Amps
[Top] [All Lists]

[Amps] 8877 stability in PAs with a few more examples TSPA

To: amps@contesting.com
Subject: [Amps] 8877 stability in PAs with a few more examples TSPA
From: "John T. M. Lyles" <jtml@lanl.gov>
Date: Wed, 31 May 2006 09:19:31 -0600
List-post: <mailto:amps@contesting.com>
I suppose Rich is referring to amateur radio 
power amplifiers that he is 'aware' of when this 
was stated:

>
>Every manufacturer that I'm aware of on this planet, except?
>Ameritron, uses a parasite suppressor on their 8877.  I'm doubting?
>that everybody else is out of step and only Ameritron got it right.??
>Just how the AL-1500's designer came to the conclusion that 5000-ohms?
>of feedback XC at the tube's max freq rating could not cause
>oscillation is a mystery.
>
>R L MEASURES, AG6K. 805-386-3734

I will provide some other examples which don't, 
to improve this awareness. Several successful 
commercial power amplifiers that I know well do 
not have suppressors between the anode and tank. 
The Broadcast Electronics FM1.5A/B fm transmitter 
which i helped design back in 1983 with Bill 
Harness and Geoff Mendenhall in Quincy, IL, ran 
8877 in class B in a transmission line circuit at 
1500 watts TPO,  and didn't require a suppressor. 
The RF plasma Products HFS-1000G 13.56/27.12/40 
MHz industrial RF generator had none either. I 
converted one of these over to 5 MHz operation in 
October of 2002 ( see archives here for my 
comments then). It worked beautifully and had no 
instabilities. With this 5 MHz amplifier, the 
idling current (Zsac) was set for class AB1 
setting. I ran it into horrible mismatches at 1 
kW output, adding an AGC loop to maintain full 
output even into mismatches, as I was testing 
ferrite cores for a new synchrotron accelerator 
design, and needed consistent voltage across it 
as it swept frequency and tune.

I built a 100 kW class A pulsed amplifier here at 
work in 1998, using a push pull pair of Thales 
TH555A tetrodes, each rated for 200 kW in the HF 
range. Using ferrite cores mounted around the 
socket (basically bolted to the screen ring) this 
amplifier was also stable without drive and 
mistuned. I originally designed a parasitic 
suppressor using ferrite tiles made by Fair Rite 
company, that sat bolted within the fold of a 
hairpin inductor in the anode to tuning cap 
strap. This worked well on the cold sweeps using 
a network analyzer, in getting rid of a pesky 
high Q peak at 17 MHz. However, under power, the 
amplifier proved that it was not needed, so the 
parts were not installed. The amplifer runs at 
2800 KHz.

I am not saying that every amplifier can be built 
without suppressors, but the blanket statement 
that every 8877/3CX1500A design has them, is 
unrealistic and also not correct. Layout and 
component selection of the final tuning capacitor 
and tank circuit, as well as input network 
design, affects amplifier sensitivity to 
parasites, as well as the G-BW product of the 
tube, as Rich also refers to - for the 8877 
triode.

Keep up the good work, I rather enjoy discussions 
over arcane topics like parasites (which are the 
bane for every tube power amplifier designer) 
instead of how to wire AC circuits in a ham 
shack. Although, I am about to embark on a new 
garage project - planning to have 220VAC single 
phase outlet in the back for you know what....

73
John
K5PRO

_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps

<Prev in Thread] Current Thread [Next in Thread>