Keyboard Shortcuts
ctrl + shift + ? :
Show all keyboard shortcuts
ctrl + g :
Navigate to a group
ctrl + shift + f :
Find
ctrl + / :
Quick actions
esc to dismiss
Likes
Search
qdx l12 inductance
|
Hi All, as I did some mistake while?doing?the trifilar, I wasted some of the supplied 28 awg wire. So ended up with not enough wire to wound the L12 and used some other ~0.35mm wire laying around. Now when I do a RF sweep on terminal I see a very wide almost non existing filter on 20m and the maximum attenuation point exactly on the 80m operating frequency. Both bands will be useless I'm afraid. 30-40m sort of ok. Squeezing or widening the spacing on it does not do much of a difference. I have some other wires on the way (could not find 0.33-0.32 but I ordered some? 0.30 and 0.35. Probably I will rewound the L12. To be able to do right this time, I need the correct inductance data for the start-end and tap points? so I can use a LC meter. Br. |
|
Sorry if I'm not understanding, but as long as you carefully count the turns, the inductance will be correct.? You're right, the spacing of the turns really don't change inductance, it's purely the number of times the wire goes through the hole.? (The wire gauge won't change the inductance either.)
Dennis |
|
On Fri, Nov 5, 2021, at 09:50 AM, Dennis Atwood wrote:
Sorry if I'm not understanding, but as long as you carefully count the turns, the inductance will be correct.? You're right, the spacing of the turns really don't change inductance, it's purely the number of times the wire goes through the hole.? (The wire gauge won't change the inductance either.)Dennis, You are correct in the points that wire size does not impact inductance and that the number of turns is the critical one.? That is assuming that you are not overlapping the turns.? As Hans has pointed out in tweaking the QCX series of low pass filters, spacing on the toroid DOES impact the inductance of the inductor, though not as much as adding or subtracting a turn.? It is also in the QDX manual to tweak the bandpass filter on the input of the receiver (page 89 of the manual). Suat, You can get the normal inductance value from this web page: You enter in the number of turns and then press calculate to get 8.24 ?H for the full 41 turns.? Press clear and then enter in the other number of turns to get the tap point inductance from the start. On Fri, Nov 5, 2021, at 07:39 AM, Suat Alper Seyhan wrote: Now when I do a RF sweep on terminal I see a very wide almost non existing filter on 20m and the maximum attenuation point exactly on the 80m operating frequency. Both bands will be useless I'm afraid. 30-40m sort of ok. Squeezing or widening the spacing on it does not do much of a difference. I have some other wires on the way (could not find 0.33-0.32 but I ordered some? 0.30 and 0.35. Probably I will rewound the L12.Would it be possible to post screenshots of your sweeps?? Based on the problems being 20 and 80 meters I suspect that the larger number of turns (the first 19) are off.? I would recount that section of L12.? This is just a guess on my part, using the circuit to figure out which taps correlate to which bands, also a guess as I do not have access to the program. Also, a high-resolution picture of your L12 would help. 73 Evan AC9TU |
|
Also, there is always the possibility of accidentally shorting turns together. 73 de Lee KX4TT On Friday, 5 November 2021, 12:28:54 pm GMT-4, Evan Hand <elhandjr@...> wrote: On Fri, Nov 5, 2021, at 09:50 AM, Dennis Atwood wrote: Sorry if I'm not understanding, but as long as you carefully count the turns, the inductance will be correct.? You're right, the spacing of the turns really don't change inductance, it's purely the number of times the wire goes through the hole.? (The wire gauge won't change the inductance either.)Dennis, You are correct in the points that wire size does not impact inductance and that the number of turns is the critical one.? That is assuming that you are not overlapping the turns.? As Hans has pointed out in tweaking the QCX series of low pass filters, spacing on the toroid DOES impact the inductance of the inductor, though not as much as adding or subtracting a turn.? It is also in the QDX manual to tweak the bandpass filter on the input of the receiver (page 89 of the manual). Suat, You can get the normal inductance value from this web page: You enter in the number of turns and then press calculate to get 8.24 ?H for the full 41 turns.? Press clear and then enter in the other number of turns to get the tap point inductance from the start. On Fri, Nov 5, 2021, at 07:39 AM, Suat Alper Seyhan wrote: Now when I do a RF sweep on terminal I see a very wide almost non existing filter on 20m and the maximum attenuation point exactly on the 80m operating frequency. Both bands will be useless I'm afraid. 30-40m sort of ok. Squeezing or widening the spacing on it does not do much of a difference. I have some other wires on the way (could not find 0.33-0.32 but I ordered some? 0.30 and 0.35. Probably I will rewound the L12.Would it be possible to post screenshots of your sweeps?? Based on the problems being 20 and 80 meters I suspect that the larger number of turns (the first 19) are off.? I would recount that section of L12.? This is just a guess on my part, using the circuit to figure out which taps correlate to which bands, also a guess as I do not have access to the program. Also, a high-resolution picture of your L12 would help. 73 Evan AC9TU |
¿ªÔÆÌåÓýHi All ???? ????Using the calculator you can also input value of series capacitor (assuming 80M=220pf,20M=22pf ) to show resonant frequency ,80m =3.73Mhz 20M=25.5 Mhz ? requires 72pf to resonate at 14.074 Mhz. ???????? Regards ????????????????????? Mike? G6GN Sent from for Windows ? From: Lee via groups.io
Sent: 05 November 2021 16:49 To: [email protected]; [email protected] Subject: Re: [QRPLabs] qdx l12 inductance ? Also, there is always the possibility of accidentally shorting turns together. On Friday, 5 November 2021, 12:28:54 pm GMT-4, Evan Hand <elhandjr@...> wrote: ? ? On Fri, Nov 5, 2021, at 09:50 AM, Dennis Atwood wrote:
Dennis,
Would it be possible to post screenshots of your sweeps?? Based on the problems being 20 and 80 meters I suspect that the larger number of turns (the first 19) are off.? I would recount that section of L12.? This is just a guess on my part, using the circuit to figure out which taps correlate to which bands, also a guess as I do not have access to the program. ? |
|
Hi Again,
First of all thanks to everyone who replied. When I was first doing the RF filter sweep qdx RF port was not connected to anything. When I connect it do a 50ohm dummy load and re-do the rf filter sweep, 80m? turned out quite fine. Please see the attachments. (80m sweep and 80 m no dummy files) But with dummy or no dummy 20m sweep still looks weird to me and I'm not hearing well on 20m. Only very strong close stations are being decoded (compared to another radio connected to same antenna differece on 20m is severe, on other bands ~same reception and decodes with the other radio)? Did a re-heat on all L12 pads. Nothing changed.? Also a hires pic of L12 attached. What might be wrong with the 20m bandpass filter? Br. |
|
I re-did the L12 quite carefully but pretty much nothing changed on the rf filter sweeps. Either I did the same mistake twice or there is something else going on. I was only looking to shape of the graph but when I now look in to numbers all the numbers are quite low compared to screenshots Hans shared. Instead of connecting a 50ohm dummy if I try it again connected to good/resonant antenna may I expect to see different results?
|
|
On 08/11/2021 18:36, Suat Alper Seyhan wrote:
Instead of connecting a 50ohm dummy if I try it again connected to good/resonant antenna may I expect to see different results?You will see different results but they will be wrong. You must use a dummy load, any signals or noise received will give false readings. 73 Alan G4ZFQ |
|
Yes, and the signal is injected through a 100k resistor (ok, well, it has a fraction of a pF parallel capacitance too ) so it will act as a "current source" when feeding the low-z antenna input. The injected voltage will be (almost) proportional to the impedance seen by the resistor and this can vary quite wildly if there is an antenna connected (instead of a known stable 50 ohm dummy load/termination).
toggle quoted message
Show quoted text
73 SM6LKM Alan G4ZFQ wrote: You will see different results but they will be wrong. |
|
Suat,
The RF sweep is measuring the total response that includes the transmitter/receiver LPF.? I would contend that most, if not all are the measurement is the response of the LPF.? The bandpass aspect of the L12/T2 combination is wide, so not sure of the impact to the RF sweep report.? IF you want to make a change to the frequency response you would need to redo L2, L3, L4, L6, L8, and L10 toroids, though by the looks of your screenshots they look fine to me.? The 20 meters curve is different because it is tweaked to take into account the use of it on 30 meters. Note: L12 and the trifler wound T2 are part of the Reciever circuit and not used during transmit.? L14 and T1 are used only during Transmit. and do not impact the receiver.? They also are not very frequency dependant.? The 3 low pass filters are used on both receive and transmit, one at a time through the PIN diode switching network. Above are my interpretations of the schematic and descriptions that I have access to.? They could be wrong and I am ALWAYS open to polite feedback and open discussion.? I am still learning. 73 Evan AC9TU |
|
Evan,?
Oh this is interesting.? You made me look closer at the schematic.? I don't think it's quite that the RF sweep has anything to do with the transmitter circuitry.? If you look, CLK2 from the si5351 goes right to the BNC connector, and is not used anywhere else.? This must be the rf generator.? It makes sense now that it goes through the LPFs so the square wave is filtered.? It must otherwise be in receive mode, though, because CLK0 and CLK1 do double-duty driving the push pull transmitter section, as well as providing clocks for the quadrature detectors.? It would need to tune along with the generated frequency in order to measure amplitude.? I wouldn't think the LPFs should attenuate the signal much in the sweep range.? They are for filtering out higher order harmonics so the rolloff doesn't happen until a bit higher.? But I'm looking at the response graphs on page 70 and 71.? 40m for example, it does look like the rolloff just starts around 8MHz.? It's close, so I suppose the LPFs could be having an affect here, especially if they aren't in spec. Still, I'm seeing in every case a rolloff on the RF sweep graph that starts before the LPFs are supposed to begin attenuating, which makes me think it's mostly the BPF.? And of course any attenuation on the lower frequencies is certainly the band pass filter. Does this make sense? Dennis |
|
Dennis,
I do not know how to figure out what to adjust based on the scan results.? I guess we will need a response from the designer as he should know. The scan is most definitely looking at the total filtering chain from the antenna, through the LPFs, through the bandpass filter, and to the detector.? ?I do not see a reason to scan the bandpass filter.? It is not really that critical to the overall operation.? Power out seems to be the most significant issue with QRP FT8 transmissions as I have always had a hard time working stations that I could hear but could not hear me.? Also, I am always more worried that I am operating legally rather than if my receiver is optimized for the band. Again, you could be correct.? I just wonder about the usefulness of a scan of the bandpass filter. 73 Evan AC9TU |
Still, I'm seeing in every case a rolloff on the RF sweep graph that starts before the LPFs are supposed to begin attenuating, which makes me think it's mostly the BPF.? And of course any attenuation on the lower frequencies is certainly the band pass filter.Dennis, Section 6.6 of the manual says it is for the BPF. It also shows a "bad" result that Hans dismisses. Obviously a low LPF will also affect it. A low LPF would show by low power output on an affected band. 73 Alan G4ZFQ |
|
Just out of curiosity and as I did not like the looks of it, I re-did the T2 trifiliar :) Numbers grew a little bit on the Y axis of the graph but almost no change at all on the filter shape I see on RF sweep. As a next step I will probably re-do the LPF's of 20/30m. For getting a decent output I had to squeze them a lot? anyway when I was doing the first tests. Which also tells me something maybe wrong with them. Also when I was winding them I measured them with my cheap Chinese Lcr meter and readings was a little bit off but I decided to stick to the number of windings in manual as I don't trust that lcr meter too much. As I'm pushing my luck and soldering skills a lot on this, I'll probably first wait for the weekend to at least enjoy a few more Qso's with QDX before I probably ruin some pads on the board while experimenting :) I deeply regret that I did not order two QDX's (one for using one for experimenting..)
|
Lee
to navigate to use esc to dismiss