On a sunny day (21 Sep 2006 08:30:25 -0700) it happened "Telstar Electronics"
wrote in
om:

Jan Panteltje wrote:
So we should not blame it on Telstar if anything is wrong with his amp.
I have to admit his PCB looks the nicest of all the ones at that site.
But whre is it heading? I dunno, but it is possible he does.

It's certainly true no design is perfect. The existing SkyWave has
proved to be a reliable design over the years. The new SkyWave
basically just adds the temperature compesated biasing. By controlling
the bias level when the amp heats up... this feature increases the
overall reliability even further by allowing the heat sink to run
significantly cooler. I don't waste resources on gimmicks like pre-amps
that just lower the overall reliability and don't really do anything.
If I add something to the design... I want it to be worth it. Where is
it heading?... a future revision already on the drawing board has both
over-temperature (using Analog devices TMP01) and excessive-swr
shutdown (using Sage Labs Wireline). When these two features are
incorporated... it should make the amplifier virtually indestructible
to burning up or antenna mismatches.

www.telstar-electronics.com


SWR bridge, could be done with PCB tracks for 27MHz,
http://web.telia.com/~u85920178/use/swrpwr.htm

I opened my Pan SWR / power meter, and it has a zig-zag going PCB track
from RF in to RF out, and on each side of these a second track as in the above
link.
So the 2 tracks form a transformer with the main track.
No extra parts other then the diodes resistors and caps...

You can also feed the diode outputs into the analog input of a PIC,
and also the temp sensor, battery voltage, and rectified input voltage,
then you have in power, out power, SWR, and temperature to play with.
In the case of the micro (PIC) you can use one with EEPROM and calibrate
with a 50 Ohm dummyload for power.
No trimpots. The SWR is a ratio, can be done in software.
A PIC is almost cheaper then a trimpot these days...

I have tried to make some pics of the PCB of the SWR meter, but could
not get close with this lens.

Big view, out of focus:
ftp://panteltje.com/pub/s/1.jpg

This one, close up lens, the wide track goes directly from RF in to RF out PL259,
the thinner tracks next to it are the other windings:
ftp://panteltje.com/pub/s/2.jpg

See through with flashlight behind PCB:
ftp://panteltje.com/pub/s/3.jpg

All this should give you the picture how to do a SWR on a PCB.

Jan Panteltje wrote:
SWR bridge, could be done with PCB tracks for 27MHz,
I opened my Pan SWR / power meter, and it has a zig-zag going PCB track
from RF in to RF out, and on each side of these a second track as in the above
link.
So the 2 tracks form a transformer with the main track.
No extra parts other then the diodes resistors and caps...
You can also feed the diode outputs into the analog input of a PIC,
and also the temp sensor, battery voltage, and rectified input voltage,
then you have in power, out power, SWR, and temperature to play with.
In the case of the micro (PIC) you can use one with EEPROM and calibrate
with a 50 Ohm dummyload for power.
No trimpots. The SWR is a ratio, can be done in software.
A PIC is almost cheaper then a trimpot these days...
All this should give you the picture how to do a SWR on a PCB.

You're right... it could be done on the PCB. The main advantages with
the wireline are a great way of saving board space and is fully
shielded. Their performance is predictable, repeatable, and would avoid
some iterations of PCB layout I think. I have used them in the past
with excellent results.

www.telstar-electronics.com

On a sunny day (21 Sep 2006 10:55:07 -0700) it happened "Telstar Electronics"
wrote in
om:

Jan Panteltje wrote:
SWR bridge, could be done with PCB tracks for 27MHz,
I opened my Pan SWR / power meter, and it has a zig-zag going PCB track
from RF in to RF out, and on each side of these a second track as in the above
link.
So the 2 tracks form a transformer with the main track.
No extra parts other then the diodes resistors and caps...
You can also feed the diode outputs into the analog input of a PIC,
and also the temp sensor, battery voltage, and rectified input voltage,
then you have in power, out power, SWR, and temperature to play with.
In the case of the micro (PIC) you can use one with EEPROM and calibrate
with a 50 Ohm dummyload for power.
No trimpots. The SWR is a ratio, can be done in software.
A PIC is almost cheaper then a trimpot these days...
All this should give you the picture how to do a SWR on a PCB.

You're right... it could be done on the PCB. The main advantages with
the wireline are a great way of saving board space and is fully
shielded. Their performance is predictable, repeatable, and would avoid
some iterations of PCB layout I think. I have used them in the past
with excellent results.

www.telstar-electronics.com

Sure, was just trying to up you profit $ by not buying into alien tech.

On a sunny day (21 Sep 2006 10:55:07 -0700) it happened "Telstar Electronics"
wrote in
om:

Their performance is predictable, repeatable, and would avoid
some iterations of PCB layout I think.

As Frank mentioned, think modular for a moment.
You would test this on a simple piece of single sided scrap board,
(well I presume you can etch your own?) just for the SWR part.
Once you have a SWR meter correctly working you can either
take the layout and paste it into the main board, OK first time,
or make some header pins and solder bridges and use it as sub-board,
'option'.
I would do the first.
Many things I designed are just a combination of little pre-tested parts.
[That way] You get a library of circuits over time that works, like stones to
build a house, or castle.