The process has been described in a 50's QST article by Bill Orr, I dont
have the details handy but a search of the ARRL archives should work. The
article also shows a tapped L suppresor which I havent seen in use, its
something Ive got to make time to test.
The simplest way to start is with a calibrated absorbtion wavemeter which is
all that was available in the early days. Fire up the amp at the lowest
level necessary for it to just barely start indicating a parasitic but at a
safe level. Then its a matter of calculating the inductor. Strays play a big
part and their effect goes up with power especially when close coupled to
the tube. This makes it hard when the tube is jammed next to a wall and a
cover is in place but its still possible to get some indication thru the
ventilation holes.
Ive not found a GDO to be of much use as there are way too many spurious
resonances that confuse the readings. However a GDO in the diode mode is a
much more sensitive indicator than a absorbtion wavemeter. Ive used a
Measurements Model 59 for decades in that mode.
A spectrum analyzer is the tool I use now as its pretty foolproof and shows
things the others cant. (When I was working I could usually bring a SA home
on weekends. Then I also obtained a 141T system in a rather creative way
when HP was offering high trade ins on obsolete SA's. I still have that but
its used in the operating position and a couple of newer ones are on the
bench.) For instance the SA can show a spurious resonance in a multiband
tank circuit building up before it reaches the point of blowing across the
switch or Tune cap. This is what Measures keeps calling a parasitic but isnt
and can be killed with a snubber cap across switch contacts as Ive mentioned
for years on here and elsewhere.
In a 3-500 6M amp you can see the parasitic as well as the 3rd harmonic
which are fairly close. Then its a matter of tweaking the suppressor so that
it works on both and you can place a brick on the key and nothing burns up.
Plus the amp efficiency is text book. A LPF brings remaining harmonics into
the noise. Dont bother with a Pi-L on 6M, they are more trouble than they
are worth, a good LPF is a better choice.
There is no magic formula that works in real life. I suppose some computer
genius can create a model but IMO there are way too many variables to make
it anything but a chore to use.
The resistor has the sole purpose of just stopping the parasitic from
starting and not absorbing any of its power or any of the wanted RF on 10 or
6M. This becomes a balancing act with glass tubes such as the 811/572B,
4-1000, 250TH, etc. External anode tubes have their parasitic well up into
the high VHF or low UHF region and the 4-400/3-500 somewhere in between.
Its a fallacy to say use a 3T coil for everything. Proven designs cover from
7 to 3 turns close wound on a 1-2W resistor, it varies with the tube type.
With Globars being expensive the 2W carbon has been the R of choice. However
they are out of production except in OEM quantities. Old stock may be way
out of tolerance and ineffective. One way to use a good 2W is to wind a 1-2
turn coil of roughly 1" diameter with the R centered, copper strap works
well here and minimizes heating. This method reduces the coupling to the R
and requires a more accurate setting of the L to the parasitic frequency.
If heating in the R occurs then vary the coil turns spacing and see if it
helps. In particulary bad cases with very strong parasitics then parallel
more R and keeping it around 50 Ohms.
An alternative to the carbon R is a 5W MOX. This has some L depending upon
construction but that can be negated by a couple of silver mica caps. Ive
described the construction previously.
To answer the basic question....some magic is still required but most can be
eliminated with proper layout, construction practices, and some common
sense.
Carl
KM1H
----- Original Message -----
From: "Alek Petkovic" <vk6apk@bigpond.com>
To: "DF3KV" <df3kv@t-online.de>; <amps@contesting.com>
Sent: Thursday, July 15, 2010 6:29 PM
Subject: Re: [Amps] parasitic suppressor voodoo
At 09:27 PM 15/07/2010, DF3KV wrote:
>Parasitic suppressors should be designed for the circuit which needs it.
OK. Just a simple question from down under.
Who, among all the subscribers on this list, knows how to do that?
I'm not talking about cut and try methods. I'm
talking about mathematics and tube spec sheets
and parasitic frequencies etc etc. I have NEVER
seen anything, anywhere, which gives the designer
or builder of an amplifier, the tools or formulas to work this stuff out.
Time and time again, we hear this Nichrome stuff
raised and time and time again, the two camps
slug it out and time and time again, nobody wins.
I don't wish to align with any side here. My question is genuine.
How would I design a parasitic suppressor from
scratch? What formulas would I use to calculate
the L, R for a particular tube or a particular rf deck layout?
Don't tell me stuff like "four turns and 50 ohms
worked for me." I am talking science and that
sort of approach is NOT scientific. Gimme calculations. Somebody must know.
73 and thanks,
Alek. VK6APK
>The parasitics frequencies vary with tube selection and amp layout.
>Most of the time I don´t even need one.
>
>
>73
>Peter
>
>-----Original Message-----
>From: amps-bounces@contesting.com [mailto:amps-bounces@contesting.com] On
>Behalf Of Randy Pence
>Sent: Donnerstag, 15. Juli 2010 14:10
>To: amps@contesting.com
>Subject: [Amps] parasitic suppressor voodoo
>
>Can someone explain the difference between the Parasitic Suppressors **
>that
>Rich Measures sells compared to the ones Jeff Weinberg sold.
>
>--
>Randy Pence
>N4JZY
>_______________________________________________
>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
>
>
>
>
>=======
>Email scanned by PC Tools - No viruses or spyware found.
>(Email Guard: 7.0.0.18, Virus/Spyware Database: 6.15440)
>http://www.pctools.com/
>=======
From sunny Binningup, Western Australia
http://www.qrz.com/db/vk6apk
http://www.qrz.com/db/vk6ap
=======
Email scanned by PC Tools - No viruses or spyware found.
(Email Guard: 7.0.0.18, Virus/Spyware Database: 6.15440)
http://www.pctools.com/
=======
_______________________________________________
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
|