? ? ? WARNING This reply is too long.
>>>1.) I'll agree that the peak power HANDLING capability of the tube can be
>>>related to its anode voltage.
>>
>>? this is not what I said.
>
>I know. I was intentionally putting words into your mouth. :-)
>>
? hardly amusing
>>> But to me the peak of a particular signal
>>>is based on the level of the input signal and the gain of the tube. If
>>>you mean to imply that the tube has less gain with less anode voltage, I'd
>>>agree, but how much less?
>>
>>? depends on HV sag.
>>
>>> Is it enough to make a difference?
>
>Again, does the gain drop enough to make any difference in output power?
? The bottom-line is that less anode-supply volts makes less suds.
>Of course perhaps I should have asked how much difference you have
>measured in peak output powers between SSB and CW. Is it watts? 10s of
>watts? 100s of watts???
? Depends on the amplifier. Depends on the microphone gain setting.
>
> > > If we hold the gain of the tube constant then all bets are off.
>>
>>? To do this, all supply potentials to the tube must be regulated.
>
> If the gain vs. supply voltage varies substantially then we cannot assume
> it to be constant.
? ¿?
> However, my gut tells me the variation is minimal and
>therefore we can approximate the gain to be constant. If the variation in
>gain of a tube is not minimal (less than a few tenths of a dB) then please
>provide what that number is.
>>
>>>If the anode supply
>>>drops then following Ohms law, current increases and power remains the
same.
>>
>>? no.
>>
>No? Huh?
? Confuse not RL with output.
>If gain is approximated to be constant, it has to be the case
>as I explained earlier. I left response below:
>
>>>In the real world, I would bet that given the few hundred volt sag
>>>in the anode supply that the gain of the tube doesn't change all that much
>>>to be noticeable. But if a tube has a gain of 10 dB and you input 100
>>>Watts, it will output 1000 Watts. If the amplifier has an efficiency of
>>>50% this means you are dissipating 2000 Watts (input power). Input power
>>>is the product of plate current and plate voltage. If the input power
>>>remains constant then as plate voltage dips, plate current increases. So
>>>I have a technical problem with your argument.
>>>
>>
>>? With light loads, the filter C essentially charges to the peak v. from
>>the transformer. The filter C maintains the charge. until increased load
>>consumes it.
>>
>Agreed. The way you wrote it, it sounded like you were saying that as the
>caps begin to discharge the increase the anode voltage.
>
>
>>>
>>>OK. So then. If the SSB duty cycle is lower, then that means that the
>>>voltage on the anode is less
>>
>>? hardly. As the load on an unregulated anode supply decreases, volts
>>out increases.
>
>Typo here by me. I meant to say,
>
>>> "If the SSB duty cycle is lower, then that means that the voltage on the
anode±
> > >is greater >>which means the average voltage on the anode
>>>will be higher. Therefore, by your argument, the higher the voltage on
>>>the anode, the more power. Then SSB should see HIGHER peak power than CW
>>>because the sag is less. This is not the case according to the original
>>>poster.
>>>
>>>Even allowing for your argument, still however, at the PEAK which is what
>>>we are talking about, the SSB and CW powers will be identical. The sag
>>>will be the same. And power out will be the same.
>>>
>>? even though the duty cycle of the SSB and CW loads on the unreg.
>>supply are different?
>
>Again:
>
>Let's say peak power out of the rig is 100 Watts. The rig will deliver
>100 Watts upon keydown on CW or on instantenous voice peaks on SSB. If
>the gain of the tube in question under FULL load (after voltage sag) is
>10, then the peak output power will be 1000 Watts.
>
>If instantenous peak power out of the transceiver is identical for SSB or
>CW, then it matters not what the duty cycle of the loads are on the
>unregulated supply. They are reaching the same value at the peak.
>
>I agree that the filter cap action of the HV supply will prevent the SSB
>peak from sagging the HV as much as for CW. But by how much?
>
? You gotta measure how much for the particular amplifier.
>Also again, I point out that if your argument is correct, then peak SSB
>power on a meter should be higher than CW peak power. The original poster
>claimed just the opposite.
>
? A measurement with a calibrated 'scope and HV multiplier probe is
needed.
>I'll grant you there *may* be a difference, but by how much?
>
? Unless one measures the potentials with the appropriate test
equipment, it's guesswork.
>
>>>Average sag might be. But again, we are talking about peak levels. At
>>>the peaks, all is the same.
>>>
>>? It takes a calibrated 'scope and x1000 probe to see what be.
>
>I'll take your word for that.
>
>
>>>Are you sure this difference that you see from the beginning of a dash to
>>>the end isn't the typical overshoot one sees on a normal CW signal?
>>
>>? With ALC action, there is sure to be overshoot. However, with the
>>drive control backed off a bit, the ALC threshold is not reached and a
>>series of dashes exhibits flat tops.
>
>Are you saying that there is no overshoot of any kind without ALC?
? In radios that have an independant Drive control, and use a regulated
Collector or Drain supply, it is possible.
> So the transceiver generates a perfect square wave?
? to my eye, the envelope shape is rectangular, but only if No ALC is
going on. At the output of the amplifier, the envelope is a tilted
rectangle.
>>
>>>If the transceiver doesn't put out a true square wave, but a ramped wave
with
>>>a little overshoot, this will show up on the output of the amplifier and
>>>consequently in your RF voltage, HV sag, etc.
>>>
>>? Quite true. However, with an oscilloscope and HV multiplier probe,
>>one can see what's what, wherever. Never leave home without one.
>
>Well, if there is overshoot on the CW wave, how do you know that the
>overshoot you see on the output of the amp is not due to the input signal??
>
? One either either applies arithmetic to the measurements, on one turns
down the Drive control until transceiver ALC stops mucking up the
envelope. .
>
- cheers, Jon
Rich...
R. L. Measures, 805-386-3734, AG6K, www.vcnet.com/measures
--
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
|