[Amps] Parasitics

[Bill Fuqua]

         There are a number of types of parasitic oscillations. In grounded 
grid amplifiers the most common is VHF. But there is a fundamental mode as 
well but more often occurs in common cathode amplifiers ( tuned grid-tuned 
plate oscillation) but does show up in poorly constructed grounded grid 
amplifiers. A very rare type is a low frequency parasitic oscillation. I 
have encountered this twice in the past 10 years. Once when tracking down a 
paging system that was getting into repeaters in the area.  Using a 
spectrum analyzer I could see the offending paging system with 2 sidebands 
drifting  4 to 6 MHz on either side of the stable signal. It was sometimes 
getting into the local police and fire repeaters as well. This was a 150 
MHz paging system with a 5 MHz low frequency ( drifting )parasitic 
oscillation. These are usually caused by resonance in the RF chokes with 
other components.
         The strangest low fequency oscillation was in a 15 MHz 300 watt 
amplifier used in our linear accelerator.  This amplifier had large  30 Hz 
amplitude oscillation. Being solid state ( copied from Motorola Application 
Notes) it used feedback to establish a constant gain. The advantage of this 
is that it did not require a regulated power supply or even a very well 
filtered one. After making some measurements I discovered that the output 
of the unregulated power supply was oscillating at 30Hz which is strange 
because it runs off a 400Hz power source.  The problem was that under the 
right conditions the LC filter in the power supply ( resonant at 30 Hz) and 
the amplifier together formed a low frequency oscillator while trying to 
amplify a 15 MHz signal.  After a bit I realized that the amplifier 
appeared to be a negative resistance to the power supply and its LC 
filter.  Think about it, the gain of the amplifier is constant ( determined 
by negative feedback) . And if the efficiency remains constant or increases 
with an increase of the supply voltage the supply current must drop  to 
maintain a constant input power, and drop even more if the efficiency 
increases with supply voltage. That is negative resistance. The solution 
was to simply reduce the Q of the LC filter ( added a smaller bleeder 
resister).

73
Bill wa4lav


At 07:58 AM 8/21/2003 +0100, peter.chadwick at Zarlink.Com wrote:
>There's a header that should wake people up!
>
>In the latest QST, the article  on maintenance of linears suggests that
>poor neutralization can lead to parasitics. Not something I've come across
>before, although obviously, it can depend on exactly what you mean by
>parasitics - personally, I tend to think of 'parasitics' being 'VHF
>parasitics' but of course, that's not necessarily the case, although
>arguably the most common.
>
>I haven't the article with me here at work, but I don't remember seeing any
>advice about oiling fan/blower motor bearings, which I would have thought
>was likely to be desirable, depending on the fan or blower.
>
>73
>
>Peter G3RZP
>
>
>_______________________________________________
>Amps mailing list
>Amps at contesting.com
>http://lists.contesting.com/mailman/listinfo/amps