[ARC5] 10 meter ARC-5s

[Kenneth G. Gordon]

On 16 Oct 2014 at 20:04, AKLDGUY . wrote:

> > I completely rewound all three inductors in the receiver following the article
> > that Gordon White wrote some time ago entitled, "Command Receivers for
> > All Frequencies", in which article Gordon described and listed the turns on
> > the coils that ARC came up with for THEIR early HF receivers up to 27 MHz.
> >
> > I used ARC's 27 MHz data, in fact, as the starting point for my coils.
> 
> OK, you rewound the coils on the basis of A.R.C.'s data.

Well, not exactly: the coils had already been rewound, very sloppily and very 
poorly. I didn't want to take the time to figure out what the values should 
have been, especially the mixer coil, so I decided to take a short-cut and 
read Gordon's article more closely. That started me on the correct road, and 
I simply followed it. That article gave me a starting point, especially for the 
mixer coil. The mixer coil is really "weird", in fact. It is suppose to tune very 
high in frequency above the frequencies of the RF and HFO coils.
 
> Q is the reactance of either the inductor or capacitance at resonance, divided
> by the series resistance, which in this case is in the coil if we ignore other
> very small losses.

Yes.

> In my suggested modification (no modification of the coil) the Q is more than
> tripled by going from 7-9 to 28-30 MHz because the coil reactance more than
> triples. The higher Q at the higher frequency means the selectivity curve is
> the same as at the lower frequency. The curve will be the same number of dB
> down at the same frequency offset at both 9 MHz and 30 MHz.

Hmmm....I'm not convinced. The R remains the same, and therefore would, I 
would think, constitute the determining factor for the "Q". But I have been 
wrong before, and will be again. The math does seem to bear out your 
contention, though.

> There would therefore seem to be no good reason for taking turns off the coils.
> Why did A.R.C. use reduced turns? Probably because they were faced with a too
> large value of tuning gang (no smaller gang available),

I don't see that as an issue at all. ARC could have had any capacitor they 
wanted or needed. I suspect that there is some OTHER reason, or several 
reasons, for their doing it the way they did. Drift? Microphonics? What? 
Those guys weren't stupid and never did ANYTHING without some extremely 
good reason.

> so they used a large
> amount of **parallel** capacitance across the gang to reduce its tuning range
> and reduced the number of turns to compensate.

What I don't understand is that if you are correct, why, then does reducing 
the number of turns also result in the proper selectivity curve and "Q"?

> > Now I WILL say that your method would most certainly eliminate the
> > necessity for removing any plates from the capacitors, but I STILL say, that
> > the inductances would have to be "adjusted" also in order to maintain circuit
> > "Q".
> 
> Yes, my method does eliminate the need to remove plates, but the selectivity
> self-adjusts by virtue of the rise in Q as described above. A Q of 100 at 9 MHz
> (easily possible) becomes a Q of more than 300 at 30 MHz, and the selectivity
> (percentage change) at both bands is the same.
> 
> I'd suggest that anyone contemplating modifying a 6-9.1 Mc/s receiver for 10m
> try my suggestion first. You have nothing to lose. It's simple and reversible.
> Use very good quality capacitors and let them cool before evaluating.

Well, I would agree with that, all right.

I suppose I should dig out my Terman's and do some reading....

vy 73,

Ken W7EKB