|
HIGH Q CAPS FOR VLF LOOP ANTENNA?
On Thu, 27 Oct 2005 20:52:45 -0400, TRABEM wrote:
That being said.... .... I have a series resonant loop with moderately large conductor wire and reasonably high Q capacitors. It's tuned to resonate at 60 KHz. Is that to mean the loop circuit consists of an inductor (including its radiation resistance and copper losses) of about j10 ohms and a capacitor (including its losses) of about -j10 ohms and a load resistance (being the 2 to 10 ohms receiver input Z) in series (ignoring transmission line for the moment)? You quote a Q figure and talk about expected bandwidth earlier in the thread. Wary of making any unwarranted assumptions, is it safe to assume that you know that it is the loaded Q that will determine the bandwidth of the circuit in operation? If you insert the rx input Z in series in the loop as described above, you don't need a calculator to see that the loaded Q cannot be 200+, and you might be lucky if it is better than 5 if the numbers you have quoted are correct. This circuit is not likely to give you much front end selectivity, is it? Perhaps you need to consult a textbook to review your understanding of unloaded Q, loaded Q, efficiency, and bandwidth, and where to apply which Q value. Owen -- |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
On Thu, 27 Oct 2005 14:55:32 -0400, "Fred W4JLE"
wrote: May I ask, what is with the almost fanitical adherence to Q? Sure, it's a fair question. I have a simple receiver with a low impedance input that is few with a toroid transformer and a tuned circuit to match the impedances and to keep out of band signals out. I want to convert the receiver from HF to VLF (60 KHz) and to use a series tuned loop of high Q as an antenna. In order to simplify the receiver input, I have mentioned as an option to eliminate the 50 ohm matching transformer and the tuned circuit in the front end of the receiver....and to feed it directly with my low impedance loop. In this way, the loops high Q would serve as the only means of preventing out of band signals from getting into the receiver. In order to make sure that actually happens, I suggested making the loop Q as high as possible. Hence my 'almost fanatical adherence to Q' Not sure if it will work, but wanted to run it past the group. Regards, T |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
On Thu, 27 Oct 2005 20:52:45 -0400, TRABEM wrote:
I have a series resonant loop with moderately large conductor wire and reasonably high Q capacitors. It's tuned to resonate at 60 KHz. A series resonant loop. First, this is a contradiction in terms and as your interpretations hinge upon this reading, it bears examination. A series tuned circuit is a low impedance circuit, so 60 KHz signals from the antenna are passed to the receiver To be in series, you have to describe the source and its common, you simply describe the loop. With this, you neglect the coupling between the actual source of power, a remote transmitter, and the antenna's Radiation Resistance. This value appears no where in your analysis and yet it is largely responsible for the atrocious efficiency of this breed of antenna. I've thrown together a quick model of your 5M on a side loop using (and being generous) #1 wire. The bottom of the loop is 10M above ground. In terms of performance relative to an isotropic antenna it is down 60dB. It displays an impedance of: Impedance = 0.05849 + J 10.37 ohms An addition of a 0.2555µF capacitor draws this down to: Impedance = 0.05851 + J 0.006073 ohms Please note that the capacitor is perfect, no ESR whatever. In a system with Zc = 0.0585, this antenna presents a 1.11 SWR. The half power points of its resonance are only 600 Hz apart. Hence a Q of 100. This is without any extraneous detector circuitry whatever. We will see where its addition leads. By simply inserting your 2 Ohms (I know full well where my 10 Ohms will take us) - in series - (again, your thesis) and re-assigning the system Zc to that same 2 Ohms, this antenna presents a 1.077 SWR. Sounds hunky-dory, right? Except when you look at the Q which has plunged to 2.9 and the antenna loss now compares to -75dB compared to an isotropic. Your receiver, in series with the loop, has just killed 15dB of gain and wiped out the Q by 95%. Not bad for a day's work. I will forgo the remainder of your questions to allow you to digest the material above. You can validate these readings by using EZNEC which in its free version is perfectly suitable to this question. ------------------------------------- schematic I sent you by email. Didn't get it. My Kill filters barely let your last schematic through. Not sure what I did to deserve an honored position in your kill file. This is not the kill file of unsophisticates simply ignoring by posting name. Agent has much more flexibility to read headers and judge what is spam. Works great and eliminates that source by - well I cannot guess the amount simply because I don't count the kills, and none survive the trash can except those at a lower level of sifting. I'm getting a lot of correspondence right now helping designers out. About half a dozen posts a day. Seems to be peak season and their mail lands in my inbox without incident. I confess to being stubborn and cranky, but I don't think I was disrespectful or made inappropriate comments. I won't email you anymore schematics. I simply pointed out I've only receive one of your emails, and the kill files put it in the trashcan - that is one step above absolutely erasing it. The presumption from the several kill-rules is that your headers appear to be spoofed. Now, tell me that you aren't doing something out of the ordinary like passing mail through an open relay. ;-) 73's Richard Clark, KB7QHC |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
Hi Owen,
Hi Owen, I am using large copper cable and relatively low loss capacitors. I expect the 9 ohms of inductive reactance to cancel out the -9 ohms of capacitive reactance leaving only the sum of the AC (or RF) resistance of the copper and the ac resistance of caps to give me my net impedance. Am I correct up to this point?? I'm trying to take this step by step so I can understand where I've gone wrong.....clearly I must have made an error somewhere. I don't know what the actual ac resistance of the caps is, but I do know the ac resistance of 20 meters of #2/0 copper welding cable is pretty damn low. Yes, the loop is series tuned, so the output is taken on the unconnected capacitor terminal and the unconnected wire end. Rjeloop3 gives estimates my Q at 221 even though it thinks I'm building a parallel tuned loop. But, I think Q is Q, and the Q of both types of loops is the same provided the same materials have been used in both loops. I am (for now) not considering the effects of hooking it to a receiver and/or the transmission line. I confess I have not tried to quantify the actual values of the ac resistance of the copper and have only rough estimates of what the esr of the caps is. I'd be pretty surprised if the dc resistance of the cable is much more than .1 ohms though, so the ac resistance should be a little higher at 60 KHz. Can you estimate what the unloaded Q of the loop is (in round numbers), and if so, can you agree that it might be around 221 (as Reg's software predicts)? Can you estimate what the impedance of the loop is (in round numbers)? Again, do not factor in the receiver input impedance as we aren't sure whether I'll keep it as is or match it's impedance with a preamp and/or toroidal transformer. For the moment, assume the receiver is mounted at the loop (which is a very real possibility since it's fairly small). Thanks for jumping in. T |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
On Thu, 27 Oct 2005 22:38:47 -0400, TRABEM wrote:
Hi Owen, Hi Owen, I am using large copper cable and relatively low loss capacitors. I expect the 9 ohms of inductive reactance to cancel out the -9 ohms of capacitive reactance leaving only the sum of the AC (or RF) resistance of the copper and the ac resistance of caps to give me my net impedance. Am I correct up to this point?? I'm trying to take this step by step so I can understand where I've gone wrong.....clearly I must have made an error somewhere. I don't know what the actual ac resistance of the caps is, but I do know the ac resistance of 20 meters of #2/0 copper welding cable is pretty damn low. Yes, the loop is series tuned, so the output is taken on the unconnected capacitor terminal and the unconnected wire end. Rjeloop3 gives estimates my Q at 221 even though it thinks I'm building a parallel tuned loop. But, I think Q is Q, and the Q of both types of loops is the same provided the same materials have been used in both loops. Forget Rjeloop3 for the moment and think about what you have. You focus on how low the resistance of the loop inductance is, and whether or not the capacitor ESR is significant... neither is when you jam a 2 ohms receiver in series with it all (ignoring the transmission line). You seem to be analysing your series circuit with part of it (the rx) replaced with a s/c. I am (for now) not considering the effects of hooking it to a receiver and/or the transmission line. I confess I have not tried to quantify Well, what good is it to know what the loop L and C do when not connected to the receiver? the actual values of the ac resistance of the copper and have only rough estimates of what the esr of the caps is. I'd be pretty surprised if the dc resistance of the cable is much more than .1 ohms though, so the ac resistance should be a little higher at 60 KHz. Can you estimate what the unloaded Q of the loop is (in round numbers), and if so, can you agree that it might be around 221 (as Reg's software predicts)? Can you estimate what the impedance of the loop is (in round numbers)? Again, do not factor in the receiver input impedance as we aren't sure whether I'll keep it as is or match it's impedance with a preamp and/or toroidal transformer. For the moment, assume the receiver is mounted at the loop (which is a very real possibility since it's fairly small). Read Richard's response, though it is more detailed and no doubt more accuracy. I think you will understand the problem when you analyse a three component series circuit (your topology), the Loop L, the Loop C and the Rx input z. (You can ignore radiation resistance, loop loss, capacitor loss, they are all much less than rx input z so the loops loss is dominated by the rx input z in your circuit.) The place this will end up is that you will come to realise that knowing how the L and C of the loop behave unloaded, and dwelling on that behaviour ignoring the effect of loading is probably why you are where you are (an assumption I know). When you have worked that out, you may understand why others are asking how you are going to couple to the loop. Your proposal to insert the 2 ohms (or whatever) rx input in series with the loop components isn't delivering what you wanted, and it won't matter how thick the loop conductor is, or how low the ESR of the capacitor is. Owen -- |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
In a series resonant circuit, at resonance it is equivalent to a dead short
(disregarding the R of the circuit). Series resonant circuits are usually used as traps. To develop a voltage one needs a parallel resonant circuit at the resonant frequency, The Q will simply determine how quickly the voltage falls off each side of resonance. Next there are two types of Q, first the calculated unloaded Q and second the in circuit or loaded Q. I think you are heading down the wrong path with the series circuit as your fighting a loosing battle. Assuming a perfect coil and capacitor you create an infinite Q circuit. Now you hook it up in your circuit. First there has to be enough resistance to develop the voltage , and here is the rub, as you increase the resistance to develop a voltage you decrease the Q. Yuk! Go with a parallel circuit like the rest of the world uses and you will be going in the right direction. TRABEM wrote in message ... On Thu, 27 Oct 2005 14:55:32 -0400, "Fred W4JLE" wrote: May I ask, what is with the almost fanitical adherence to Q? Sure, it's a fair question. I have a simple receiver with a low impedance input that is few with a toroid transformer and a tuned circuit to match the impedances and to keep out of band signals out. I want to convert the receiver from HF to VLF (60 KHz) and to use a series tuned loop of high Q as an antenna. In order to simplify the receiver input, I have mentioned as an option to eliminate the 50 ohm matching transformer and the tuned circuit in the front end of the receiver....and to feed it directly with my low impedance loop. In this way, the loops high Q would serve as the only means of preventing out of band signals from getting into the receiver. In order to make sure that actually happens, I suggested making the loop Q as high as possible. Hence my 'almost fanatical adherence to Q' Not sure if it will work, but wanted to run it past the group. Regards, T |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
I have a series resonant loop ......
A series resonant loop. First, this is a contradiction in terms and as your interpretations hinge upon this reading, it bears examination. OK, now we're making progress. I knew there had to be an explanation for your insistence that the antenna presented a 2K impedance to the feedline! I absolutely knew it could not be correct and you were equally determined:: Let me describe exactly what I hope to build, and you can enlighten me regarding what the proper term is. Thanks for hanging in there, the road was a little bumpy.... I'm thinking feedline attached to one end of the wire. The other end of the wire is attached to the capacitor bank. The other side of the capacitor bank is attached to the other feedline terminal. Or, stated another way, the cap is in series with the wire and the 2 transmission line terminals are connected to the left over cap and the unused wire end. A series tuned circuit is a low impedance circuit, so 60 KHz signals from the antenna are passed to the receiver To be in series, you have to describe the source and its common, you simply describe the loop. With this, you neglect the coupling between the actual source of power, a remote transmitter, and the antenna's Radiation Resistance. This value appears no where in your analysis and yet it is largely responsible for the atrocious efficiency of this breed of antenna. OK, but I hadn't thought this entered into the calculations of the loop.....it is what it is and we all know it's too damn short and too damn close to the ground to be efficient. At 60 Khz, there is so little difference between 5 feet off the ground and 50 feet off the ground, that I guess I never thought it mattered much...and, so tended to skip over these details. Any antenna that I can build with my budget will never be efficient:: Is this a fatal error, or do I really need to look at this issue to proceed? I understand these types of shortened antenna are often less than .001 percent efficient because they are so short relative to the wavelength being transmitted or received. I've thrown together a quick model of your 5M on a side loop using (and being generous) #1 wire. The bottom of the loop is 10M above ground. In terms of performance relative to an isotropic antenna it is down 60dB. It displays an impedance of: Impedance = 0.05849 + J 10.37 ohms An addition of a 0.2555µF capacitor draws this down to: Impedance = 0.05851 + J 0.006073 ohms Please note that the capacitor is perfect, no ESR whatever. In a system with Zc = 0.0585, this antenna presents a 1.11 SWR. The half power points of its resonance are only 600 Hz apart. Hence a Q of 100. OK, are you telling me my receiver would need to have an (impossibly low) input impedance of .06 ohms to work well with the antenna I've planned? This is without any extraneous detector circuitry whatever. We will see where its addition leads. By simply inserting your 2 Ohms (I know full well where my 10 Ohms will take us) - in series - (again, your thesis) and re-assigning the system Zc to that same 2 Ohms, this antenna presents a 1.077 SWR. OK, I'm not completely understanding the last paragraph. Have we abandoned the .06 + j10 real loop impedance and/or are we talking only about feeding a 2 ohm impedance loop into a 2 ohm impedance receiver (which seems to be a match made in heaven at first glance)? If we aren't considering the .06 j10 anymore, are you trying to impress upon me that even a perfectly matched 2 ohm antenna connected to a 2 ohm impedance receiver knocks the Hell out of the Q by that large of a factor?? If so, I can understand the need for the buffer follower you suggested earlier!! Send me a sign, I sense an incoming lightening bolt:: Another user just suggested I think of Q as stored energy and that anything that consumed that energy severely lowers the Q. I can understand that since the goal of the impedance matching is to transfer as much of the stored energy as possible. Further, if this is the case, it seems an active antenna matching buffer amp (impedance shifter) is necessary when using a loop of the type I planned. Not sure whether I am getting it or going off on another tangent....like I said above, send me a sign:: Sounds hunky-dory, right? Except when you look at the Q which has plunged to 2.9 and the antenna loss now compares to -75dB compared to an isotropic. Your receiver, in series with the loop, has just killed 15dB of gain and wiped out the Q by 95%. Not bad for a day's work. I will forgo the remainder of your questions to allow you to digest the material above. Yes, please do forgo, for the moment anyway...... Please specify whether you are explaining a 2 ohm impedance loop hooked to a 2 ohm impedance front end (in the preceding paragraph). Then, we can proceed I think..... I'm trying to assume NOTHING. Please forgive me if I seem to need a lot of clarification. Not sure what I did to deserve an honored position in your kill file. This is not the kill file of unsophisticates simply ignoring by posting name. Agent has much more flexibility to read headers and judge what is spam................. OK, understand. When you made the original comment, I thought you were saying I had done something purposely inappropriate. I use Eudora in order to avoid the Bill Gates problem, so I understand the difference between sorting to the trash and sorting to the junk mailbox! Trash is trash, gets emptied forever when I close the program. Junk is possibly trash, but doesn't fit all the criteria, so it's saved (just in case it really isn't garbage). Got yah. your headers appear to be spoofed. Now, tell me that you aren't doing something out of the ordinary like passing mail through an open relay. Well, I'm not sure how and what happens after I hit the send button. To the best of my knowledge, I am not doing anything of that nature. It's (my email) a paid service, so it should be on the up and up. Is it possible it might be a DSL issue, where the IP address is masked to some extent? Thanks for hangin in there. T |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
I think you are heading down the wrong path with the series circuit as your fighting a loosing battle. Assuming a perfect coil and capacitor you create an infinite Q circuit. Now you hook it up in your circuit. First there has to be enough resistance to develop the voltage , and here is the rub, as you increase the resistance to develop a voltage you decrease the Q. Yuk! Go with a parallel circuit like the rest of the world uses and you will be going in the right direction. I think I'm starting to get it. Am I cutting off my foot to spite my face:: Comments made by you and a few others have nudged mein the right direction..... The higher I make the series resonant Q, the lower the impedance goes, hence it's almost impossible to get a lot of voltage out of it?? Not sure why it matters that much. But, I was under the impression that a perfectly matched antenna and front end would only decrease the Q by a factor of 2. Follow along with Richard's comments if you like and add your comments as I check here often and read everything, sometimes many mant y times:: Regards, T PS:I had begun thinking that the higher imedance presented by a parallel loop was harder to match with a balun, which is why I started thinking of a series loop. I'm gettin there, thansk again. |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
I've been trying to follow this thread because I like to play with tuned loop antennas for broadcast band reception. I'm not sure you missed that much, I'm possibly on the right track though. I missed the part about how much this 20 meters of #2 copper cable with its support weighs. Your project sounds Serious. That antenna must weigh close to 500 pounds No, not all. It's a little heavy, but we have big tall hard wood forest here and my main concern is not weight, it's the ice that happens in Winter. So, lighter wire would never survive. The loop antennas I've been building are large diameter coils of smaller wire. I recognize that the type antenna I build arent acceptable for your consideration. But, I do have some experience with using a low freq loop in the city. If you are located near man made noise, it is very likely that resonating the loop doesnt result in highest Signal/Noise ratio. I understand this. But, I live quite a distance from any city and the distances to my neighbors is measured in hundreds of feet. It's quite rural. If I suspect there is a source of noise in the house, I have access to the high voltage (pole mounted disconnect) and can shut the power off to my house from a switch 900 feet from the house. If the noise persists, it isn't a local noise source:: Perhaps you already have experience with Low Freq loops and can tell me about your experiences. I am interested in learning. I'm still learning too......and look forward to getting the big loop up. It's starting to look like i need to go back to the books though. Right now, Richard is trying to get me back on track as I've apparently led myself astray. I'm definitely still learning:: T |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
On Fri, 28 Oct 2005 03:00:00 GMT, Owen Duffy wrote:
I think you will understand the problem when you analyse a three component series circuit (your topology), the Loop L, the Loop C and the Rx input z. (You can ignore radiation resistance, loop loss, capacitor loss, they are all much less than rx input z so the loops loss is dominated by the rx input z in your circuit.) So... if you did this, and you want the loop to give you front end selectivity, and you want the bandwidth to be xxx which led you to want the LOADED Q to be 100 or more (whatever), you now know that the load introduced by the receiver into the series loop you have dictated needs to be better than () XL/Qloaded or 0.1 ohms (not twenty or more times that value). How can you deliver a load impedance to the loop derived from the rx input circuit and its transmission line that is efficient and less than 100 milliohms (at the loop)? Is your single turn series loop idea practical at 60KHz? Owen -- |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
On Fri, 28 Oct 2005 00:15:03 -0400, TRABEM wrote:
OK, now we're making progress. I knew there had to be an explanation for your insistence that the antenna presented a 2K impedance to the feedline! I absolutely knew it could not be correct and you were equally determined:: You are not off the hook yet. Let me describe exactly what I hope to build, and you can enlighten me regarding what the proper term is. Thanks for hanging in there, the road was a little bumpy.... I'm thinking feedline attached to one end of the wire. The other end of the wire is attached to the capacitor bank. The other side of the capacitor bank is attached to the other feedline terminal. Or, stated another way, the cap is in series with the wire and the 2 transmission line terminals are connected to the left over cap and the unused wire end. Fine, but it doesn't really matter. The unloaded Q is one thing, the loaded Q that is seriously depressed is another. OK, but I hadn't thought this entered into the calculations of the loop.....it is what it is and we all know it's too damn short and too damn close to the ground to be efficient. You have so little efficiency, that the proximity of ground hardly matters (you already know this) - until we look at the lobe pattern, balance, and that side of the coin. or do I really need to look at this issue to proceed? You do your best - you are not that far from it. The loop allows you to end load it better than using a dipole with coils. I've thrown together a quick model of your 5M on a side loop using (and being generous) #1 wire. The bottom of the loop is 10M above ground. In terms of performance relative to an isotropic antenna it is down 60dB. It displays an impedance of: Impedance = 0.05849 + J 10.37 ohms An addition of a 0.2555µF capacitor draws this down to: Impedance = 0.05851 + J 0.006073 ohms Please note that the capacitor is perfect, no ESR whatever. In a system with Zc = 0.0585, this antenna presents a 1.11 SWR. The half power points of its resonance are only 600 Hz apart. Hence a Q of 100. OK, are you telling me my receiver would need to have an (impossibly low) input impedance of .06 ohms to work well with the antenna I've planned? Just compare that to the 2 Ohms you hoped for and see what happened. OK, I'm not completely understanding the last paragraph. Have we abandoned the .06 + j10 real loop impedance and/or are we talking only about feeding a 2 ohm impedance loop into a 2 ohm impedance receiver You have injected 2 Ohms into the series circuit - it is now part of the antenna even if it inside the receiver. (which seems to be a match made in heaven at first glance)? Simply a slight of hand, mathematically. You have to pick a reference you call Zc (characteristic impedance) of something, and your resistance is wedded to the antenna. However, the reality of it is that the native loop compared to the actual load is severely mismatched. You don't lose 15dB gain on the stairway to heaven. Most would parallel a loop's capacitor with a really big resistive input Z match to preserve Q. You chose a path that is leading to grief. Send me a sign, I sense an incoming lightening bolt:: ZAP A series load in a circuit that is heavily laden with current requires a very, very small value. Your switches are deadly elements even if they are the best in the market. That same circuit, sans series detection, also supports high voltages courtesy of Q. At this frequency you have a world of components that can tap them without seriously loading the circuit and killing the Q, and the voltage, and the efficiency. Hence a follower ACROSS the capacitors and AT the loop driving the line back to the detector. A follower that presents 20K or even 2000K Ohm load to the loop at 60KHz is a walk in the park. You aren't going to recover that initial 60dB plunge from the size of the antenna, but you don't have to grease the slide for another 15dB plummet. Still, and all, there's always the practical reality that the design will work out of the box. Wrist watches do it. It won't be the best, but there's not much competition and this isn't DX we are talking about, unless you want to work Europe's VLF (and even then you may hear signals). You still haven't bitten the bait on what makes the "I" and "Q" channels useful beyond cramming them into black boxes to strip away the only really useful angle (pun intended) they offer - phase information. This is in fact the single most fascinating aspect of the detectors. 73's Richard Clark, KB7QHC |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
TRABEM wrote in message ... I think you are heading down the wrong path with the series circuit as your fighting a loosing battle. Assuming a perfect coil and capacitor you create an infinite Q circuit. Now you hook it up in your circuit. First there has to be enough resistance to develop the voltage , and here is the rub, as you increase the resistance to develop a voltage you decrease the Q. Yuk! Go with a parallel circuit like the rest of the world uses and you will be going in the right direction. I think I'm starting to get it. Am I cutting off my foot to spite my face:: Comments made by you and a few others have nudged mein the right direction..... The higher I make the series resonant Q, the lower the impedance goes, hence it's almost impossible to get a lot of voltage out of it?? Not sure why it matters that much. But, I was under the impression that a perfectly matched antenna and front end would only decrease the Q by a factor of 2. Follow along with Richard's comments if you like and add your comments as I check here often and read everything, sometimes many mant y times:: Regards, T PS:I had begun thinking that the higher imedance presented by a parallel loop was harder to match with a balun, which is why I started thinking of a series loop. I'm gettin there, thansk again. ======================================= Trabem, This discussion is getting you nowhere very fast. So let's summarise. I don't have your exact dimensions but the following are good enough. L = 27uH, Reactance = j10 ohms, Conductor loss = 0.05 ohms, ESR = 0.01 ohms, Radiation ohms = 0. Receiver input = 10 ohms, Ground loss ohms = 0.01 The intrinsic Q of the loop is 10 / 0.05 = 200. The working Q of the loop, when series connected, is Reactance divided by the SUM of all resistances including the receiver. Working Q = Reactance / ( 0.05 + 0.01 + 10 + .01 ) = 10 / 10.07 = 0.993 Take note of the ridiculous low value of working Q due to the loop being in series with the receiver. ---- Reg. |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
OK Richard, it looks like the lights are beginning to come on and I'm
on a different road (a road that doesn't end in a fiery disaster). At the risk of jumping to far ahead...... Let me know if this is correct or not...... Let's dump the series loop, it doesn't sound promising:: So I'm (now) using the same wire and the same resonating C, but it's in a parallel configuration. My loop is 2K impedance, my receiver is 50 ohms (more conventional although maybe not necessarily tuned, TBD). Is my mission to purposely isolate the loop so that anything happening in the receiver doesn't impact the loops Q? To maintain the best possible Q, would I mismatch the antenna to preamp impedance on purpose, providing a 50K or even 100K input impedance buffer, so as not to suck power out of the loop? And, should the output of the buffer present an approximate 50 ohm output impedance, therefore matching the input impedance of the RX (for best power transfer)? If this is the case, an active buffer amp seems inevitable. You gave me numbers for a 2 ohm loop fed into a 2 ohm RX. If I do not buffer the loop from the RX, wouldn't a 2K loop fed into a 2K RX also cause similar loss of Q (just like the 2 ohm over 2 ohm example you gave previously)? I'm absolutely sure I'm not out of the woods yet, but does it look like I'm on the right road yet?? It's a big change for me, can you give me a brief indication please.... Thanks. And, I am definitely not avoiding the I/Q issue. I know of successful hardware handling examples of the I/Q and also of successful software handling methods. I just haven't decided which one to use yet. The stand alone hardware is a bit harder to implement, complex filters, much more hardware to build, and a wideband phase shifting network are needed. But the advantage is that it's a stand alone solution. Software works well also, but leaves one dependent on the programmer and requires a computer. Not sure which way I'll go, but it's not important yet as I don't have a viable front end and antenna yet. So, am I on the right road, or am I still very far off track with regard to the antenna Q issue? Regards, T |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
Dear Trabem,
The input impedance seen looking into the series-connected loop is the RF loss resistance of the loop, in your case about .05 ohms. If 0.05 is impedance-matched to a 10-ohm receiver then the working Q only falls to about 100. But it is not an easy matter to match 0.05 ohms to 10 ohms at 60 KHz. ( I do not know the precise input resistance of your receiver but you get the idea.) The working Q of any tuned circuit, either series or parallel connected, when impedance-matched to a load, always results in the working Q becoming equal to half of the tuned circuit's intrinsic Q. This is rather obvious because the loss resistance of the tuned circuit and the load (after being transformed to the tuned circuit value) are equal to each other. Of course, impedance-matching also results in maximum voltage and maximum current being developed in a given load (or receiver). Which is also a desirable condition. It is a serious mistake to think in terms only of volts-input to the receiver. Or only current-input to the receiver. Receiver S-meters are POWER meters. That's why they can be calibrated in decibels or in terms of 6dB per S-unit. Or S9 plus so many decibels. For example, with a 50-ohm receiver, the reference level S9 = 50 pico-watts receiver input power. Please accept my apologies for digressiing from 5-metre square loops at 60 KHz. ---- Reg. |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
On Fri, 28 Oct 2005 11:02:18 -0400, TRABEM wrote:
Is my mission to purposely isolate the loop so that anything happening in the receiver doesn't impact the loops Q? The advantage of Q is that it multiplies I and V giving you sensitivity. As I have pointed out before, your current design could work without any changes. I cannot answer this for myself much less you and the advice I would have to offer is that you build your receiver with flexibility in mind. We are not talking big changes in components. That is the long answer. The short answer is yes. If this is the case, an active buffer amp seems inevitable. Easy enough to include, or remove depending on need. If I do not buffer the loop from the RX, wouldn't a 2K loop fed into a 2K RX also cause similar loss of Q (just like the 2 ohm over 2 ohm example you gave previously)? Certainly, but not similarly. The Q is not going to plunge to 2 or 3. And, I am definitely not avoiding the I/Q issue. I know of successful hardware handling examples of the I/Q and also of successful software handling methods. I just haven't decided which one to use yet. That is the point of my questions. They are veiled implications, not tests of knowledge. No one in your list of links, much less those I've read over the years knows the PRACTICAL implication of the "I" and "Q" channels. So, I may as well drop the other shoe. One does the demodulation of AM signals, the other provides demodulation of FM and SSB signals. I'm not sure which and what particular arrangement of supporting circuitry is required beyond simple AM amplifiers because my construction for that application was back in 68-69. Building tube models and guaranteeing design considerations was not as simple as the Tayloe circuit offers now. However, one of the fascinating characteristics of this style of detector is that you can feed each channel to the earpieces of a stereo headset. "I" for one, "Q" for the other earpiece. This gives you the chance to use your wet-ware instead of someone's software and hardware. The brain does all the necessary fourier analysis automatically and in real time. The upshot of it is that when listening to a CW signal, and hearing the field of signals around it, you perceive those signals in a mind-space. The signal that is center tuned sounds like it is between your ears, in the middle of your, as I described it, mind-space. Those signals that are above it in frequency sound as though they are coming from the right, and those signals that are below it in frequency sound as though they are coming from the left. The advantage of this detector, in this configuration, with this kind of perception, is that your mind is separating the signals psychologically. Even though the signals you hear on the left and right are in equal amplitude to the center, you can exclude them mentally. Imagine taping a conversation in room full of people and the microphone is not at your, or your partners lips, but between you, and you are both standing off a couple of feet talking over the crowd. You full know that you could understand your partner at the time of the recording, and you probably know that the tape would be a bitch to make sense of, even though it makes a faithful record of the conversation in that free-for-all. The difference is that your binaural perception with its phase separation capability could be brought to bear to ignore the field of noise to concentrate on your partner. When you hear the mono recording, the phase information is lost and your partner's conversation merges with the background noise. I cannot personally vouch for this effect because the payoff in my construction back then didn't come down to finally evidencing this effect for myself. This wet-ware characteristic was reported to me to be one of the attractions of building for my professor. I have also played with bucket-brigade delay lines to create this effect. At one time Paul McCartney was using it with his music. Aural phase relationships have a strong psychological information content that is taken for granted. 73's Richard Clark, KB7QHC |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
The mixed-up confusion along this extended thread is due to the inability of contributors to describe in plain English exactly what they mean about a relatively simple matter. It's a breeding ground for baffle-gab, confusing nonsense and old wives. To avoid wasting more time I respectfully suggest Trabem obtains a big bunch of capacitors of various values and gets on with the job. We will all be very interested in the outcome. Now perhaps we can return to which part of a 1/2-wave dipole does the most radiating - the middle bit or the ends? ---- Reg, G4FGQ. |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
OK, maybe you're beginning to understand. Q can be calculated as
reactance (at resonance) divided by the effective series resistance, or as effective parallel resistance divided by reactance at resonance. For a loop where you know the series resistance, it's easiest to use that first relationship. If you put your 10 ohm receiver input in series with your 10 ohm reactance loop, you've ruined all that effort to get to a very low loop conductor resistance and obviated the need for high-Q capacitors. And we're all having a very hard time seeing how you will couple your 10-ohm receiver input to EITHER the parallel-tuned loop OR the series tuned loop, without having nasty consequences for your holy-grail Q. You might think it's best to impedance match ("conjugate match") to your load, so you transfer the most power to the load. However, that may not be optimum from a system design standpoint. If you already have enough signal (along with atmospheric noise) that the receiver doesn't contribute significantly to the overall SNR, then you may be better off by intentionally mismatching so that the Q remains high, if that's important to you. (I personally think you've overrated it, but that's up to you to decide.) But even if you're wanting to get the lowest noise contribution from your electronics, the appropriate match is generally not the conjugate impedance match that results in highest power transfer. For example, an MMBT2222 NPN transistor running at about 100uA collector current in a common-emitter configuration with no feedback will have a low-frequency (e.g. 60kHz) input resistance around 50kohms, but the optimal source resistance from a noise standpoint--the source resistance which will yield the lowest noise figure for the amplifier--will be about 2kohms. At optimal source resistance, you can get a noise figure well below 1dB from an MMBT2222--and from many other bipolars. One reason that people like to use FET amplifiers across their "parallel-tuned" loops is that the amplifier input resistance is quite high, but (using appropriate FETs) the noise contribution of the amplifier is negligible. And with proper design, the distortion contribution can be considerably lower than the distortion of your detector. For high source impedances, JFETs can give noise figures that are a small fraction of a dB. Cheers, Tom |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
OK Reg,
Hi Reg, Read both of you responses, and it's very clear now that I was seriously out in a fantasy world with respect to the topic. I feel a lot closer to reality now. If you can, check the latest comments between Richard and myself. I think I'm getting it, or at least the first approximation:: You provided a key piece of info when you gave me the verbiage about the loaded Q formula in a series tuned loop. When I started out, I had no idea that the loaded Q could possibly drop so much when connected to a receiver! The working Q of any tuned circuit, either series or parallel connected, when impedance-matched to a load, always results in the working Q becoming equal to half of the tuned circuit's intrinsic Q. I knew this already! My big problem was in realizing that the loop impedance was so very very low. Once Richard got me a closer approximation of the actual number, it became VERY clear to me that there was no impedance match in my original configuration! Richard suggested the impedance of the loop was 2K, I guessed it was 2 ohms, but the actual figure was in the milliohm range. I feel SO MUCH better now and I think I'm much better off thanks to your (and Richard's) patience. Thank you so much for helping me to get to this point! T |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
By the way, I consider the most sensible and understandable
contributions to this thread have been the questions asked by the originator, Trabem. I am now half-way down a bottle of South African red plonk. It's supposed to be good for the arteries. ---- Reg, Hic. |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
"TRABEM" bravely wrote to "All" (28 Oct 05 00:42:15)
--- on the heady topic of " HIGH Q CAPS FOR VLF LOOP ANTENNA?" TR From: TRABEM TR Xref: core-easynews rec.radio.amateur.antenna:219360 This should require a matching transformer otherwise not enough current can flow to take advantage of the potentially huge Q. Without current flow there will be no energy storage in a series circuit. In a parallel circuit the current flow is internal between the parallel reactances but in a series circuit it must be external. A series circuit is naturally current driven so the transformer would basically be converting current into a voltage that the receiver input can use. If you measure your loop with a test signal you should find it requires a lot of current and little voltage. From your investigations, calculate the effective resistance and use this as your Rs to find the actual Q and the matching turns ratio required for an autotransformer, for example. Good luck, A*s*i*m*o*v .... Why is Brassiere singular and Panties plural??? |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
Hi Richard,
I feel SO MUCH better now!!!!!!!! Just read and replied to Reg's latest comments a few minutes ago and it seems I am on the right track-my major error was not realizing that my series loop impedance as originally suggested was very much lower than I thought it to be ...and I've finally realized that it COULD NOT POSSIBLY be matched to a receiver that had 2 (or 10) ohms input Z. Now that I've realized my originals series loop antenna had an impedance in the milliohm region, the other explanations that you gave made perfect sense. At least, I feel like I made it to the first level:: I'm a little hesitant to make this suggestion, but let me ask the question(s) at the risk of taking to large a step forward and stumbeling:: If you would, just give me a yes or no answer to these questions so I can make sure I don't have seriously flawed remnants of the old thinking..... ------------------------- With the current model of parallel loop... If my receiver was 2K input impedance (whether tuned input or not), could I connect it to my loop with a piece of 2000 ohm open wire line and expect the net (or loaded) Q to be around 100 (assuming a 2K impedance open wire line could be built and that my unloaded antenna Q was 200 to start with). Yes or No? ------------------------- With the current model of parallel loop... If I elected to use a buffer amp with megohms of input impedance, would I preserve the unloaded Q and end up with a net Q of about 200 because I haven't loaded the loop? Yes or No? ------------------------- I believe my receiver is microvolt sensitive and that the loop will deliver a relatively good signal to the receiver even though the loop isn't terribly efficient. If I build selectivity into the front end of the receiver, do I really need high Q (200)?? I think the answer is NO..... Since my receiver is quite sensitive (characterized at uV before I convert it to VLF), I think I could save a lot of money by sacrificing some antenna Q and building modest selectivity into the front end. Or, I could elect to use the antenna as planned (impedance matched, but no front end tuned circuit) and instead convert the receiver to an untuned input (allowing the antenna Q to be the sole form of tuning). Is this reasoning basically correct or seriously flawed? ------------------------- However, one of the fascinating characteristics of this style of detector is that you can feed each channel to the earpieces of a stereo headset. "I" for one, "Q" for the other earpiece. This gives you the chance to use your wet-ware instead of someone's software and hardware. The brain does all the necessary fourier analysis automatically and in real time. The upshot of it is that when listening to a CW signal, and hearing the field of signals around it, you perceive those signals in a mind-space. OK, I am an avid cw operator, often operating as a hired gun at m/m HF contest efforts. So, I completely understand the concept of having the brain do the processing. The brain is a very seriously viable filter that is adaptive with regard to the audio spectrum sent to it by a conventional receiver. I haven't tried actual binaural operation, but have heard others talk about it. The users claim it is a different world from the very first second of listening to it and are constantly amazed at the effect and improvement. I have a friend who isn't quite local...but we chat from time to time although we don't see each other that often. He has a high frequency hearing loss in one ear and has a great deal of difficulty with cw. He built a binaural project from a QST article and was stunned to hear the results. He was sold ont he idea in short order! But I never though much about hooking up a stereo headphone to the i/q audio streams. The difference is that your binaural perception with its phase separation capability could be brought to bear to ignore the field of noise to concentrate on your partner. When you hear the mono recording, the phase information is lost and your partner's conversation merges with the background noise. OK, I am with you with respect to the brain filtering out unwanted conversations to let you focus on your conversational partners distant voice. But, I thought traditional binaural receiver meant that the frequencies higher than a certain point went to one ear and that the all the frequencies lower than the same frequency went to the other ear. In this manner the listener has a feeling of 'depth' or 'richness' that isn't present in a mono setup. This is interesting though. But, I never thought that the brain could process the I and Q to provide opposite (unwanted) sideband rejection...which is why I thought the primary function of the I/Q precessing was about (whether it be hardware or software based). Are you suggesting that the brain can also process the I/Q output streams and provide opposite sideband rejection as well as selective frequency and adaptive filtering? I have also played with bucket-brigade delay lines to create this effect. At one time Paul McCartney was using it with his music. Aural phase relationships have a strong psychological information content that is taken for granted. I'll investigate this over the Winter season, which is long and hard here. Thanks for planting a bug in my ear (no pun intended) about this. I probably would not have thought of it otherwise. Regards, T |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
Trabem,
Without wishing to detract you in any way from your objective of a matched series tuned loop I would like to describe how I would do a similar job with the usual parallel tuned, multiturn loop. I do not understand the type of receiver you propose and I am not seriously interested. But I should say the theoretical working bandwidth of my proposal is about 1/2 of yours. Actual bandwidth of both your and my proposals is indeterminate because of the uncertainty of ground proximity and nearby environmental loss. The working bandwidths could be very similar. Using similar size loop dimensions to yours, ie., 5.3 metres square - Frequency = 60 Khz. 5 turns of close wound 2mm diameter enamelled wire. Inductance = 710 micro-henrys. Tuning capacitor = 0.01 uF approx. Reactance of L and C = 268 ohms. Conductor resistance loss = 2.5 ohms. Intrinsic coil Q = 107 Matched working Q =53 3dB working bandwidth = 1.12 KHz. Impedance match to 50-ohm receiver obtained via small coupling loop, in the same plane, about 1 metre square. Working Q = 53 or less depending on height above ground. The working Q may not be high enough for your particular application. I describe the antenna for you to see what is possible in comparison with your series-tuned proposals. ---- Reg, G4FGQ. |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
By the way, I consider the most sensible and understandable
contributions to this thread have been the questions asked by the originator, Trabem. I think that's a high compliment, considering how totally messed up I was at the start of this. I am now half-way down a bottle of South African red plonk. It's supposed to be good for the arteries. If a little is good, is more better? And a very BIG + THANKS from me. Thanks for hangin' in there. Regards, T PS:Read your previous example, which closely parallels some existing real life loops I found on the Internet last evening. Thanks for the example as well and I think it's time to start soldering. |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
On Fri, 28 Oct 2005 09:02:14 -0700, Richard Clark
wrote: That is the point of my questions. They are veiled implications, not tests of knowledge. No one in your list of links, much less those I've read over the years knows the PRACTICAL implication of the "I" and "Q" channels. So, I may as well drop the other shoe. Didn't Don Stoner describe a synchronous detector way back. I think I remember reading an article in the mid sixties in "The Sideband Handbook" or similar. I was about 15 then, so a detector that had something like 17 bottles in it seemed overkill when I was copying CW and SSB on an AM receiver (ie diode detector) with BFO. The appeal being an all-mode detector (including DSBSC), but synchrounous detectors didn't seem to catch on in comms receivers, well not until DSP detection... well I don't recall coming across them anyway. Owen -- |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
On Fri, 28 Oct 2005 14:19:14 -0400, TRABEM wrote:
With the current model of parallel loop... If my receiver was 2K input impedance (whether tuned input or not), could I connect it to my loop with a piece of 2000 ohm open wire line and expect the net (or loaded) Q to be around 100 (assuming a 2K impedance open wire line could be built and that my unloaded antenna Q was 200 to start with). Yes or No? You have too many suppositions to give a straight answer. First, there is no such thing as a 2000 Ohm open wire line. As for the gist of the question: Yes. With the current model of parallel loop... If I elected to use a buffer amp with megohms of input impedance, would I preserve the unloaded Q and end up with a net Q of about 200 because I haven't loaded the loop? Yes or No? Yes. However, as Tom has pointed out separately, this may not be the optimal solution. I believe my receiver is microvolt sensitive and that the loop will deliver a relatively good signal to the receiver even though the loop isn't terribly efficient. If I build selectivity into the front end of the receiver, do I really need high Q (200)?? I think the answer is NO..... Well, this is a good opportunity to examine that tumble down the slope to the Q = 2 (caused by the severe loading of your proposed design). The correlative to this is, how much selectivity do you need in a field where stateside VLF is relatively rare? Further, by the action of the strong filtering that usual attends the "I" and "Q" channel processing, you could easily repair any shortfall. However, back to that Q = 2. That still offers respectable (not fantastic) selectivity against signals out at the bottom of the AM band which is 10f (one decade) away. OK, I am with you with respect to the brain filtering out unwanted conversations to let you focus on your conversational partners distant voice. But, I thought traditional binaural receiver meant that the frequencies higher than a certain point went to one ear and that the all the frequencies lower than the same frequency went to the other ear. In this manner the listener has a feeling of 'depth' or 'richness' that isn't present in a mono setup. Binaural is what the fellows in white coats mean by listening with two ears. This is interesting though. And rarely reported for this style of detection. What a pity. Are you suggesting that the brain can also process the I/Q output streams and provide opposite sideband rejection as well as selective frequency and adaptive filtering? The rejection is psychological, not actual. It is what I meant by "mind-space." The vectors do not add up to zero, the mind simply ignores the off-band content like you would at a party listening to that cute office temp's whispers when your wife is yelling across the room at you. Listen to a recording of that same scenario in mono and you WILL hear your wife! The brain reassembles all delay/phase information content at the party to sort out what to pay attention to. When that same information is passed through a monaural channel, the phase information is lost and everything competes equally lousy given the S/N ratio. 73's Richard Clark, KB7QHC |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
On 28 Oct 2005 09:44:00 -0700, "K7ITM" wrote:
OK, maybe you're beginning to understand. Q can be calculated as reactance (at resonance) divided by the effective series resistance, or as effective parallel resistance divided by reactance at resonance. For a loop where you know the series resistance, it's easiest to use that first relationship. If you put your 10 ohm receiver input in series with your 10 ohm reactance loop, you've ruined all that effort to get to a very low loop conductor resistance and obviated the need for high-Q capacitors. And we're all having a very hard time seeing how you will couple your 10-ohm receiver input to EITHER the parallel-tuned loop OR the series tuned loop, without having nasty consequences for your holy-grail Q. It appears to me that a 10 ohm receiver input impedance is dead center with regard to any loop configuration I could come up with...it doesn't work with anything! It's easy to see now:: You might think it's best to impedance match ("conjugate match") to your load, so you transfer the most power to the load. However, that may not be optimum from a system design standpoint. If you already have enough signal (along with atmospheric noise) that the receiver doesn't contribute significantly to the overall SNR, then you may be better off by intentionally mismatching so that the Q remains high, if that's important to you. (I personally think you've overrated it, but that's up to you to decide.) But even if you're wanting to get the lowest noise contribution from your electronics, the appropriate match is generally not the conjugate impedance match that results in highest power transfer. For example, an MMBT2222 NPN transistor running at about 100uA collector current in a common-emitter configuration with no feedback will have a low-frequency (e.g. 60kHz) input resistance around 50kohms, but the optimal source resistance from a noise standpoint--the source resistance which will yield the lowest noise figure for the amplifier--will be about 2kohms. At optimal source resistance, you can get a noise figure well below 1dB from an MMBT2222--and from many other bipolars. Is the MMBT2222 the same as a 2N2222, which I already have in my junk box? Also have 3904's, maybe they are just as suitable? Should I tune the output, or rely on a modest input tuned filter in the front end and just do a no tune in and out common emitter? Assuming a 50 ohm receiver input impedance, is it ok to take the output across a 50 ohm emitter resistor? One reason that people like to use FET amplifiers across their "parallel-tuned" loops is that the amplifier input resistance is quite high, but (using appropriate FETs) the noise contribution of the amplifier is negligible. And with proper design, the distortion contribution can be considerably lower than the distortion of your detector. For high source impedances, JFETs can give noise figures that are a small fraction of a dB. JFET's sound good too. Richard, which type of buffer amp has a lower distortion and better linearity in the presence of (potentially) large out of band signals? If I do use a buffer to preserve the Q, it has to be 'clean' in order to preserve the performance of the QSD receiver. Appreciate any enlightenment along these lines. I'm not opposed to using an op amp IF it provides cleaner output signals and if the power consumption is manageable. I must say it seems much cheaper (and probably more practical) to use a buffer to preserve the existing Q than it is to build higher Q into the loop...only to have it cut in half by the addition of a receiver front end. Regards, T |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
On Fri, 28 Oct 2005 20:12:48 GMT, Owen Duffy wrote:
Didn't Don Stoner describe a synchronous detector way back. I think I remember reading an article in the mid sixties in "The Sideband Handbook" or similar. I was about 15 then, so a detector that had something like 17 bottles in it seemed overkill when I was copying CW and SSB on an AM receiver (ie diode detector) with BFO. The appeal being an all-mode detector (including DSBSC), but synchrounous detectors didn't seem to catch on in comms receivers, well not until DSP detection... well I don't recall coming across them anyway. Hi Owen, 17 bottles indeed. That seems to strike a resonant chord in the ganglia because my construction was on a utility box of about 3" x 9" x 15" (not counting power supply requirements). We were working from a printed article certainly; and to confirm your recollection, there was a list of modes that could be detected that was long. My perception of the resurgence of interest in synchronous detection (it seems to have many names) is that a considerable body of knowledge evaporated in the 70s and 80s to leave only fragments of what this detector was useful at. 73's Richard Clark, KB7QHC |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
On Fri, 28 Oct 2005 13:35:19 -0700, Richard Clark
wrote: The rejection is psychological, not actual. It is what I meant by "mind-space." The vectors do not add up to zero, the mind simply ignores the off-band content like you would at a party listening to that cute office temp's whispers when your wife is yelling across the room at you. Listen to a recording of that same scenario in mono and you WILL hear your wife! I understand the Bell Labs explored this effect (which they referred to as the "cocktail party effect") when exploring the nature of conversation for the purposes of novel approaches to telephony multiplexing. I don't think they developed a technology solution to exploit the cocktail party effect, but they did incorporate their knowledge of the statistical / syllabic / sentence characteristics in their Time Assignment Speech Interpolation (TASI) equipments, and TASI was quite successful. I think the term we would use for the cocktail party effect on a phone channel is "a crossed line", and you may be right in that the loss of spatial information because of the mono channel may have been the reason it didn't work. -- |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
On Fri, 28 Oct 2005 21:10:04 GMT, Owen Duffy wrote:
I don't think they developed a technology solution to exploit the cocktail party effect, but they did incorporate their knowledge of the statistical / syllabic / sentence characteristics in their Time Assignment Speech Interpolation (TASI) equipments, and TASI was quite successful. Hi Owen, This mimics Paul McCartney's use of phase mixing in his music in the late 70s. Earlier, I was using Reticon bucket-brigade chips to develop a frequency independent delay line such that I could mix input and output (much like the FIR/IIR topology of today's DSP technology, except I was doing it in analog rather than digital) to obtain a variable phase. I still have the breadbox sitting on the shelf. I was anticipating using this device to place "voices" in the stereo-space of program content that I was mixing for. The concept was much too cerebral for the producer who simply wanted matched levels. 73's Richard Clark, KB7QHC |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
I believe my receiver is microvolt sensitive and that the loop will deliver a relatively good signal to the receiver even though the loop isn't terribly efficient. If I build selectivity into the front end of the receiver, do I really need high Q (200)?? I think the answer is NO..... Well, this is a good opportunity to examine that tumble down the slope to the Q = 2 (caused by the severe loading of your proposed design). The correlative to this is, how much selectivity do you need in a field where stateside VLF is relatively rare? Further, by the action of the strong filtering that usual attends the "I" and "Q" channel processing, you could easily repair any shortfall. However, back to that Q = 2. That still offers respectable (not fantastic) selectivity against signals out at the bottom of the AM band which is 10f (one decade) away. OK, I'm not sure how we got back to the Q=2 scenario. I think my proposed design is the old series tuned loop, which I have firmly rejected. With no tuning in the front end, the Q of the receiver would approach 1.......Yes, I understand that. I understand the answer you gave initially, which was 'I think the answer is NO....'. I understand the degree of protection against signals in the bottom of the bc band. When you said "Well, this is a good opportunity to examine that tumble down the slope to the Q = 2 (caused by the severe loading of your proposed design)"...................did you mean to say or to infer '(caused by the severe loading of your old (now defunct) series resonant loop design"??? All is in agreement above EXCEPT in not sure why the reference to the old design. Thanks again. T |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
On Fri, 28 Oct 2005 18:33:23 -0400, TRABEM wrote:
When you said "Well, this is a good opportunity to examine that tumble down the slope to the Q = 2 (caused by the severe loading of your proposed design)"...................did you mean to say or to infer '(caused by the severe loading of your old (now defunct) series resonant loop design"??? Yes of course. You have asked a number of questions outside of this old (now defunct) design, but you haven't, as far as I can tell, formalized a replacement. The ancillary point that I've made is that the original could work. However, you've never stated any operating specification to test that against. I've offered that all components need to be scrutinized in the face of your goal. We saw where that lead. You've only specified your desire for High Q capacitors (properly, low D capacitors). I offered that ESRs vary widely and could easily derail your goal. The presence of an ESR equal to the 0.06 Ohm of the loop is very well within being guaranteed. It still is. Reg dismissed this as inconsequential. So be it, but being that it is easily remedied through selection, then why toss away half your Q to casual indifference? What Reg actually meant, and he has a hard time with that given he can often be found on both sides of an argument, is that such loss may not matter. There I agree, but this does not advance the topic of High Q Caps for VLF Loop Antenna when they can be obtained. 73's Richard Clark, KB7QHC |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
On Fri, 28 Oct 2005 17:03:55 -0700, Richard Clark
wrote: On Fri, 28 Oct 2005 18:33:23 -0400, TRABEM wrote: When you said "Well, this is a good opportunity to examine that tumble down the slope to the Q = 2 (caused by the severe loading of your proposed design)"...................did you mean to say or to infer '(caused by the severe loading of your old (now defunct) series resonant loop design"??? Yes of course. Agreed! Just wanted to be sure I didn't miss something, so I asked. And, your're right, I haven't formulated a replacement. I bought a 250 foot poll of cable, and it has not been cut. So, it's still in one piece and returnable if I decide not to use it. I was just pondering the alternative of allowing the front end to be untuned. The receiver is susceptible to harmonics, each harmonic of the tuned frequency is down 6 db though. Since the loop would be resonant somewhere on HF, it is probably a bad idea to leave the front end untuned as HF can be unpredictable. so, I am thinking I need some front end selectivity. Reg gave me an example of what he might do. And his antenna came out much cheaper to build and probably easier to put up. I'm also thinking about the method of feeding the signal to the house. It will be around 70 feet from the house, so 90 feet of cable of some sort is needed. I can easily go 700 feet in any of 3 directions, but there is probably no practical need to go that far out into the woods. Since the house is a noisy place for LF and VLF, I have to be concerned about how I feed the antenna. I also think I'd like to have it fed with balanced line to minimize the possibility of the feed line acting as an antenna. If a preamp is used, I have to feed power to the antenna as well, so I will have to wind a common mode filter to do that job as well. So, I got a lot to think about. Although I haven't formulated a plan for a replacement antenna, the series loop is 99.99 percent history. So, I am thinking about it. I have a reading session planned for the late night here so I can reinforce the lesson(s) you and Reg have taught me. And, will probably work on the actual antenna design tomorrow. My caps are on order from Mouser, should be here next week There are a few preamp designs around the web, but none of them seems very well thought out...although they might be well planned. It's possible they are solidly designed, but that the authors haven't shared all the gory details in their web presentation(s). Thanks again. I'll keep you posted if you like......I'd appreciate sending up a red flag if I attempt to commit additional acts of stupidity with regard to whatever I come up with for a design. Regards, T |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
Hi Jim,
I ran some estimates earlier of a much smaller multiturn look with smaller wire. The Q held up much better than I expected and I'm also considering smaller diameter multiturn loops. Appreciate the quotes. Regards, T wrote: To emphasize one idea that Richard Clark has presented: Here is a direct quote from page 6-1 (section 6.2) of the chapter on loop antennas in the (first edition) Antenna Engineering Handbook edited by Jasik (1961): "The radiation pattern of a small loop is identical with that of a small dipole oriented normal to the plane of the loop with the E and H fields interchanged." The use of "magnetic" or other buzz words notwithstanding, a small loop is the dual of a small dipole. Practical considerations might cause one to select a small loop over a small dipole. 73 Mac N8TT |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
TRABEM wrote I ran some estimates earlier of a much smaller multiturn look with smaller wire. The Q held up much better than I expected and I'm also considering smaller diameter multiturn loops. ======================================== With multi-turn loops don't be tempted to neglect wire diameter. And it's well worth while spacing the wires by 1 or 2 wire diameters. Spaced wires reduce RF proximity-effect loss and increase Q. Neatness of construction is of little consequence. Wires need be supported only at the corners of the loop. Keep wires taut enough to prevent them flopping about in the wind. Electrically isolate the main loop. Don't connect it to anything. It will then maintain its good directional properties including its sharp null. Don't even think about screening or shielding the loop. It will serve no useful electrical purpose. It will cause proximity effect to reappear. Connect the smaller coupling loop to the unbalanced receiver via a thin twisted pair of whatever length you like without fear of it picking up any signal of consequence. Remember sensitivity depends ONLY on the enclosed loop area and NOT on the number of turns. So don't be tempted to reduce loop size just because it is multi-turned. Big signals never did anybody any harm. Whereas . . . . . ! I assume your QTH is not adjacent to WWV's antenna. ;o) ---- Reg, G4FGQ. |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
Hi REg,
Looks like I have a plan for a new loop. New loop design: I have a plan for a more appropriately dimensioned loop. At 60 KHz, 4 turns of 2 mm diameter wire, spaced 4 wire diameters apart. 2 meters per side 123 uH, 60,000 pF to resonate 4.7K across the loop. 300 ohm feed impedance at single turn loop feed Q (unloaded) = 101 This allows me to feed the loop with 300 ohm balanced line, which I can easily transform to 50 ohms at the receiver. I'm not sure what the impedance of twisted wire is though, which would be even cheaper than 300 ohm twin lead. Also, my Q will be slightly higher as I can stagger the turns some, so that the wires won't run parallel to each other for the entire length. I was never quite thrilled with a big loop threaded through the trees and supported in that manner, it makes it hard to rotate:: Being able to rotate the loop is a good thing:: Is the 1 turn pick up loop critical??? And, I have another question....... I used rj2loop3 and ran the same numbers as above, except that I separated the wires by 10 wire diameters instead of 4. Instead of seeing the Q improve, it was reduced (from 101) to 93. I expected the Q to improve, not get worse. It seems odd to me. Is there a good reason for this?? Ran it with a very low number and the Q also get worse. So there appears to be an 'optimal' wire winding pitch for optimizing Q? Byr the way, nice software package, thank you for it's use! I assume your QTH is not adjacent to WWV's antenna. ;o) Temporarily, it is on the East Coast of the US. But, at home it is high in the mountains in Northern EU (with no commercial power for miles around). Thanks, T |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
Trabem,
I will try to answer your remaining questions in the order at which they occur. There must be a compromise between size of loop, receiving sensitivity, and the ability to rotate it. Only you can decide. I can suggest only that you bias your opinion towards size matters more than the ability to rotate the thing. Just make sure that the broadside null does not correspond to the direction from which your favourite transmissions come from. The receiving lobes in the polar diagram are very broad. The polar diagram is a figure of 8, like a pair of touching circles. You won't need Eznec. The impedance of the line from the coupling loop to the receiver doesn't matter two hoots. At 60 KHz it is just a pair of wires. A twisted pair of wires has an impedance of very roughly 130 ohms. But at 60 KHz the line length is so short in terms of wavelengths it doesn't matter what its impedance is. The coupling loop can be considered to be directly connected to the 50-ohm receiver. And even with an extremely long line any impedance mismatch loss will be negligible. So forget about 300-ohm balanced line and just use a simple not-tightly-twisted pair. NO IMPEDANCE MATCHING REQUIRED at either end. The size of the small coupling loop inside the main loop matches 50 ohms to a 50-ohm receiver. So ideally the line to the receiver could be 50-ohm coax. But, as I say, it doesn't matter. The size and shape of the small coupling loop is not critical. It can be circular or square. Theoretically, to match the loop to a 50-ohm receiver, it should have an area about 1/25th of the main loop area. To simplify construction the coupling loop can be made self-supporting. Electrically, the thickness of the wire in the coupling loop need be no greater than the wire in the line which connects it to the receiver. The only wire diameter which matters is that of the main loop itself. As the spacing between wires on the main loop increases the RF proximity loss in the loop conductor (related to skin effect) decreases and Q increases. But other things happen when the width of the loop increases with spacing. For example, loop inductance decreases. We are not comparing like with like. And in any case maximisation of Q is not the primary objective. There are other things to be considered. For example, if you want to increase Q then don't bother to increase spacing between turns, just increase wire diameter. But with given wire diameter, the optimum spacing between the wire centres of adjacent turns, to maximise Q, is very crudely about twice the wire diameter. But, as I say, it is very non-critical and you might be better off by increasing wire diameter as it simplifies loop construction. Then, once again, you will have the option of increasing spacing between turns. Compromises are never ending. ;o) Whatever you end up with I can see from your enthusiasm you are enjoying your efforts and will continue to do so. ---- Reg, G4FGQ. ===================================== TRABEM wrote in message ... Hi REg, Looks like I have a plan for a new loop. New loop design: I have a plan for a more appropriately dimensioned loop. At 60 KHz, 4 turns of 2 mm diameter wire, spaced 4 wire diameters apart. 2 meters per side 123 uH, 60,000 pF to resonate 4.7K across the loop. 300 ohm feed impedance at single turn loop feed Q (unloaded) = 101 This allows me to feed the loop with 300 ohm balanced line, which I can easily transform to 50 ohms at the receiver. I'm not sure what the impedance of twisted wire is though, which would be even cheaper than 300 ohm twin lead. Also, my Q will be slightly higher as I can stagger the turns some, so that the wires won't run parallel to each other for the entire length. I was never quite thrilled with a big loop threaded through the trees and supported in that manner, it makes it hard to rotate:: Being able to rotate the loop is a good thing:: Is the 1 turn pick up loop critical??? And, I have another question....... I used rj2loop3 and ran the same numbers as above, except that I separated the wires by 10 wire diameters instead of 4. Instead of seeing the Q improve, it was reduced (from 101) to 93. I expected the Q to improve, not get worse. It seems odd to me. Is there a good reason for this?? Ran it with a very low number and the Q also get worse. So there appears to be an 'optimal' wire winding pitch for optimizing Q? Byr the way, nice software package, thank you for it's use! I assume your QTH is not adjacent to WWV's antenna. ;o) Temporarily, it is on the East Coast of the US. But, at home it is high in the mountains in Northern EU (with no commercial power for miles around). Thanks, |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
On Fri, 28 Oct 2005 13:45:17 -0700, Richard Clark
wrote: On Fri, 28 Oct 2005 20:12:48 GMT, Owen Duffy wrote: Didn't Don Stoner describe a synchronous detector way back. I think I remember reading an article in the mid sixties in "The Sideband Handbook" or similar. I was about 15 then, so a detector that had something like 17 bottles in it seemed overkill when I was copying CW and SSB on an AM receiver (ie diode detector) with BFO. The appeal being an all-mode detector (including DSBSC), but synchrounous detectors didn't seem to catch on in comms receivers, well not until DSP detection... well I don't recall coming across them anyway. Hi Owen, 17 bottles indeed. That seems to strike a resonant chord in the ganglia because my construction was on a utility box of about 3" x 9" x 15" (not counting power supply requirements). We were working from a printed article certainly; and to confirm your recollection, there was a list of modes that could be detected that was long. My perception of the resurgence of interest in synchronous detection (it seems to have many names) is that a considerable body of knowledge evaporated in the 70s and 80s to leave only fragments of what this detector was useful at. I think the appeal of it in the early days of suppressed carrier exploitation by amateurs lay in its application to DSBSC demodulation. It probably fell by the wayside when filter method SSB transceivers became lower in cost. Here is an interesting hypothetical. Australian amateurs at the unrestricted licence grade are subject to the following power restrictions: 16 Transmitter output power (1) Subject to section 15, the licensee must not operate an amateur unrestricted station, using a transmitter output power of more than 400 watts pX, if the emission mode of the station includes: (a) C3F; or (b) J3E; or (c) R3E. (2) The licensee must not operate an amateur unrestricted station, with an emission mode not mentioned in subsection (1), using a transmitter output power of more than 120 watts pY. Since the emission mode for DSBSC is not one of those mentioned in 16(1), then the power limit is 120W pY. If the pX/pY (PEP/Average) power ratio for radiotelephony is somewhere around 12dB to 15dB, that suggests that (using a worst case of 15dB) that the 120W pY DSBSC telephony transmitter is around 3800W pX (PEP), whereas you will note that if we use SSBSC (J3E) we are limited to 400W pX. Doesn't make sense, does it. Did I get the maths wrong? Owen -- |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
On Sat, 29 Oct 2005 21:36:07 GMT, Owen Duffy wrote:
Doesn't make sense, does it. Did I get the maths wrong? Hi Owen, Gad! I left all that computation behind when I finished the exam for FCC 1st Class. (Actually when I finished the Radar portion as I took the entire suite of tests, 3rd, 2nd, 1st, and Radar endorsement in one sitting.) However, I do recognize a strategy when I see it, and I doubt, beyond slipping a decimal the wrong direction, you may have found a loophole. 73's Richard Clark, KB7QHC |
HIGH Q CAPS FOR VLF LOOP ANTENNA?
OK Reg,
You are very patient and your time and expertise is appreciated. Last night I looked at many web articles on loops, many of which had preamps. I was a little discouraged. Many were fed with coax, which seems a little odd, especially since some of these loops had secondary resonances in the MF or HF bands. It seems like a preamp with a little non linearity, some big MF signals nearby, and a feed line that isn't balanced is a bit of a disaster waiting to happen. I know many of the HF radios in use have preamps that can be turned off (actually they are 20 db pads). But, regardless of my receivers susceptibility to out of band signals, I would want to feed my receiver with balanced line...especially since it's much cheaper than coax and should help prevent the feed line from picking up signals. So............followed everything in your last message and it all seems so simple (now).... Getting back to last nights study session. Spent a couple of hours in my 1987 ARRL Handbook and the remainder on the web looking at real life loops published there. The web aspect was really disappointingly devoid of technical jargon, it seems like most of the loop authors just threw something up and it seemed to work-the end:: Do people just throw stuff up without understanding what they're doing, or do they understand and just fail to document the theory?? The Handbook tour was almost as bad. Very little was said about loops except that which applied to the full wave resonant loop and how it can serve as a driven element in a 1 lambda 'guad' type radiator. Other than the theoretical wavelength, the correction factor for wire diameter, there was not more than 2 paragraphs written with useful information on short loops such as I am trying to put up. ------------------------- There was a question I had, perhaps I'm reading between the lines here tho.... YOU SAID: And even with an extremely long line any impedance mismatch loss will be negligible. So forget about 300-ohm balanced line and just use a simple not-tightly-twisted pair. NO IMPEDANCE MATCHING REQUIRED at either end. The size of the small coupling loop inside the main loop matches 50 ohms to a 50-ohm receiver. So ideally the line to the receiver could be 50-ohm coax. But, as I say, it doesn't matter. The size and shape of the small coupling loop is not critical. It can be circular or square. Theoretically, to match the loop to a 50-ohm receiver, it should have an area about 1/25th of the main loop area. To simplify construction the coupling loop can be made self-supporting. With your statement above in mind, are you telling me (or inferring) that any loop can be made to match a 50 ohm line by controlling it's size (area inside the main loop compared to area inside the coupling loop)?? In rjeloop3, the different sized loops have different feed line impedances, ranging from about 200 to 4K ohms. I made my final selection based on the model that matched my existing 300 ohm twin lead. I know you said my line impedance doesn't matter much, and I agree, except that it upsets the tuned front end in the receiver a bit. If it's possible to control this impedance of the small loop, what are the parameters I need to know to adjust mine to different values (other than those given by the software)?? I know you gave me the formula for matching to 50 ohms already. But, if this value can be controlled to customize the impedance, what do I need to know in order to make a match for other values? For my planned 2 meter square loop, my area is 2 X 2 X 4 (2 meters by 2 meters times 4 turns), sm my total loop area is 16 square meters. From the information you gave me (above), I can feed the receiver with 50 ohm feed line if my coupling loop is 16/25 (or about .8 square meters)? I know the matching isn't critical, but I'd like to have a good match for the sake of the front end tuned circuit in the receiver. If you can't say without writing a book, please feel free to decline to answer. If you can give the answer briefly, I'd appreciate some comments however. Hey Reg, was down in the basement where I keep the junk box and found a couple hundred feet of 4 conductor #24 solid copper cable. It's used for connecting telephones indoors. It's even cheaper than 300 ohm twin lead and I was wondering if I could/should use that as a feed line to the house?? Today I surveyed the loop location in the back yard, I have my solder and my 5/16" double weave rayon rope spool out and ready to go. My loop resonating capacitors should be here Wednesday. If all goes well, I could have a working installation by this time next week! Thing's are looking up. Also, Richard suggested a book about loops which I did not find in the library, or even in the reference section. I didn't price it yet, but I need to get some additional technical info. Do you have any suggestions for what to buy for books?? I don't need anything other than short loop theory (without heavy math). Thanks to you and Richard and everyone else who commented in this thread, I learned a great deal. Regards, T At 60 KHz, 4 turns of 2 mm diameter wire, spaced 4 wire diameters apart. 2 meters per side 123 uH, 60,000 pF to resonate 4.7K across the loop. 300 ohm feed impedance at single turn loop feed Q (unloaded) = 101 |
| All times are GMT +1. The time now is 10:22 PM. |
Powered by vBulletin® Copyright ©2000 - 2025, Jelsoft Enterprises Ltd.
RadioBanter.com