W4EF wrote:
Is Wes' raw impedance data for the conventional and
nichrome suppressors published anywhere? Without any
data to study, this debate ends being more like a
political battle than scientific inquiry. Seems like
letting everyone see the impedance data for themselves
would settle the matter.
Mike, W4EF...........................................
----------------
Mike, as things would have it, the answer is "yes" but...as in "yes" but you
might not be too pleased.
It appears that the suppressors that Wes tested are not the same suppressors
that Rich provides today. Rich advises that the length of wire in today's
supressors is the same as the length of wire used in the unit Wes tested.
But the one Wes tested was wound about a resistor, whereas "today's"
suppressors apparently are hairpins. Rich appears to believe that the
length of wire is what makes the inductance. Length of wire, of course, is
secondary to the area enclosed by the wire and also secondary to mutual
coupling effects in a coil. Thus, a hairpin today made of the same length
of NiCr is a distinctly different component than a coil that is merely made
of the same length of NiCr. This is fundamental physics that we do not yet
have permission to alter.
I have some concerns, too, about the approach taken to this matter. We must
seriously consider what our intent is when using a suppressor, or any
component. Very recently, Rich provided this quote:
"The combination of both resistance and inductance is very effective in
limiting parasitic oscillations to a negligible value of current. F.
E. Handy, 1926-edition of The Radio Amateur's Handbook, page 72."
I find this to be an extraordinary statement. When one observes writings
about that time, it becomes evident, if not proved, that parasitic
oscillations were an accepted norm for the day. They existed, and the goal
was merely to keep them from causing troubles. Thus, this statement holds
for 1926. However, with today's better components, and our enhanced
knowledge of the need for short connections, it should be considered very
poor engineering practice, indeed, to merely pursue the reduction of
parasitic oscillations to a negligible level of current. The only useful
goal with today's technology is to not allow the parasitic oscillation to
start in the first place. This is a very different matter than reducing the
levels. Therefore, prudence suggests that we attenuate suggestions based on
such early work significantly, as our goals have changed dramatically.
This other item is one of some delicacy, and I regret presenting it, but I
hope the level of stridence on the reflector may abate as a result. I
engaged in an email exchange with N7WS a time ago. He advised me that he
had, indeed, placed the results of his measurements on the web, but only
after he had numerous assurances that Rich would never quote the results in
an effort to demonstrate that supressors were beneficial. I shall leave it
to the readers of the reflector to determine whether Rich has violated that
agreement.
However, you can all see for yourselves this item: Wes' webpage indicates
that the material is copyrighted and can be freely distributed <only> for
non-commercial purposes. Unless Rich's suppressor sales activities are
legally established under IRS code 501(c) as a non-profit enterprise, he is
in violation of copyright law every time he mentions Wes' measurements in a
manner that makes his devices appear favourable.
I am really quite disturbed by the level of acrimony that appears to
accompany this topic. Although it appears I have targeted Rich, that is not
the intent. Inflammatory statements are made on both sides; Rich is simply
the only emoter on his side of the issue, and his postings are easier to
identify because of the frequent re-use of unusual phraseology.
Valves are amenable to modelling by the likes of SPICE, and circuit
parasitic reactances are, as well. There is no need to pound repetitively
on the nail of suppressor Q when a few hours work with a nearly free
software package will show whether it makes sense to do so or not. If the
amount of time spent pounding the keyboard to make postings here (I shudder
to consider what things look like on the un-moderated usenet lists) were
spent performing responsible analyses, this discussion would not be taking
place.
Wes' measurements are at:
http://www.azstarnet.com/~n7ws/supp-2a.htm
For myself, once I began designing with valves whose dimensions were less
than 1/8 wavelength at my operating frequency, I determined that all
instances of parasitic oscillation were due to poor component layout. Pay
particular attention to where ground currents flow. That tuning capacitor
has a grounded end. Where does that ground current need to flow? To the
grid of the grounded-grid valve! Is that distance short? What about the
load capacitor? Do you know where that ground current must flow? Yes, to
the output connector! You see, star grounding is not necessarily correct
all the time. Just because a component has a short connection to the
chassis does not mean it's well grounded.
Whew. Enough for me for now, I'm at risk of being called the Deuteronomist.
FF
________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.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
|