RF choke and coupling capacitor to direct RF to the next stage, the
regenaformer.
The whole assembly is built on a
flat aluminium chassis. That was an
expensive metal back then, reflected
in the US$85 retail price for the set.
Circuit details
A close-up of the regenaformer section and the detector valve V2 of the
Browning-Drake model 6A.
the CX112A and CX340 were made
by Cunningham Inc, New Jersey, USA.
The first knob to the left is the on/
off switch which disconnects the valve
filaments from the A battery. The second is the ‘sensitiser’, claimed to pull
in far distant stations, according to
one advertisement. This controls a
variable capacitor of about 100pF (C2;
see Fig.1), which is in parallel with
the first RF tuning capacitor (C1) and
really adjusts tracking between both
ganged tuning capacitors C1 and C5.
The third lower centre knob provides single-point tuning, making
station selection user-friendly, which
was not common with mid-1920s
receivers. The fourth to the right is the
usual wire-wound rheostat controlling
the valve filament voltage, which is
adjusted as the ‘A’ battery voltage
drops with usage. On the far right is the
control for the rotation of the feedback
coil within the regenaformer.
The model 6A was a well-thoughtout design. It has other cutting-edge
innovations for the time, such as the
completely separate shielding of the
first and second RF sections plus
the rear audio sub-assembly, and the
use of resistance-capacitive coupling
between stages.
I could not find a circuit diagram for
this set, so I drew my own, shown in
Fig.1. B-D receivers used similar first
and second RF stages. A UX199 valve
was used in the first stage because it
was easier to neutralise due to its lower
internal capacitance.
Hazeltine neutralisation was implemented using C3 at a few picofarads.
Medium-size variable capacitor C2 is
in parallel with the large ganged variable tuning capacitor C1 and tuning
coil L1; its purpose is to adjust tracking as ganged tuning capacitors C1 and
C5 tune across the dial.
RF choke L5 in the plate circuit
blocks RF from the B+ 90V line to
redirect through C4, an Aerovox
500pF capacitor, and through L2, the
primary of the regenaformer. L2 is
wound on a thin former that has been
glued and fitted under the Earthy end
of the larger tuned winding, L3. L4 is
the rotatable feedback coil providing
regeneration.
The grid leak detector is made using
V2, a 200A triode, together with R1
Fig.1: the circuit diagram for the model 6A radio. As there wasn’t any existing circuit online for this radio one was drawn
up by tracing and testing the components by hand.
96
Silicon Chip
Australia's electronics magazine
siliconchip.com.au
Viewing the model 6A chassis from above shows the RF and AF shielding partitions. The first RF amplifier is right, while
the regenaformer-detector section is left. Four audio valves are shown below (the rear of the set).
and C6. R2 blocks RF from the detector B+ 45V battery tap, while R4 and
R6 block the audio signal from the B+
90V to be redirected through coupling
capacitors C9 and C10.
R3, R5 & R7 are part of the negative
grid biasing circuit of the four audio
valves. The configuration of C7, C8
& L6 is a bit unusual. That section
appears to block and bypass RF from
the grid of V3, the first of four audio
valves mounted on the sub-assembly
to the rear of the shielding cans.
There are no audio coupling transformers; instead, resistance-capacitive
coupling is used throughout. The
audio sub-assembly is a separate,
closely packed unit that was difficult
to access while tracing the circuit.
Two ‘equalisers’ (made by Amperex)
are used to limit the current drawn by
the valve filaments (providing a measure of protection similar to an NTC
thermistor).
After some probing, I determined
that the plates of V4 & V5 are connected together, as are both grids; so
V4 & V5 are in parallel. Valves are
usually connected in series to provide
more voltage gain. So my first thought
upon seeing this is that they needed
more current drive than a single triode
could provide.
After reassembly, I removed V5 to
see what difference it made. There
was no difference in the sound output,
nor was there any difference when I
replaced V5 and removed V4. So the
siliconchip.com.au
need for the extra valve is a bit of a
puzzle. Perhaps some CX340s had
weaker drive than others, and this
was a ‘crutch’ to allow them to get
away with using the weaker valves.
Or maybe there is another reason...
A bit of a puzzle
It had taken almost a century, but the
designers were finally caught out. Why
weren’t the four audio valves operated in series? I am not sure. All TRF
receivers of this general type I have
known have had not more than three
audio stages, and I can only suggest
that adding more could have caused
instability.
I have another neatly-constructed,
home-built TRF set that had an extra
valve paralleled experimentally to the
audio output valve (both UX201s).
Still, from my experience, that does
not improve the sound output.
Firstly, the human ear has a logarithmic sensitivity; doubling the
sound output power would give only
The RF section of the model 6A, which incorporates a UV199 valve (V1) and the
variable capacitors C1 (ganged tuning), C2 (centre), tuning trimmer and valve
balancing trimmer.
Australia's electronics magazine
April 2023 97
a small increase in the maximum perceived volume level. Secondly, there
could be an impedance mismatch to
the speaker load when two valves are
used in parallel.
Further thoughts
Unusually, this B-D model 6A will
operate well with reduced volume
with the first RF valve (a UV199)
removed. I have sometimes pondered
the actual gains achieved by placing
the primary winding in a close-wound
slot fitted under the Earthy end of the
secondary.
I have three examples of factory-
made regenaformers, and all seem to
have the primary turns wound sideby-side on a separate thin former slid
inside the main tube at the opposite
end to the rotating feedback winding.
In each case, all turns are close-wound
with fine wire.
The problem is that the former cannot easily be removed to check the
effect on performance without damaging the unit. This is the technology
of 100 years ago and is now part of
the history of vintage radio. Battery
TRF sets had a short lifespan before
becoming redundant by the end of the
1920s due to advances in valve technology and the rise of the superheterodyne set.
Conclusion
If I were an adult living around 1927
and were given the choice of any of the
TRF battery-powered receivers from
that period that I have in my collection,
I would choose my Browning-Drake
The audio side of the model 6A contains valves V3-V6. Clips within the two
subpanels hold removable resistors. From left-to-right the valves are: UX201A,
CX340, CX340 and CX112A.
model 6A. Connected to its original
Utah horn speaker, it gives a good
sound level from the two remaining
AM broadcasters in Hobart.
It is lightweight, reasonably easy to
set up and tune in once you get the
knack, and it is very stable in operation. I now know why those pioneering
old-timers working in the bush would
get excited when they were talking
about their Browning-Drake wireless.
Would I remove that extra paralleled
audio valve to conserve battery current. Maybe not, would I have known?
To power my set, I use the Universal
Battery Eliminator designed by Peter
Lanksheer from Invercargill, NZ and
published in Electronics Australia,
March 1990. That design has proven
invaluable in powering my battery
receivers.
After I work out the connections
for a particular set, I wire it up to
an eight-way connector from Jaycar
that matches a connector in the supply lead from the battery eliminator.
That enables me to use the one Battery
Eliminator for multiple radios, with
quick and correct connections to each
SC
wireless set.
U Cable Tester
S
B
Test just about any USB cable!
USB-A (2.0/3.2) USB-B (2.0/3.2)
USB-C Mini-B Micro-B (2.0/3.2)
Reports faults with individual cable
ends, short circuits, open circuits,
voltage drops and cable resistance etc
November & December 2021 issue
siliconchip.com.au/Series/374
DIY kit for $110
SC5966 – siliconchip.com.au/Shop/20/5966
Everything included except the case and batteries. Postage is $10 within Australia, see our website for overseas & express post rates
98
Silicon Chip
Australia's electronics magazine
siliconchip.com.au
|