Remote Antenna Switches and K2 Questions

[Edward A. Dauer]

Two questions about the K2 and remote antenna switches – particularly, the Ameritron RCS-4.

The K2 I finished building three months ago is finally on the air and, so far, working reliably, with great thanks to Don Wilhelm for his knowledge, skill, and patience.

It is now on its fourth set of PA transistors, plus a raft of other components in the PA circuits.  The story until now had been the same – the rig works fine on Don’s bench; he returns it to me where it works fine with a dummy load on my bench at my city QTH where real antennas are forbidden.  I then take it to my operating QTH where my antennas are, hook it up, and within 20 minutes or less a loud pop and a puff of smoke precede the darkness.  PA transistors blown away again.  Three times, all the same.  Replaced power supplies twice, added cascades of surge suppression on the AC mains, no dice.  Sent the rig back to Don for another round of radiology and surgery.  Thrice.

This time, on a hunch, I removed one thing that differed between the operating QTH and the test bench in the city – an Ameritron RCS-4 remote antenna switch.  With that out of the circuit, so far the rig seems happy.  No pop; no smoke; no darkness; no tears; no round trip back to North Carolina.

Don suspects the problem has been some mismarked component in the base K2 which, though it looked right, was of the wrong value and caused an imbalance between the two matched PA transistor circuits.  On the rig’s last trip to NC Don replaced virtually everything in those circuits.  And now it works OK, so far.

I wondered, though, if the RCS-4 could also be a culprit.  That switch inserts into the coax feed line the AC and DC voltages that energize its remote relays out where the antennas are.  According to the schematic there is no way DC voltage could appear at the input side of the switch console.  But with a DMM reading the Ameritron’s RF input side, I see some.  There are DC transients as high as 5 volts, negative and positive, occurring at odd times and of very short duration.  They have never seemed to affect the K3/KPA/KAT, nor the KX3/KXPA when these rigs are connected to the same RCS-4.

So, two questions:  First, is it possible that these weird transients from the control box could have been the cause of the PA failures?  Has anyone else ever experienced anything like this?  Second, rather than test that hypothesis by hooking the RCS-4 up to the K2 again, a test that could be expensive, I may just look for another remote switching system.  Any suggestions about a replacement?  If I need to use a separate DC voltage cable I’ll have to wait until the spring thaw in order to bury it – but maybe the DC voltage insertion system just isn’t a good idea?

Tnx for any advice anyone can offer,

Ted, KN1CBR
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

[pa0pje]

Hi Ted,

I just looked at the schematic, downloaded from the Ameritron site, and
wondered whether or not there are any protection diodes across the 3
relay coils in the remote box. They do not show on the drawing.

I can imagine these spikes could be much bigger than the +5 and -5 volts
you quote.

73,
Peter - PA0PJE

Op 2016-03-26 16:28 schreef Dauer, Edward:
> Ameritron RCS-4 remote antenna switch
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

[Edward A. Dauer]

Thanks for the note, Peter.  I haven’t opened the relay box, not wanting
to break whatever weather seal it still has (besides,  it’s snowing and
blowing out there); but I don’t know why a schematic would leave out
anything that’s actually in the equipment.

I’ve had a half-dozen off-list replies to my original post, almost all of
them noting the three parallel caps that pass the RF input and
theoretically isolate it from the relay voltages.  If any of them failed,
there could be DC showing up at the RF input.  Don Wilhelm pointed out
that because the LPF in the K2 is grounded on one side the presence of DC
itself wouldn’t be a problem, but that rapid changes in any voltage
getting into the RF input side could be. (Do I have that right, Don?)  The
curious thing, though, is that I have two Ameritron RCS-4s, and these
voltage anomalies appear in both of them in almost identical ways.  Maybe
a design flaw rather than a component failure?  In any case, something
caused multiple failures when the K2 was here, but not when the K2 was
anywhere else; and having the RCS-4 in the line here is the last such
difference I could think of that hasn’t already been ruled out.  So maybe
it has been a K2 component problem, but maybe not - three nearly identical
failures make that seem less probable than only one would, but snot
unlikely.

What I’ve measured with the DMM are short bursts - and, in fact, at one
point I saw a reading of over 18 volts as your note suggests might be
possible.  I am bringing one of the units back to the city with me Monday,
where my oscilloscope is, so that I can get a better picture of what these
transients look like.

For now I am just being wary of a system that inserts the relay energizing
voltages into the coax.  The disadvantage of the alternative, meaning a
separate low-voltage cable from the control box to the relays, is that
it’s another cable.  For reasons of good domestic relations the coax from
the house to the common antenna feed line site is buried.  Any other
cables running that way would have to be as well.  From the looks of
things today it may be June before the ground at this altitude (8,600
feet) thaws enough to allow for even shallow trenching.

My thanks to everyone who replied with observations and suggestions.  This
list is a never-fails resource . . .

Ted, KN1CBR








On 3/26/16, 4:42 PM, "Peter Eijlander (PA0PJE)" <[hidden email]> wrote:

>Hi Ted,
>
>I just looked at the schematic, downloaded from the Ameritron site, and
>wondered whether or not there are any protection diodes across the 3
>relay coils in the remote box. They do not show on the drawing.
>
>I can imagine these spikes could be much bigger than the +5 and -5 volts
>you quote.
>
>73,
>Peter - PA0PJE

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

[Don Wilhelm-4]

Ted,

Yes, you have that right.
Capacitors or diodes across the relays would be a good thing to swallow
the voltage kick when the relay is de-energized.

If your DMM is showing voltages in excess of 13 volts, then the peak
voltages will be much higher.
The diodes or capacitors would have to go into the remote box across the
relay coils.  If diodes are used, observe the polarity carefully - the
cathode must go the the side of the relay with the more positive voltage
- in other words for those relays driven by negative voltages, the
cathode goes to the side with the negative voltage closer to zero volts.
Diodes will do a better job than capacitors.  If you use capacitors, I
would suggest 0.1uF although 0.01uF may be sufficient - it all depends
on the inductance and resistance of the relay coil.

73,
Don W3FPR

On 3/26/2016 7:48 PM, Dauer, Edward wrote:
> Don Wilhelm pointed out
> that because the LPF in the K2 is grounded on one side the presence of DC
> itself wouldn’t be a problem, but that rapid changes in any voltage
> getting into the RF input side could be. (Do I have that right, Don?)

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

[Brian Hunt]

Looking a the schematic it appears they use 0V, +V, -V and ACV for the 4 switching states but only + or - voltage is applied to the relays. The big cap is used to smooth out the AC ripple.  I would put a back diode across each relay coil and add a MOV of appropriate voltage across C203 (V to ground).  Reference the QST article in January 2013 issue. I built one with those guidelines and it's been fine for a couple of years.

73,
Brian, K0DTJ



______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]