[HBR] Another one-week (okay, probably not!) HBR -- Part 1

[Walt Hutchens]

Pete said:

> How about some photos as the project progresses? You could open an
> account on PhotoBucket, and post the links as you progress with the
> project. Photos would be neat!!

I'll do photos when it gets a little farther.  As of this evening the
panel is done (except for whatever I forgot ...) and the chassis is
mostly laid out and centerpunched.

The Eddystone dial is just as much fun to install as I expected.  
Assembling it on the panel is a sort of four-hands-needed job.
However, it is a very good dial, AND good looking.

I discovered one design mistake: you can't really drive a dual
triode mixer with a triode LO in this way without getting a little bit
of pulling on the higher frequencies. I've been spoiled by doing it at
1700 kcs in dual conversion sets with a tunable second IF, but with
the LO going to 13 Mcs or something like that in this design, it
wouldn't work.  With a little jiggering of things I was able to switch
to a pentode ECO which will give enough isolation.

Might get the chassis holes done tomorrow.  I'll post some pics when
that's done.

> I'm playing around with the R-23 80kHz transformers. Boy, they don't
> appear to be very high Q, and use a high L to C ratio.  I assume this
> was done to control the bandwidth for AM reception? Also, I have a
> question about the button silver mica caps and the old dogbone style
> ceramic caps that were used.  I know these were good caps back in the
> late 30s and early 40s, but I wonder if it it would be good practice
> to replace them with newer 180pF dipped micas since these components
> are now 60 years old?

I seem to remember that there have been discussions of the Q of those
transformers before; maybe a couple of years ago.  You could try
searching the archives.  When you consider that they operate at 85kcs,
you wouldn't want a very high Q: Q=40 would give you a bandwidth of 2
kcs with one coil.

The ARC-5 manual has curves of the selectivity for the receivers, as I
recall. My previous HBR project with 4 85 kcs command set IFs is about
as sharp as you can use on an SSB signal with all the rods pulled up.
It's really too sharp to be ideal for AM reception, even by tuning to
one side -- when the voice signal is peaked, there's not enough
carrier to demodulate it properly.

As to replacing the caps inside the transformers, if they seem
okay, I don't think I'd bother. The MOLDED micas from those days are
bad actors because the bakelite shrinks due to heat and the passage of
time, pulls away from the leads (or cracks open) and moisture gets in,
but the ceramic and button silver micas are excellent quality; I don't
recall ever running into one that had failed unless there was obvious
mechanical damage.

Among the other command set parts, the sealed-in-the-can paper caps
are also fairly common failures, due to moisture getting in.  Most
sets need one of the bypass cans replaced.

Walt
KJ4KV




-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qth.net/pipermail/hbr/attachments/20081001/013de7fa/attachment.htm