Amps
[Top] [All Lists]

Re: [Amps] QSK switching time

To: "'Paul Christensen'" <w9ac@arrl.net>, "'k5jv'" <k5jv@kingwoodcable.com>, "'Amp Reflector'" <amps@contesting.com>, "'W5PVR'" <alpha76@triton.net>
Subject: Re: [Amps] QSK switching time
From: "Joe Subich, W4TV" <lists@subich.com>
Reply-to: lists@subich.com
Date: Fri, 18 Dec 2009 14:15:50 -0500
List-post: <amps@contesting.com">mailto:amps@contesting.com>

> One factor oftentimes neglected when reporting diode-switched, 
> QSK performance is the degree of IMD product produced. 

The problem with Alpha and some of the others is that the diodes 
are not operated with FORWARD bias when in the "on" state.  This 
creates momentary crossover distortion on each RF cycle.  In 
order for a diode T/R switch to be truly "clean" any diode in 
the RF path needs to be turned on hard enough that there is no 
switching (rectification) during the RF cycle.   

73, 

   ... Joe, W4TV 
 


> -----Original Message-----
> From: amps-bounces@contesting.com 
> [mailto:amps-bounces@contesting.com] On Behalf Of Paul Christensen
> Sent: Friday, December 18, 2009 1:07 PM
> To: k5jv; Amp Reflector; W5PVR
> Subject: Re: [Amps] QSK switching time
> 
> 
> Lon,
> 
> One factor oftentimes neglected when reporting diode-switched, QSK 
> performance is the degree of IMD product produced.  For 
> example, the Alpha 
> 86 had approximately 10-15 dB worse 3rd order IMD than the 
> comparable Alpha 
> 87A and 89 as a result of non-optimized PIN diode biasing.  
> The RF deck and 
> components off all three Alpha amps are nearly identical.
> 
> Paul, W9AC
> 
> ----- Original Message ----- 
> From: "k5jv" <k5jv@kingwoodcable.com>
> To: "Amp Reflector" <amps@contesting.com>; "W5PVR" 
> <alpha76@triton.net>
> Sent: Friday, December 18, 2009 12:56 PM
> Subject: [Amps] QSK switching time
> 
> 
> > Greetings to all,
> >
> >        I would like to share a great discussion that I was 
> evolved in
> > this
> > morning on an 80 meter net about QSK,  with some of my own comments.
> >
> >        The basic argument is that because PIN diodes switch 
> faster than PN
> > diodes,  PIN's must make a better QSK switch.  Well, do 
> they?   If you 
> > take
> > the entire QSK circuit into consideration, you will see 
> that overall 
> > switching time is not determined by the diodes, but rather by a 
> > sequence of, time consuming, switches, the bias switching 
> transistors 
> > being the major consideration.
> >
> >        Both types of diodes switch in microseconds, with 
> Pin's being 
> > noticeably faster.  Meaningful QSK switching times are measured in 
> > milliseconds, not microseconds (a magnitude of 1000, or 
> so).  Since a 
> > diode switch can not switch any faster than it's bias is 
> switched, the 
> > QSK switching speed is really controlled by the switching 
> time of the 
> > bias switching transistors.  These transistors are many 
> times slower 
> > than either
> > type of diode. Therefore, unless you just like to spend 
> more money than 
> > you
> > need to, PIN's are not necessary (an expense of $5 vs $200 
> for a set of
> > diodes).
> >
> >    The other factor that is  really misunderstood is SWR with a QSK
> > system.
> > The basic argument here is that diode QSK will not handle 
> SWR.  Not so, at
> > lease with in reason.  The maximum SWR that a QSK board can 
> operate with,
> > without self destruction, is determined by the value of 
> reverse bias 
> > placed
> > to the diodes during transmit.  This is usually a fairly 
> low DC value,
> > sometimes as low as 250 volts, and seldom over 400 volts.  
> The HeathKit
> > boards and early Ameritron boards had switching transistors 
> that were 
> > fairly
> > reliable, at these voltages.  The switching transistors in 
> the current
> > boards will work reliable with 400 volts reverse bias, most 
> of the time.
> > However, do not try to increase reverse bias with these 
> transistors.  I 
> > have
> > had them literally explode off the board with only 600 
> volts reverse bias.
> > I replace all four switchers with 1200 volt devices.  These 
> devices switch
> > reliable with 900 volts reverse bias applied (the highest I 
> have tested).
> >
> >        With these new switching transistors installed, 
> along  with the 
> > above mentioned PN diodes, and 900 volts reverse bias, I have run 
> > repeated, two hour long tests, with 2500 watts and SWR  of 
> 4 - 5 to 1 
> > with absolutely
> > no failures. (yes, I did reverse the SWR)  In, fact, taking frequent
> > temperature readings, on 160, 80, 40, and 20 meters, with 
> an inferred, 
> > LASER
> > guided, thermometer, I found that the PN diodes actually 
> run between one 
> > and
> > three degrees C cooler  then PIN's.
> >
> >        Not  a week goes  by that I do not buy one, or two, of these
> > boards,
> > sometimes the complete stand alone units, that the owner 
> can not make 
> > work,
> > or in unsatisfied with.   I do not understand why anyone 
> would remove one 
> > of
> > these boards from an amp.  With just a little updating, 
> they are extremely
> > reliable and a pleasure to operate on any mode.
> >
> >
> > 73 de Lon, K5JV
> >
> >
> > _______________________________________________
> > 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

_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps

<Prev in Thread] Current Thread [Next in Thread>