John Larkin wrote:> On Sun, 25 Mar 2007 22:27:06 -0400, John Popelish <jpopelish@rica.net> > wrote: > >> John Larkin wrote: >> >>> In a superhet, high Q will make it that much harder to track the LO, >>> so you may well lose signal with a q=1000 rod. Why do you want a high >>> antenna Q? In the AM band, gain is cheap and s/n is dominated by >>> ambient noise, so it won't matter much. >> I think the main point to keep in mind is that it is easy to >> throw Q away, but hard to make it if the L and C don't have >> it, to start with. You might want to calculate the ideal Q, >> and then use an antenna coil construction technique that is >> pretty sure to exceed that requirement. > > At 800 KHz, q=1000, the resulting audio bandwidth will be 400 Hz!So what would be a reasonable Q for the tuned antenna? Something close to 100, I suspect.
Litz wire for AM ferrite Rod Antenna?
Started by ●March 24, 2007
Reply by ●March 25, 20072007-03-25
Reply by ●March 26, 20072007-03-26
On Sun, 25 Mar 2007 22:27:06 -0400, John Popelish <jpopelish@rica.net> wrote:>John Larkin wrote: > >> In a superhet, high Q will make it that much harder to track the LO, >> so you may well lose signal with a q=1000 rod. Why do you want a high >> antenna Q? In the AM band, gain is cheap and s/n is dominated by >> ambient noise, so it won't matter much. > >I think the main point to keep in mind is that it is easy to >throw Q away, but hard to make it if the L and C don't have >it, to start with. You might want to calculate the ideal Q, >and then use an antenna coil construction technique that is >pretty sure to exceed that requirement.At 800 KHz, q=1000, the resulting audio bandwidth will be 400 Hz! John
Reply by ●March 26, 20072007-03-26
On Sun, 25 Mar 2007 22:45:35 -0400, John Popelish <jpopelish@rica.net> wrote:>John Larkin wrote: >> On Sun, 25 Mar 2007 22:27:06 -0400, John Popelish <jpopelish@rica.net> >> wrote: >> >>> John Larkin wrote: >>> >>>> In a superhet, high Q will make it that much harder to track the LO, >>>> so you may well lose signal with a q=1000 rod. Why do you want a high >>>> antenna Q? In the AM band, gain is cheap and s/n is dominated by >>>> ambient noise, so it won't matter much. >>> I think the main point to keep in mind is that it is easy to >>> throw Q away, but hard to make it if the L and C don't have >>> it, to start with. You might want to calculate the ideal Q, >>> and then use an antenna coil construction technique that is >>> pretty sure to exceed that requirement. >> >> At 800 KHz, q=1000, the resulting audio bandwidth will be 400 Hz! > >So what would be a reasonable Q for the tuned antenna? >Something close to 100, I suspect.Or maybe a bit less... 50? Again, for a superhet, the tracking problem isn't trivial. John
Reply by ●March 26, 20072007-03-26
On Mar 25, 5:35 pm, John Larkin <jjlar...@highNOTlandTHIStechnologyPART.com> wrote:> On 24 Mar 2007 18:53:56 -0700, "Bill Bowden" <wrongaddr...@att.net> > wrote: > > > > >On Mar 24, 8:22 am, John Popelish <jpopel...@rica.net> wrote: > >> Bill Bowden wrote: > >> > How much improvement can I expect using Litz wire to wind a AM ferrite > >> > Rod antenna as opposed to using solid copper wire? > > >> > Is it worth the trouble to obtain Litz wire, or can I expect almost > >> > the same response at say 1 MHz using regular solid enamaled copper > >> > wire? > > >> If you would like to see some comparative experimental data, > >> Ben Tongue has performed some experiments and posted the > >> data to his web site.http://www.bentongue.com/xtalset/29MxQFL/29MxQFL.html > > >Thanks John, > > >Yes, it looks like Litz wire has a significant advantage. If I read > >the results right, the unloaded Q factor at 943Khz is 141 using solid > >copper wire verses a Q factor of 1030 using Litz wire. > >That's quite a significant difference. Am I reading the results right? > > In a superhet, high Q will make it that much harder to track the LO, > so you may well lose signal with a q=1000 rod. Why do you want a high > antenna Q? In the AM band, gain is cheap and s/n is dominated by > ambient noise, so it won't matter much. > > JohnIt's just a little portable AM radio I've been wanting to build for years. I took a radio class in 7th grade 50 years ago and never got around to finishing the superhet design. But I got an A anyway. We used tubes in those days. I'm using the Signetics NE602 balanced modulator IC that produces about 13dB gain. The antenna rod is buffered with a JFET so there is minimal load on the antenna rod yielding another 12 to 18 dB. The front end is pretty hot. But as you say, the bandwidth is narrow with a high Q coil, so I'm using a switch to short a couple turns on the antenna rod to load the antenna for local strong stations. Local/DX select. The biggest problem is separating a distant station 40KHz away from a strong local 50KW station 5 miles up the road. -Bill
Reply by ●March 26, 20072007-03-26
On Sun, 25 Mar 2007 17:42:10 -0500, "amdx" <amdx@knology.net> Gave us:> >"MassiveProng" <MassiveProng@thebarattheendoftheuniverse.org> wrote in >message news:il9d03tohd4u68ili8u6oa7on5q69i6lhu@4ax.com... >> On 24 Mar 2007 18:53:56 -0700, "Bill Bowden" <wrongaddress@att.net> >> Gave us: >> >>> The proximity >>>effect, but not the skin effect loss may be much reduced if the wires >>>are space-wound. >> >> Use insulated wire, and another good sub for litz is SPC (silver >> plated copper), as you get a slightly better skin, and the insulation >> gives the space winding. >> >>A twisted group of smaller SPC wires can >> give a slight Litz effect as well, like 7 32 ga SPC wires in teflon, >> or other sheathing twisted together evenly. Not true litz, but better >> than a single conductor. Particularly if the space winding effect are >> the main desire. > > I'd be interested in seeing the results of that experiment. Ben's best >Q is 431 using a single #31 wire. The results shown in Table 7 suggest >that, getting the wires close to each other reduces Q. Twisting 7-#32 >wires (with teflon) together and winding with that bundle would probably >end up with no space between turns. > Mike > >Of course there would be space between the turns. The insulation alone edicts it. Can you really be that thick? The solid wire nests right next to itself from turn to turn because it is mag wire, dipshit.> > > > >Let the name calling begin, but try to use something new.Like your 8 blank line dumbshit, followed by yet another retarded remark after your reply?
Reply by ●March 26, 20072007-03-26
"MassiveProng" <MassiveProng@thebarattheendoftheuniverse.org> wrote in message news:mpse03lnu0i71ehco4soelsh44a7fa13q7@4ax.com...> On Sun, 25 Mar 2007 17:42:10 -0500, "amdx" <amdx@knology.net> Gave us: > >> >>"MassiveProng" <MassiveProng@thebarattheendoftheuniverse.org> wrote in >>message news:il9d03tohd4u68ili8u6oa7on5q69i6lhu@4ax.com... >>> On 24 Mar 2007 18:53:56 -0700, "Bill Bowden" <wrongaddress@att.net> >>> Gave us: >>> >>>> The proximity >>>>effect, but not the skin effect loss may be much reduced if the wires >>>>are space-wound. >>> >>> Use insulated wire, and another good sub for litz is SPC (silver >>> plated copper), as you get a slightly better skin, and the insulation >>> gives the space winding. >>> >>>A twisted group of smaller SPC wires can >>> give a slight Litz effect as well, like 7 32 ga SPC wires in teflon, >>> or other sheathing twisted together evenly. Not true litz, but better >>> than a single conductor. Particularly if the space winding effect are >>> the main desire. >> >> I'd be interested in seeing the results of that experiment. Ben's best >>Q is 431 using a single #31 wire. The results shown in Table 7 suggest >>that, getting the wires close to each other reduces Q. Twisting 7-#32 >>wires (with teflon) together and winding with that bundle would probably >>end up with no space between turns. >> Mike >> >> > Of course there would be space between the turns. The insulation > alone edicts it. Can you really be that thick? The solid wire nests > right next to itself from turn to turn because it is mag wire, > dipshit. >> >>Oh, I thought you might have read the article. The length of the windings of all coils in Table 7 is 1.624". His High Q coil has about .019" between each turn. I doubt you could even duplicate his experiment with your suggested wire. But as I said, "I'd be interested in seeing the results of that experiment.">> >> >>Let the name calling begin, but try to use something new. > > Like your 8 blank line dumbshit, followed by yet another retarded > remark after your reply?Come on Massive, dipshit, dumbshit and retarded, is that the best you can do? Entertain us with something original. You use the same old tired names over and over. They've lost their oomph, their punch, their pizzazz. Space intentionally left blank for new improved original name calling. Mike
Reply by ●March 26, 20072007-03-26
Bill Bowden wrote:> On Mar 25, 5:35 pm, John Larkin>> In a superhet, high Q will make it that much harder to track the LO, >> so you may well lose signal with a q=1000 rod. Why do you want a high >> antenna Q? In the AM band, gain is cheap and s/n is dominated by >> ambient noise, so it won't matter much. >> >> John > > It's just a little portable AM radio I've been wanting to build for > years. I took a radio class in 7th grade 50 years ago and never got > around to finishing the superhet design. But I got an A anyway. We > used tubes in those days. > > I'm using the Signetics NE602 balanced modulator IC that produces > about 13dB gain. The antenna rod is buffered with a JFET so there is > minimal load on the antenna rod yielding another 12 to 18 dB. The > front end is pretty hot.I take it this is connected to pins 1 and 2? How did you do your oscillator on 6 and 7?> But as you say, the bandwidth is narrow with a high Q coil, so I'm > using a switch to short a couple turns on the antenna rod to load the > antenna for local strong stations. Local/DX select. > > The biggest problem is separating a distant station 40KHz away from a > strong local 50KW station 5 miles up the road.What are you using as an IF filter and AM detector? I've been tinkering around trying to build a 10MHz WWV receiver and I have an NE602 that I'm injecting my "PIC locked" 9.545MHz LO into (pin 6) and using a 10MHz tank circuit ( 2.8uH and ~100pF) as a preselector with the antenna coupled to the coil. I only have that much done so far, but it's time to do some filtering and detection and I'm not sure what direction to take on that. Just playing around, nothing critical. I always wanted to build a crystal set using an oatmeal box sized coil form with a sliding tap. Maybe when my daughter gets a little older she'll wind it for me. ;-)
Reply by ●March 26, 20072007-03-26
On Sun, 25 Mar 2007 22:40:34 -0700, Bill Bowden wrote:> On Mar 25, 5:35 pm, John Larkin >> On 24 Mar 2007 18:53:56 -0700, "Bill Bowden" <wrongaddr...@att.net> >> >On Mar 24, 8:22 am, John Popelish <jpopel...@rica.net> wrote: >> >> Bill Bowden wrote: >> >> > How much improvement can I expect using Litz wire to wind a AM ferrite >> >> > Rod antenna as opposed to using solid copper wire? >> >> >> > Is it worth the trouble to obtain Litz wire, or can I expect almost >> >> > the same response at say 1 MHz using regular solid enamaled copper >> >> > wire? >> >> >> If you would like to see some comparative experimental data, >> >> Ben Tongue has performed some experiments and posted the >> >> data to his web site.http://www.bentongue.com/xtalset/29MxQFL/29MxQFL.html >> >> >Yes, it looks like Litz wire has a significant advantage. If I read >> >the results right, the unloaded Q factor at 943Khz is 141 using solid >> >copper wire verses a Q factor of 1030 using Litz wire. >> >That's quite a significant difference. Am I reading the results right? >> >> In a superhet, high Q will make it that much harder to track the LO, >> so you may well lose signal with a q=1000 rod. Why do you want a high >> antenna Q? In the AM band, gain is cheap and s/n is dominated by >> ambient noise, so it won't matter much. > > It's just a little portable AM radio I've been wanting to build for > years. I took a radio class in 7th grade 50 years ago and never got > around to finishing the superhet design. But I got an A anyway. We > used tubes in those days. > > I'm using the Signetics NE602 balanced modulator IC that produces > about 13dB gain. The antenna rod is buffered with a JFET so there is > minimal load on the antenna rod yielding another 12 to 18 dB. The > front end is pretty hot. > > But as you say, the bandwidth is narrow with a high Q coil, so I'm > using a switch to short a couple turns on the antenna rod to load the > antenna for local strong stations. Local/DX select. > > The biggest problem is separating a distant station 40KHz away from a > strong local 50KW station 5 miles up the road.I've read somewhere that the idea is to build the selectivity into the IF part. Have Fun! Rich
Reply by ●March 26, 20072007-03-26
On Mon, 26 Mar 2007 22:27:55 GMT, Rich Grise <rich@example.net> wrote:>On Sun, 25 Mar 2007 22:40:34 -0700, Bill Bowden wrote: >> On Mar 25, 5:35 pm, John Larkin >>> On 24 Mar 2007 18:53:56 -0700, "Bill Bowden" <wrongaddr...@att.net> >>> >On Mar 24, 8:22 am, John Popelish <jpopel...@rica.net> wrote: >>> >> Bill Bowden wrote: >>> >> > How much improvement can I expect using Litz wire to wind a AM ferrite >>> >> > Rod antenna as opposed to using solid copper wire? >>> >>> >> > Is it worth the trouble to obtain Litz wire, or can I expect almost >>> >> > the same response at say 1 MHz using regular solid enamaled copper >>> >> > wire? >>> >>> >> If you would like to see some comparative experimental data, >>> >> Ben Tongue has performed some experiments and posted the >>> >> data to his web site.http://www.bentongue.com/xtalset/29MxQFL/29MxQFL.html >>> >>> >Yes, it looks like Litz wire has a significant advantage. If I read >>> >the results right, the unloaded Q factor at 943Khz is 141 using solid >>> >copper wire verses a Q factor of 1030 using Litz wire. >>> >That's quite a significant difference. Am I reading the results right? >>> >>> In a superhet, high Q will make it that much harder to track the LO, >>> so you may well lose signal with a q=1000 rod. Why do you want a high >>> antenna Q? In the AM band, gain is cheap and s/n is dominated by >>> ambient noise, so it won't matter much. >> >> It's just a little portable AM radio I've been wanting to build for >> years. I took a radio class in 7th grade 50 years ago and never got >> around to finishing the superhet design. But I got an A anyway. We >> used tubes in those days. >> >> I'm using the Signetics NE602 balanced modulator IC that produces >> about 13dB gain. The antenna rod is buffered with a JFET so there is >> minimal load on the antenna rod yielding another 12 to 18 dB. The >> front end is pretty hot. >> >> But as you say, the bandwidth is narrow with a high Q coil, so I'm >> using a switch to short a couple turns on the antenna rod to load the >> antenna for local strong stations. Local/DX select. >> >> The biggest problem is separating a distant station 40KHz away from a >> strong local 50KW station 5 miles up the road. > >I've read somewhere that the idea is to build the selectivity into the >IF part. >Yes, but you have to be careful that nothing overloads in the front-end, or in the first IF stage, from that 50 KW monster. A decent antanna Q helps some. Fets are really good for avoiding nonlinearity. So, low-gain jfets or mosfets in the front end and maybe the first IF, and pile up selectivity and gain in later IF stages. Manual stage gain pots, rather than AGC, would be fun. John
Reply by ●March 27, 20072007-03-27
On Mar 26, 7:07 am, "Anthony Fremont" <spam-...@nowhere.com> wrote:> Bill Bowden wrote: > > On Mar 25, 5:35 pm, John Larkin > >> In a superhet, high Q will make it that much harder to track the LO, > >> so you may well lose signal with a q=1000 rod. Why do you want a high > >> antenna Q? In the AM band, gain is cheap and s/n is dominated by > >> ambient noise, so it won't matter much. > > >> John > > > It's just a little portable AM radio I've been wanting to build for > > years. I took a radio class in 7th grade 50 years ago and never got > > around to finishing the superhet design. But I got an A anyway. We > > used tubes in those days. > > > I'm using the Signetics NE602 balanced modulator IC that produces > > about 13dB gain. The antenna rod is buffered with a JFET so there is > > minimal load on the antenna rod yielding another 12 to 18 dB. The > > front end is pretty hot. > > I take it this is connected to pins 1 and 2? How did you do your oscillator > on 6 and 7? >The oscillator is the Hartly version with a tapped coil. Pin 7 goes to the tap through a 0.1uF cap. Pin 6 goes to the high side of the coil through another 0.1uF cap. Low side of coil goes to ground and tuning cap goes across the coil. But, I had to rewind the oscillator coil because the tap on the regular (red slug) coils is too close to one end for good oscillation. I put the tap about 1/3 the way up and used a resistor (1K) in series to get it about right.> > But as you say, the bandwidth is narrow with a high Q coil, so I'm > > using a switch to short a couple turns on the antenna rod to load the > > antenna for local strong stations. Local/DX select. > > > The biggest problem is separating a distant station 40KHz away from a > > strong local 50KW station 5 miles up the road. > > What are you using as an IF filter and AM detector?I have 2 IF stages planned, but just using one right now. It just uses the normal (black slug) coil with a 2K secondary to drive the detector. The detector is a Schottky diode that feeds the gate of a JFET. There is a 2 Meg resistor from gate to +V to bias the diode at a few microamps and a 1000pF cap from gate to ground to filter out the RF. Audio comes off the source of the JFET. I'm not sure exactly how this works, but it does the best job of all the other configurations I tried. The audio is good from a weak station. I have an article about using a zero bias JFET so that no diode is needed, but I haven't tried that one.> I've been tinkering > around trying to build a 10MHz WWV receiver and I have an NE602 that I'm > injecting my "PIC locked" 9.545MHz LO into (pin 6) and using a 10MHz tank > circuit ( 2.8uH and ~100pF) as a preselector with the antenna coupled to the > coil. I only have that much done so far, but it's time to do some filtering > and detection and I'm not sure what direction to take on that. Just playing > around, nothing critical. > > I always wanted to build a crystal set using an oatmeal box sized coil form > with a sliding tap. Maybe when my daughter gets a little older she'll wind > it for me. ;-)-Bill
