MW Band Attenuation to improve S/N up and down band

KA7U

I'm looking for an inexpensive filter to attenuate across the MW AM Broadcast band without blocking out the LW and lower frequencies. I have success in improving the S/N ration and eliminating what appear to be AM Broadcast station spurs up the frequency by using a 2 pole Chebyshev Hi-pass filter that starts attenuating below 1800 Hz. The attenuation curve is substantial starting at 1 KHz and completely eliminates most stations below 500 KHz. The first 2 attached images show the attenuation characteristics first with filter, then without. The 2nd 2 attached images attempt to show the difference in S/N ratio between the filter and without the filter. The first image being with filter and the second being without filter. I do not have experience with filter circuits that would achieve this result without using switches. Generally, I want to reduce the AM Broadcast MW band by about 15 dB and not affect the LW band or the frequencies above 1800 Hz.

Ron - KA7U

Attachments:
https://forum.kiwisdr.com/uploads/Uploader/88/2ee564fcf83e382a0feb6aa58b9b2c.png
https://forum.kiwisdr.com/uploads/Uploader/05/caf606650da8e9cdead28ba35e1e1b.png
https://forum.kiwisdr.com/uploads/Uploader/3e/7aad4188b1387ac11815b253663a73.png
https://forum.kiwisdr.com/uploads/Uploader/98/396d416af682dd80a2ecb5ccb876f5.png

jks

Some Googling around found this: http://www.qsl.net/kp4md/mwfilter.htm

Of course the real solution is to use a Clifton Laboratories Z10020. But sadly they are no longer in business.

KA7U

John,

Thank you for the link to  Dr. Carol F. Milazzo's filter. She certainly puts thought into her projects and is very good to share them. I subscribe to your youtube videos and have tried a few of her setups.  I am wondering if a band stop/notch type filter could be designed with opams to do this function. The LC circuits don't have a sharp roll off and have a fairly high insertion loss depending on frequency. I'm thinking an opamp circuit that attenuated signals between 580KHz and 1700KHz with steep side curves might be a better overall solution. Specifically I'm thinking it would have little or no insertion loss, and hopefully still be a very low noise device. Would something like this be practical, 10KHz through 30MHz??? Ever seen a practical example of such a thing?

Ron - KA7U

jks

Don't know how modern op-amps would work at MF. My guess is that it might be difficult to get the necessary Q? But that's just a guess. The only thing I know is that they can be used to emulate large value inductors for use at audio (or VLF/LF). See: https://en.wikipedia.org/wiki/Gyrator

KA7U

John,

That gyrator discussion is a bit intense. I think I understand it well enough to think about it though. This web site is still active: http://cliftonlaboratories.com/z10020_am_medium_wave_band_reject_filter.htm 

and I've written to Jack Smith asking if he will make a filter for me to attenuate the frequencies between 550KHz and 1700KHz by 25-30 dB. I'll let you know if I get a reply.

Ron - KA7U

KA7U

John,

I removed the low pass filter and replaced it with a series LC circuit, 4000pF-3.3µH, and tuned the coils on the T50-2 toroid until it pretty much centered at 1380KHz. During the day that is the very strong station close by to my antenna. 15KW about 15 miles south. I used mylar +/- 5% capacitors and so it was pretty close to right on first try. It dropped the level of that station from -15dB to -40dB. No more images from it up and down the band. Apparently that one station was the problem. The remaining signals seem to be handled by the KiwiSDR. 670KHz is now the strongest station @50KW about 70 miles West of here. It is peaking @ -35dB. So now I have all 30MHz, and more experience with filtering problems. The notch wave trap is much better than a high pass filter. I've checked for harmonics on the stronger stations, and they are all at or below the noise. 

I read up on the 70MHz opamps made by Texas Instruments and it would seem that they would be useful in a notch filter/band reject gyrator type circuit. I can imagine a guy like you building them into a receiver or active antenna to attenuate problem signals. From what I read the Q should be quite sharp over 10-20KHz.

Thanks for your help.

Ron - KA7U