As an addendum to Paul’s informative post, my own QSK circuit (QST June
2015, and http://www.w8zr.net/homebrew/) has a “lock-on” feature that
keeps the amplifier locked on-line until RF from the exciter is off, even
if the PTT line opens. Similarly, it keeps the amplifier locked “off-line”
when the PTT line closes if RF is present at the input. The purpose of both
features is to prevent hot switching of the vacuum relays.
73,
Jim w8zr
Sent from my iPhone
On Mar 16, 2021, at 5:32 AM, Paul Christensen <w9ac@arrl.net> wrote:
"Anything under 2ms (2 milliseconds = 2000 microseconds, which is a whole
bunch of nano seconds) is WAY MORE then enough for most transceivers.
(Which was the point I was trying to make here)."
A diode's forward bias transition time is one important parameter but when
considering PIN diodes for use in the Alpha 87A, the diode's minority
carrier lifetime may be an even more important parameter. The purpose of
the intrinsic region between the P and N wafers is to store a charge long
enough to keep the diode from rectifying RF during a complete cycle. For
us, the worst- case band of operation is 160m (excluding the two LW
bands). For 160m, the carrier lifetime computes to approx. 0.5 us. But,
typically much more time is required to minimize IMD products.
When used as an HF T/R switch, Microsemi recommends 1A of DC current to
fully bias the UM2100 series diodes -- and nearly 1KVDC reverse bias on the
receive PIN when HF power levels reach 1KW.
I believe ETO/Alpha originally spec'd the MA4P4006D for the transmit PIN
which was superseded by the identical UM4006 when Microsemi acquired the
product line. I hear rumors that Alpha has been recommending 1N4007 diodes
as transmit PIN replacements, but I've never seen an equivalent carrier
lifetime spec that would give me comfort if a replacement is needed in my
amps. And, if these work reliably, I would like to know how close they are
to the ragged-edge of acceptance. And, are all 1N4007 diodes created the
same and offer identical performance? Even diodes from decades ago?
BUT if you use an IC7000 or similar, you'll have trouble. But I think Alpha
compensates for crappy radios that still transmit RF AFTER the PTT line is
released.
To my knowledge, the Alpha 89 is the only ETO/Alpha amp that prevents a
return from T=>R when RF is present. Possibly the 9500 also protects this
way but I've never tested it.
A big deal at the moment is Icom's amp sequencing in the IC-7300, 9700, and
705 models. The amp key line releases approx. 3 ms BEFORE RF decays to
zero. This is a fault on amp break, not amp make. As such, no amount of
fiddling with a transceiver's menu helps to prevent hot-switching with fast
amps. The Elecraft amps DO have a menu selection to add delay on break.
Icom has known about the problem since the 7300's release in 2016. Myself
and others have given up hope that they care enough to do something about
it. To that end, I created the new S-QSK Mini device that can be seen on
my QRZ page. It's a tiny 1X1 sequencer that's smaller than a book of
matches. Precise timing is set by an Adafruit Trinket MO microcontroller
and it completely compensates for Icom's problem through software. The
S-QSK Mini is powered exclusively from a Micro B USB cable. For those with
an interest, here's a quick link to my QRZ page:
www.w9ac.com
Paul, W9AC
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|