[AMPS] Re:

Jon Ogden jono@webspun.com
Tue, 12 May 98 19:34:17 -0500

>>derived through proper DOE (Design Of Experiement) procedures, would 
>>interested parties ante up?
>AMEN Arlen! Count me in.

I agree, Carl.  I would be very interested in this.  It would put to rest 
all of the "religion" that all of us experience with regard to this 

However, I agree as you do that it would be difficult to consistently 
recreate an oscillatory condition w/o some work.  DOEs typically need to 
be controlled experiments with the correct variables picked to do the 
proper analysis.  If the wrong variables are picked, or the wrong limits 
on the variables or the experiment is not properly controlled, you can 
compile all the statistical data you want and it won't mean a thing.  
Believe me - been there; done that.

So Arlen and Carl, what sort of DOE do we want to do?  Keep it simple and 
do a 2x2?  Or should we make it more complex.

A simple 2X2 could consist of:

Variable One:  Oscillation - Yes or No
Variable Two:  Supressors - Rich's or Carl's

Then we could create a 2 x 2 matrix with these variable.

However, I would suspect that this isn't enough and we may have to go 
larger like a 4x4 as we haven't even taken the gas arcs into account, let 
alone any other modifications that are done to the amps.

Also, you would want to do this on multiple amplifiers.  How many 
repetitions are we going to have? 3?  4?

I am not being funny or sarcastic here.  I am totally serious.

All this is needed for a successful DOE.  I have texts and papers on this 
subject if anyone needs info..........
>I do not think that a methodical approach will work very well...at least
>with the SB-220. I have tried numerous times over the past few years to
>recreate "events".

Well, then a true DOE probably won't work.  It's gotta be done 

I guess trying to catch parasitic events is like trying to find Big Foot 
or the Loch Ness Monster.  Some people claim all sorts of stuff!  :-)

>As has been previously pointed out, there is no guarantee that the series
>cap will work on every antenna combination. With a Bird dummy load there
>is no problem. With the average feedline to say a tribander being 100' or
>so the flat loss in the coax will flatten out the VSWR antenna response
>at VHF/UHF.

If the output Pi-network of the SB-220 has a low-pass response as most 
output networks do, then what difference at VHF/UHF does the antenna make 
anyhow?  Do we really need some sort of a load?  After, everyone talks 
about how the output Pi-Net works as a filter to remove the harmonics, so 
if it removes (or reflects - The S11 of the filter at the frequency of 
the harmonic energy is high) harmonics, won't it also reflect VHF/UHF 
energy?  Why worry about how the HF load looks at VHF/UHF?




Jon Ogden


"A life lived in fear is a life half lived."

FAQ on WWW:               http://www.contesting.com/ampfaq.html
Submissions:              amps@contesting.com
Administrative requests:  amps-REQUEST@contesting.com
Problems:                 owner-amps@contesting.com
Search:                   http://www.contesting.com/km9p/search.htm