Push-Pull Output Transformers - Part III, The Final Countdown:

Discussion in 'Amateur Radio Amplifiers' started by KD2NCU, Sep 28, 2017.

ad: L-HROutlet
ad: l-rl
ad: MessiPaoloni-1
ad: DLSpec-1
ad: L-MFJ
ad: Left-2
ad: Left-3
ad: Subscribe
  1. WA1GFZ

    WA1GFZ Ham Member QRZ Page

    The best advice I got was from ON9CVD. He told me to look at the second harmonic output. I didn't get the second harmonics below -30 dBc until I eliminated T2.
  2. KD2NCU

    KD2NCU Ham Member QRZ Page

    I am actually very interested in moving on to the circuit you posted but you have been using some "slippery" language in your answers recently either deliberately or inadvertently, don't care which, so let's understand exactly where you stand on a couple of issues.

    1. In all of my discussions, I have made it very clear that I'm talking about the current going into the center tap of the coil after all the bypass capacitors. In fact I have explicitly stated in previous posts that it is the bypass capacitors that are doing the low pass filtering, not the coil. I put the current probe where I did because we are talking about what the coil does or does not do. If you put the current probe between the bypass capacitors and the DC supply you are not seeing what the coil does, you are seeing what the bypass capacitors are doing. Any measurement between the bypass caps and the DC supply tells you nothing about what the coil is doing because the measurement is corrupted by the presence of the bypass caps. I already know what the bypass capacitors are doing, I want to know what you think the coil is doing. So the measurement point is right at the feed point of the coil after all the bypass capacitors and I have never been vague about this. Now, given that the measurement point IS after the bypass capacitors and just before the entry into the center tap of the coil what are you saying the current will be if measured at that point given that the collector current is 16 amp half-wave current pulses. It's not a trick question. In the past you've said dead flat DC with minor ripple. Is this still your answer or not? Or are you now saying that the current at that point will be essentially a replica of the alternate pulses of current of the transistors, ie; a full wave rectified sinewave of current peaking at essentially 16 amps. A virtual replica of the transistor currents except for minor/negligible leakage current through the capacitance of the OFF transistor?

    2. I never asked whether inductors can store energy. I will try to make this very clear. You said the current going into the center tap would be a flat DC equal to the average value of the collector current pulses. I asked where the disparity in amps and electrons were going. WW1WW said they were going into stored energy in the magnetic fields of the coils. When asked if you agree with this, it sure sounded very much like you agreed that electrons could jump out of a wire, stay in a magnetic field for while, thus explaining how the current going into one end could be different from the current coming out of the other end of the very same wire just a few inches away with no escape path in between as illustrated in the Goo-on plasma theory above. (A bit more detail here than in the original question, but I want to make sure I understand exactly what question you are answering and exactly what your answer is.) So to restate exactly what I am asking: The collector current is 16 amp half sine pulses. What do you say the current going into the center tap of the bifilar coil is when measured at the point of the current probe in the previously posted diagram? [By now, you will have answered this in question 1 above.] If your answer to question 1 above is anything other than the current as measured at current probe location shown in the previous diagram will essentially be full wave rectified sinewave of current peaking essentially at 16 amps and essentially an exact replica of the collector currents except for minor leakage currents through the off transistor then I am asking you where the those extra electrons and amps are going when the difference between the current going into the center tap of the coil is 10 amps flat DC ("Pure DC") but the current of the on transistor has only risen to few milliamps or whatever its minimum is at the start of its "on" cycle and thus there is essentially a 10 amp discrepancy between the current entering the center tap and the current leaving the circuit through the on transistor collector. Not to sound redundant, but, to be clear, I am not asking you whether energy can be stored in magnetic fields. I am asking you if you think electrons can be stored in a magnetic field and whether that explains the current discrepancy and if you say no they can't, then where ARE the extra electrons and amps going when the collector/drain current is much much lower than the 10 amps of "Pure DC" going into the center tap.
  3. KD2NCU

    KD2NCU Ham Member QRZ Page

    A few other points I need clarified so that I understand exactly what you are saying.
    1. Why are you so anxious to leave the discussion of Push-Pull amps and the bifilar coil used in Push-Pull amps to explain how a bifilar coil used in a push-pull amp works? We're talking about how a bifilar coil works in a push pull amp. You really shouldn't need to go to a circuit that is not push-pull and does not have a bifilar coil to explain how a bifilar coil in a push pull amp works, should you? The push pull and these concepts really are not all that complicated. Either you think the current directly entering the center tap of the coil is pulsating or you don't. And if you say the current is "Pure DC" entering the center tap, then either you can explain where the discrepancy of up to 10 amps is going or you can't. Either you believe electrons are jumping out of wires and being stored in magnetic fields or you don't. These questions are not complicated and they're not "trick". By the way, if you've changed your mind and you now agree that the current directly entering the center tap is a series of full wave rectified sine pulses essentially replicating the transistor collector/drain pulses then we probably don't actually need to go to a circuit that is not push pull and which does not have a bifilar coil in it to explain how a bifilar coil in a push pull amp works do we? But either way, I'll be glad to go to that circuit after I know what you are actually answering to the questions I've asked.

    2. The circuit I clipped and pasted is from an actual circuit. Why do you suddenly need "values"? You did not need values when you very adamantly said the current entering the center tap was "Pure DC". Why do you suddenly need values? We know the current flowing in the transistors of a class AB or Class B amp is essentially half sine pulses except for some crossover effects don't we? If you would like to use a more exact representation of the transistor collector/drain currents to include cross over effects, be my guest. The collector/drain current will still essentially be half sine pulses and the current just entering the center tap of the coil will still be either "Pure DC" or it will be pulsating essentially like the collector current. No, you really don't need values to answer these questions. Go ahead and assume the transistor currents are half sine or close, peaking at 16 amps, and that the bifilar coil was designed and made properly. What other values could you possibly need? There are no other components in the output circuit other than the transistors, output transformer, and the bifilar coil. There are no capacitors or resistors or anything else in the output circuit. It's a real simple circuit. OK, OK, if you really really need values, tell you what. Pick any class AB or class B push-pull circuit you want that has whatever values you seem to need, has a bifilar feed coil essentially connected like the one in the circuit I provided, and a simple output transformer like the circuit I provided, no capacitors, no additional chokes, striplines, baluns, doo-dads, trinkets, etc. and in which the collector current is large, say at least 10 amps or more, tell me what you are saying the collector current will be, tell me what you think the current going into the center tap of the bifilar coil looks like at the point where I showed the current probe in the previous diagram. This is not a trick question. Don't try to be tricky or slippery with your answer. I'm basically saying choose any push pull circuit you want that is essentially exactly like the circuit I provided but which has actual values that you think you need. There are thousands and thousands to choose from.

    3. I need to know exactly what your thinking is at this point about DC power supplies as a result of your recent answers. I think that at this point you are saying that a DC supply can, in fact, supply pulsating current, AC current, backwards current, and that the DC supply doesn't care what the current is as long as it's within its current limit and that the DC supply does not in any way determine what its current is, but rather the connected circuit alone determines what current flows for a given DC. Please confirm that this is your understanding. Because the other day you were saying something completely different. Do you remember what you said? I do, I can tell you what you said. You said, "You're wrong on so many levels here. But a question I have for you, that has never been answered is how can the current be pulsing at a DC point, when that point is an AC short? The amplifier is fed with pure DC to the input of the choke. The current at that point is pure DC, it is not pulsing at the rate of the RF. Once you can acknowledge this, then maybe we can go onto some other points." Now we know you were talking about the DC supply. So is your understanding different now? So maybe NOW, you should instead be saying something like, "Yes, I was very confused and wrong about how DC supplies operate. I no longer have this misconception about how they work. Thanks, KD2NCU for hanging in there and clearing up my misconception."

    And you shouldn't need to reference a show about people using mind altering drugs to explain the discrepancy of 10 amps either. Hmmm.
    Last edited: Oct 9, 2017
  4. KD2NCU

    KD2NCU Ham Member QRZ Page

    Good morning WA1GFZ: Click on my "QRZ Page" under the icon at the left and this should take you into my profile page where you should be able to see my BIO, email address, and/or send me a private message via this website with attachments. Let me know if that works or not. I've been able to see other peoples' BIO's, if they put them up there, as well as email addresses and I've been working "off line" with a handful of engineers via the private messaging with attachments.

    But WA1GFZ, you didn't answer my question: Can electrons jump out of wires into a magnetic field, hang around for a while, then jump back into the wire again a little later like the diagram below right? And could this explain how the collector or drain current could be real low, just a few milliamps or so but the current entering the center tap is 10 amps? Please do give me your thoughts on this. Thanks; KD2NCU.

  5. K7JEM

    K7JEM Ham Member QRZ Page

    The current measured there would be DC with a pretty good ripple at the AC frequency. It would not look the same as at the transistors, due to the inductive reactance of the choke, coil, or transformer.

    The extra electrons are going into charging the inductor. If you look up electrical charge, you will see that is is excess electrons on one side, compared to the other. A capacitor and an inductor can both store a charge, which means they can have excess electrons, depending on current passing through them, or voltage across them, and their specific inductance or capacitance.
  6. K7JEM

    K7JEM Ham Member QRZ Page

    As explained above, the electrons are going to charge the inductor, so I guess they are being stored as magnetic fields. But the amount being stored is not 10 amps, it is 10 amps divided by the frequency of operation, since the inductor is charging an discharging on every single cycle. The reason to change to a simple circuit is to simplify things. If we can't agree how a single transistor with a single choke behaves, why are we trying to determine how two transistors and a bifilar winding behave? The reason for posting an actual circuit is to show that there are actual values of inductance and capacitance given. I don't care if it's that particular circuit, but we need to have one that is simple, has values, and is operating at a given frequency in order to make some determinations as to what is going on.

    See answer above. The question about the DC current draw is especially misleading is no operating frequency is given, and no parameters of the choke are given. The answer will vary wildly depending on specific frequencies and inductors being used.

    Again, to answer the question, if we strip out the bypass capacitors (which are needed for proper operation), the DC current input will still not exactly mirror the current shown at the transistors. This is due to inductance of the choke, and the fact that it is charging. This will give a majority value to DC current, and a smaller value to the AC (ripple) current at that point.

    Well, first off, a DC supply is not going to provide AC current. When AC current hits the DC supply, it is shunted to ground. Any AC current is not coming from that supply, it is a result of something attached to it causing DC current to flow at an AC rate (pulsating DC). But that current will not exist at the DC supply, since it acts as a short at higher frequency AC, if we are talking about RF. If we are talking about a switch closure at 1 Hz into a resistor, then yes, that pulsating DC will be coming from the supply, since the supply is not a short circuit at 1 Hz.

    This is why I like to look at actual circuits, rather than some made up representation of a circuit with generic value parts, or no values at all. It is more clear to see why bypass capacitors at the feed chokes are needed, and why the DC current is measured at that point. These capacitors and series inductors will eliminate the RF current flow back toward the DC supply, to the extent that the current is "pure DC" with the smallest ripple of RF.

    The easiest thing for you to do to prove this to yourself is to insert a current probe at the output of the DC power supply, and measure how much AC ripple is making it back to the supply, when the amplifier is running.
  7. WA1GFZ

    WA1GFZ Ham Member QRZ Page

    Jumping electrons are way over my head.
  8. KD2NCU

    KD2NCU Ham Member QRZ Page

    Electrons can be stored in capacitors, however, there are no capacitors inside the circuit we've been discussing.
    Electrons cannot be stored in magnetic fields. Don't say this is not a new concept or this is common knowledge because it's not, and if it is, then produce evident and references to back this up. The current into one end of a coil will equal the current coming out the other end exactly at all times. Unless it is arcing to ground or unless it's a poor design and the coils are too close to ground or other components allowing excessive capacitance to ground or excessive capacitance to other components in which case it's just a result of poor design. Now there will be turn to turn parasitic capacitance in a coil but that doesn't leak electrons out of the coil. With turn to turn capacitance electrons are essentially bypassing portions of the coil turns but still coming out the other end of the coil wire. The current in one end still equals the current coming out the other end.

    This is garbage. If not, produce a valid reference that says the current going into a coil doesn't have to equal the current leaving the other end because electrons are being stored in the magnetic field before we can go any further. You have not accounted for the huge discrepancy between current going into the center tap and the current going into the collector if you say the current going into the center tap is essentially flat DC. Your argument is completely based on two things: 1. Laughable garbage science that electrons leave a coil, get stored in magnetic fields and later return to the coil as electrons again, and 2. An assumption that the coil is providing inductance (with absolutely nothing to support this assumption) and subsequent circular logic to say that since the coil is is providing inductance the current into the center tap must be flat DC and since the current is flat DC then the device must be a choke. You've produced no valid analysis to support your assumptions of inductance and no valid reference to support storage of electrons in magnetic fields.

    Produce some actual valid physics or circuit analysis reference that says the current going into a coil doesn't have to equal exactly the current leaving the coil assuming the coil isn't arcing or has excessive leakage capacitance to ground or other components due to poor design and that electrons can be stored in magnetic fields or we can't "move on". We're not "moving on" based on garbage science. If it's not garbage science, then you should be able to prove it with technical references. And you don't need any values real circuits or anything else. Either electrons can leave a coil and go into the magnetic field or they can't and either the current coming out of one end of a wire can be less than the current entering the same wire because electrons are leaving the coil and hanging out in the magnetic field or it can't, and either you have a gross misconception about this or you don't.

    Now if you want to throw in the towel and quit because I won't accept garbage science unless you validate the garbage and you cannot validate the garbage, then quit and run like everyone else seems to have. I posted some analysis, you said I'm wrong, I said prove it, you're trying to use garbage science to prove me wrong, I'm saying this is garbage and I don't accept the garbage unless you can cite valid scientific references to back it up, if you want to prove me wrong, then validate the garbage, otherwise, quit if you want to. If it's not "Pure Garbage", then validate it.
    Last edited: Oct 9, 2017
  9. K7JEM

    K7JEM Ham Member QRZ Page

    Wow. I give up. Have a great day.
  10. AA7QQ

    AA7QQ Ham Member QRZ Page

    Darn... I was starting to enjoy this thread...

    K7JEM likes this.

Share This Page