>
>Peter Chadwick wrote:
>>
>>Colin says:
>>
>>>I believe most of the modern film resistors are made with a spiral design
>>>on a surface. Thus, they will have some inductance.
>>
>>So where's the difference between them and nichrome inductor, except that the
>>resistance is higher?
>>
>Inductance is smaller.
? And it's getting smaller. A few years ago, Matsushita 1-ohm, 3w
resistors had c. 36nH of inductance because the conductive film had a
number of turns. Currently, such resistors have c. 12nH of L because
they have many fewer turns. My guess is that the film is now more
resistive - so fewer turns are needed to arrive at 1.0 ohms. The 100-ohm
units we use in our suppressors have c. 10nH. Normally, two are
paralleled.
>......
>By using a bundle of several higher-value resistors in parallel, you can
>get the inductance down to quite low values.
Not all that low. This is a logarithmic function. Two, 10-nH units in
parallel do not = 5nH.
>This works as a grid load
>for 50MHz and even 144MHz, so it will work for parasitic suppressors
>too. (There's also a way to get the inductance down to exactly zero, but
>I won't steal someone else's lines.)
>
"exactly" sounds like Tom
>>Let's not lose sight of the fact a nichrome inductor will have a Q that
>>increases with frequency - at least, until the frequency is reached at which
the
>>self capacity affects the Q.
>>
>Let's not get too hung up on the Qs of individual components. What
>ultimately matters is the damping load that is presented to the tube at
>the frequency of the potential VHF parasitic.
>Calculating this load
>requires a lot of parallel<>series impedance transformations,
? (principally one) The suppressor is the main game in the anode
circuit because all of the anode current must pass through it.
>which
>involve ALL the other circuit values: tube capacitance, RF choke
>inductance and self-capacitance, stray series inductance and even the
>setting of C1 in the pi-tank.
Indeed. In Henry 2K4s and 3Ks, the output compartment appears to be
resonant near the parasitic freq. . . Murphy was undoubtedly right.
>
>All these other circuit values affect VHF stability - not just the
>parasitic suppressor - which incidentally explains why a particular
>suppressor design may work in one amp but not in another.
>
>Until I got down and crunched the numbers, I hadn't fully realised that
>the value of the suppressor Q is literally *useless*. To do the
>calculation, you need both the L and the R of the suppressor network -
>the two separate numbers. Rolling them together into a single value of Q
>creates a "dumbed-down" number that you can't actually use. (In other
>words, it's a mathematical proof of something we already knew - that the
>arguments about "my suppressor Q is better than yours" are completely
>vacuous.)
Q is directly related to the parallel equivalent R - which Wes calls
''Rp''. This is important because it directly affects VHF gain. Higher
Rp means higher VHF gain. At 100MHz, the copper-wire suppressor had an
Rp of 166-ohms, and the resistance-wire suppressor had an Rp of 101-ohms.
Of course, the Rp of the copper-wire suppressor could easily be reduced
to 101-ohms by simply increasing L-supp. The sticky wicket is that
dissipation in R-supp increases exponentially. If R-supp burns out on
10m, how effective is the suppressor going to be?
During the parasitics debate I kept bringing up the issue of R-supp
dissipation - but Wes and Tom stonewalled me.
>
>.....
cheers, Ian
- Rich..., 805.386.3734, www.vcnet.com/measures.
end
--
FAQ on WWW: http://www.contesting.com/FAQ/amps
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
|