[HBR] Eddystone Three Section Variable Capacitor Question

[Walt Hutchens]

Whitebear1122 said:

> I'm thinking of mounting the capacitor to an aluminum plate using
> your two 1/4" screws, then cut several larger pieces and attach the
> whole assembly to the chassis using standard 6-32 1" screws.

When the HBRs were first built, AM was standard and tiny bits of
backlash in a dial were unlikely to be noticed, let alone be
considered a problem. I'm sure that mounting a tuning cap on screws
into an aluminum chassis was good enough and I doubt anyone ever
noticed that expansion of that chassis led to detuning as it rotate
the capacitor frame relative to the shaft attached via the dial to the
front panel.

With SSB almost universal now (unless you want to stick to strictly
vintage round tables) one needs an order of magnitude better stability
and setting precision.

I have two finished HBR projects: One is based on the Heath HR-10B and
uses the original cap which is string drive and mounted flat on a
steel chassis: NO issues with backlash or heat-induced mechanical
tuning on that one, though the string drive slide rule dial is perhaps
a notch less precise than the Eddystone and resetability/resolution
considerably less.

The other is an original design on a heavy aluminum chassis. The
tuning cap is mounted in a 90 degree rotated position on the vertical
surface of a 5" x 4" x 3" Minibox screwed along its flanges at the
rear and near the right side of the main chassis. This SEEMS plenty
rigid and backlash is negligible with the Eddystone dial but at the
start it would tune a couple of kcs on 80 as the chassis got warm. A
'get well' program consisting of measures to keep the chassis cooler
and a bulkhead screwed under the front flange of the box (inside the
chassis) and to the right chassis wall cut that to well under 1 kcs --
tolerable if not great.

Doing that one again I'd think how to mount it to the panel with
little or no connection to the chassis. OR make the rotational
stiffness considerably better (what stresses will result?) OR use a
steel chassis. OR at least mount it with the shaft at the chassis
center so rotation would be minimized.

Aluminum has a much greater temperature coefficient of expansion than
steel. It has many good points but you get shape/dimensional changes
in exchange.

Moral: Mounting the tuning cap isn't as easy as I used to think, if
one wants the full performance that the electronics of these designs
would allow.

I would definitely use something more rigid than corner posts to an
aluminum chassis. Try to visualize how the chassis will change shape
as it heats -- the usual effect would be ballooning of the top surface
-- and how you'll make that a non-event, tuning wise.

Of course such effects are invisible but when you consider that 500kc
is tuned in less than 180 degrees and you want less than 0.1% of that
even on 80M, it doesn't take very much.

Someone with a machine shop setup could put a dial indicator on the
top of a chassis and apply a hair dryer, and convert the result to
rotational degrees and then to kcs, assuming mounting of the cap (say)
25% of the way from left or right edge of the chassis.

I did not at first believe this was an issue, but a few experiments
with a soldering iron tip and an ice cube convinced me otherwise.
Precision instruments like a ham band SSB receiver are tougher than
they look!

Screws will be in the mail tomorrow.

Walt Hutchens
KJ4KV