On Thu, 11 Sep 1997 08:46:57 +0100 "Ian White, G3SEK"
>
>The N7WS measured Rp and Lp values for Rich's suppressor at 100MHz are
>Rp = 103.2 ohms, Lp = 123.9nH. With ideal components, those are the R
>and L values you'd connect in parallel to EXACTLY reproduce the
>behaviour of Rich's suppressor at 100MHz.
>
>But that simply begs two more questions:
>
>1. What about real-life components?
>2. What happens at other frequencies?
>
>You can't "design" a real-life suppressor entirely on paper, because
>of
>the stray properties of the components at VHF: the copper-wire
>inductor
>has very significant self-capacitance and some series resistance, and
>the resistor has some self-inductance.
>
>What you CAN do is sit at an impedance/network analyser with a bunch
>of
>real resistors and some copper wire. Using the idealized values as a
>starting-point, you can cut-and-try to find a combination of values
>that
>will EXACTLY reproduce the measured parameters (Rp and Lp) of Rich's
>R/NiCr suppressor. Note that the analyser impartially measures the R-X
>properties of whatever is connected to its terminals - it doesn't know
>or care how the network is constructed.
>
>With a wide enough choice of R and L values, it is ALWAYS possible to
>reproduce Rp, Lp and therefore Q, at ONE frequency. There is
>absolutely
>no doubt about that. Naturally you would choose the frequency of the
>VHF
>parasitic resonance in the PA that you're trying to suppress.
Which is roughly determined to be the frequency of the highest amount of
G-P feedthru...is that a valid assumption Ian? If not, then how would YOU
determine the parasitic frequency? Preferably this would be in a home
brew amplifier and prior to a smoke test in order to be more valid to the
purpose of this reflector.
>The second question is: what is the difference between the two types
>of
>suppressor at other frequencies?
>
>Plotting the N7WS measurements on a (log Q) versus frequency scale
>showed that the Q values of the two different types of suppressor
>track
>quite closely across the VHF range. The measured Q values were about
>40%
>different at 100MHz, but the two curves run pretty much parallel
>across
>the VHF range. On a log scale, that means that they pretty much
>maintained a 40% ratio between them.
>
>So, if you make a conventional R/L suppressor that mimics the Rp-Lp
>behaviour of the R/NiCr suppressor EXACTLY at one chosen VHF
>frequency,
>you'd then find that the two curves remain very, very close across the
>VHF band.
>
>But N7WS's measurements show that the Q of the conventional suppressor
>rises FASTER at HF - in other words, at HF it looks less like a
>resistor
>and more like a small inductor, which is exactly what it's supposed to
>do. The Q of the R/NiCr suppressor also rises at lower frequencies,
>but
>more slowly. In other words, the significant difference between the
>two
>types of suppressor is not at VHF, where their performance can be made
>almost identical. The main difference is at HF.
Contrary to other comments, I could not measure any meaningful power
difference between either type on 10M in either a SB-220 or a TL-922.
Have you run any actual tests at power Ian?
Another point that disturbs me greatly is that the AG6K suppressor in
the N7WS article does not come even close to matching what is shipped in
Rich's kits or in his 2-11-96 5 page flyer. Rich....am I
missing something basic here??
Wes wound 4.5 turns of .25" id and adjusted for 90-110nh at 10MHz.
Nowhere in any AG6K paperwork can I find that configuration. It appears
to be more in line with a 572B/811A suppressor and NOT the supposedly
apples vs apples 3-500Z type since the W8JI supplied one was from the
AL-80B.
Yet the AG6K versions I have here in my hand and supported by the
accompanying paperwork are a 3.5" long U shaped loop with a stated value
of 60-70nh. (L Meter freq not specified). They are also shunted by two
paralleled 100 Ohm resistors, not just one as in the review.
IMO, this appears to throw a big WHOA into the discussion. Just what did
Wes measure???
At this moment I think we need to consider a new evaluation from a less
involved and seemingly biased source. I would suggest a test at the ARRL
lab that is witnessed by at least a representative of both sides.
I would appreciate some meaningful commentary from multiple sources
please.
Tnx, Carl KM1H
>
>OK, that's it for a second time, in a different newsgroup and
>hopefully
>explained even more clearly.
>
>
>73 from Ian G3SEK
--
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
|