>
>a follow up...
>
-
>: From: 2 <2@vc.net>
>>>The swamp resistance value would have to be
>>>changed. The indicated swamp R value appears
>>>to be similar to the average dynamic tube
>>>impedance.
>-
>>: 50-ohms might fly.
>-
>Just to be sure the point is not lost... The alpha input
>circuit parallels a swamp resistor network with the
>tube(s) cathode(s).
>-
I am aware of this.
>The swamp R values are/were probably chosen to
>allow convenient matching to the rf input transformer
>secondary, to "trade" a higher drive requirement well
>in line with many of the common transceivers ...
When Dick Errorhorn designed this amplifier, tube-output radios were
common. Since they had a tuned Pi output, they worked acceptably driving
g-g amps that lacked a tuned Pi input.
>and to add a measure (no pun intended) of stability.
>Most of the time, it is mucho betta practice to use a
>proper "LC" circuit.
>
Indeed
>The Alpha type circuit provides low swr to the exciter,
Not low emough for transistor-output radios unless a tuner is used.
>is also relatively broadband, is low in cost and physical -
>mechanical part requirements and easy to install.
>
The trade-off of having no flywheel is not small. -
>It is by no means the best, but it works much better
>than nothing. Distortion is present, but not considered
>excessive by the mfgr. A "trade off is made."
>-
>>: The sticky wicket is that the cathode impedance is far
>>: from constant. It varies from virtually infinity to approximately
>>: one-half of the mfg's spec impedance. A transformer is not
>>: going to work very well in impedance hell.
>-
>In one half cycle of operation, we have a resistor network
>and a parallel cathode impedance equal to the xmfr
>secondary. All seems well for this half cycle...
>-
Two 8874s in parallel have a driving Z of roughly half of 26-ohms at the
negative V peak. This requires matching with a Pi-network/flywheel.
>The alternate half cycle is (for the example) pretty much
>just the resistor network and some C. So the toroidal rf
>xmfr secondary "load" halves.
What is the swamper resistance?
>Not great, but much better
>than nothing.
Much better would be not having the swamper in the circuit during the
negative half of the cycle.
>The swamp R network remains in circuit,
>some of the secondary fixed loading is maintained. The
>average swr back to the exciter is "reasonable."
With solid-state radios, sans-tuner, the SWR usually needs to be 1.2:1 or
better to avoid power foldback. A swamper/transformer is not capable of
that.
>>There is
>>no flywheel effect provided by the rf input transformer.
>-
>My interest in your suggestion was the simple series
>circuit to provide the missing tube impedance
>in the alternate half cycle. Still, no flywheel effect is
>provided. It is still not the best choice, but possibly a
>better option if it works. Contributions by the additional
>series D&R circuit would probably have to be compared
>and measured.
>-
>>>Your suggestion taken to the next step, would
>>>be a series R & UF Diode paralleled with
>>>the current circuit. The R value chosen to be
>>>similar to the tubes parallel drive impedance
>>>value.
>-
>>: I do not see how such a circuit would do the job.
>>: Perhaps a road map would help.
>-
>It would not provide flywheel, just "better" exciter
>matching in the alternate half cycle.
Exactamente
>Impedancehell would potentially be less hell in the alternate
>half cycle.
>-
>>>The current amplifier drive requirements well
>>>suit the output of many common Amateur
>>>Transceivers. Most Amateurs will probably
>>>over drive the amplifier with the swamp
>>>R circuit removed or revised to provide only
>>>half cycle loading.
>>
>>: A fool and his money are soon parted.
>-
>The over drive protection might be or have been
>a method of real world manufacture self preservation.
>-
3-500Zs and other filament type tubes are virtually immune to overdrive
because the emissive ditungsten carbide layer is not frangible and it
limits emission to little more than than the anode current rating. All
of the 3-500Z autopsies that I performed showed that the dearly parted
died from:
1. Bent filament helices shorting against the grid.
2. Gas caused by leaky glass-metal seals.
3. Anode cooler shorting against grid due to mfg defect (Eimac).
4. Loss of emission from excessive fil. V (mostly four models of Henry
amps).
>A major draw back of an amplifier warranty period,
>is that it allows many operators to plug in and hit the
>gas before they read and understand care and
>feeding.
>-
Destruction by over-drriving an oxide cathode tube in g-g config happens
- even when grid current is within recommendations. In other words, a
grid trip is as useful as the tits on a boar hog.
>The bullet proof valve amplifier has yet to be
>made en'masse.
>
The 8169, 8281, 4CX1500A, 8170 & 8171 can apparently stand all but being
dropped on a concrete floor.
>>>Some of the relative gratis stability
>>>provided by the lower existing swamp R circuit
>>>might go away. But it might not matter with this
>>>ground grid layout.
>>-
>>: However, if series cathode resistors are added, things
>>: might be less squirrelly.
>-
>The Alpha circuit seems to work well for exciter
>matching, trading drive power for low swr and moderate
>but increased distortion. Alpha claims the distortion meets
>requirements for Type Acceptance.
FCC type acceptance regarding IMD is a joke because of paragraph D. (no
splatter is allowed outside the ham bands)
>-
>ciao for now
>skipp
>
cheers, Skipp
PS - answering your reply was a R-5 pain in the posterior because the
attribution marks were largely wrong and had to be corrected by hand
(mine).
- R. L. Measures, 805.386.3734,AG6K, www.vcnet.com/measures.
end
--
FAQ on WWW: http://www.contesting.com/FAQ/amps
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
|