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: Left-2
ad: Subscribe
ad: L-MFJ
ad: MessiPaoloni-1
ad: DLSpec-1
ad: Left-3
  1. KD2NCU

    KD2NCU Ham Member QRZ Page

    WA1GFZ, I think you are saying that you can always pick up a feedback signal with a third winding and that this indicates unbalance and/or the actions of the loss monster. I don't agree that this necessarily proves unbalance or loss monster. There is a third mechanism at work that will always make it possible to pick up a feedback signal with a third winding even if everything is perfectly balanced and the loss monster is asleep. I had not thought of this till just now thinking about your post above. Here's what's going on. As you said, the on transistor is essentially placing a portion of the power supply directly across the upper winding which is acting like the primary of a transformer. This induces a voltage in the secondary winding which is acting like the secondary of a 1:1 transformer. I think we agree so far. Now whatever impedance the lower coil is looking into is directly reflected as the same impedance in the upper coil because of the 1:1 turns ratio of the two coils. Here's why I think you can put a third winding in the transformer to pick up a signal and use it for feedback and it doesn't necessarily mean there is imbalance or the loss monster is responsible. Here's my assessment of why you can do this. I'm betting that you know about the magnetizing current that exists only in the primary winding of a power transformer (actually all transformers), but like me, never applied that thinking to an RF transformer although it does apply and you have probably designed RF transformers taking this into account but may not have realized that you were. Anyway, in any and all transformers that are acting like transformers the primary winding has a magnetizing current that the secondary does not have.

    Take the case of a 1:1 isolation power transformer. With the secondary open circuited, the primary still has a current flowing. This current is partially due to leakage inductance and partly due to a magnetizing current that has to flow in order to produce a flux so that there is a secondary voltage even when open circuited. Ignore the leakage component on the primary for now. When we first apply power to the primary, a current begins to flow as it would in any inductor. At equilibrium, the applied voltage and the flux always have to satisfy Faraday's law that the applied voltage always equals the time rate of change of the flux times the number of turns in the primary. The current that it takes to set up this magnetizing flux is the magnetizing current and by design is usually a small fraction of the rated load current of the transformer. Now when we apply a load to the secondary winding, a current flows in the secondary and Lenz's law says the current will produce a flux that opposes the flux that generated the voltage in the secondary thus starting to cancel the flux caused by the primary magnetizing current. At this point you can look at what happens several ways. I like to over simplify it a bit and say that since the secondary current tends to cancel or reduce the flux in the core, Faraday's law is no longer satisfied on the primary winding and the primary current begins to increase to bring the flux back up to a value that satisfies Faraday's law for the applied voltage. So now the primary winding has two current components (actually more if we look at all effects). One is the original magnetizing current that is producing the flux in the core that produces the voltage in the secondary, and the other is the primary load current that results from the secondary load current. The net result is that when we connect a load to the secondary, the flux in the core pretty much ends up right where it started except for some relatively minor differences due to leakage and resistance, etc. As the secondary current increases or decreases due to load changes, the primary current increases or decreases in response always keeping the flux in the core essentially constant at whatever value satisfies Faraday's law on the primary winding.
    The point is, even in a high power transformer, the flux produced by the secondary current is cancelling the flux produced by the primary LOAD CURRENT component and the flux in the core is essentially the same as when the secondary is open circuited. IE; for all practical purposes, the flux in the core does not change when going from no load to full load. If you were to calculate the flux produced by either the secondary or primary winding by looking at the current and the number of turns, you would come up with an enormous flux possibly as much as ten times the saturation level of the core. This is because the flux due to the LOAD current components of the primary and secondary windings ACTUALLY are cancelling one another and what remains is the original magnetizing flux. This never goes away no matter how well balanced everything is.
    My guess is that you probably know all this and/or have seen equivalent models for real transformers including the magnetizing currents and such.
    This all applies to RF transformers as well. Classic design criteria for RF transformers say to establish the number of primary turns based on the criteria of making the inductive reactance of the primary winding at least 4 or 5 times the load impedance expected to be reflected into the primary winding at the lowest frequency of interest. What this is doing is setting the magnetizing current of the primary to be 1/4th to 1/5th of the actual signal current of the primary winding. It can be made a smaller fraction of the load current by adding more turns. In the case of power transformers, the magnetizing current can be made to be a very small fraction of the load current. In the case of RF transformers, adding turns adds resistance very fast because of the skin effect so you reach diminishing returns quickly.

    In the case of a power transformer, we don't change the primary voltage so the magnetizing current and flux is constant and does not change with load. However, in the bifilar feed coil in a linear amp, we do change the amplitude of the applied voltage according to the amplitude of the signal so the magnetizing current and resultant magnetizing flux in an RF transformer WILL be proportional to the RF signal strength even if everything is perfectly balanced.

    What's the point of all this? We are alternately placing the power supply across one coil then the other. Even if the load currents are perfectly balanced and synched, there is always always always a small magnetizing current and flux in the primary that is never never never completely cancelled by the secondary current even when everything is perfectly balanced. If you follow typical RF transformer design guidelines, this current and resultant flux can be as much as 1/4th to 1/5th of the actual load currents.

    So you will always be able to pick up a signal with a third coil even with perfect balance of load currents, identical transistors, although this may not be the only reason you are able to pick up the feedback signal with a third winding.

  2. KD2NCU

    KD2NCU Ham Member QRZ Page

    WA1GFZ, below is the classical model of a transformer which I'm pretty sure you are probably familiar with. I circled the magnetizing reactance Xm. This represents current that flows to set up a magnetic field in the core even when the secondary is open circuited. When we use a design criteria to establish the number of primary turns on an RF transformer, we are establishing Xm in the model below. The magnetizing current is Im in the model below and note that it is modeled as a parallel current (although it's not really) to reflect the fact that it does not depend in any way on what's going on in the secondary and it depends only on the primary voltage and primary winding properties. IE; for a given transformer design and applied voltage, this magnetizing current is fixed and the flux it produces is not cancelled by anything else anywhere.
    The flux from this component is not balanced by any current anywhere so it always exists as is obvious when the secondary is open circuited but we still have a voltage on the secondary which obviously means there has to be a current in the primary to produce the flux that is producing the voltage in the secondary.
    So it will always be there and a third winding would always pick up on it just like the secondary winding picks up a voltage even when the secondary is open circuited.
    This unbalanced magnetizing current can be made arbitrarily small by adding turns to the primary (and increasing turns on the secondary accordingly), however, this comes at a cost of a larger transformer, and/or higher resistance, etc., so there are practical limits on how small to make this.

    Speaking of which, what set of guidelines or design criteria do you typically use to establish the absolute number of turns on the primary of an RF transformer? Something like what I stated above for example?

  3. WA1GFZ

    WA1GFZ Ham Member QRZ Page

    I totally agree. I shoot for 200g at the lowest operating frequency. I have 2X2 turns of 25 ohm coax in my drain transformers. I think the core area is 1.76 cm^2. The primary is 4 turns CT using the shields to carry the DC current. This also takes care of the stuff going on in T2 and is just as balanced as T2. The nice thing is I do it all on one core. You also need about 5 uH of primary inductance (MOT AN749) on 160 running 48 volts and 300 watts. Type 43 core gave me well above that. Type 61 would have been close on the inductance for 160 meters but lower loss on 6m. I was worried about type 43 on the higher bands but at 50 MHz I'm running about 4g. The proof of the pudding is temperature rise. I do not feel any temperature rise in the core. The Wire warms up just a bit after a long AM transmission. I got to my design which isn't my invention since Harris, Helge and others have implemented the same configuration because of limited space on the MRI boads I used. An even more efficient transformer can be built with the BN 43 7051 two hole balun core. That can be built with 2X1 turn of coax through both holes. The core area is close to 3 cm^2 and again runs cold. The shorter the cable the better the high frequency performance. My case I didn't have a clean way to mechanically mount the baluns so went back to the sleeve.
    The best advice I got was from 0N9CVD. Bob told me that you have it right when your second harmonic energy is low. He worked for Philips. I think their App notes are superior to the Motorola app notes. T2 just complicates the circuit and I never got good performance out to 6 meters. gfz
  4. KD2NCU

    KD2NCU Ham Member QRZ Page

    Wow, that was a mouthful. I will take some time and digest all that. Thanks.
    I have most of the Motorola app notes. I've found a few Philips app notes on line but I know there are a lot more than I've found. Do you have a good link to the "secret stash" mother lode?
  5. KD2NCU

    KD2NCU Ham Member QRZ Page

    Today’s Class – Bifilar Coil Provides a 1:4 Impedance Transformation
    Welcome back, what did we learn last time?
    Suzie: We learned a process. Understand how the individual components work first. Then put them in the circuit diagram and analyze the circuit using established circuit analysis methods and established physics principles.
    Bobby: It’s like a riddle. Any description we propose for how the circuit is working has to satisfy all constraints.

    What are the constraints that this solution must satisfy?
    •The collector current will always equal IQ1 = I1 + I2 (Kirchhoff’s current law)
    •IQ1 is half sine pulse because it’s a linear amp.
    •The output transformer primary current IL is half sine pulse because it’s a linear amp
    •If there are AC components in I1 and I2, they must be predominantly equal and opposite. (action of common mode choke inside bifilar coil.)
    •The leakage current through the off transistor is small and probably negligible (unless we use a transistor rated much higher than needed).

    Timmy: You mean we can’t just make up any solution we want?
    Suzie: Dork
    Billy: No whatever we think the circuit is doing has to satisfy all the constraints and can’t violate Kirchhoff’s laws or physics, or even common sense, otherwise, it’s junk science
    Alex: So it seems like there is only one solution that satisfy all those constraints: IL is a half sine pulse and I2 and IL will essentially be equal so I2 has to be a half sine pulse.
    So I1 will also be a half sine pulse because of the action of the common mode choke at the center of the bifilar coil.
    We know that IQ1 is a half sine pulse and IQ1 = I1 + I2 so now we also know that IQ1 is a half sine pulse twice the amplitude of I1 and I2.
    This satisfies all the constraints above.

    Timmy: Whoah, Dude! Did you just say that the half sine current pulse of the transistor is twice the amplitude of I1 and I2 half sine pulses?
    Alex: Yes I did.
    Timmy: Whoah, Dude! Then that means the amplitude of the current pulse of the transistor is twice the amplitude of the current pulse in the primary of the output transformer.
    Alex: Correct.
    Timmy: Whoah, Dude! But that means there is an impedance transformation happening doesn’t it? The transistor is the source, the output transformer primary is the load, and the source sees twice the current of the load. Isn’t that a change of impedance?
    Suzie to Bobby: See, told ya he’s a frickin’ genius!

    Any questions before we get started?
    Alex: Yes, is everyone out there a troll?
    No, as a matter of fact, most are not. I think I only counted about four trolls.
    Alex: Boy it seemed like they were all trolls.
    Suzie: Yeah, the trolls make so much noise with their junior high trash talk they tend to dominate the discussion and drive away anyone trying to learn anything or ask real questions or bring any real science to the table.
    Alex: What about that one guy who seemed helpful at first and gave you that link to the Ludens article.
    Yeah he’s obviously conflicted at this point. He’s going to have to make some decisions. He’s starting to realize that he doesn’t know anywhere near as much as he thought he did, and more importantly, he’s starting to realize that his buddies don’t either.
    Alex: Seems like their troll behavior drives away anyone who really knows anything and what’s left is trolls with their misconceptions.
    Exactly. They create these little “safe spaces” where they are secure in their belief that they know a lot and they shield themselves from anything that might threaten this belief.
    Suzie: Yeah, if the guy that linked the Ludens article really wants to learn anything, he’s gotta get away from the trolls that drive knowledgeable people away, open his ears, and listen to people that use real science.

    Bobby: Well why do they get so porky when you challenge them with real physics and circuit analysis?
    Suzie: Because they don’t really understand physics, electronics, and circuit analysis, or haven’t had the training and education that others have or whatever and they need to use bully tactics and shouting and insults to protect that dirty little secret. Along comes someone who challenges them with real physics and circuit analysis and isn’t impressed or intimidated by their antics and asks them to back up their junk science and suddenly their lack of knowledge is exposed AND their bully tactics no longer work so they’re threatened and ultimately run away.

    Alex: Are they bad for not knowing stuff and not having the training and education and experience that others might have.
    We already answered that and said no. They’re bad for spouting junk science, not listening to those who have more knowledge than they do on a particular topic, hurling insults, and all their other 6th grade trash talk.
    Timmy: Can we get back to the technical stuff, I think I’m actually starting to understand.
    You bet.

    Let’s redraw the circuit slightly in a way that makes it easier to see what’s going on with the voltages.
    You mean the diagram below is the same circuit?
    Yes, exactly the same.
    OK, we learned that the current in the output transformer primary is twice the current of the transistor.
    Let’s look at the voltages present.
    What can we say about the voltage across the primary of the output transformer?
    Alex: It has to be a half sine pulse of voltage.
    Timmy: Why
    Alex: It’s a LINEAR amp so it better be a half sine pulse or it’s not a linear amp.
    Anything else?
    Alex: Yes, the half sine voltage pulse is in phase with the half sine current pulse.
    Timmy: Wrong on so many levels!!!
    Suzie: We started this discussion assuming everything downstream is well matched meaning that the load is overwhelmingly resistive and line impedances are matched, etc., so the voltage and current will be in phase.

    How big will this output voltage pulse be?
    Bobby: Couldn’t we use Kirchhoff's voltage law and the physics of transformers to determine that?
    Yes we could.
    Suzie: So let’s say the drive to the transistor takes the transistor just short of saturation. At this point, the transistor is almost a short so it is placing just about the entire 12 volts across Coil 1, probably about 11.8 volts to be exact with the polarity shown in the diagram below. Now the coil acts like a transformer producing 11.8 volts across the other coil with the polarity shown.
    Timmy: How’d you get that.
    Suzie: As long as the core is not saturated, anything you do in one coil causes something to happen in the other coil. You go back to the three dimensional diagram, use Faraday’s law, Lenz’s law, and the right hand rule to determine the polarity of the voltage induced in the second coil and it will be as shown below.
    Timmy: You mean we have to use real science and physics to determine what the second coil does? We can’t just pull something out of our … (Suzie: Shut up Timmy.)

    Bobby: So then using Kirchhoff’s voltage law we would write a loop equation and say -V1 –V2 + VL = 0
    Timmy: What the heck is this Kirchhoff thing and what is a loop equation?
    Bobby: The sum of the voltages around any closed loop is zero.
    Alex: That sounds very useful for analyzing circuits. Do the trolls know about Kirchhoff's voltage and current laws and loop and node analysis?
    Apparently not, or maybe they do but they ignore it because it gets in the way of their junk science explanations.
    Timmy: Well I kinda agree with the trolls. It's much easier to just wave hands and talk than to actually use all this analysis stuff and physics to figure out what a circuit is doing.
    (Crickets chirping)

    Bobby: OK, so we have -V1 –V2 + VL = 0.
    So now we can say VL = V1 + V2
    Timmy: But aren’t V1 and V2 essentially equal and with the polarities shown and didn’t we say V1 was about 11.8 volts at the peak of Q1 conduction?
    Timmy: Well then that would mean that the voltage across the primary of the output transformer is 23.6 volts!
    Yes it would.
    Timmy: Well then that would mean the voltage swing of the output transformer primary is twice the voltage swing of the transistor that caused it!
    Yes it would.
    Suzie: Izzit gettin’ hot in here again, or is it just me?

    Timmy: Whoah Dudes!!! The transistor is seeing twice the current of the output transformer primary and half the voltage swing of the output transformer primary. Wait one minute here. Doesn’t …… that …… mean …….. that the device is providing a 1 to 4 impedance transformation?
    Suzie: Good god, it’s like a frickin’ sauna in here! Someone open a window now dammit!

    Yes, that’s pretty much the definition of an impedance transformation. The transistor is the source and the output transformer primary is the load and the load is seeing twice the voltage and half the current of the source so the load impedance is 4 times the impedance being seen by the transistor.
    Timmy: Awesome!
    OK, that’s enough for today. Any questions?

    Alex: Isn’t this going to upset the trolls?
    You bet. Big time. Prolly pop their gizzards!
  6. WA1GFZ

    WA1GFZ Ham Member QRZ Page

    I'll shoot you an email with some Phillips stuff. I'm hoping you continue this to my configuration to show how the extra complication degrades bandwidth. gfz
  7. KD2NCU

    KD2NCU Ham Member QRZ Page

    Today’s Class Topic: Is the bifilar coil acting like an inductor to the current coming into the center tap DC connection and keeping AC or RF out of the DC bus? NO, it’s Not.
    Timmy: But it has squiggly lines.
    Suzie: Tim-bo, we’ve been over and over this. Not everything that looks like an inductor is necessarily acting like an inductor and especially when it’s interacting with another coil.
    Timmy: Well how do you know it’s not being an inductor.
    Well actually we don’t really “KNOW” anything until we analyze and prove it. And if we make an assumption about something, we have to justify and prove the assumption or else we really haven’t proven anything.
    Alex: Don’t some people assume something and then use that assumption to prove the same thing in a big circle?
    Yes, trolls.

    Below is our circuit. Prior to today we demonstrated the following:
    1.The common mode choke inside the device tries to keep the currents equal and opposite.
    2.So if there are any AC current components in the wires, they will be predominantly equal and opposite.
    3.Any currents that are equal and opposite produce flux that opposes and cancels.
    4.Therefore large AC currents can flow through the wires experiencing no inductive reactance as long as they are equal and opposite.
    5.We demonstrated that all of the currents in the circuit are essentially half sine pulses during any one transistor’s cycle.

    Kirchhoff's Current Law:
    Alex: Let’s use Kirchhoff’s current law at the center tap to see what the current coming off the DC bus looks like. It says the sum of the currents entering a node has to equal zero due to conservation of charge. So at the center tap we would say Is – I1 – I2 = 0.
    Bobby: Why the minus signs?
    Alex: Cause those currents as defined on the diagram are not entering the node they are leaving so we give them a minus sign.
    Alex: So this says the current coming from the DC bus has to equal the sum of I1 and I2 whatever they are. Is = I1 + I2.

    Bobby: These Kirchhoff laws seem like really useful tools for figuring out what’s going on in a circuit. They seem to take all the mystery, guesswork, and hand waving out of the picture. So why don’t the trolls like to use these techniques?
    Because they take all the mystery, guesswork, and hand waving out of the picture.

    Alex: So we established that I1 and I2 are half sine pulses and their polarity is additive as defined on the diagram and they are essentially equal and opposite in the coil wires as imposed by the common mode choke.
    Timmy: So then doesn’t the current Is have to also be a half sine pulse twice the amplitude of I1 and I2?
    Suzie: Aw geez, again with the hot flashes!
    Timmy: And other than some insignificant leakage current through the off transistor, doesn’t this say the current Is coming into the center tap is essentially an exact replica of the on transistor collector current?
    Correct on all counts Timmy.

    Alex: So basically what’s happening is the current pulses from the DC source are splitting equally into the red and black wire where they are now equal and opposite in the transmission line, so they experience no attenuation by the coil, then they exit the coil, one current goes through the output transformer primary, they recombine as the collector current and exit to the power supply common.
    Yes, exactly.
    Suzie: It really seems very simple and logical. Why do the trolls need to make up all this convoluted junk science?
    Good question.

    Effect on the DC Supply Bus
    The coil provides no attenuation of AC current components going to or from the coil into the DC bus as previously demonstrated several ways.
    The coil is not “attempting” to introduce any AC voltage components onto the DC bus, however it is causing a pulsating current to flow on the DC bus.
    The DC supply itself will keep the voltage of the DC bus relatively constant but it does nothing whatsoever to smooth out or attenuate the current pulses. It is more than happy to supply current pulses if that’s what the circuit is asking for.
    Since the coil provides no smoothing of the full wave rectified sine current pulses, the current in the entire bus back to the first capacitor or the DC supply will be current pulses which will cause radiation and interference with other circuits on the same bus if not mitigated.
    The only way to stop the current pulses from propagating along the DC bus and radiating and causing interference in other circuits is to put local bypass capacitors or other low pass filtering as close as possible to the center tap of the coil. The length of wire or trace between the center tap of the coil and the nearest bypass capacitors or other low pass filtering is a radiating antenna and should be kept as short as possible.

    Some Common Sense Thinking:
    Alex: So let’s think about this. We know the currents I1 and I2 are pulses, we know that ac in the coil wires will be equal and opposite, and we know that equal and opposite currents produce flux that cancels and we know that if there’s no flux, there’s no inductance. So that’s one way of knowing there is no significant series inductance presented to the current coming in from the DC bus.

    Bobby: And we just proved that the current coming off the DC bus is pretty much an exact replica of the on transistor collector or drain current. Given that it’s an exact replica of the collector current, doesn’t that sorta indicate that there’s no significant inductance present?
    No, that alone doesn’t prove that there’s no inductance but it does prove that the coil is NOT keeping AC or pulsating current out of the DC bus.

    Timmy: Now if the current into the center tap was flat DC, that would mean that the current going into the output transformer primary was also flat DC and even I know that DC into the primary doesn’t create AC on the secondary so I would think it would be impossible for the current into the center tap to be flat DC. Even I can see that that defies common sense.

    Suzie: Seems kinda trivial, actually. There is basically one wire going in, the center tap, and one wire going out, the collector of the on transistor, and maybe some small leakage through the off transistor. So it would seem that simple common sense would tell you the current into the center tap pretty much has to be a replica of the current leaving the circuit through the on transistor. The electrons coming in have to equal the electrons coming out.

    "That Guy", Junk Science
    Timmy: What about what “that guy” said about electrons bunching up, getting stored in capacitors, getting stored in the magnetic fields and such and so the current going into the coil could be “Pure DC” and the current in the transistor and output transformer could be AC.

    Suzie: That’s all made up junk science. Electrons do not bunch up at these voltages and in these devices. For every electron that goes in one end of a wire or coil, an electron leaves the other end. They do not build up or get stored in the magnetic field. The same is true about capacitors. For every electron that enters one terminal, an electron leaves the other terminal. The current in one lead equals the current leaving the other lead, always. We’re never adding electrons to capacitors like an air tank, we’re merely moving them from one plate to another.

    Bobby: What about capacitors to ground or common?
    Suzie: Capacitors or anything else to common would provide additional “escape paths” for the electrons but there are none in this circuit except the off transistor so that’s a red herring.
    Bobby: So basically, in this circuit, every last device, wire, etc, the current into the device has to equal the current leaving the device in all cases and at all times.

    Didn’t “that guy” try to say you couldn’t have AC pulses of current coming out of a DC supply?
    Yes he did. He seemed to be very confused about this. He seems to be confusing voltage and current some how. He seemed confused about the fact a DC supply is a short to AC current and that since it’s an AC short then AC pulses can’t come out of it he thinks. Again, I think he’s completely confused about current and voltage and a lot of other concepts as well.

    The End Of Segregation:
    Didn’t he and others also make statements/questions like how can you have AC on the DC side?
    Yes, again they have huge misconceptions and assumptions about how things work. There is no such thing as an AC and DC side to a coil or transformer. The coil is completely dumb and unaware of such concepts as the DC side and AC side. The coil blindly reacts to what ever voltages and currents are applied to it by the external circuit and carries out whatever the laws of physics say it should do with these voltages and currents.

    Didn’t “that guy” try to say that since the voltage on the DC bus is constant that the coil must be keeping AC off the bus?
    Yes he did. Again, he is confusing current and voltage. The DC supply is keeping the voltage constant, not the coil. The coil is not attempting to put any AC voltage on the DC bus. It is merely drawing huge pulses of current from the bus. The power supply doesn’t care about current. It will provide whatever current the external circuit asks for. It will fight hard to keep the voltage constant but does not care a bit about the current.

    Doesn’t “that guy” still say that a DC supply cannot supply pulses of current? Yes, it certainly seemed like he was still suck on this.

    But I Measured It!
    Didn’t “that guy” say he measured the current going into the center tap and it was “Pure DC”? Why don’t you believe him?
    He seemed to be confusing current and voltage, he seemed very confused about where to measure the current, he seemed very confused about how a DC supply works, he used circle logic, he thinks electrons can jump out of coils into fields and on and on. As confused as he is and with his mind full of junk science, would you trust his ability to measure the right thing the right way and to interpret what he thinks he saw correctly?
    Yeah, good point.

    Additionally, he might be measuring the current in a circuit that does have capacitors or components to common between the coil and the collectors which would explain why the current at the center tap would be different from the current of the collectors.
    Can you show us an example?

    Apples and Oranges:
    Below is an output stage from Philips Application Note AN98030.
    Note that there are two chokes after the center tap measurement point and then C12 and C13 are connected between the incoming DC and common acting like bypass capacitors but on the collector side of the DC feed point. So in this circuit, of course you would expect the current at the indicated measurement point to look quite a bit different and smoother than the collector current pulses. This is a completely different circuit.
    If a circuit has these additional “escape paths” for current to common then the coil center tap current will look different from the collector current.
    Also, if there is excessive stray capacitance to common, that will allow the currents to be different and the center tap current to be smoother than the collector current.

    In any case, if there are these extra paths to common and the current at the center tap is smoother than the collector current, it’s not the coil that is causing this, it’s the capacitance.

    What about the fact that the coil is not made perfectly, imbalance, leakage flux in the coil, etc.
    None of that matters, you have one wire coming in, and one wire going out, and some small leakage through the off transistor, so the current in the center tap has to replicate the collector current unless you have “escape paths” as in the circuit above from the application note. The circuits we’ve been looking at have no such escape paths.

    Does the End to End voltage/current impact this analysis? (NO)
    •There is a voltage across the entire coil and as RICHS correctly pointed out, an end to end current will attempt to flow from one side to the other.
    •Let’s see what happens to this current and if it changes any of our analysis so far.
    •Note that for an end to end current to flow, the currents in the paired wires are not equal and opposite, They produce flux that does not cancel so they experience the wrath of the common mode choke at the center of this device and are severely choked.
    •The inductance seen by this end to end current depends on the core and the number of turns of the pair of wires and this current is generally deliberately kept large by design to minimize any collector to collector current. Sometimes a capacitor is placed from collector to collector to resonate with this and any other collector to collector inductance to cause this end to end current to be zero.
    •As long as the core does not saturate and is operating in a relatively linear region, we can completely ignore this end to end current according to the principle of superposition (look it up) and analyze this current completely separately from the other currents we’ve been discussing and vice versa.

    Summary Of Operation:

    1.The device provides a 1 to 4 impedance transformation. The transistor sees twice the current and half the voltage swing of the output transformer primary winding, therefore, the transistor “sees” 1/4th the impedance presented by the output transformer primary winding.
    2.The device does not keep AC components or RF out of the DC supply. It draws large full wave rectified sine pulses of current from the DC bus, essentially a full wave rectified version of the currents in the transistors. These currents are loaded with higher order harmonics. Other lowpass filtering needs to be provided as close to the center tap as possible to mitigate interference and radiation caused by these large current pulses.
    3.The length of the conductor from the center tap to the first lowpass filter element on the DC bus is an antenna broadcasting a signal that is twice the RF frequency and lots of nice harmonics so it needs to be kept a short as possible.
    4.The device is not smoothing these current pulses to any significant extent, rather, it is directly causing them.
    5.The device is causing the current pulses entering the center tap to split equally into equal and opposite currents in the transmission line where they do not experience any significant attenuation.
    6.The device is a Ruthroff 1:4 transmission line transformer in all respects and as such has been extensively analyzed in the literature.
    7.In these circuits, the current going into any area must equal the current leaving the same area at all times. See below about electrons bunching up and getting stored in electric and magnetic fields.

    Summary of Myths and Wives Tails:

    1.Myth: “The current into the center tap is flat DC equal to the average value of the collector currents.” The current into the center tap of the device is not flat DC, and is not flat DC current equal to the average value of the collector currents. It is full wave rectified sinewave pulses of current essentially replicating the collector currents. Think of a high power amp whose collector current during an on cycle is peaking at 16 amps. At the start of a cycle, the collector current is only a few milliamps. If the current into the center tap were a flat DC equal to the average current, where is all this extra 10 amps going? Certainly not through the off transistor.
    2.Myth: “The coil has an AC side and a DC side.” The coil does not have an AC side and a DC side. The coil merely obeys the laws of physics according to the voltages and currents applied to it by the external circuit.
    3.Myth: “The AC signal does not pass through the device because it’s a choke.” In fact, every bit of signal current that flows through the output transformer has come directly through the coil windings.
    4.Myth: “The device only handles DC, not AC.” See above. In addition, the device is providing a 1:4 impedance transformation.
    5.Myth: “A core cannot saturate if it’s not grounded.” The coil can saturate if a net DC voltage is impressed upon it regardless of whether it is grounded.
    6.Grounding something conveys some sort of magic upon things such as preventing core saturation. There’s nothing special or magic about grounding something. Ground or circuit common is just another conductor.
    7.Myth: “Electrons can bunch up thus explaining how the current into the center tap can be 10 amps while the collector current is only a few milliamps.” Electrons do not bunch up like air going into a pressure vessel. Not even in capacitors. In every last device in these circuits, for every electron entering one terminal of a device, an electron leaves another terminal of the device. The current into every device and every wire has to equal the current leaving the device at all times.
    Myth: “Electrons can jump out of wires and get temporarily stored in electric and magnetic fields and then come back into the wires later as electrons thus explaining how the current into the center tap can be 10 amps while the collector current is only a few milliamps.” No.

    Attached Files:

  8. WA1GFZ

    WA1GFZ Ham Member QRZ Page

    Look at AN749. The output impedance is VCC^2/2PO so in a typical 300 watt stage 1 FET
    (Optimum Voltage is around 43 volts for 300W. Check out Helge's 1KW Bipolar amp where he talks about operating voltage) 43^2 / 2 (150)=6.16 ohms so two in push pull will be 12.3 ohms. So I don't see the need for impedance transformation in T2. I'm asking the question not stating fact here. When one FET is on, it sinks current from T2 plus sinks current from T3 primary since they are common to the drain. The induced current in the other winding of T2 is now sourcing current to the other end of the T3 primary and charging the drain stray C of the FET that is turned off. I look at T2 as a 1:1 phase reversing TLT.
    I've thought about this a lot and wonder what you think? Think in terms of DC.
    At 50% efficiency 300W/43V =6.9A 43/6.9=6.16 ohms.
  9. KD2NCU

    KD2NCU Ham Member QRZ Page

    Good morning WA1. I think you know I have never advocated for or against the T2/T3 arrangement. I started these threads trying to understand more about why the T2/T3 arrangement was used at all and what T2 is and does that T3 cannot. I got so many garbage answers that defied common sense and physics, I decided I had to analyze it myself & then when I posted what I came up with for review the trolls tried to tar and feather me rather than actually look at the analysis I had presented. Anyway, I don't know enough yet to have an opinion on the T2/T3 vs T3 alone, but I'd like to keep working on this to learn and get a better understanding although I do think I have a solid understanding of what T2 is and does now.

    I've only heard two statements that actually make 100% sense to me, and they're actually related.
    1. If using T3 only, as you go higher and higher in frequency, the primary winding of T3 wants less and less turns. At some point you can't effectively center tap it so you introduce T2 as the center tap. Makes perfect sense. But there are other ways as well.
    2. If using T3 only, if you need a large impedance transformation you end up with a lot of turns on the secondary resulting in a large transformer, large wires to keep the resistance down because the resistance goes up very quickly with longer wires at RF due to skin effect and so on. So the use of the T2 provides a 1:4 impedance transformation and not as many turns are needed on the secondary.

    So here's what I'm going to do. First I'm going to review all the answers I got when I first asked what T2 does that T3 alone cannot and try to sort out what was garbage and what seemed valid. I will also review the references you provided.

    As an example, one person said that T2 only handles DC while T3 only handles AC and therefore T3 can be much smaller and T3 won't saturate and so on. Well, T2 "handles both AC and DC very clearly, actually pulses", and I've not seen a good explanation as to why T3 can be much smaller in this arrangement, T3 is actually handling pulses as well, and I would think that T3 can still saturate if the pulses are unbalanced, and from the application notes and on line reading I've done, it certainly looks like T3 is generally ENORMOUS compared to T2 right next to it, & so on so I don't know if I believe that T2 allows a "much smaller" T3, and some say T3 will saturate if it has both DC and AC in it, yet I see a hell of a lot push pull circuits with centertapped T3 only so I'm assuming they're all saturated ;-] . Maybe the fact that T2 provides a 1:4 impedance transformation allowing less turns on the T3 secondary does allow a somewhat smaller T3, but I've not really looked at that yet. Now I can't say any of this is wrong because I haven't analyzed it yet, but it won't surprise me if a good percentage of what I've heard is junk science and doesn't stand up under scrutiny.

    Anyway, send me whatever you have, AN's, links, references, etc., that seem to be explaining or analyzing the differences and pros and cons of the T2/T3 vs T3 only arrangement and I'll start working on this with you.

    Also, thanks for all the "bedtime reading" material. Nothing like curling up with some good reading material on a cold fall night!

  10. WA1GFZ

    WA1GFZ Ham Member QRZ Page

    I have a couple things to mail you from home later. Just a couple comments. As the frequency is increased both T2 and T3 require less reactance to isolate the power supply. Now to plant a seed. Say you wind your T2 with two coax cables and wire the shields as you would in the typical T2 with the CT going to DC power. There is your T2 function. Now take the two center conductors at the FETs and cross connect them to the opposite drains as the shields. So the phase of the RF current is opposite between each center conductor and shield. Next the two remaining center conductors at the DC ends of the shields become the amplifier output. Add series capacitors to isolate the DC from the outside world and here is your balanced output. You need a balun to convert balanced output to single ended coax. Everything on one core. You have a transmission line transformer with balanced (well almost) DC just like T2. The center conductor is the step up function to get to 50 ohms. Some people prefer to use two cores one for each coax but that changes things. That works at the expense of high flux density in the core due to unbalanced dc offset. Also you need more turns to get to minimum reactance. W6PQL uses 4 but I think 3 will work. When you only use one core you can get away with 2 turns of each coax on a common core to get enough reactance for 160m. Remember the shorter the coax the better your high frequency performance. (Check out Jerry Servick's book #4 on TLTs showing the advantage of using 1 core) W6PQL and I use a similar core but he uses 4 of them while I use 2. His output transformer is 1:3 so he needs T2 because there isn't an easy way to apply DC to the transformer.
    I never played with the 1:3 transformer but I would be inclined to use a pair of cascaded 1:2 transformers and maybe two big FETs in push pull parallel. Legal limit easy with 1 stage. This is covered in Helge's book in the output transformer section..
    I'm going to send you a couple application notes showing BB VHF amplifiers to think about. gfz

Share This Page

ad: wmr-1