[AMPS] Re: Parasitics

Rich Measures measures@vc.net
Fri, 22 May 98 11:12:11 -0800


>Rich Measures wrote:
>>>Rich Measures wrote:
>>>........
>>>>In Wes' 
>>>>measurements, he uses the term "Ls".  What is "Ls",  Mr. White?  
>>>
>>>Copied and pasted directly from Wes's own page: "the measured effective
>>>(series) inductance".
>>>
>>Mr. White:  You, sir, are correct.  Sorry for the confusion.   
>
>OK!
>
>>Henceforth,  I propose that "Lsu" designate the vhf parasitic 
>>suppressor's inductor, and that "Rsu". designate the suppressor's resistor.  
 
>>
>Rather than use "s" at all (which could still be confusing), 

which is why I proposed using su as the subscript.  "su" is hardly the 
same as "s".  


>could I propose simply "R" and "L", AKA "the resistor" and "the inductor".
>
Seems a bit chancy.  

>>As you see it, at 100MHz: 
>>What was the inductance of the copper-wire Lsu?  
>>What was the inductance of the resistance-wire Lsu?  
>>
>Remember that the Impedance Meter can only measure the *effective*
>values that it sees between its binding posts. 

agreed

>In network theory jargon,
>what it  measures are always *equivalent* values. Internally the
>instrument measures something like magnitude and phase, and then it
>calculates either Rs and Xs (or Ls), or Rp and Xp (or Lp). It's entirely
>your choice, and all the answers are equally valid at the measurement
>frequency.
>
>There isn't much point in looking for the "true" values of R and L in
>the suppressor, because the amplifier circuit sees the same *effective*
>values that the Impedance Meter measures.
>
I was talking about the values that the HP  Z-Analyzer measured.  As I 
recall, Wes measured one of the 100 ohm suppressor resistors we use for 
Rsu and found that that the intrinsic L was around 11nH, which is not at 
all bad at 100MHz.  .  

>Anyhow, to pick a few more bones out of Wes's tables...
>
>For a perfect inductor with no losses and no self-capacitance, Rs=0,
>Rp=infinity, Ls=Lp. 

ok
>
>You can see this in Wes's tables, where Ls and Lp are identical for the
>copper-wire inductor because the losses in the inductor itself are very
>low. The value down at 10MHz is 105.7nH, and that's as close as you'll
>come to a "true" inductance measurement. The *effective* inductance
>increases at higher frequencies because of self-capacitance, which Wes
>estimated to be about 2pF.
>
seems reasonable.  Does it seem reasonable that one should try to build a 
lossy device with a very low loss inductor?

>The nichrome inductor has built-in resistive losses, so Ls and Lp are
>different, even at low frequencies. The skin effect makes Ls and Lp move
>apart quite quickly at higher frequencies, but they are both valid
>quantities.
>
>>At 100MHz, the measured Q of the copper-wire suppressor was 2.2, and the 
>>measured Q of the resistance-wire suppressor was 1.5.  How do you account 
>>for VHF-Q being c. 46% higher in the copper-wire suppressor?  
>
>Because of the resistive losses in the nichrome wire. When you think of
>the lossy inductor as its parallel equivalent, the losses appear as an
>equivalent resistance in parallel with the physical resistor. This makes
>Rp lower and so Q is lower too.
>
agreed
>
>>What change 
>>would you make to the copper-wire suppressor to decrease its VHF-Q to 1.5?
>>
>Reduce Rp by adding physical resistance in parallel with the existing
>resistor (or decreasing the resistor value). 
>
Paralleling the c. 109-ohm Rsu with 1-ohm would undoubtedly lower Q, 
however, doing so would undoubtedly destroy the equal division of current 
that is apparently necessary to build a successful VHF suppressor.  In my 
opinion, Q can be reduced by increasing Lsu -- provided that Rsu is not 
burnt-out at 28MHz, and provided that semi-equal VHF current division 
takes place between Rsu and Lsu.  (see below)

>However, one thing this discussion has taught me is that the Q of the
>suppressor is *not* a magic number; it doesn't tell you everything you
>need to know about suppressor performance. To get the optimum
>performance in a real-life amplifier, you have to optimize several
>different things at the same time... and that's where the hands-on
>experience comes in.
>
Surely.  Obviously, a 1-ohm Rsu sucks.  It is my unexpert opinion that to 
achieve the desired staggered-resonance effect in a VHF suppressor, Rsu 
should be semi-close to the reactance of Lsu at the anode's VHF resonance 
-- i.e., fairly equal amounts of VHF current need to pass through Lsu and 
Rsu.  This was one of the major points of contention during the grate 
suppressor debate -- i.e., Mr. Rauch asserted that virtually no VHF 
current passes through Lsu, and that virtually all VHF current passes 
through Rsu, so it makes no difference whether Lsu is made out of 
resistance-wire or made out of copper-wire.   (pp. 71-4, Sept., 1994 
*QST*) 



Rich...

R. L. Measures, 805-386-3734, AG6K, www.vcnet.com/measures  


--
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