Re: {Collins} Question re 30S-1 to 32S( ) connection

Nice comments and contribution Peter.   Thanks, and I look forward to
talking with you.


In this day of "appliance" radios and people who are new to the Collins and
tube rig experience, I would - and always - encourage people new to the
"sport" to get to know their radio or amplifier a bit more (as in read the
manual and do a little research) and understand what they are doing when
they twist those two black round things on the front of the box.


The use of those round things to optimize output power is a dangerous and
most often misunderstood technique for running/tuning any tube PA, whether
it be in the transmitter/exciter, or in the amplifier. Poorly tuned and
monitored, MAX OUTPUT POWER often equates to messy output.


Simply put, both the output stage of a transmitter and the amplifier are
supposed to be a (Class AB1) linear stage and IF YOU DO NOT KNOW WHERE YOU
KNOW NOTHING ABOUT THE LINEARITY and the legality and cleanliness (not to
mention safe operating point for the rig) of your signal.


There are now a number of pretty good articles on our website about how to
tune, and your comments about monitoring are VERY WELL TAKEN.


And, by the way Peter, for your info, the issue that caused the (some)
Collins engineers to conclude they needed to make that cable 20.5 feet, was
a vain attempt to get the last percentage of IMD reduction, interestingly,
in the driver S-Line output stage of  the dual 6146s.  The IMD was always
"GOD" at Collins and they thought that was the thing to do.  That thinking
ended abruptly when Art Collins saw what had happened (Already in manuals
and in use in both the 30L-1 and the 30S-1) and told them that "If there was
a problem, to FIX IT, and if there wasn't to quit chasing the numbers and
get that cable length instruction out of the manual and out of use".  End of
story.  It was IMD they were chasing and not instability - although it turns
out that it does indirectly effect it.


Warren Bruene (of RF Power fame) developed it reluctantly with some other
engineers and Gene Senti, the engineering manager at the time, did not
support it much.


Now, as Paul Harvey was want to say, "now you know the rest of the story".




From: Peter Hall [mailto:P.Hall@xxxxxxxxxxxxx] 
Sent: Sunday, October 15, 2017 3:04 AM
To: Bill Carns <wcarns@xxxxxxxxxxxxx>; 'Lou Fiore' <ltfiore@xxxxxxx>
Cc: COLLINS@xxxxxxxxxxxxx
Subject: RE: {Collins} Question re 30S-1 to 32S( ) connection


Hi Bill and Lou


As an experienced RF engineer,  but relative newcomer to Collins collecting,
I'd certainly echo Bill's comments regarding the value of the CCA archives;
they are of immense value in bringing the motivated new-comer up to speed
and are always the first port of call when queries arise. In the case of the
30L-1, there are several great articles, including at least one on the cause
and treatment of medium and high frequency instabilities in some amplifiers.


As one might expect given the vintage, the 30L-1 is not a plug-and-play
device and it behoves operators to familiarize themselves with some of the
foibles of the amplifier.  This is of course a foreign to many new hams but
I've also spoken with a number of older owners who do little or no testing
of their equipment.  One of the things that's surprised me since returning
to amateur radio is that people now do less monitoring than ever, despite
the abundance of better and cheaper test equipment.  In addition to my bench
equipment, I use a monitor scope (a nice old Kenwood SM220)  permanently in
my transmit line and, based on my experience , 30L-1 instabilities are
clearly visible to the practiced eye.  Naturally, more sensitive equipment
can tell you more but I had no problem picking both high frequency and
medium frequency instabilities in my 30L-1, a RE model probably manufactured
in 1968-69.  These were actually at a fairly minor level and I was getting
good signal reports when I first put the linear on the air.  The message is
that if you don't look, you might never know.


With the help of the CCA archives the mf oscillation was pretty easily dealt
with but my amplifier had a stubborn, low-level signal-frequency (or near
signal-frequency) instability on 10m when connected to anything other than a
dummy load.  Out of interest, I replaced my 3' drive cable and inserted an
~20' length of RG58 to the exciter and, interestingly, the 10m instability
vanished convincingly on a spectrum analyser.  The exact length was not
critical; any similar length worked.


The case for the long 30S-1 cable is clear and well written up, most
especially in The Signal article referenced by Bill.  But the 30L-1 is
treated as a bit of an afterthought and, to date, I have found no single
convincing analysis.  I suspect that in fact there is none:  the 30L-1 is
conditionally stable and a variety of causes, including the somewhat motley
collection of tubes many of us have to use these days, various source and
load configurations, etc.,  could all contribute to a given amplifier
benefitting from something that changes the magnitude and/or phase of a
feedback signal.     


So, my take would be to start with a convenient length of exciter cable but
be an active, responsible ham and check thoroughly for both high frequency
and medium frequency oscillation.  If you find residual instability on the
higher bands, it'd be at least worth trying the longer exciter cable, along
with any other required (and more conventional) fixes. 


Incidentally, regarding the more insidious medium frequency oscillation (~1
MHz), I undertook the CCA-recommended treatment of L3 and its swamping
resistor, although I had earlier found that excessive external impedance or
poor de-coupling in the keying line triggered the original problem and, when
corrected, the 30L-1 was actually fine with its original components.  As a
further aside, there is a Polish eBay supplier currently selling cheaply
stocks of the replacement Epcos 56 uH choke having a high self-resonance
frequency (70 MHz).


Best 73,





From: Collins [] On
Behalf Of Bill Carns via Collins
Sent: Sunday, 15 October 2017 9:39 AM
To: 'Lou Fiore' <ltfiore@xxxxxxx <mailto:ltfiore@xxxxxxx> >
Cc: COLLINS@xxxxxxxxxxxxx <mailto:COLLINS@xxxxxxxxxxxxx> 
Subject: Re: {Collins} Question re 30S-1 to 32S( ) connection


There need not be another thread on this. It has been discussed and
researched ad nauseum and the results are in the reflector archives and also
in the RX for your Collins on our website. You should always check there
for answers when having any issue.

RX for Your Collins: 

The short answer is Yes, you can shorten it all you want!

Here is the link to the RX article,

and scroll down to page 16 of the Signal issue in which it appeared.

Hope this helps, and to others, let'
Use the archives and not redo this issue..

Thanks, Bill

Bill Carns, N7OTQ (Trustee K0CXX)
Past President, Collins Collectors Association
Founding Board, Collins Radio Heritage Group
Editor, Signal Magazine (Retired)
Wimberley, TX
512 618 2762 (Cell)
512 847 7010 (Home)

-----Original Message-----
From: Collins [] On
Behalf Of Lou Fiore via Collins
Sent: Saturday, October 14, 2017 4:12 PM
To: collins@xxxxxxxxxxxxx <mailto:collins@xxxxxxxxxxxxx> 
Subject: {Collins} Question re 30S-1 to 32S( ) connection


A question for the group:

Figure 1-2 Note 5 of the 30S-1 manual warns against shortening the
originally supplied 20.5 foot cable connecting the RF output of a 32S( )
transmitter to the RF input of a 30S-1. Does anyone have any thoughts on

Thank you for your input.

73, Lou W2LTF

*** You are subscribed to Collins as wcarns@xxxxxxxxxxxxx
<mailto:wcarns@xxxxxxxxxxxxx> . If you wish to
unsubscribe, or modify your preferences please visit ***

*** You are subscribed to Collins as p.hall@xxxxxxxxxxxxx
<mailto:p.hall@xxxxxxxxxxxxx> . If you wish to unsubscribe, or modify your
preferences please visit ***

This archive was generated by a fusion of Pipermail (Mailman edition) and MHonArc.