5 ms·

> But random wires end up with impedances that vary considerably from crazy high to crazy low across relatively small changes in frequency, and this in turn suc

by myth2018 5y ago

> But random wires end up with impedances that vary considerably from crazy high to crazy low across relatively small changes in frequency, and this in turn sucks. Could you send a link with a practical case? I'm not implying I don't believe you, it's just that I've never seen such complain. Personally, I don't use random wires. But I never see someone complaining like you about how wildly the characteristic impedance varies. Granted, it varies, but well within the range of most couplers. Besides, this is another source of myths: you often encounter folks excited about how a given coupler "tune everything" and "turn any piece of metal into an antenna", but they are never concerned about things like radiation patterns, radiation resistances, losses from the antenna, the cables, the coupler etc. They just see a SWR close to 1:1, celebrate and go to the internet complain about how empty are the bands. > And if your impedance is 15k ohms, you're going to need to reach a rather ridiculous voltage to get any power out. This is another case I haven't faced so far, neither myself or in the literature. 15k ohms? If the impedance really reached that high, maybe it's for the lack of a counterpoise, even if a small one? In a real-world end-fed, if you don't go high power (low power DX is perfectly feasible on days with good propagation and narrow bandwidth modes), you can keep the voltages at the feed point within the limits of commercial connectors, so that shouldn't be a problem. > Yes, which is fine as long as your signal comfortably fits in the bandwidth I can't remember of any mode demanding more than 30KHz (again, I'm not implying they don't exist). Sure, you have Amateur TV, but this mode is reserved for clubs and, if they are really meaning business that hard, they will certainly not use a magloop. Ah, I've also seen some guys transmitting signals to generate pictures at waterfalls of web SDRs. They were using a HackRF and probably a very high bandwidth signal (I'm not sure), but again, if you are getting into experimental modes, you are probably not going to use a small, portable antenna anyway

mlyle 5y ago
> Personally, I don't use random wires. For a long time I had a long random wire up and had a lot of success on it. But it would not tune on small segments of the bands because the impedance was very high or very low. And it was relatively deaf near those points, too. I shortened it slightly until those dead zones were in places I didn't care about. As the wire gets longer, those scary multiples of half wavelength become more common (though narrower). E.g. if I have a 150m electrical length antenna, around 17.987MHz, 18.986MHz, 19.986MHz, 20.985MHz, 21.984MHz, are really bad. This is if the wire is straight and far above the ground. If it also has random bends, you get even more places where funny things happen. And then, of course, really crazy things happen when you start trying to use this as a receive antenna up on UHF-- 399.7, 400.7, 401.7, 402.7, etc, not to mention the unpredictable, strong, and varying directionality. > This is another case I haven't faced so far, neither myself or in the literature. 15k ohms? If the impedance really reached that high, maybe it's for the lack of a counterpoise, even if a small one? https://vk6ysf.com/longwire_antenna.htm https://vk6ysf.com/longwire_antenna.htm " A multi-band end fed wire antenna presents a potential problem in that a ¼ wave length antenna cut for say the 80 metre band will present a feed impedance of about 70 ohms however when the same antenna is used on the 40 metre band the feed point impendence will be several thousand ohms and therefore difficult to achieve a match even with a quality tuner. The trick is to making this type antenna an easier match by avoiding the lengths that are 1/2 wavelength and harmonically related to 1/2 wavelengths or simply multiples of 1/2 wavelengths that will present the most extreme rang of impedances." 4NEC2 graphs are in the article. Admittedly, 15k ohms is an extreme case: at some point, earth losses, etc, prevent you from getting all the way to infinity, and ohmic losses of the wire and feedline stop you from getting all the way to 0 ohms. Of course, neither of these mechanisms makes you better at radiating power or receiving. > I can't remember of any mode demanding more than 30KHz (again, I'm not implying they don't exist). Lots of exotic, interesting modes, but also --- AM radio. 15KHz is in the human audible range (of younger people) and some AM stations carry audio this high. Occupied bandwidth of an AM radio station is 2 times highest audio frequency conveyed. Also, you probably are a few KHz off in tuning of your magloop, so you may have e.g. 20KHz on one sideband and 10KHz on the other within the bandwidth of the magloop. In turn you get the weird phasing artifacts and attenuation of higher frequencies that comes from receiving only 1.5 sidebands.
- myth2018 5y ago
  Thanks for the references. As I suspected. You take an antenna cut for a band, try to use it on others and find its limits -- nothing abnormal there. Still, maybe something you could circumvent with additional devices to transform the impedance? I can't check the article now, but maybe some changes to form a counterpoise, or feeding the antenna at a different point, could adapt it without big investments -- if you're really willing to spend resources on such improvisation. And about AM, it doesn't transmit the full human-audible spectrum, so the channel is not that wide. Currently standards result in ~20KHz IIRC. BTW, some amateurs use receiving tunnable loops to listen to distant AM stations -- the antenna's narrow bandwidth is actually an advantage in such cases. And, again, someone interested in transmitting in AM or using some of those exotic modes have no reason whatsoever to worry about loop's bandwidths -- just look for an appropriate antenna and stop worrying. It doesn't make sense to try to hammer a nail with a slipper and complain about its sole not being hard enough.
  - mlyle 5y ago
    > You take an antenna cut for a frequency, try to use it on others and find its limits The whole idea with a long, random wire is that it's not cut for a frequency, and happens to work well for most. This is about the most common case that exists, IMO :P There's nothing you can do for it to make it work well on those places where it's worst-- other than shorten or lengthen it. Yes, you can make it tune up by e.g. increasing ground losses, but that is not a win. (E.g. the T2FD does this deliberately, adding a shorting resistor to make the antenna tune anywhere). > Currently standards result in ~20KHz IIRC. AM is enormously variable depending upon band and operator. Yes, MW AM in the US is about 16KHz wide. Ecos del Torbes was frequently 35KHz wide, and sounded fantastic (better than broadcast FM) when it was strong and there weren't terrible atmospheric crashes. CRI is usually 30KHz wide. Witness the R390A, with a bandwidth KC (audio bandwidth) switch that goes to 16KHz. That's the -3dB point, too, and it rolls off relatively gently past where you have it switched. High fidelity AM was and is a thing. https://upload.wikimedia.org/wikipedia/commons/1/12/R390A.png https://upload.wikimedia.org/wikipedia/commons/1/12/R390A.pn... > It doesn't make sense to try to hammer a nail with a slipper and complain about its sole not being hard enough. No one has said that magnetic loops or long wire antennas are bad. I like them. I just pointed out the limitations and you disagreed.

mlyle 5y ago

> Personally, I don't use random wires. For a long time I had a long random wire up and had a lot of success on it. But it would not tune on small segments of the bands because the impedance was very high or very low. And it was relatively deaf near those points, too. I shortened it slightly until those dead zones were in places I didn't care about. As the wire gets longer, those scary multiples of half wavelength become more common (though narrower). E.g. if I have a 150m electrical length antenna, around 17.987MHz, 18.986MHz, 19.986MHz, 20.985MHz, 21.984MHz, are really bad. This is if the wire is straight and far above the ground. If it also has random bends, you get even more places where funny things happen. And then, of course, really crazy things happen when you start trying to use this as a receive antenna up on UHF-- 399.7, 400.7, 401.7, 402.7, etc, not to mention the unpredictable, strong, and varying directionality. > This is another case I haven't faced so far, neither myself or in the literature. 15k ohms? If the impedance really reached that high, maybe it's for the lack of a counterpoise, even if a small one? https://vk6ysf.com/longwire_antenna.htm https://vk6ysf.com/longwire_antenna.htm " A multi-band end fed wire antenna presents a potential problem in that a ¼ wave length antenna cut for say the 80 metre band will present a feed impedance of about 70 ohms however when the same antenna is used on the 40 metre band the feed point impendence will be several thousand ohms and therefore difficult to achieve a match even with a quality tuner. The trick is to making this type antenna an easier match by avoiding the lengths that are 1/2 wavelength and harmonically related to 1/2 wavelengths or simply multiples of 1/2 wavelengths that will present the most extreme rang of impedances." 4NEC2 graphs are in the article. Admittedly, 15k ohms is an extreme case: at some point, earth losses, etc, prevent you from getting all the way to infinity, and ohmic losses of the wire and feedline stop you from getting all the way to 0 ohms. Of course, neither of these mechanisms makes you better at radiating power or receiving. > I can't remember of any mode demanding more than 30KHz (again, I'm not implying they don't exist). Lots of exotic, interesting modes, but also --- AM radio. 15KHz is in the human audible range (of younger people) and some AM stations carry audio this high. Occupied bandwidth of an AM radio station is 2 times highest audio frequency conveyed. Also, you probably are a few KHz off in tuning of your magloop, so you may have e.g. 20KHz on one sideband and 10KHz on the other within the bandwidth of the magloop. In turn you get the weird phasing artifacts and attenuation of higher frequencies that comes from receiving only 1.5 sidebands.

myth2018 5y ago
Thanks for the references. As I suspected. You take an antenna cut for a band, try to use it on others and find its limits -- nothing abnormal there. Still, maybe something you could circumvent with additional devices to transform the impedance? I can't check the article now, but maybe some changes to form a counterpoise, or feeding the antenna at a different point, could adapt it without big investments -- if you're really willing to spend resources on such improvisation. And about AM, it doesn't transmit the full human-audible spectrum, so the channel is not that wide. Currently standards result in ~20KHz IIRC. BTW, some amateurs use receiving tunnable loops to listen to distant AM stations -- the antenna's narrow bandwidth is actually an advantage in such cases. And, again, someone interested in transmitting in AM or using some of those exotic modes have no reason whatsoever to worry about loop's bandwidths -- just look for an appropriate antenna and stop worrying. It doesn't make sense to try to hammer a nail with a slipper and complain about its sole not being hard enough.
- mlyle 5y ago
  > You take an antenna cut for a frequency, try to use it on others and find its limits The whole idea with a long, random wire is that it's not cut for a frequency, and happens to work well for most. This is about the most common case that exists, IMO :P There's nothing you can do for it to make it work well on those places where it's worst-- other than shorten or lengthen it. Yes, you can make it tune up by e.g. increasing ground losses, but that is not a win. (E.g. the T2FD does this deliberately, adding a shorting resistor to make the antenna tune anywhere). > Currently standards result in ~20KHz IIRC. AM is enormously variable depending upon band and operator. Yes, MW AM in the US is about 16KHz wide. Ecos del Torbes was frequently 35KHz wide, and sounded fantastic (better than broadcast FM) when it was strong and there weren't terrible atmospheric crashes. CRI is usually 30KHz wide. Witness the R390A, with a bandwidth KC (audio bandwidth) switch that goes to 16KHz. That's the -3dB point, too, and it rolls off relatively gently past where you have it switched. High fidelity AM was and is a thing. https://upload.wikimedia.org/wikipedia/commons/1/12/R390A.png https://upload.wikimedia.org/wikipedia/commons/1/12/R390A.pn... > It doesn't make sense to try to hammer a nail with a slipper and complain about its sole not being hard enough. No one has said that magnetic loops or long wire antennas are bad. I like them. I just pointed out the limitations and you disagreed.