"nearly 60 dB over so probably want at least 20 dB reduction ?"
No. You really only need about 3 dB to get out of OV. This is an SDR not an analog receiver, level doesn't matter much until you run over the top of the ADC and then it totally breaks. Well, there's a subtle lift in the noise floor from ADC non-linearity mainly when the top bit is toggling so maybe 9 dB, but there's little value in more than that and you can easily get that much out of the single LC already described.
lol I'll try not get carried away, looks simple enough, will take time to get all the parts.
How many turns is reasonable for one of those T50-2 ?
I've read that silver Mica are the best to use in these applications, gives higher Q but getting one with the right value vs reasonable turns on the T50-2 is a challenge.
See my previous post. FT50-61 and about .012 uF is one solution. Make the C from parallel .01 uF and a ~.002 uF trimmer. You can probably see where the center of the notch is by loking at noise floor. Tweak the .002 uF as necessary. Make it from multiple C's if helpful against your local stock. Don't worry about Q of the C, L will dominate and in any case you don't care all that much.
Looking at your KiWi, I think you only need about 10dB attenuation at 252kHz, so this should be easy to achieve, even if you decided to use a pre-wound axial epoxy coated inductor.
As the others have already said, don't overthink it. RF filters don't always match their theoretical characteristics due to stray capacitance and inductance, plus there is the self resonant frequency and Q factor to take into consideration too. The more sections / parts their are in a filter, the more difficult it becomes to align and get working properly.
Ok thanks for the advice lads, will hopefully get around to ordering parts at the weekend.
If I were to go with a premade inductor can anyone send me a link via this site as to which would be best for this please ? https://ie.rs-online.com/web/
There are probably the only suppliers locally, and at that they might not even have all the parts, it's difficult to get a lot of stuff here on this little Island, the T50-2 forget that, that will take a week or more to get over seas, this is why something on one of these sites might work just as well and I can get it a lot faster if available locally.
Farnel.ie has stock and you don't have to buy a minimum of 10 or 100
You may have to add a 10 ohm series resistor to reduce the theoretical notch depth from 25dB to somewhere around 10dB.
Note that component tolerances mean that you may have to 'fine tune' the capacitor value.
Alternatively don't add a resistor and the notch depth will still be greater than 10dB over a bandwidth of 40KHz around the centre frequency, so even allowing for some variation in component values, the chances are the standard values will be close enough.
Have to say I really am struggling to find suitable capacitor and inductor combination on these sites. These are local suppliers, anything else has to come from over seas with probably long wait times.
Finally got the parts, rough but hey it works ! there's a nice bit off attenuation on 252 Khz, roughly 10 dB reduction which is fine and it's now as strong as BBC Radio 4 which is sadly due to close I think next year as the BBC continue their assault on AM radio.
But I am very happy I finally got 252 Khz under control.
Not sure, all I hear and see on the waterfall is powerline noise which is quite mild now but it can be quite bad especially from around 9 Mhz and up. This 10 Kva power line runs over the bottom quarter of my garden.
I must try get on to the regulator to see if they can get the power company to fix it they have ignored me completely for 5 years but I'm afraid that if I get them out and the noise is not there they will just ignore me in future.
I have filed a complaint to COMREG in relation to the mad QRM I have between around 650 Khz and 10 Mhz particularly noticeable around 80m.
Had to give them "evidence" and evidence it's not coming from my QTH so I went out on the road yesterday with my portable Tecsun PL-680 and of course got some weird looks. The noise has got really bad, I traced it to an Electricity pole so hopefully it will get sorted but I won't hole my breath.
I have two sources of noise but concentrating on this first as it's by far the worst as it's more constant.
If the pole was wood.... while listening to your radio, rap the pole with a small sledge hammer and listen for a change in the noise. Even better on a rainy day!
I can only imagine what the people that saw me walking around with a portable radio while recording on my phone would think, can't imagine what they'd say if they saw me going around whacking electric poles with a sledge lol. It's a tiny community where I live, bad enough if they spotted me shaking the pole. 😁
Shaking the pole make no difference.
I guess I'll find out soon enough if anything is done about it, I gave Comreg samples of audio and screen recording of my Kiwi and gave them my Callsign + youtube videos where I'm walking around and the noise can be heard a little bit at my QTH and it's loudest at that pole.
Comments
"nearly 60 dB over so probably want at least 20 dB reduction ?"
No. You really only need about 3 dB to get out of OV. This is an SDR not an analog receiver, level doesn't matter much until you run over the top of the ADC and then it totally breaks. Well, there's a subtle lift in the noise floor from ADC non-linearity mainly when the top bit is toggling so maybe 9 dB, but there's little value in more than that and you can easily get that much out of the single LC already described.
Don't get carried away...
lol I'll try not get carried away, looks simple enough, will take time to get all the parts.
How many turns is reasonable for one of those T50-2 ?
I've read that silver Mica are the best to use in these applications, gives higher Q but getting one with the right value vs reasonable turns on the T50-2 is a challenge.
See my previous post. FT50-61 and about .012 uF is one solution. Make the C from parallel .01 uF and a ~.002 uF trimmer. You can probably see where the center of the notch is by loking at noise floor. Tweak the .002 uF as necessary. Make it from multiple C's if helpful against your local stock. Don't worry about Q of the C, L will dominate and in any case you don't care all that much.
The number of turns you can fit on a toroid core depends upon the wire size used.
However the more turns you add the lower the self resonant frequency becomes, which may be problematic in some applications.
For high values of capacitance in tuned circuits like this I tend to use Polystyrene, for lower values up to around 1nF I use ceramic C0G (NP0).
Don't bother with Silver Mica unless you are building a transmitter filter, as they are costly.
Looking at your KiWi, I think you only need about 10dB attenuation at 252kHz, so this should be easy to achieve, even if you decided to use a pre-wound axial epoxy coated inductor.
As the others have already said, don't overthink it. RF filters don't always match their theoretical characteristics due to stray capacitance and inductance, plus there is the self resonant frequency and Q factor to take into consideration too. The more sections / parts their are in a filter, the more difficult it becomes to align and get working properly.
Regards,
Martin
Ok thanks for the advice lads, will hopefully get around to ordering parts at the weekend.
If I were to go with a premade inductor can anyone send me a link via this site as to which would be best for this please ? https://ie.rs-online.com/web/
Or here https://ie.farnell.com/?CMP=KNC-GIE-PFB-exact-Control-816&mckv=sPLr5Mbg3_dc|pcrid|540346620940|kword|farnell|match|e|plid||slid||product||pgrid|83386991448|ptaid|kwd-133401130|&gclid=CjwKCAjw14uVBhBEEiwAaufYx8WhaCiAqPwGHxjbJKOADStcwz0sYrud5LTuNvvPtGcADKjSUuMe_hoC1ykQAvD_BwE
There are probably the only suppliers locally, and at that they might not even have all the parts, it's difficult to get a lot of stuff here on this little Island, the T50-2 forget that, that will take a week or more to get over seas, this is why something on one of these sites might work just as well and I can get it a lot faster if available locally.
Looking at stock levels and suitability.
Your best bet may be to use a 33uH inductor and a 12nF capacitor.
This type has a Q of 40 which is OK.
B82144F2333J000
RF Choke, LBC+, 33 µH, 1.9 A, 0.22 ohm, ± 5%, 9 MHz
Farnel.ie has stock and you don't have to buy a minimum of 10 or 100
You may have to add a 10 ohm series resistor to reduce the theoretical notch depth from 25dB to somewhere around 10dB.
Note that component tolerances mean that you may have to 'fine tune' the capacitor value.
Alternatively don't add a resistor and the notch depth will still be greater than 10dB over a bandwidth of 40KHz around the centre frequency, so even allowing for some variation in component values, the chances are the standard values will be close enough.
Regards,
Martin
Thanks Martin, will get ordering tonight hopefully.
Cheers
Mark
struggling to find Polystyrene caps, can I use ceramic instead ?
if you use ceramic, you better use NPO type as they are stable
Thanks will watch out for NPO.
Have to say I really am struggling to find suitable capacitor and inductor combination on these sites. These are local suppliers, anything else has to come from over seas with probably long wait times.
There is nothing that says you can't use multiple caps in parallel. Also, NP0 is equiv. to C0G
https://uk.rs-online.com/web/p/mlccs-multilayer-ceramic-capacitors/1220709
Thanks Jim. Going to check it out now.
Parts ordered. Might have them by Wednesday.
Had to order from Radionics due to capacitor supply issues at farnell so got different 33 uF inductor but same Q of 40.
Still waiting on parts, the joys of living on a small Island!
Finally got the parts, rough but hey it works ! there's a nice bit off attenuation on 252 Khz, roughly 10 dB reduction which is fine and it's now as strong as BBC Radio 4 which is sadly due to close I think next year as the BBC continue their assault on AM radio.
But I am very happy I finally got 252 Khz under control.
http://emeraldsdr.ddns.net:8073/
I have my other Kiwi back up now too this time connected to my EFHW.
http://emeraldsdr1.ddns.net:8074/
Thanks a lot for all the help lads, greatly appreciate it ! :-)
Just taken a listen and that's a lot better now so well done.
What's the source of the noise around 2MHz ?
Is it possibly from a variable speed motor drive or something like that ?
Regards,
Martin
Not sure, all I hear and see on the waterfall is powerline noise which is quite mild now but it can be quite bad especially from around 9 Mhz and up. This 10 Kva power line runs over the bottom quarter of my garden.
I must try get on to the regulator to see if they can get the power company to fix it they have ignored me completely for 5 years but I'm afraid that if I get them out and the noise is not there they will just ignore me in future.
which cap did you use in the notch?
I used this one you recommended.
https://uk.rs-online.com/web/p/mlccs-multilayer-ceramic-capacitors/1220709
I have filed a complaint to COMREG in relation to the mad QRM I have between around 650 Khz and 10 Mhz particularly noticeable around 80m.
Had to give them "evidence" and evidence it's not coming from my QTH so I went out on the road yesterday with my portable Tecsun PL-680 and of course got some weird looks. The noise has got really bad, I traced it to an Electricity pole so hopefully it will get sorted but I won't hole my breath.
I have two sources of noise but concentrating on this first as it's by far the worst as it's more constant.
If the pole was wood.... while listening to your radio, rap the pole with a small sledge hammer and listen for a change in the noise. Even better on a rainy day!
I can only imagine what the people that saw me walking around with a portable radio while recording on my phone would think, can't imagine what they'd say if they saw me going around whacking electric poles with a sledge lol. It's a tiny community where I live, bad enough if they spotted me shaking the pole. 😁
Shaking the pole make no difference.
I guess I'll find out soon enough if anything is done about it, I gave Comreg samples of audio and screen recording of my Kiwi and gave them my Callsign + youtube videos where I'm walking around and the noise can be heard a little bit at my QTH and it's loudest at that pole.