To: <amps@contesting.com>
> Date: Thu, 21 May 1998 12:14:36 +0100
> From: Peter Chadwick <Peter_Chadwick@mitel.com>
> Subject: [AMPS] parasitics
> To: 'amps' <amps@contesting.com>
Hi Peter,
I appreciate your reply.
> Now, I don't know with an 8877 - never used one.
They don't change much Peter. Let me use a tube we recently have had
gain variation in, the 3CX1200A7. When gain changes as little as 0.2
dB, they fall out of our limits and we return them.
An exceptionally poor tube will have less than 1 dB gain difference,
most tubes are within .05 dB and require no circuit changes from
unit to unit. Idling current does vary, but not greatly.
>New RCA 6146B show a
> variation in plate current for fixed plate, bias and screen voltages from
> 4mA to 27mA for new tubes, with a standard deviation of about 11mA.'
Sure, but that's at the knee of the curves. in a grounded grid
triode PA the output is in series with the input, and feedback is
mainly controlled by the ratio of driving impedance to output
impedance. Transconductance has a very small effect on gain, other
than how it might slightly affects input impedance.
> 4CX250B's show similarly high variations. Then there's the variation in
> interelectrode capacity. I looked into the variation side when I did a
> marine tx with 6146B's in the 70's to decide whether it was better (ie
> cheaper) to use individual bias pots, or select tubes. We went for the
> pots...................
I agree. Tetrodes are more of a problem, because ALL internal
tolerances add up to affect performance.
> >Two more questions. How did you know the amps had parasitics?
> >What lead running near the glass?
>
> 'Cos with a 50 ohm input 'load' and a 50 ohm output load, the plate current
> varied as you varied the tuning, as would the grid current. Some tuning
> points, just standing plate current and no grid current. Others, lots of
> both. A neon lamp on a stick offered up to the tank circuit had that sickly
> bluey - yellow - orange sort of glow that goes with VHF energy, and at the
> tank capacitor, it didn't glow as brightly - lots of variation as you move
> it around the leads. That, to me, (and, I hope, to you) is adequate evidence
> of VHF parasitics. LF parasitics caused by chokes resonating in plate and
> grid usually give a very deep red colour in a neon on a stick, and the
> intensity doesn't vary much as you move it along the leads.
Did you measure the frequency? Is there some reference I can read
that indicates what color is what frequency? I'm not being critical,
I just have never used a neon bulb to measure frequency, and I have
no idea how accurate it is as a frequency measuring device.
I have found the bulbs change color with voltage.
> The lead running near the glass was the bypassed HV lead as I remember it.
> Trouble is that memory fades, and it was about 25 years ago, but to this
> day, I can picture that amp and the glow of the neon.
Most 572B amps I've seen oscillate at about 60-100 MHz or so at VHF,
and also at upper HF (20-35 MHz). Neutralization or swamping
cures the HF stuff, but the VHF stuff is a bear at times. They
require a very high suppressor ESR to lower Q, since the anode leads
are very long and inductive.
> But I still wouldn't go for an HF amp without suppressors without a lot of
> evaluation and development - and probably not even then. Even if I don't
> agree with Rich about bending filaments with parasitics!!!
Doing something because it makes people "feel comfortable" is ok.
Myself, I think parking the anode at 200 MHz and the grid at 490 MHz
is a good enough spread.
I look at it this way Peter.
MANY VHF amps work quite well with the 8877 , and they intentionally
have the anode and cathode at the same frequency in a high Q circuit.
The only time I've seen a problem is when the anode has a resonance
up on 450-550 MHz.
73, Tom W8JI
w8ji.tom@MCIONE.com
--
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
|