Re: Twiddlesticks

Cursitor Doom <cd@notformail.com> wrote:

On Wed, 18 Jun 2025 22:27:15 +0100, liz@poppyrecords.invalid.invalid
(Liz Tuddenham) wrote:
 

Cursitor Doom <cd@notformail.com> wrote:
>

Gentlemen,
 
Many of us are familiar with the all-too-common problem of reciever
re-alignment. Dealing the vintage radios with fragile ferite cores in
their IF stages is unavoidable and can prove to be a major PITA. The
cores are brittle and eager to shatter at anything much above mm/mg
torque levels. The key is to use a tool which is not only non-metallic
but also a perfect fit for the slot it must engage with in the top of
the core. Given the range of sizes one encounters in this concern,
it's often necessary to 'roll your own' tool. Well I have a suggestion
to make. Wood makes the best tool for this job and opinions vary as to
which particular wood is ideally suited to this task. Lemon and orange
are often cited. But they're expensive and hard to source typically.
However, I've found a marvellous alternative: yellow heart. I don't
know what the technical name is for this wood, but that's what it's
informally called. It's cheaper and more readily obtainable than
orange or lemon and AFAIC, *better* than either with it's incredibly
fine grain and viceral hardness. It files and sands readily and when
dipped in cellulose sealer, will not fray in use.
Just thought I'd share that with you guys.

>
You must be psychic!
>
Last weekend I bought an Eddystone EC10 at a radio rally.  It suffered
from the usual problem of 'whiskered' OC171s causing random bangs and
crashes so I decided to change all the transistors to silicon and make
it negative earth.
>
When I came to align it, most of the ranges were within specification
(not bad for ba set that is 50 years old) but Range 3 was so badly out
of alignment that I thought someone must have adjusted the local
oscillator high-side at the bottom calibration point and low-side at the
upper one.  It turned out that this wasn't the case but the Range 3
oscillator core was cracked and had obviously suffered from 'screwdriver
drift'.
>
I couldn't find my Neosid nylon hexagonal trimming tool, so I used a
plastic knitting needle filed approximately to a hexagon - if that
hadn't worked, I was going to use a wooden barbecue skewer.  Luckily the
bits of the core unscrewed with a bit of patience and - even more
luckily - at the radio rally I had been given a box of old scrap coils,
one of which had an identical hexagonal-holed ferrite core in perfect
condition.
>
Range 3 still wouldn't line up and I discovered that the tracking
capacitor was out of spec.  That proabably accounts for why the previous
owner had been messing around trying to re-align it.  Everything is
working now and I never cease to be amazed at the sensitivity and low
internal noise level of these sets - even though, in 1968, I worked in
the factory on the production line of its companion receiver, the EB35.
>
My job? - aligning them!

 
Yes, a properly aligned AM rx with a decent loudspeaker in a
properly-designed wooden cabinet can sound fantastic. What's your
preferred methodology for IF stage re-alignment? I have a Knight
signal generator/wobbulator made in the late 60s which I like to pair
up with a spectrum analyzer. Some folks seem happy just to tweak 'n'
peak the audio level, but I don't think that comes out as well and
risks osciallation of course. Then there's the 'S' curve brigade (not
my cup of tea personally). There are probably other methods as well,
which I'm sure someone will chip in with!

For this one I used a Marconi Sig.Gen. connected to a frequency counter
and set exactly to the recommended IF frequency.  With 30% modulation at
1 Kc/s, I repeatedly tweaked each transformer for maximum A.F..  That is
the method recommended by the designer and published in the handbook
(and used on the production line)  I didn't have the recommended
moving-coil AC meter to measure the audio output power, so I used an
oscilloscope instead. 

As the sensitivity came up, the signal was reduced to keep the AGC from
operating.  Eventually, at about 2uV, the signal and noise became almost
equal.

The IF tuning is fairly broad, so nothing more sophisticated is
necessary, the RF side is a bit more touchy and many passes are needed
to get it optimised.


--
~ Liz Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk