// There is a 107.9 MHz FM transmitter about 3 miles from my house. There is a 107.3 MHz FM transmitter about 45 miles from my house. See any problem? Do you have any suggestions for a notch or lowpass filter to kill the
107.9 MHz signal (Hip-Hop format)? I believe it is causing intermod distortion on 107.3 and other FM stations.
I already tried a 5-element directional antenna at various orientations (vert/horiz, tilt up/down, etc.)
I was looking at the following:
1) Coax stub notch filter.
2) Filters from Tin Lee Electronics.
3) Filters from Par Electronics.
4) Complaining to 107.9 and getting a free filter. \\\\
I wonder if anyone here has experience with 2 or 3? I suspect 1) will not have enough Q, and the 4) people won't have enough clue. [Anyone here remember back when radio stations had Chief Engineers with a First Class License?]
--
A host is a host from coast to coast.................wb8foz@nrk.com
& no one will talk to a host that\'s close........[v].(301) 56-LINUX
Unless the host (that isn\'t close).........................pob 1433
is busy, hung or dead....................................20915-1433
Did you try a simple attenuator at the input? Are you sure the interferror gets through the antenna? May be it is picked up directly by the receiver PCB.
Coax stub made from a reasonable cable will not have sufficient Q at
100MHz.
I do not have particular experience with those, however the required notch filter is definitely doable with the lumped components.
5) Moving to the different location?
Vladimir Vassilevsky DSP and Mixed Signal Design Consultant
Obviously, a better receiver might do the trick (one with better adjacent channel rejection). Depending on the output power of 107.9, you could be in the Blanketing Contour (roughly 115dBm), and if so, the receiver itself may not work in that strong a field, but that's probably not your situation - just thought I'd mention it.
Antennas make poor filters. Don't bother.
You could try to filter the 107.9, but the desired signal is reasonably close (and weaker in amplitude), so you have to be careful.
If you can control a phase-stablized cable (environmentally, etc...) you could try a stub-tuned filter. Take a half-wave at 107.3, and on each end of this, stub a 1/4 wave at 107.9. You'll probably need a network analyzer to get it built right. Plus a calculator and lots of patience!!
My advice would be to reverse feed an isolator into a selective load. That way, you won't have all the insertion loss of the filters (just the 0.3dB or so for the Isolator). And you can tighten the skirts on your 107.9 filter as strong as needed.
I can draw this up if it doesn't make sense to you, (my ASCII art skills are horrible). Basically, you pass everything through the isolator EXCEPT what gets sucked into the 107.9 port. (This would be the port the dummy load usually goes to, if the isolator were used in a ferrite combiner arrangement.) Oops. Forgot to mention, you do need to terminate the
107.9 selective load.
After that, then you can start looking at directional antennas to
Phil made several creative suggestions, but this one is (a) clean and (b) probably the best idea.
Building VHF traps with Qs of several hundred isn't completely trivial. (Just for a start, over 30 degrees C or so tuning drift due to inductor tempco could easily be 600 kHz at 100 MHz if you use a plastic coil form.)
You might try a folded dipole made of twin-lead, with the wires pointing along the line between you and the nearer station. You can wire that up in about 5 minutes. There are fancier phasing things you can use if it's worth it to you, but that would be a good place to start.
You'll have a tough time getting a sufficient Q with a straight filter approach. I won't say it can't be done, but my swag is that it would be tough.
Take note of the things Vladimir said. Especially knowing how the interferer is getting into your system. You must have control of that.
Also, there are techniques to null interferers, but hardly trivial/ inexpensive.
The basic nulling principle: You need two good directional antennas, one for each transmitter. In your "main" system path for the desired signal, you need to add in the undesired signal in the correct amplitude and phase such as to null the pickup of the interferer (picked up by the antenna pointed to the desired station). This means, of course, that the undesired signal, from the antenna "pointed" at the undesired transmitter is adjusted to the correct phase and amplitude such that it cancels the undesired station in the "main" path when "summed in."
None of this is easy. It will probably require frequent retuning. Moreover, the interferer station is close by and the field strength in your neighborhood is probably very high. Because the antenna directionality and null responses can by used to your advantage, this is possible. Even co-channel interference can be reduced this way. (I've done this before.)
What would I do? Turn off the radio and read a book. Neither this or combined technique with a filter is guaranteed to provide you acceptable reception, even with money and time investment. If the station is operating legally, you are probably sol.
I had a First Class license. I have the lifetime General Class now. It is worthless/meaningless anymore.
A long time back, cable TV injected an interfering beep into pay TV channels. Those that paid for the service got a notch filter to remove the beep. Those that didn't pay, could buy a kit and build their own filter. The kits were tuneable and provided about a 40db notch 1 mhz wide. If you search google for TV notch filters, there still might be information available. With a bit of tweaking, I'm sure you could get one of those kits to work. bg
It would be good, before leaping to confusions and building a filter, to make sure that the problem really will be solved with a filter or other means to reduce the bad guy relative to the good guy. For example, the 107.9 station may be radiating enough energy on 107.3 to be a problem, or there may be intermod generated in non-linear junctions (corroded metals in contact, for example) external to the receiver.
An antenna may help, but you'll probably find that a narrow-band Yagi optimized for the desired station will work a lot better than a typical broad-band Yagi or log-periodic used to cover the whole FM band (or more?). You can optimize the design, or at least it is possible to optimize the design, to give a high ratio between the responses in the direction to the good guy and the direction to the bad guy. Nulls _can_ be deep, but there is typically not a null but rather a minor lobe directly to the rear from the main lobe. Actual realization of the predicted pattern also depends on decoupling the antenna from the feedline and from other metal structures in the area, so that the other metal doesn't act like part of the antenna.
If you want to try a filter, you'll need resonators with very high Q. Based on a quick hack I just did, I'd plan on resonator Q of 2000 or above. Consider that the Q of a coax stub is at best about
100*sqrt(freq,MHz)*line_OD_inches, that suggests a line at least 2 inches in diameter. At 100MHz, you can make a more compact filter using coils, but they still need to be moderately large--things wound with 1/4" copper tubing to a nominal 2" coil diameter work about right. You may be able to get by with a circuit which puts a notch on
107.9MHz and specifically passes 107.3MHz, but even just three such coupled resonators on 107.3 to form a band-pass filter could give you about 20dB improvement in the ratio of signal levels.
If the strong station is putting out energy on 107.3--which it undoubtedly is, to some extent--and if that's the root of the problem and not overload in the receiver--then a filter won't help. A directional antenna could, however. For that matter, a better receiver could also help, if it is a receiver overload problem.
Extremely unlikely.!!! (Although 107.9 power to the front end could be excessive.)
Before I continue this discussion, and in all fairness to Tom, we really do need more information in order to provide a meaningful reply
- but let's go with what we know so far.
Possible. Not likely. The RF field is likely very strong, and the receiver selectivity is probably quite poor.(?) Do all receivers in your house have this problem? I could tell you HORROR stories....!!!
FM
Almost a total waste of time... (but you could get lucky) No Yagi-Uda antenna is going to have a sufficient cutoff at 600 kHz. Or enough azimuth gain to matter. 107.9 is probably circularly polarized anyway. Also, not sure the orientation, but obviously if 107.3 and 107.9 are located on the same relative azimuth to your position, then antennas alone are even less likely to solve the problem.
Not with most antennas. And even 20dB is unlikely to help.
ual
Don't forget the 1.4 ground bounce. No lobe is going to fix that.
What!? (Nevermind, I think I see where you're going. Bigger is better, to a point.)
ch on
Such an arrangement will have so much insertion loss on the third adjacent that you will be shooting yourself in the foot trying to receive the station 45 miles distant. I asked for the call signs so I could estimate the dBu fields. (I need the station classes & antenna patterns, etc..)
To the OP, distance to 107.9 is critical. Power will roll off inverse as the square of the distance, so estimating 3 miles, when it's really
2.6 miles is a big deal. Generally speaking... Lat/Lons would be a big help to telling you (in advance) whether filters will do the job. Or you could just try 'em.
Not likely. (Unless its a POS CCA Transmitter, then you're lucky its even radiating!) By law it has to be 43 + 10 Log10 the Power (in watts), or -80dB, whichever is less.
Agreed.
Having done this for a living, I strongly caution the OP against trying to resolve this issue with antennas - and certainly not antennas alone. You can easily get 30-40dB of isolation for less than
1dB insertion loss by using the reverse isolator technique described above. You will lose 107.9 & 107.7 and maybe a good bit of 107.5 too (as if that matters?), but (if you tune the selective load properly!!) should be good to go by the time you hit 107.3 (600 kHz away). Also, the nice part about the isolator approach is you can keep the rest of the band response relatively flat - depending on the quality of the components used.
Now, 100 MHz cavities aren't exactly cheap, or compact - but this approach will definitely work. How important is reception to you??? Cheaper than moving..? :)
You could try to stub-tune a reject load on the isolator port,?? but that sounds messy... A line stretcher would work great, but who has those lying around?
And you could certainly try inline traps. With a little luck, it will work. And it would definitely take up less real estate....
If it wasn't mentioned earlier, bandpass filters inline won't roll off sharp enough (no matter how tight you set the skirts). Ditto for notch filters. You could try crystal filters (band reject or notch), but at 115dBu (assuming blanketing), you might not get enough signal attenuation??. Cascaded filters might work, but there goes the downstream NF. And rising insertion losses for the 3rd adjacent (107.3). I've even seen this problem with no antenna - signal radiating directly into receiver RF section. (Keep that in mind when you place your bets on EXTERNAL filters..)
Best of luck. I'll have to tell you about a 9th harmonic problem I had with 90.9 FM someday....
Oh, I didn't mention it before. 107.9 is obviously in the Commercial band, and a lot of these (though not all) are high power class-C stations (C3 through C0). It would be very helpful to your situation if the 107.9 in question here is a lower-power class station. The less power you have to filter, the better.
First, see if the desired radio station has an internet stream.
Given the closeness of the frequencies, filtering before the radio doesn't look promising. [A coaxial cavity resonator would be a good conversation piece.] There are plenty of old tuners with wide/narrow IF settings. On ebay, you need to go to electronics>home audio>tuners
You could get an old radio and pop in different bandwidth ceramic filters (10.7MHZ), but it has been my experience that it is hard to buy specialty ceramic filters in small quantities. It is easier to get them sampled since such ceramic filters are expected to be sold in the
100k to millions per customer.
I found this crystal filter on ebay, but no specs.
Without getting into any of the rest of this (except to say that 20dB f/b ratio on a 5 element narrow-band Yagi is pretty straightforward, over typical ground), do you seriously think that relative attenuation equivalent to moving the interfering station TEN TIMES further away (line-of-sight) is unlikely to help?? And if you do, why do you say that the difference between 2.6 miles and 3.0 miles is really significant?
Doesn't look like we're going to hear from the OP again on this, and we all know what the thread degenerates into at that point...
Sheeeeesh! You mean we can't make political hay out of it ?:-)
...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC\'s and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
Obama snickers, claiming McCain is running for Bush\'s third term,
While he thinks running for Carter\'s second term is a good thing?
I'm reading along, but don't have anything to add at present. He lacks network analyzers etc necessary for some of the suggested approaches.
I agree that the stub is pointless. I think he needs some tuned cavities, but I suspect he'd balk at Sinclar's price.
Not sure how audio rectification in rusty gutters works with FM, but then I learn something every day.
I agree with the on-line streaming suggestion, except RIAA just killed that deader than dead with the latest Copyright Rights Board rates.
--
A host is a host from coast to coast.................wb8foz@nrk.com
& no one will talk to a host that\'s close........[v].(301) 56-LINUX
Unless the host (that isn\'t close).........................pob 1433
is busy, hung or dead....................................20915-1433
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.