Amplifier ALC Theory questions and help

Discussion in 'Amateur Radio Amplifiers' started by N2RDQ, Jun 29, 2018.

ad: L-HROutlet
ad: l-rl
ad: MessiPaoloni-1
ad: K5AB-Elect-1
ad: Left-3
ad: L-MFJ
ad: Subscribe
ad: Left-2
  1. G0GSR

    G0GSR Ham Member QRZ Page

    And remains so....
    OK, simple half wave rectifier. 10k load, 1uF smoothing.
    DC volts out against frequency, input level constant:
    60Hz = 7.48v (smoothing is not quite sufficient here)
    1kHz = 7.74v
    10kHz = 7.3v
    100k = 5.67v
    1MHz = 2.4v
    10MHz = 0.3v
    Pretty constant above this frequency, rectifier output is an asymmetrical sine wave.

    Yes, it still "rectifies" but how badly???

    Phil, I am not disputing your findings that your diode does work but it must certainly be something special if it is a 4000 series.

    Frank (Last word ;-)
     
  2. G3YRO

    G3YRO Ham Member QRZ Page

    I think you have proved the point Frank !

    Incidentally . . . I use some power diodes as Varicaps to resonate my 160m Receiveing Loop aerial . . .

    Some US stations expressed concern that those diodes might cause problems, as they live near a MW broadcast transmitter.

    But presumably they are also likely to be bad as RF rectifiers . . . so are unlikely to cause a problem?

    Roger G3YRO
     
  3. SM0AOM

    SM0AOM Ham Member QRZ Page

    It is not unlikely that even a 1N40o7 would provide some form of peak rectification when back-biased for the vast part of the RF cycle,
    good enough to enable ALC action.

    When reading this thread, I find that a lot of misconceptions about various ALC circuits are around.
    The original ALC sensed grid current, and acted as a peak-reading threshold circuit, which furthermore was AC coupled,
    in the case of the KWM:s through a coupling capacitor and in the case of the 30S-1 using an AF transformer in the grid return.

    They had a high-pass characteristic which generated more ALC voltage when the time-derivative of the RF envelope was
    large.

    This is exactly the characteristic you want to prevent peak clipping.

    Later equipment such as the 30L-1 used DC coupled ALC circuits, where the output voltages become independent on the rise-time of the RF envelope.

    In both the 30L-1 and 30S-1 there are loading comparators which provide a means of establishing the proper load-line when tuning
    by comparing the input and output voltages. These are entirely unrelated to any ALC circuits.

    A loading comparator is an indispensable part of a servo-tuned linear amplifier, and it is likely that Collins expertise in the servo-tuning field for professional radio was carried over to amateur gear.

    I have yet to see an ALC circuit in a conventional amateur tube or solid state amplifier that works on comparing the input and output envelopes and adjusting the system gain to prevent distorsion. Digital adaptive pre-distorsion is one modern way to accomplish this, but has not yet been commonly implemented.

    Some ISB-rated HF tube amplifiers from the 1960s used the comparison method of distorsion reduction as a complement to RF feedback, called "Envelope Feedback", or in German literature "H├╝llkurvenr├╝ckkopplung". The method originates from AM modulators where a feedback loop equalised out larger deviations from the linear transfer function.

    It can however not be used as a means for decreasing the peak-to-average ratio of the emission, as its very purpose is to "straighten" the transfer function. Such actions must take place before the linearised power amplifier.

    73/
    Karl-Arne
    SM0AOM
     
  4. AC0OB

    AC0OB Subscriber QRZ Page

    I noticed that Frank, our "Friendly Diode Denier" :) used a 1k load resistor in his first circuit and a 10k load resistor with a 1 uF cap on the last one. This means that for the second circuit the impedance seen at the diode's Cathode, when switched on, would have been on the order of Z = 10k - j0.04 ohms.

    The simplified circuit I presented has an Impedance at 3.75 MHz as seen at the cathode of 47k - j42.5 ohms, a big difference which amounts to a 5 fold increase in impedance as seen by the source compared to Frank's circuit.

    In the original Dentron ML-2500A/B ALC circuit, there is no loading resistor across the filter cap but only a 27k series resistor going to the ALC port, so my circuit should present a worst-case scenario for diode loading.

    Regardless, the 1N4007 or 1N4004 works marvelously well in an ALC circuit, whether it be the Dentron ML-2500A/B, or in some of the Henry series linear amps. :p


    Phil
     
    Last edited: Jul 9, 2018
  5. K9AXN

    K9AXN Premium Subscriber QRZ Page

    Don't recall Henry using the 4000 series diodes in the ALC circuits in their amps. Must have been an early one that they found the ALC not to work on the high bands; like Franks testing indicates. Be great if you would test your theory and post a scope trace at 20MHz and numbers like Frank took the time to measure.

    Had a great Saturday riding endless thermals to 6200 feet. The SPF30 worked fine under the wing.

    Have a great day Jim
     
  6. G0GSR

    G0GSR Ham Member QRZ Page

    I've been called worse...
    No, I don't think so.
    On switch-on with the capacitor discharged, yes but 10k || -j0.04 not in series.
    Once the capacitor is charged, it almost "disappears" from circuit as very little current is required to keep it charged. Reactive current is much reduced.
    So virtually 10k resistive load. I don't think that a higher load would have made much difference.
    Now should I just try that??

    "Frank the FDD" (Actually, those are almost my initials!)
     
    Last edited: Jul 9, 2018
    SM0AOM likes this.
  7. AC0OB

    AC0OB Subscriber QRZ Page

    It was the 1N4000 series we are discussing.

    Henry found the 1N4004 works great in the 2K-4 ALC circuit, I know, I still have mine.

    And it would be great if you could even remotely get a grasp of the physics of thermionic devices (such as tetrode TUBES;

    https://forums.qrz.com/index.php?threads/replacing-6146a-with-6146b.601117/page-12#post-4601485 ) so I doubt you have even a remote understanding of the physics of solid-state devices.

    I did measure the circuit with a scope but since you always "comment before reading" the other posts, I will re-post the circuit for your benefit. Source impedance of Agilent signal Generator was 50 ohms. impedance of scope probe and DMM is 1Mohm.




    Phil
     

    Attached Files:

    Last edited: Jul 9, 2018
  8. K9AXN

    K9AXN Premium Subscriber QRZ Page

    Strange how the others used RF diodes ---must have discovered a better way. Looking for a scope trace to verify the levels at 3.7 and 28MHz. and how smooth the line is. Your numbers don't square with Franks measurements.

    As for the 6146A and B thread, I don't care what you think, I care that the issue is answered completely. I'll revisit it to provide folks with explanations of the differences in the tubes, how to measure the capacity, and explain the effect of the 2:1 ratio of plate current. As for my understanding of vacuum tubes, I'll leave that to the information that I provide explained in a way that most can understand and the folks that read it. As you recall, I stated that I would not leave that thread until the question was resolved. Your notion that the cathode was the only difference is incorrect.

    Regards Jim
     
    Last edited: Jul 9, 2018
  9. AC0OB

    AC0OB Subscriber QRZ Page

    AXN: "...Looking for a scope trace to verify the levels at 3.7 and 28MHz. And how smooth the line is. Your numbers don't square with Franks measurements."

    I and Frank did the scope traces, now it is your turn. You did not see that Frank's circuit and mine were different? ? How typical of you.

    AXN: "As for the 6146A and B thread, I don't care what you think, I care that the issue is answered completely. I'll revisit it to provide folks with explanations of the differences in the tubes, how to measure the capacity, and explain the effect of the 2:1 ratio of plate current. "

    Please don't revisit it because it showed you had a complete lack of understanding of basic tube theory and you still do not understand the 2:1 ratio statement, in which Carl gave a clear and accurate narrative of the testing situation of the Characteristic Range Values and its associated statistics.

    KM1H: "The 2:1 issue is a standard bell curve showing acceptable production variances; nothing more. The end effect is a variance in the "apparent" tube gain over that range, aka the amount of driving power needed to overcome the bias for a specific output. "

    Yet you still don't get it but keep pushing the mythical AXN scenario.


    AXN: "As for my understanding of vacuum tubes, I'll leave that to the information that I provide explained in a way that most can understand and the folks that read it. As you recall, I stated that I would not leave that thread until the question was resolved. Your notion that the cathode was the only difference is incorrect."

    Many of us resolved resolved many questions using published, non-mythical vacuum tube theory, but you only added confusion because of your many misunderstandings.

    Your Modus Operandi is: long silences broken up by nothing new.

    My recommendation for you is to leave this thread before your waders fill up with quicksand.o_O


    Phil
     
    Last edited: Jul 9, 2018
  10. WB2WIK

    WB2WIK Platinum Subscriber Platinum Subscriber QRZ Page

    Parts list from my Henry 3K:
    Henry 3K parts list.PNG

    ALC rectifier (D2) is not a 1N4004, it's a 1N458 which is an ultrafast switching signal diode.

    1n458.PNG


    There are obviously exceptions but I've never seen a 1N400X work in any kind of RF detector application.
     

Share This Page

ad: UR5CDX-1