CQ-Contest
[Top] [All Lists]

Re: [CQ-Contest] PGXL in SO2R mode? (RIP K7BV)

To: CQ Contest Reflector <cq-contest@contesting.com>, "Mike & Coreen Smith" <ve9aa@nbnet.nb.ca>
Subject: Re: [CQ-Contest] PGXL in SO2R mode? (RIP K7BV)
From: "Bob Wilson, N6TV" <n6tv@arrl.net>
Date: Mon, 13 Jan 2025 19:54:52 -0800
List-post: <mailto:cq-contest@contesting.com>
Mike,

A recent (or the latest?) release of the PGXL firmware requires 25 ms (!)
of PTT Lead Time, otherwise you get an "RF Early < 25ms!" message from the
amp if RF input is detected less than 25 ms after the RELAY input is
switched to ground.  This means you will have to use a PTT Lead Time
setting in logging software of 30 ms or so, and never use the internal
keyer of the radio with VOX (most radios provide far less PTT lead time;
the IC-7410 default PTT lead time is worth checking with a scope since the
menu has no TX Delay adjustment).  SSB VOX may be OK, but it may generate
RF Early as well.  K3 has a TX DLY setting but it can make external keying
sound poor, especially if set to the max value.  If you enable CW QRQ mode
in the K3, the PTT lead time drops to 4 ms (way too short for amp), and the
TX DLY setting is ignored.  Again, use keyer or software PTT to get around
this.

Were you planning to add the Tuner Genius (TGXL) too?   The PGXL is a
broadband solid-state amp, so it really needs an external tuner unless all
of your SWRs are pretty low.

The TGXL tuner wants to have its PTT IN / OUT connectors in series between
the rig and the PGXL, to open the amp's key line when the TGXL is tuning,
but this delays amp relay closure by an additional 6-8 ms for *every* PTT
closure, so you get even more RF Early warnings unless you increase the
logging software PTT delay to 35 ms or so.  Note that a dot at 40 WPM is
only 30 ms long.  And if you don't use the PTT jacks of the TGXL, it
complains "No PTT!" when it detects RF.

The PGXL amp does not support ALC feedback to the rig at all, so it cannot
tell the rig to back off the power via fast analog feedback.  It does have
an APC (Automatic Power Control) system that reduces amp gain when output
power exceeds 1800W.  I don't think it's smart enough to send commands to
tell an Icom or Elecraft rig to back off the drive power.

The IC-7410 does not have an RS-232 connector.  Connect the PGXL CI-V jack
to IC-7410 REMOTE jack via standard audio cable (3.5mm male to 3.5mm
male).  The PGXL does not have a legacy Icom "Band Voltage" (Icom Band
Data) input, so switching bands may be a bit slower than you would like.

PGXL band changes by K3 or K3S Band Data pins will be much faster, but even
though the K3 ACC and PGXL BCD jacks use the same 15-pin connector, they do
not use the same pins, so you'll need a custom cable (I can provide this if
needed).  But, the TGXL tuner needs *frequency*, not just band, so it will
not auto tune well unless it gets frequency via RS-232 (RF sensing is too
slow).  The K3 RS-232 line will be in use by your logging software, so you
have to use RS-232 "listen only" mode to switch the TGXL (TXD pin open).
The Serial Box (S-BOX) handles this if you don't want to build your own
special RS-232 Y-cable.   Unfortunately, when connected this way, the TGXL
will stop tracking whenever the logging software stops running.

Another option to consider is pair of Elecraft KPA1500s, which include a
built-in automatic tuner, automatically track Icom via CI-V, or Elecraft K3
via AUXBUS, and can be keyed with as little as 5 ms of PTT delay, with
completely silent QSK.  The IC-7410 power overshoot problem could still
cause it to back off drive power (soft "overdrive" fault) or fault into
STBY if overshoot is excessive.

More comments interspersed below.

73,
Bob, N6TV
https://bit.ly/S-BOX
https://bit.ly/Y-BOX

On Mon, Jan 13, 2025 at 6:29 PM Andy Blank <andyn2nt@gmail.com> wrote:

> > IE: I run a pair of IC-7410 Icoms (currently, someday I'll upgrade,
> > probably
> > to original K3's(non-S's)) and these 7410's are known to have a small
> > amount
> > of initial overshoot. How bad, I don't know. It's no longer fixable by
> > Icom.
>

See the detailed report by AB4OJ (SK) here:

https://www.ab4oj.com/icom/ic7410/7410notes.pdf

(Search for "overshoot".)

> (related) : TX delay is NOT adjustable like in most modern transceivers.(I
> > use N1MM&YCCC SO2R+ to do the TX delay)?
>

Correct.  You may need to use "PTT by Radio Command" in N1MM rather than
relying on the WinKey emulation.


> > Second question.  Is it possible to connect radio "A"  to an antenna
> switch
> > to feed 6 various antennas, but radio "B" only gets a single antenna or
> is
> > there any kind of funny antenna swapping done inside the PGXL like
> another
> > popular SS amplifier which won't work for me because of this quirky
> > feature?
>

The PGXL only has one antenna output per radio.  So connect PGXL Output A
to your switch, and PGXL Output B to your single antenna.  But then Radio B
will not be able to use any of the other antennas, unless you add more
external antenna switching such as the Antenna Genius (AGXL), which allows
any unused antenna to go to either amplifier output.  The AGXL also has a
built-in band decoder, same connectors as PGXL (more splitting required).


> > SO2R is not really discussed in the manual.  I am not clear on how the
> > radio(s) and the PGXL would share the same com port, say to N1MMLogger+
> for
> > example?
>

Each radio has its own USB virtual COM port (Icom), or legacy RS-232 port
(K3).  So you tell N1MM which Port is Radio1 and which port is Radio2. The
RELAY inputs (PGXL has two of them) determine which radio gets amplified.
If you try to key both radios at the same time, one side is bypassed.

> They're a lot of money to a Canadian, so I want to be sure before I finish
> > saving up $$ and purchase one, it's what will work here, otherwise, I'll
> > just stick to a pair of smaller amps for the time being.
>

Please let us know what you decide to do.
_______________________________________________
CQ-Contest mailing list
CQ-Contest@contesting.com
http://lists.contesting.com/mailman/listinfo/cq-contest
<Prev in Thread] Current Thread [Next in Thread>