Hi Bruce, like you say with a high swr the forward power increases along with
the reflected power so the easiest and quickest way to shut
down the amp is to monitor the forward power output and trigger a shut down
when it goes above a certain value which is what I plan to do.
With most new rigs the output ramps up at a certain rate when it switches to
transmit and the internal swr protection cct. has time to read
the values and engage the fold back or shut down of the output before things
get out of hand, but some rigs overshoot on output or have a
fast rise time. The protection cct of the amp has to almost marry the
transmitter or have its own ramp up of the output to make it co-operate
with the worst offenders. Just remember that 1 ms is eternity for a transistor
that is being punished with high voltage on its collector or
drain and they always wave at you when they say good bye. That's why I still
use tube amps. They take a licking and keep on ticking hi hi.
73 Steve
"Bruce R. Knox" wrote:
> Absolutely agreed, Dave.
>
> For microprocessor coding using integers, there is an easier algorithm that
> leaves out the square root (and makes it compute WAY faster):
>
> ((Forward A/D + Reflected A/D) * 100) / (Forward A/D - Reflected A/D) = VSWR
> * 100
>
> so, for example, a VSWR of 2.3:1 would yield 230.
>
> This is how our wattmeters do it.
>
> Bruce R. Knox W8GN
> RF Applications, Inc.
>
> -----Original Message-----
> From: Dave D'Epagnier [SMTP:DAVED@ctilidar.com]
> Sent: Friday, December 01, 2000 12:17
> To: Steve; amps@contesting.com
> Subject: RE: [AMPS] SWR Protection
>
> << File: ATT00000.html >> Steve, why do you want to know the instantaneous
> forward and reflected
> power? You really wouldn't need this to implement an effective SWR
> protection function, unless you expect that your SWR varies extremely
> rapidly which probably is not the case. It should suffice to use one A/D
> converter and measure the forward and reflected power at different times and
> then do the math. For instance, measure the forward power 10 times at
> 32us/sample. Then measure the reflected power 10 times at 32uS per sample.
> Then average all the forward samples, and the reflected samples for a solid
> measurement. Now do the math:
>
> SWR = (1+SQRT(Pr/Rf))/(1-SQRT(pR/pF))
>
> Trip the relay if the SWR is too high.
>
> The whole operation could be performed in less than 1ms.
>
> --Dave
> K0QE
>
> > -----Original Message-----
> > From: Steve [SMTP:ve3tu@rac.ca]
> > Sent: Friday, December 01, 2000 10:03 AM
> > To: amps@contesting.com
> > Subject: [AMPS] SWR Protection
> >
> >
> > Talking about more interesting aspects of amps, I was wondering if
> > anyone knows of a single chip microprocessor or pic that has 4 or 8 A/D
> > converters in the chip that can monitor 4 input simultaneously. I am
> > planning on building a swr protection board for a solid state amp using
> > a 68hc11 chip that has a single A/D converter that has selected inputs.
> > It takes 32 micro seconds to do each conversion so you don't really get
> > a snapshot of the forward and reflected power at the same time. I know
> > the you can fudge it with software and get it close but I an interest in
> > simultaneous reading on the A/D input or a simple outboard A/D chip
> > would do.
> >
> > tu es 73 Steve VE3TU
> >
> >
> > --
> > FAQ on WWW: http://www.contesting.com/FAQ/amps
> > Submissions: amps@contesting.com
> > Administrative requests: amps-REQUEST@contesting.com
> > Problems: owner-amps@contesting.com
>
> --
> FAQ on WWW: http://www.contesting.com/FAQ/amps
> Submissions: amps@contesting.com
> Administrative requests: amps-REQUEST@contesting.com
> Problems: owner-amps@contesting.com
--
FAQ on WWW: http://www.contesting.com/FAQ/amps
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
|