[AMPS] Another arc question
km1h@juno.com
km1h@juno.com
Sat, 09 May 1998 12:42:32 EDT
On Sat, 9 May 1998 07:19:31 +0100 "Ian White, G3SEK"
<G3SEK@ifwtech.demon.co.uk> writes:
>Carl wrote:
>>
>>On Sat, 9 May 1998 00:31:47 +0100 "Ian White, G3SEK"
>><G3SEK@ifwtech.demon.co.uk> writes:
>>
>>Lots of SNIPs
>>
>>
>>>increases faster as the frequency is reduced into the HF region.
>>
>>
>>Ian, I have a problem understanding your last statement. If the L was
>>properly selected for a particular tube AND circuit why would the Q
>>always detrimentally increase at a lower frequency? I look at the
>>suppressor network as a series trap and would expect the response to
>be
>>somewhat uniform on either side of its optimum frequency. The ideal
>>suppressor would then be only frequency dependent over the
>"parasitic"
>>range of the particular amp.
>
>As the words "series" and "trap" are normally used, that would imply
>that the suppressor is resonant
Wrong choice of words on my part Ian. But since the L is self resonant at
some point would this not be a consideration? After all it is simply a
RFC with a Q damping resistor across it. No different IMO than the old
grid bias choke-resistor combination from ancient Class C days. In fact
Heath uses the same combination in the SB-200 bias circuit.
The L in a SB-220 suppressor shows a self resonance of 210 MHz on my
trusty old Measurements 59 GDO. With the resistor attached I can not see
a dip which at that frequency is to be expected.
- which it ain't. It's a resistor
>paralleled by an inductor, and the network analyser measurements show
>that the conventional LR suppressor behaves very close to theoretical.
>
>As you increase the frequency, the network behaves progressively more
>like a resistor; as you decrease the frequency, it behaves
>progressively
>more like an inductor.
I dont remember hearing of a network analyzer test. I thought Wes used a
vector impedence analyzer. If I am mistaken...sorry. I have little free
time to play on WWW these days.
>
>Truly resonant suppressors do exist, particularly for mode killing in
>cavity amplifiers, but here we're talking exclusively about "lowpass"
>suppressors which are intended to be lossy at VHF but transparent at
>HF.
There is no way to get around the stray C in the L; therefore it must be
resonant at some point. In the SB-220 it just happens to be slightly
above the parasitic frequency of that particular amp. Also, this is
unloaded so I would then expect it to lower somewhat in circuit. Now...is
this a miraculous coincidence OR design?
>
>
>>I know of no modern tube that requires any excessive R dissapation at
>>30MHz. Indeed, the parasitic frequencies are well above 50MHz and if
>the
>>suppressor was designed for the parasitic it should then run cold at
>>28MHz.
>>
>Sure - no problem with that. The higher the VHF resonance is that you
>need to suppress, the more chance there is of designing a suppressor
>that looks lossy at VHF but also runs cool at 30MHz.
>
>>One very good argument for a lossy L is that it works equally
>well...or
>>poor over a much wider frequency delta. The "classic" suppressor may
>very
>>well be far superior at some particular narrow range of
>frequencies...but
>>what is its characteristics outside of that area?
>
>Check the graph of measured Q from N7WS. On a log-log plot, the
>classic
>LR suppressor gives an almost straight sloping line (just as it
>should)
>with absolutely no sign of resonance effects below 200MHz where the
>measurements stopped.
But the L and the R are critical and one size does not fit all tubes or
even the same tube in all layouts. Sloppy layout 4-1000A amps are a
good example of the wide range of suppressor requirements. The 572B,
6146 and other tubes require considerably more L, at least in classic
commercial products. This tells me that reasonance, call it whatever you
like, must be part of the design calculations.
I believe it was Tom who showed that there must be a certain dB level of
gain reduction at the parasitic frequency before the tube would be
stable. So therefore that sloping line plot you mentioned must be
shifted to place the attenuation where it is needed.
>
>The nichrome suppressor has significant losses in the inductor as well
>as in the resistor. That's the only difference. The question is
>whether
>this is an advantage or not...
I believe it has an advantage where the parasitic is not sufficiently
suppressed in the original network. It makes it easier for "Joe Sixpack"
who has neither the knowledge or equipment to design his own.
One thing that this discussion must make clear is that you can not lump
all of the failure events under one heading. You have gas, parasitic
resonances in the tube and VHF/UHF resonances in the tank circuit. Three
completely seperate issues that have to be addressed individually....even
though there certainly may be interactions. Thats what make amplifiers
fun.
73 Carl KM1H
>
>
>>The issue, as I see it, is a "monkey see, monkey do" attitude.
>Someone
>>published a suppressor design back in ancient days and it has been
>copied
>>ever since as a cure all for one particular tube in one particular
>layout
>
>How true!
>
>>. Ian, you have a similar problem trying to educate people about
>tetrode
>>screen supply design.
>
>Yes - but please, Please, not in this thread!
>
>
>73 from Ian G3SEK Editor, 'The VHF/UHF DX Book'
> 'In Practice' columnist for RadCom (RSGB)
> http://www.ifwtech.demon.co.uk/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
>
_____________________________________________________________________
You don't need to buy Internet access to use free Internet e-mail.
Get completely free e-mail from Juno at http://www.juno.com
Or call Juno at (800) 654-JUNO [654-5866]
--
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