[Heathkit] Heathkit Digest, Vol 127, Issue 9

[Drew P. via Heathkit]

Robert Myers wrote:

"Question about the SB-303 / SB-401 interconnection cables *required* for the
HFO, BFO and LMO.

As I am sure most are aware, the SB-303 and SB-401 manuals call for the
following cables to be connected between the various SB-303 oscillator
output sockets and the corresponding SB-401 oscillator input sockets:

SB-303 ?BFO OUT? -> 24? RG-62/U? -> SB-401 ?RCVR BFO?
SB-303 ?HFO OUT? -> 24? RG-62/U? -> SB-401 ?HET OSC?
SB-303 ?LMO OUT? -> 24? RG-174/U -> SB-401 ?RCVR LMO?

It seems to me that the designers, for some reason, decided to incorporate
the various cable capacitances (that result from their specified lengths)
and their characteristic impedances, along with the input impedances of
those various oscillator inputs of the SB-401, into what amounts to
?essential operating circuitry? of the corresponding oscillators based on
the SB-303.  (But why?  To save money?)

Is the requirement for 'cable weirdness' due to the various oscillators?
output circuitry within the ?303, or is it due to the requirements of
various oscillators? input circuitry within the ?401?? Or is it a
combination of both?

I mean, if the ?303 requires a particular capacitance ?hanging off? the
LMO?s output socket (such being provided by exactly 24 inches of RG-174/U)
in order to function properly, as well as a 50 Ohm load (inside the 401
during transceiver), would it not have been better to do something like
provide all those necessities *within* the ?303 (right at the LMO
oscillator) and follow that by an isolating buffer/voltage-follower (still
inside the ?303) with a 50 Ohm output impedance?? You could do the same for
the HET, and BFO oscillators? circuitry as well ? thus allowing for
arbitrary lengths of, say, RG-58/U cables (not different cables) for
interconnection of the HET, LMO, and BFO sockets between the two units.

Is there a fix for this cable 'weirdness,' allowing the cables to be the
same type (say, 50 Ohm RG-58 since you need that for the antenna
connections, for sure), and their lengths to be not critical?"

In this application, the cables are not being operated at anywhere near their characteristic impedance and in this case they present a capacitive reactance.  It would boost cost and complexity considerably to match all interconnections to the cable's impedance. The logical way out is to do what the Benton Harbor folks did, and incorporate the capacitance into the tuned circuits by specifying a type and length of cable, thereby creating a repeatable capacitance.

RG-62 was the choice probably because it was readily available and had a lower capacitance than most other common cables, allowing for a longer cable run.  You could just as well use a shorter run of RG-58, so long as the length is chosen to present the correct capacitance, and so long as the cable will reach, of course. Let the "weirdness" begin! 

Drew