HFU HF Underground
Technical Topics => The RF Workbench => Topic started by: syfr on June 03, 2021, 1651 UTC
-
I built a broadcast band bandstop filter to put in front of my Kiwis, and the results are pretty cool, as measured by the NanoVNA. I got about 55dB attenuation at the frequency of interest and minimal loss at the rest of HF.
Hope some may find it interesting/useful. NanoVNA is awesome
SCROLL to the right >>>>>>>>>>>>>>>>>>>>
(https://i.imgur.com/Qd5PVo7.jpg)
-
(https://i.imgur.com/iqoo8fL.jpg)
Less than half the rejection . This was a "DISTILL AM" filter seen on Ebay
-
Schematic
(https://i.imgur.com/q6qZpjn.jpg)
Prototype
(https://i.imgur.com/9I6Dti8.jpg)
-
Nice work on the filter. In a previous thread ( https://www.hfunderground.com/board/index.php/topic,80208.msg262629.html#msg262629 (https://www.hfunderground.com/board/index.php/topic,80208.msg262629.html#msg262629) ) I evaluated the NooElec Flamingo AM and Flamingo FM filters. Both filters are well built and assembled and performed as advertised by the manufacturer. Below you will find the filter response charts based on the tests that I performed using my HP/Agilent instruments.
The NooElec products are manufactured in North America and can be purchased directly from their web site: https://www.nooelec.com/ (https://www.nooelec.com/)
(http://www.milspec.ca/board/img/nooelec-flamingo-am-filter-frequency-responce-performance.png)
(http://www.milspec.ca/board/img/nooelec-flamingo-fm-filter-frequency-responce-performance.png)
-
I built a broadcast band bandstop filter to put in front of my Kiwis, and the results are pretty cool, as measured by the NanoVNA. I got about 55dB attenuation at the frequency of interest and minimal loss at the rest of HF.
Hope some may find it interesting/useful. NanoVNA is awesome
Syfr --
Very nice work on the design and construction of your bandstop filter. 55dB attenuation is amazing!
A couple of questions:
- What software did you use to capture the nanoVNA data?
- what core material was used for the toroids?
Thanks,
Stendec
-
Unless you really need that much attenuation, that filter is too good. I have a MW band stop filter in front of my KIWI's but I added a variable resistor from input to output so I can fine tune the loss (20-30 db was about right in my case ) so I am safely out of the near ADC clipping area. Anymore than that and you are throwing away MW energy that someone may be interested in.
73 Don
VE6JY
-
That is pretty good results for a homebrew 5th-order filter. Good job!
Taking the slacker approach with an off-the-shelf solution, I have a PAR BCST-HPF, which I admit works nicely when (rarely) needed. Not cheap, though. IIRC, like about $70 years ago. Could be more now I suppose. It is a 7th-order elliptical highpass. VNA plot here:
http://www.parelectronics.com/bcst-hpf-specs.php
-
Here's some answers on the filter.... I got it boxed up and at work for good in the Kiwi.
* I needed this filter (and I need more of it) as there is a commercial AM transmitter not far from me on 1090 kHz. Artifacts of this daytime overload (it's off the air at night) are all over the Kiwi spectrum. I don't do any AM listening, so I'm not concerned about over attenuating things. I might cascade another to see if I can knock down the 1090 signal some more.
* The software I used was VNASaver, which is really a great way to control the VNA from the PC UI, rather than donking around with the little rockers on the screen. I'd really recommend this as it transforms the VNA into a very easy machine to use. Installation was easy, as was the calibration via the GUI.
* I'm not sure which all the core materials were. I know -37 and there was one more. I have a large junkbox and I just tried values which ended up measuring closely on my impedance bridge. I tweaked turns and core size until I got close (very ) to the design values in both L and C. I use an old (and great!) AADE LC bridge for this. I suppose the VNA would work great too. UPDATE: I used type 43 and type 2 cores
* Getting Chinesium standard extruded enclosures and precut Pride Of China (thanks Chris!) copperclad made housing this a simple affair (it's now in a box with BNC's at either end)
I wish a 1/4 wave coaxial stub cut for 1090 kHz wasn't so darned big!
PS: https://toroids.info/ is really handy to get you close. You can also use this data + the empirical measurements to determine the core material. Great reference page !
-
Have you considered an LC notch filter instead to knock down 1090? You could get some decent attenuation there, while preserving the rest of the MW band.
-
Have you considered an LC notch filter instead to knock down 1090? You could get some decent attenuation there, while preserving the rest of the MW band.
Thst sounds like a better solution.
-
I have a LC notch. It's just a *wide* notch :-) Seriously, I don't think I've ever used the Kiwis for MW, so it's not a loss for me.
I'm pretty sure I have a tunable notch in one of the various "tuners" I have here. That's probably the easiest to try....
-
Built the LC notch and added it to the filter . 1090 is still pounding in.