Receiver Type W5737
The following pictures
show this old CNY2 Naval receiver from 1944 with no valves, no
outer case and missing mechanical parts. I understand the latter
were removed by the last owner to make good a second receiver,
but the overall condition looks good enough to attempt to restore
this example to a working state. As I also
have the matching power supply this makes an attractive project.
Above the black tuning
knob has two steel grub screws which grip the protruding bush.
Unfortunately both grub screws are seized. I managed to back
these off but I'll need to drill a new hole for a third grub
screw. Dial markings include the outer (log scale) from zero
to 200 and the three RF wavebands. The three 8BA nuts secure
the slow motion drive. Slow motion tuning is via the fold-out
lever and fast tuning by gripping the brass part under the knob.
The slow motion assembly comprises
three pairs of discs which grip the outer ring and are driven
from a bush carrying a lip which is also gripped by the pairs
of discs.The missing gear would be fitted to a quarter inch shaft
passing through the slow motion assembly and secured by the two
4BA screws. A puzzling feature.. three countersunk holes with
shadows of fittings. Below I temporarily fitted three posts but
these prevent the drive from turning through more than a small
portion of the travel. You can also see the outside edge of the
ring has strangely shaped slots. It's possible these could be
bent outwards to provide friction to keep receiver tuning from
shifting under vibration.
I was fitting one of the parts having
unscrewed the three retaining nuts (which fit on the threads
.. picture above left) used to hold the drive to the dial, but
should be in place to prevent the drive from falling apart..
which it did! I set the various pieces to one side, because a
day job had been delivered, and after several days went to reassemble
the thing so I could continue the restoration. Imagine my surprise
when I'd put it back together to discover it was correctly assembled
and previously I'd put it together back-to-front... the pictures
above show the drive incorrectly assembled.
As an illustration I temporarily
fitted a gear (above right) to indicate the general part required.
The driven gear is rivetted to the tuning condenser shaft and
will mate with a spur gear carried on a quarter inch shaft which
passes through the dial assembly. Below, if you study the front
panel you'll notice here are no fixing holes for the dial assembly.
I suspect that this was due to anti-vibration mounts on the tuning
condenser. Because of these the slow motion dial needs to be
secured not to the front panel, but to the tuning condenser frame.
There are two holes in
the top of the tuning condenser frame (one is visible left) and
I suspect a plate might have been fitted, into which the end
of the quarter inch shaft locates. This would accurately position
the spur gear to mate with the driven gear. The whole tuning
assembly would therefore be fastened to the tuning condenser
frame and be subject to slight wobbling from its anti-vibration
It's not easy to determine the
size of the spur gear, but it might be about three quarters of
an inch diameter. I counted the 68 teeth on the driven gear and
the tuning condenser rotates through 180 degrees. The log scale
dial markings cover 360 degrees of the dial periphery so the
spur gear will have 34 teeth to give full condenser rotation.
I can't see exactly why there
are three countersunk holes in the slow motion outer ring.
I fitted three tapped bushes
which are clearly wrong. Maybe a bracket assembly might have
been fitted, angled clear of the rotating parts, to secure the
spur gear in some way and these then bolted to the condenser
A peg, lower right looks like
it will hold a dial locking catch.
The new mounting plate
held temporarily in place (three fixing screws are used). The
plate will secure the three steel straps mounting the slow motion
drive. The bottom of the plate needed to be filed to fit around
tuning condenser projections and a slot cut needed to be cut
from one side to allow it to slide under the gear wheel. The
slot was then filed to fit over the centre bush holding the ball
Wartime radio equipments generally
used imperial screw sizes (predominantly BA) rather than metric but
it was common to find US sizes which are neither imperial or
metric. I'll assume the CNY-2 used imperial sizes and that includes
Assuming the missing metalwork
can be readily fabricated, I looked for a new spur gear. The
first step was to make measurements. The driven gear which I'd
already measured has teeth at 16 per inch and an overall diameter
of about 1.4 inch. I counted 68 teeth equating to a circumference
of 4.25 inches and a diameter of 1.352 inches. At this point
I need to change from inches into mm because even though I'm
assuming the gears are imperial sizes most gear wheels (including
imperial) are sold by metric measurements more often than imperial.
The driven gear is therefore 34.34mm in diameter. I'll take note
of the tooth height because the gear pair will overlap. Allowing
1mm for tooth height the existing driven gear will be about 35.3mm
dameter. My measurement of 1.4 inches equates to 35.56mm so should
be an imperial size known commercially as G48-68. A mating gear
should then be a G48-34 but I need to prove this. See the sketch
below made from measurements of the receiver.
The impression shows the gear has
16 teeth per inch of circumference. The gear, which I assume
is an imperial size, has a diameter of about one and three eighths
of an inch and is constructed from two similar sections which
are spung to provide anti-backlash tuning.
From measurements you
can see that for the driving gear to be dead centre in the front
panel aperture it needs to be about 17.24mm overall diameter
and of course with 34 teeth to produce the 2:1 step down ratio.
Looking at the G48-34 gear spec.. this is around 18mm to 19mm
in diameter. Measurements tell me it should be 17.24mm. Because
of the difficulty in making measurements I'm sure 1mm is within
an acceptable margin. I ordered an H534 gear for £3.50
which seems to have about the same spec as a G48-34 having a
tip-to-tip diameter of 18.1mm, length 9.5mm and with a bore of
3mm. When it arrived I found it was exactly the right size I
needed and the teeth mated perfectly with those on the anti-backlash
gear. Because the hole is 3mm I had to make a spindle having
a quarter inch diameter to fit the dial and a 3mm section to
fit the new gear. As the new spindle needs to be precisely in-line
to allow constant meshing of the gears I used a pre-drilled aerial
fixing adaptor. This has a 3mm hole drilled through it for a
3mm stainless steel whip and an oversize outer section which
I turned down to a quarter of an inch. I used an electric drill
for this, but if I'd got a lathe I could have turned and drilled
a complete new fitting from scratch. I found a brass rod for
the gear spindle threaded at one end for 4BA. I cut it to length
and turned it down to exactly 3mm to fit the new gear.
Gear fitted to the new
The next step is to design
a mounting bracket. The bracket is extremely difficult to make
because the tolerances involved are so critical. The fixed part
of the slow motion assembly (call it the ring) needs to be secured
to the backplate fitted to the front of the tuning condenser.
This backplate needs to be drilled to hold the spindle for the
new gear. The gap between the ring rear face is only 1.5mm from
the three rotating discs. I used metal strips to hold the ring
and each of these is fastened via a countersunk hole that neatly
fitted a tiny M3 screw. I used metal strip about 1mm thick to
give me sufficient rigidity. Each securing screw thread length
needs to protrude no more than say 0.3mm otherwise it will foul
the rotating discs. Also, the three steel strips need to be precisely
bent so that on one hand the rotating discs clear them and on
the other hand they cannot touch the inner surface of the dial
centre boss. I found that I had to make the first one by bending
the end of the strip, drilling the hole for the M3 screw then
tap it then hammer the right angle in a vise to ensure it was
dead centre allowing similar clearances to the moving parts.
Once one was made and fitted perfectly I made the other two exactly
match it . To maximise clearances I filed the surfaces of the
strip flat to mate with the ring surface, then filed the inside
edges to clear the inner edge of the ring where the three rotating
discs ran. Finally I filed the fixing screws to their correct
Showing the steel straps in
place on the slow motion drive indicating the small tolerances
necessary. The three straps will be anchored to the mounting
pate fitted to the front of the tuning condenser. turning the
knbo will rotate the three discs which in turn rotate the dial. The
picture shows the straps in position but their fixing screws
need to be shortened to avoid the discs which pass over them.
You can see that the centre
area has been soldered at some time. An alternative to using
screws might be to solder the three straps in place.
As I mentioned before.. these
pictures show the slow motion drive assembled wrongly. In fact
the clearances that I'd taken so much effort to maintain were
totally unecessary because the thee discs should be facing the
inner edge of the dial.
I've left the pictures because
they may help if someone else tackles one of these drives.
The next step will be
to fit the slow motion drive and dial assembly to the mounting
plate. This needs to be done very accurately so the the gears
will mesh properly. The most important step is to drill the hole
for the 3mm gear spindle. I might make this adjustable. Once
this is done the dial needs to be fitted in place so that the
spindle is exactly in line. The three steel strips will need
to be bent and drilled. One will mate with one of the existing
backplate fixing screws but the two others will be fitted to
new holes drilled in the backplate. I might make a temporary
blanking plate over the 2.75 inch aperture in the receiver front
panel and drill a hole dead centre to match the position of the
quarter inch end of the gear spindle.
Here's the slow motion drive correctly
assembled. There's now no difficulty attaching the three brackets
because their securing srews no longer foul the three discs.
This picture shows the three straps
bent to the correct length for fitting to the back plate.
Below is a picture of the backplate
now drilled to accept the three straps. I've shown this in its
The dial fitted to the drive
Below with the tuning knob in place.
Because the tuning knob grubscrews had
seized and broken I drilled a fresh hole through the knob, drilled
through the brass centre and tapped this 4BA for a new securing
screw (visible below).
It remains to fit the pointer to the
holes above the dial. The clearance between the dial and panel
is slightly greater than original so the pointer fixing spacer
will need to be increased in size to match.
The new dial fittings
worked.. almost... perfectly, but once the tuning condenser was
either fully meshed or unmeshed the knob could continue to turn
the outer dial thus making the markings irrelevant. I'm not sure
what was slipping but I suspect it was the shaft onto which is
secured the new gear. Because the gear has only a 3mm mounting
hole the mechanical advantage of the knob over the 1.5mm radius
of the shaft overcame the friction at one of the fixing screws.
The radius of the circle described
by the tuning knob is 40mm so the mechanical advantage is 26.
Most recevers use a method of
restricting movement beyond the scale limits and no doubt the
original fittings (which were missing) included this.
My solution was to use the hole
where the dial lock screw must have been fitted. I fitted a 2BA
bolt carrying a 2BA solder tag secured with a filed down nut
and lock washer. I bent the solder tag away from the front panel
and twisted it so that it acted as a buffer. I then superglued
a 4BA nut the the inner edge of the dial at the 180 degree position.
This corresponds to a dial setting of 0/360 degrees at the top.
Now as the knob turns the dial the latter stops and similarly
after rotating the other way after 360 degrees stops at the other
side of the buffer. There is a very slight loss in rotation,
because of the width of the 4BA nut, but this is insignificant.