A friend of mine sent me this quote from the 20th Edition of Editors and
Engineers:
*****
Page 21.15 Distortion Products
"Laboratory measurements made on various tubes in the circuit of figure
10A show that a distortion reduction of the order of 5 to 10 decibels on
odd-order products can be obtained by operating the tube in cathode
driven service as opposed to grid-driven service. The improvement in
distortion varies from tube type to tube type, but some order of
improvement is noted for all tube types tested. Most amateur-type
transmitting tubes provide signal-to-distortion ratios of -20 to -30
decibels at full output in class-AB1, grid-driven operation. The ratio
increases to approximately -25 to -40 decibels for class-B grounded-grid
operation.........
In general, slightly less power output and efficiency is observed with
the untuned-cathode circuit, odd order distortion products run 4 to 6
decibels higher, and the circuit is harder to drive and match to the
exciter than is the tuned-cathode circuit of figure 10A. Best results
are obtained when the coaxial line of the driver stage is very short a
few feet or so. Optimum linearity requires cathode-circuit Q that can
only be supplied by a high-C tank circuit.
Since the single-ended class-B grounded grid linear amplifier draws grid
current on only one-half ( or less) of the operation cycle, the sideband
exciter "sees" a low impedance load during this time, and a very
high-impedance load over the balance of the cycle. Linearity is therefor
affected and the distortion products of the exciter enhanced. Thus the
driving signal is degraded unless the unbalanced input impedance can be
modified in some fashion. A high-C tuned circuit, stores enough energy
over the operating r-f cycle so that the exciter "sees" a relatively
constant load at all times. In addition, the tuned circuit may be tapped
or otherwise adjusted so that the SWR on the coaxial coupling the
exciter is relatively low. This is a great advantage, particularly in
the case of those exciters having fixed-ratio pi-network output circuits
designed expressly for a 50-ohm termination.
Finally, it must be noted that removal of the tuned cathode circuit
breaks the amplifier plate circuit return to the cathode, and r-f plate
current pulses must return to the cathode via the outer shield of the
driver coaxial line and back via the center conductor! Extreme
fluctuations in exciter loading, intermodulation distortion , and TVI
can be noticed by changing the length of the cable between the exciter
and the grounded-grid amplifier when an untuned cathode input circuit
and a long interconnecting cable are used."
*****
Ok, here we have finally gotten a documented number; 4 to 6 dB HIGHER
with an untuned cathode driven amp than a tuned input. 6 dB is a
significant amount in my book.
Now it would be nice to see validation of these numbers on real amps
that we find in use today.
It is also very interesting reading in the last paragraph regarding the
coax used to connect the exciter to the amp.
73, Tony W4ZT
Tony King - W4ZT wrote:
> Friends,
>
> There are plenty of claims that it is wise to use tuned input circuits
> on Grounded Grid amps and that the reasons are reduced IMD, better
> efficiency and better matching to modern day exciters. It is easy for us
> to see the matching benefit. There doesn't appear to be any published
> data to demonstrate the differences of IMD and efficiency in a real amp,
> or several real amps between using tuned input and not using a tuned input.
>
> Have any of you documented this kind of test before? Have you published
> the data? Would you care to? If you have, would you share it? I have
> searched the web but have found no published comparison between running
> with and without a tuned input circuit on a Grounded Grid amp. I don't
> have the equipment to run this kind of test but it certainly looks like
> it would be worth while doing. We could argue theory all day long (and
> we have!) but a practical demonstration would be worth a thousand arguments.
>
> Obviously it would be more meaningful if tests were done on more than
> one amp but even one good demonstration could prove the point. I would
> suggest using the same drive power, not overdriving the amp in any case,
> showing the output power difference and the IMD difference between the
> same amp with and without the tuned input circuit.
>
> Anyone care to comment, run the tests or offer their data?
>
> Thanks and 73,
> Tony W4ZT
>
> _______________________________________________
> Amps mailing list
> Amps@contesting.com
> http://lists.contesting.com/mailman/listinfo/amps
>
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|