|
Current through coils
Wes Stewart wrote:
The VVM probes are comprised of a quad diode sampling bridge followed by an FET amplifier. They are nominally coaxial, although without the BNC adapters, they have an exposed pin (very delicate) and at lower frequencies they can be used much as a high impedance scope probe is used. Thanks, Wes. When you say "lower frequencies", does that include 4 MHz? The instrument uses a phase-locked oscillator to drive the samplers with the "A" probe being the reference. One meter can be switched to display the amplitude of either channel and the second meter reads the phase difference between them. I was planning to use toroidal pickups and a Lissajous figure for the phase measurement. Did you know "Lissajous figure" is described in my 1957 ARRL Handbook but not in my 2000 ARRL Handbook? My main concern is how to ensure there are no reflections present during the measurement. I need to put the 75m bugcatcher coil in an RF loop where current is flowing in only one direction. That's easy to draw on paper but I'm concerned about it. How would you set it up? -- 73, Cecil http://www.qsl.net/w5dxp |
Current through coils
Wes Stewart wrote:
Cecil Moore wrote: "Your graphs show standing wave current which doesn't flow...blah blah" When I show otherwise, snip, gone without reply. Sorry, I completely missed it. I'll go back and try to find it. Sorry, "A bunch of IEEE PhD's" impresses me less than a handful of the guys posting here. Have you looked at the articles on the web pages I posted? Here's a funny quote, not previous quoted from: http://www.ttr.com/corum/index.htm "What frequency did you get in step 5? ... Is the difference within engineering accuracy ... less than 5%? If the answer is yes, then you may confidently use lumped-element modeling. However, if the answer is no, then, from the halls of Valhalla, old Wotan, himself, is thundering out over the battlements, '#*@&%!! ... Thor, you dumdum! You CAN'T use lumped circuit modeling!' ... [The coil has standing waves and is behaving as a distributed resonator.]" In case you missed it, here's what Walter Maxwell had to say about the subject: "If an inductance is in series with a line that has reflections, the current will NOT be the same at both ends of the inductor." "Consequently, circuit analysis will not work when both forward and reflected currents are present in a lumped circuit." -- 73, Cecil http://www.qsl.net/w5dxp |
Current through coils
Cecil Moore wrote:
In case you missed it, here's what Walter Maxwell had to say about the subject: "If an inductance is in series with a line that has reflections, the current will NOT be the same at both ends of the inductor." "Consequently, circuit analysis will not work when both forward and reflected currents are present in a lumped circuit." Cecil, I really think you should let Walt speak for himself. You have a history of distorting facts and taking statements out of context, and may be discrediting Walt. Walt is too nice a person for me to stand by and let that happen. If anyone really thinks that as a stand-alone statement, it is not correct. I suspect he didn't get the full story or wasn't following a discussion closely, or you have snipped something out of context. It's very easy to take small areas out of context and make it seem like someone is saying something they are not. Any circuit analysis will work so long as the load impedances used in the analysis are the same as the load impedances presented at that point by an antenna. The behavior of any small two-terminal component REQUIRES currents to be essentially equal. It's only when the component has a third significant path to the outside world that currents can be unequal. If I have a small capacitor, current flowing in one lead is equal to current flowing out the other and the phase of each current is exactly equal. Same for an inductor. That's not a guess, that's a rule of how things always behave. I'm wondering if the real problem is some people spend too much time with transmission lines and antenna and not enough time with circuit components, and become rusty? In any event, you do enough damage to people's reputations Cecil. Please leave Walt alone. He will speak for himself if he likes. 73 Tom |
Current through coils
Cecil Moore wrote: The generator sees a reactive load. When the generator sees a reactive load, current and voltage are no longer in step. This is true all through the system from source to load. I didn't ask or say anything about voltage. The fact that you refuse to answer my technical questions speaks volumes. The fact you can't understand simple direct answers does the same. You asked how what I measured could happen, I answered. You either are choosing to ignore the answer becuase you don't like it, or you don't understand it. Please define "compact" in terms of the number of degrees of phase shift measured using a traveling wave. Phase shift in what Cecil? The measured phase shift is in a traveling wave through a 75m bugcatcher coil. How long does it take the traveling wave current to flow from one end of the coil to the other? Your lumped-circuit model presupposes instantaneous current flow for traveling waves. Let's measure the current delay in a traveling wave to see if your model is correct. If it is not correct, it is useless. When Roy measured current (and I did the same) using inductive coupling in a current trasformer, a method that requires a time-varying current to excite the secondary, you dismissed Roy's measurements with some odd response about him measuring current that doesn't flow. I already measured the phase of current, and it is nearly zero degrees. It seems obvious to me that when someone gives you and answer you don't like, you either personally attack that persona and call them a liar or you make up some lame excuse like "you measured current that doesn't flow". I don't know what others think, but it is starting to look to me like you either don't understand the basics of measurements or you are just unwilling to learn. You cannot even begin to understand the problem if you don't know that basic phase shift. I'm willing to bet that my 75m bugcatcher coil has at least a 40 nanosecond delay on 4 MHz which is a 60 degree current phase shift. I can measure that. My network analyzer measures time delays. The problem I see is if I take time from my busy schedule and measure it, you will either call me a liar or say I measured current that doesn't flow. Before measuring anything specific I'm going to warn you that I've measured group delays many times before, and the group delay in an inductor is significantly less than the group delay in a transmission line of the same conductor length. I know that from past experience. But if you promise to control yourself and not dismiss a measurement with personal attacks or insults, and promise to not do an about-face like you did with Roy and say "you really didn't measure current that moves with your thing that only measures changing current", I will do that. I really wish some of your ideas were correct. If they were correct, I would not have thousands of feet of coaxial cables coiled under my bench. I would not be forcing customers to cut long delay lines when their equipment could just use a simple wound up piece of enameled wire. If that measured delay is in the ballpark of 40 nanoseconds or more, it proves that your lumped-circuit model has failed and your invalid proof is presupposed in the invalid model. The only potential problem is your reaction to measurements. You keep trying to define the "inductor" in terms of degrees related to standing waves ... Not true, Tom, and just shows how confused you are about what I have said. For the Nth time: The phase of the standing wave current doesn't change up and down the entire length of a 1/2WL thin dipole. Why would anyone expect it to change at the ends of a loading coil? As far as I am concerned we can drop any discussion of standing wave current phase. It is meaningless. The phase that Roy measured was standing wave phase. It was already known and is completely irrelevant. I asked Roy to measure the traveling wave phase shift. He didn't. Does ANYONE on this newsgroup understand Cecil? I need help here. I have done it and told you how, you ignore it. Roy has done it and told you how, you ignore it. You guys are measuring standing wave current that doesn't flow and doesn't change phase. Your measurements are completely meaningless and your flawed model has you hoodwinked. What a silly statement. We are measuring a time-varying current that doesn't flow or change! The only way to get confused on that is if someone doesn't understand behavior of the basic component, gets in over his head and confuses himself trying to use a tool that doesn't work, and then lashes out at others and refuses to listen. That's an exact description of you and your lumped circuit analysis in a standing wave environment. Do you disagree with Walter Maxwell? Walt wrote: "If an inductance is in series with a line that has reflections, the current will NOT be the same at both ends of the inductor." "Consequently, circuit analysis will not work when both forward and reflected currents are present in a lumped circuit." Yes, if he wrote what you quoted and you didn't lift something out of context I totally disagree with him. The component is not the problem, Tom. The problem seems to be your feigning of total ignorance of the laws of reflection physics in order to avoid discussing the real problem. I don't think most qualified experienced people would think I am the ignorant one. There you go again! Back to traveling and standing waves. Yes, you are never going to understand what I am saying about standing-wave antennas until you discuss traveling and standing waves on the standing-wave antenna. Your lumped-circuit model is known to fail in the presence of standing waves. Nonsense. There you go again, back to the lowest form of debate. If you can't understand something or get trapped, just call the other guy a liar. No, it's a lot simpler than that. When you lie about something I said, I call you a liar. There you go again. Do you have any idea how statements like that make you look to others? You very clearly said current in each terminal of the inductor has a different phase shift several times in your posts. One more time. The standing wave current does NOT change phase at the ends of the coil. The standing wave current essentially does not change phase unless a dipole is longer than 1/2WL. The phase of the standing wave current is totally irrelevant. The forward traveling-wave current experiences a delay through the coil. The reflected traveling-wave current experiences a delay through the coil. This delay can be measured on the bench. If the delay is not negligible, your lumped-circuit model is useless because it presupposes a delay of zero. I can't understand what you are saying or what your point is, ... Please don't insult my intelligence or yours. Every one of us performed those experiments on the bench in college. Exactly what is it about bench measuring the RF current delay through a coil that you don't understand? I understand it fine. I don't think the problem is on my end. If it is, someone besides you will chime in and tell me. I'm afraid I don't trust your opinions very much. Maybe someone else can help me with your last statement. Do you even know what a standing wave current loop is? Do you? Maybe someone else on this group can explain or understand what you are trying to say. You must have missed EE203. :-) What is it about a continuous exchange of energy between the E-field and H-field at a fixed point on an antenna wire that you don't understand? That's just a characteristic of standing waves. Roy has used the same argument in the past to try to prove that reflected energy doesn't flow. But's it's the standing wave energy that doesn't flow. Are you confusing energy and current? Or are you just joking again? 73 Tom |
Current through coils
wrote:
Cecil Moore wrote: In case you missed it, here's what Walter Maxwell had to say about the subject: "If an inductance is in series with a line that has reflections, the current will NOT be the same at both ends of the inductor." "Consequently, circuit analysis will not work when both forward and reflected currents are present in a lumped circuit." Cecil, I really think you should let Walt speak for himself. Sorry, I don't care what you think. You and I (and Walt) know exactly who distorted the facts. If anyone really thinks that as a stand-alone statement, it is not correct. I suspect he didn't get the full story or wasn't following a discussion closely, or you have snipped something out of context. It's very easy to take small areas out of context and make it seem like someone is saying something they are not. Those are Walt's exact words, not mine. If you don't believe me, send him an email and ask him. The behavior of any small two-terminal component REQUIRES currents to be essentially equal. It's only when the component has a third significant path to the outside world that currents can be unequal. Wrong! Wrong! Wrong! Your lumped-circuit model presupposes that the currents are equal so you are begging the question. YOU CANNOT USE YOUR MODEL TO PROVE ITS OWN PRESUPPOSITIONS. I see you haven't yet read what Dr. Corum had to say on that subject. http://www.ttr.com/corum/index.htm If I have a small capacitor, current flowing in one lead is equal to current flowing out the other and the phase of each current is exactly equal. Same for an inductor. Sorry, that's just not true for inductors. In the real world, there is a traveling wave current delay through the coil that can easily be measured on the bench. That delay converts directly to a phase delay. You are simply mistaken, hoodwinked by your lumped-circuit model, which presupposes the proof of what you say above. You are once again, begging the question and assuming the proof without having proved anything. That's not a guess, that's a rule of how things always behave. BS, Tom. That's a rule from a model known to fail in the presence of standing waves. Models existing in your mind don't dictate reality. It is supposed to be just the opposite. I'm wondering if the real problem is some people spend too much time with transmission lines and antenna and not enough time with circuit components, and become rusty? The real problem is that you are looking for your keys under the streetlight instead of in the dark where you lost them. The real problem is that you are doing the same thing as the naive ham who tries to measure feedpoint impedance with an ohm-meter. The real problem is that you are using a tool known to fail under the conditions in which you are trying to use it. THE LUMPED-CIRCUIT MODEL FAILS IN THE PRESENCE OF STANDING WAVES! I know that. Walt knows that. Dr. Corum knows that. A number of lurkers on this newsgroup know that. Nikola Tesla obviously knew that in his 1897 patent application. In any event, you do enough damage to people's reputations Cecil. Please leave Walt alone. He will speak for himself if he likes. Please mind your own business. I have Walt's permission to quote his stuff. If he ever asks me to stop quoting him, I will. One wonders if your attitude would be different if Walt agreed with you? :-) -- 73, Cecil http://www.qsl.net/w5dxp |
Current through coils
On Fri, 10 Mar 2006 11:51:51 GMT, Cecil Moore wrote:
Wes Stewart wrote: The VVM probes are comprised of a quad diode sampling bridge followed by an FET amplifier. They are nominally coaxial, although without the BNC adapters, they have an exposed pin (very delicate) and at lower frequencies they can be used much as a high impedance scope probe is used. Thanks, Wes. When you say "lower frequencies", does that include 4 MHz? When I wrote last, I was nursing a big toe that had just suffered the trauma of having a 4' x 8' sheet of 3/4" plywood dropped on it edge on. So I didn't want to hobble into the shack to search for the manual. Now I have it before me. The nominal impedance of the probes is 100 Kohm shunted by 2.5 pF. If this doesn't upset your measurement then you're good to go. I think that when I was remembering probing circuits with the bare probes I was thinking of the HP Vector Impedance Meter more than the VVM. It was a lower frequency instrument designed for that purpose. The instrument uses a phase-locked oscillator to drive the samplers with the "A" probe being the reference. One meter can be switched to display the amplitude of either channel and the second meter reads the phase difference between them. I was planning to use toroidal pickups and a Lissajous figure for the phase measurement. Did you know "Lissajous figure" is described in my 1957 ARRL Handbook but not in my 2000 ARRL Handbook? Well, they gotta leave something out so they can included the latest PIC controlled-automatic-rig-to-computer-interface and coffeemaker doodad. My main concern is how to ensure there are no reflections present during the measurement. I need to put the 75m bugcatcher coil in an RF loop where current is flowing in only one direction. That's easy to draw on paper but I'm concerned about it. How would you set it up? Can't help you. |
Current through coils
Wes Stewart wrote: Cecil wrote: My main concern is how to ensure there are no reflections present during the measurement. I need to put the 75m bugcatcher coil in an RF loop where current is flowing in only one direction. That's easy to draw on paper but I'm concerned about it. How would you set it up? Can't help you. What Cecil needs to do is bias the coil with a DC bias current that safely exceeds the peak RF current. Then he would have RF current flowing in only one direction. 73 Tom |
Current through coils
"Cecil Moore" wrote in message et... Wes Stewart wrote: The VVM probes are comprised of a quad diode sampling bridge followed by an FET amplifier. They are nominally coaxial, although without the BNC adapters, they have an exposed pin (very delicate) and at lower frequencies they can be used much as a high impedance scope probe is used. Thanks, Wes. When you say "lower frequencies", does that include 4 MHz? The instrument uses a phase-locked oscillator to drive the samplers with the "A" probe being the reference. One meter can be switched to display the amplitude of either channel and the second meter reads the phase difference between them. I was planning to use toroidal pickups and a Lissajous figure for the phase measurement. Did you know "Lissajous figure" is described in my 1957 ARRL Handbook but not in my 2000 ARRL Handbook? My main concern is how to ensure there are no reflections present during the measurement. I need to put the 75m bugcatcher coil in an RF loop where current is flowing in only one direction. That's easy to draw on paper but I'm concerned about it. How would you set it up? -- 73, Cecil http://www.qsl.net/w5dxp Hi Cecil I have the Technical Manual for the HP 8405 (36 MB) and could send it to you with "Usendit" or "Skype". Jerry |
Current through coils
wrote: The fact you can't understand simple direct answers does the same. I love simple answers, Tom. What I don't like are simple-minded answers based on an invalid model. When Roy measured current (and I did the same) using inductive coupling in a current trasformer, a method that requires a time-varying current to excite the secondary, you dismissed Roy's measurements with some odd response about him measuring current that doesn't flow. The inductive coupling does NOT require a time-varying current. All it requires is a time-varying H-field. That standing wave H-field is indeed varying but it's not because current is moving laterally up or down the wire. That H-field is fixed at a point on the line exchanging energy with the E-field which is also fixed at the same point. If the H-field is not moving laterally up or down the wire (it isn't) then the current is NOT flowing. You must have missed that day in your fields and waves class. Take a metal rod. Slip a string through a washer and tie it. Loop the string onto the metal rod. Put a grommet on the rod on each side of the string to keep it in one place in the X dimension on the wire. Now, keeping the X dimension fixed, swing the loop in the plane of the Y and Z dimensions and look at it on edge. You are looking at a physical analogy of the standing wave current at a point on a wire. Is the string moving? Not in the X dimension which is constant and fixed by the grommets. At any point on a wire with standing waves, the E-field and H-field are not moving laterally up and down the wire. They are *stationary* at a point on the wire. All that is happening at that point is the E-field and H-field are swapping energy at the RF frequency. The current probe naturally picks up those stationary oscillating fields. You and Roy still don't understand what it was that was being measured.The current that you and Roy measured was not flowing. It was just standing there. That's why they call it a *standing* wave. The currents that are required to be constant through the coil are the traveling-wave currents. A standing wave is not at all a wave in the classic definition of EM waves. It is simply a superposition of two classic EM waves flowing in opposite directions. Here's an optical example of what is happening to you. The yellow light coming from your TV is an interference pattern between red, blue, and green light. You are measuring yellow light thinking that's a primary color. It is not. But you could use your yellow light measurement to estimate the strength of the primary colors. The standing-wave current is an interference pattern caused by superposition of forward and reflected current waves. Like the yellow light you are seeing, it is not primary, and like the yellow light, it is an artifact of interference.. In a wire in which one amp is flowing in one direction and one amp is flowing in the opposite direction, there is no net flow of current. Therefore, standing wave current has no net flow. That is obvious from its constant, fixed phase angle which doesn't change (much). I already measured the phase of current, and it is nearly zero degrees. The measured phase of the net standing wave current is near zero degrees whether a coil exists or not. All it means is that the net standing wave current is standing still. Basing your conclusions upon measurements of a current that is not even flowing is foolish. I don't know what others think, but it is starting to look to me like you either don't understand the basics of measurements or you are just unwilling to learn. You have been seduced by your model that is known to fail in the presence of standing waves. Why you cling to such a false prophet in the real world is beyond me. I can measure that. My network analyzer measures time delays. The problem I see is if I take time from my busy schedule and measure it, you will either call me a liar or say I measured current that doesn't flow. If you measure a traveling wave current, you will be measuring a current that is actually flowing. Your S12 phase shift measurement showed a -60 to -70 degree phase shift in a 100uH coil at one MHz. That measurement of yours has already proved that your lumped-circuit model is invalid. Why didn't you just use the zero degrees predicted by the lumped-circuit model instead of measuring it? :-) Before measuring anything specific I'm going to warn you that I've measured group delays many times before, and the group delay in an inductor is significantly less than the group delay in a transmission line of the same conductor length. I know that from past experience. I know that, Tom. The point is: If there is any appreciable delay through the coil, that fact violates the presuppositions of the lumped- circuit model. Therefore, a lumped-circuit model cannot be used to explain the characteristics of that real-world coil and especially not in a standing wave environment. But if you promise to control yourself and not dismiss a measurement with personal attacks or insults, and promise to not do an about-face like you did with Roy and say "you really didn't measure current that moves with your thing that only measures changing current", I will do that. I appreciate that and I would also appreciate it if you didn't pencil whip the results before reporting them. Please just be honest. I assume we are both after the truth. And be sure to measure a coil something of the size of a 75m bugcatcher coil. I think a 75m bugcatcher coil would show more of a delay than a toroidal inductor of the same inductive reactance. I really wish some of your ideas were correct. If they were correct, I would not have thousands of feet of coaxial cables coiled under my bench. I would not be forcing customers to cut long delay lines when their equipment could just use a simple wound up piece of enameled wire. Surely, you are familiar with helical transmision lines with a very, very small velocity factor. And Intel does use simple coils as delay lines in some of their PCB designs. Does ANYONE on this newsgroup understand Cecil? I need help here. They are there, Tom. But they just don't want to tangle with a junk yard dog. Most people don't have a thick enough skin to withstand your onslaughts. I get a couple of emails a week from those guys. One distinguished gentleman and well known ham said that you have never lost an argument, even when you were wrong. I know exactly what he means. What a silly statement. We are measuring a time-varying current that doesn't flow or change! It's magnitude changes but it indeed doesn't flow or change phase. It's magnitude changes because the E-field and H-field are continuously exchanging energy at the frequency of operation. If you understood the implications of a constant, fixed, unchanging phase, you would know that. Yes, if he wrote what you quoted and you didn't lift something out of context I totally disagree with him. So be it. Your lumped-circuit model is known to fail in the presence of standing waves. Nonsense. YOUR LUMPED-CIRCUIT MODEL IS KNOWN TO FAIL IN THE PRESENCE OF STANDING WAVES! What is it about that statement that you don't understand? Your lumped-circuit model presupposes conditions that don't exist in a standing wave environment. Therefore, it is invalid and another more powerful model must be chosen.Because your chosen model is invalid, the validity of everything you say is questionable. The lumped-circuit model is a subset of the distributed-network model. The distributed-network model is a subset of Maxwell's equations. If you don't understand the limitations of the model, you will choose to use it under the wrong circumstances. That's what you, Roy, and others have done. -- 73, Cecil, W5DXP |
| All times are GMT +1. The time now is 09:46 AM. |
Powered by vBulletin® Copyright ©2000 - 2025, Jelsoft Enterprises Ltd.
RadioBanter.com