ways to stop bleedover

[w2rea]

Does anyone have any suggestions on how to stop bleedover form other frequencies? The busy Sheriff frequency on my feed is covered by bleedover from a paging service especially during buisness hours. My scanner is a Uniden 800xlt that is about 18 years old. Would a newer model scanner help?

 

[fineshot1]

Does anyone have any suggestions on how to stop bleedover form other frequencies? The busy Sheriff frequency on my feed is covered by bleedover from a paging service especially during buisness hours. My scanner is a Uniden 800xlt that is about 18 years old. Would a newer model scanner help?

1 - Identify the offending paging service frequency.

2 - Obtain a quality notch filter tuned to above frequency and install it inline with the antenna cable.

Trying a new scanner is a shot in the dark and could be expensive and yield negative results.

 

[w2rea]

Thanks, I found a lot of info after I googled notch filters.

 

[w2rea]

I was looking for an excuse to give my wife for buying a new scanner though!

 

[fineshot1]

Thanks, I found a lot of info after I googled notch filters.

You may want to focus on "Par Electronics" as they will custom make a filter for your needs.

 

[torsionbar]

Newbie question: What causes this "bleedover"? Is it the sending equipment that is transmitting in a sloppy manner? Or is it the receiver that is not picking up the signal properly?

 

[fineshot1]

Newbie question: What causes this "bleedover"? Is it the sending equipment that is transmitting in a sloppy manner? Or is it the receiver that is not picking up the signal properly?

It is usually the later or more accurately the inability of the receiver to reject adjacent or inband strong signal
overload. You can make receiver comparisons via one of the receiver specs called "selectivity" and perhaps another called "image rejection". This type of problem is more prevalent in double conversion receivers but can also happen in triple conversion receivers with less occurances if the overload is severe enough.

 

[kb2vxa]

Let's set the record straight here, but first you must define your terms as I often see many used incorrectly and/or applied inappropriately and off the helpers go on a wild goose chase.

Used correctly "bleed over" means a strong signal on an adjacent frequency is "bleeding over" or leaking into the channel you're listening to and has nothing to do with overload, intermodulation or intermediate frequency (IF) images but rather insufficient receiver selectivity to reject it. In this case an external filter of any sort will not help, the frequencies are too close together to reject the unwanted one without severely attenuating the wanted one.

Sometimes a few tricks may help like setting the channel to narrow band (NFM) which narrows down the channel width (increased selectivity). Another is programming the frequency slightly offset up or down away from the offending signal but first you must determine whether the offending one is above or below the wanted one. You may try one or the other or both, no guarantees here but it has worked for myself and others.

 

[gmclam]

... Used correctly "bleed over" means a strong signal on an adjacent frequency is "bleeding over" or leaking into the channel you're listening to and has nothing to do with overload, intermodulation or intermediate frequency (IF) images but rather insufficient receiver selectivity to reject it. In this case an external filter of any sort will not help, the frequencies are too close together to reject the unwanted one without severely attenuating the wanted one.

Sometimes a few tricks may help like setting the channel to narrow band (NFM) which narrows down the channel width (increased selectivity). Another is programming the frequency slightly offset up or down away from the offending signal but first you must determine whether the offending one is above or below the wanted one. You may try one or the other or both, no guarantees here but it has worked for myself and others.

I agree with kb2vxa. You can not use a "notch filter" to alleviate bleed-over. As I see it your choices are:
- Reduce the total signal level coming in to the scanner. But that reduces the desired signal as well as the undesired signal.
- Use a better quality receiver. However, even the best quality receiver may not be good enough.
- Add a CT/DC tone to the desired channel's settings. If they do not use a tone, you are out of luck. If they do, this will keep the scanner from stopping when there is only bleed-over, but will not stop the bleed-over from interfering when both signals are present.
- If the source of the bleed-over is from a specific direction, you might be able to physically block signals coming from that direction to your antenna. This gets a little tricky, and require that the offending signal is not also bouncing off an object and coming in to your antenna from a different direction.
- If the desired signal is from a specific direction, and the undesired signal from a different direction, you might be able to use a directional antenna to "focus" in on the desired signal.

My gut says that even if you implement several of the above techniques, you will still get SOME bleed-over. Add the CT/DC option first, it requires the least effort and see what happens.

 

[fineshot1]

Guys - before you all go assuming its as the op has said(bleedover) why don't you re-read
my first response and let the op identify the source of the problem and get back to us on it.

This way we can all take a better shot at helping him out with his problem.

 

[w2rea]

Thanks for all the replies. The frequency that I am getting the interference on is 156.210. I am pretty sure that the interference is coming from 157.860 a paging service for a hosptal that is very close by. There is also another paging service at 158.550 that does not seem as strong.

I asked the Par Electronics people about the problem and they said targeting the pager at 158.550 with a notch filter might help.

I tried entering 156.205 into the scanner and it did help a little.

Like I said the scanner that I have is 18 years old, maybe with the tax return I will try and upgrade.If you want to hear the problem first hand my feed is the Chautauqua County NY police & fire feed.

 

[jackj]

One thing you might try, if you have a well filled junk box, is making your own notch filter. You will need a short length of RG58, a 'T' connector and 3 matching connectors for the 'T'. The feed from the antenna goes in one side of the 'T', the scanner in the other side and a 1/4 wave length of coax in the third side. The 1/4 wave length stub will reflect back the opposite of whatever it is terminated with at it's resonate frequency. You will need to know the exact frequency of the interfering signal, the wave length of it and the velocity factor of the coax you use.

Once you have the wave length of the signal you want to notch out, divide by 4 to get the 1/4 wave length and then multiply the result by the velocity factor of your cable. That will tell you how long the stub needs to be. Cut it a little bit longer and then cut very small sections off as you tune it. Be sure to measure the stub from the center of the 'T' connector.

This will have a pretty high Q and should be able to notch most of the interference out without too much effect on the frequency you want to monitor.

 

[kb2vxa]

I'll go with upgrading to a modern scanner. 18 years ago narrow FM was what we now consider standard FM so it's pretty much a given your scanner lacks sufficient selectivity to reject the interference. Meanwhile just for the heck of it try dropping the frequency down another 5KHz. Read on, I hope you can understand it without TOO much confusion. (;->)

Older scanners used a rather wide IF passband to accommodate the shift encountered between the VHF and UHF bands because during the frequency conversion process they landed on slightly different intermediate frequencies (IFs). The short explanation is VHF ended up on one side of the selectivity skirt and UHF on the other and it was wide enough to accommodate both. This led to a lot of bleed but it was what made that programming trick I mentioned possible, you may have enough left to shift a bit more before the wanted signal slides out of the passband.

Not to argue but to clarify, that 1/4 wave coaxial stub filter is not in any way a notch filter. A notch filter is extremely narrow which makes it possible to "notch out" a very narrow band of frequencies centered on the one you wish to eliminate. The coaxial stub lacks such selectivity and will greatly attenuate a much broader band making it useless for the purpose. It does however make an effective band stop filter, I made one once to eliminate interference from a ham nearby operating on 2M (146MHz) who was causing severe intermod on the 167MHz band I was trying to monitor. Considering how close he was and the strength of his signal I'd say it worked very well.

 

[fineshot1]

Thanks for all the replies. The frequency that I am getting the interference on is 156.210. I am pretty sure that the interference is coming from 157.860 a paging service for a hosptal that is very close by. There is also another paging service at 158.550 that does not seem as strong.

I asked the Par Electronics people about the problem and they said targeting the pager at 158.550 with a notch filter might help.

I tried entering 156.205 into the scanner and it did help a little.

Like I said the scanner that I have is 18 years old, maybe with the tax return I will try and upgrade.If you want to hear the problem first hand my feed is the Chautauqua County NY police & fire feed.

OK - based on that info my original suggestion still stands as this is not bleed over, it is in band strong
signal overload and a notch filter will most likely help. But if your looking for an excuse to purchase a
new scanner this is as good as any.

 

[jackj]

Huhhh?

Older scanners used a rather wide IF passband to accommodate the shift encountered between the VHF and UHF bands because during the frequency conversion process they landed on slightly different intermediate frequencies (IFs). The short explanation is VHF ended up on one side of the selectivity skirt and UHF on the other and it was wide enough to accommodate both. This led to a lot of bleed but it was what made that programming trick I mentioned possible, you may have enough left to shift a bit more before the wanted signal slides out of the passband.

Carrier Freq......Local Osc......Intermediate Freq.
155.340 Mhz - 144.640 Mhz = 10.7 Mhz
462.450 Mhz - 451.750 Mhz = 10.7 Mhz

Where is the frequency shift due to the conversion process?

 

[kb2vxa]

"Where is the frequency shift due to the conversion process?"

Don't ask me, ask the Cushman Station Monitor I used to align an old Electra Bearcat BC-101 after replacing an IF can. When working at the bench you don't have time for theory and math problems, fix it, shove it out the door and get on to the next job or you're wasting time and not earning your pay.

For what it's worth my observations were the IF center frequency was 10.8MHz with a sharp skirted passband of ~14KHz using a 3 stage crystal lattice filter. The signal appeared on one side for VHF and the other for UHF, never did any show up in the center. I'm just telling you what the 'scope told me, I didn't take the time to analyze it to death.

Some years later I noticed the same thing was happening during some programming experiments with my old Rat Shack scanner and drew the same conclusion. I was having a bleed over problem so I shifted the channel frequency, it worked just fine in one direction but when I went the other way the signal slid right out of the passband. Again I didn't analyze it, when it comes to practical applications I don't bother with the book, whatever works is good enough for me.

 

[gmclam]

Receiver math - PRO-10

Carrier Freq......Local Osc......Intermediate Freq.
155.340 Mhz - 144.640 Mhz = 10.7 Mhz
462.450 Mhz - 451.750 Mhz = 10.7 Mhz

Where is the frequency shift due to the conversion process?

Here's the math for my PRO-10 scanner. For UHF, let's use 460.200 MHz as the desired receive frequency. The signal is mixed with 416.200 in the 1st mixer to derive a 44.000 MHz IF. That signal is mixed with 54.7 MHz in the 2nd mixer to derive a 10.7 MHz IF. That signal is mixed with 10.245 MHz in the 3rd mixer to derive a 455 kHz signal.

Now with VHF, let's use 154.800 MHz. It is only double conversion (UHF is triple conversion), so it is mixed with 144.100 MHz to derive the 10.7 MHz IF. The remaining process is the same as for UHF.

Because the local oscillator frequency being mixed with the signal is typically below the signal frequency, you get the image on one side. But note that the 2nd local oscillator frequency for UHF (54.7) is ABOVE the desired frequency. This puts the image on the other side.

This wouldn't hold true if UHF is only double conversion. Note that many modern scanners are single or direct conversion.

 

[jackj]

The old BC-101 does us a 10.8 MHz IF instead of 10.7 MHz that most others use. But if you are getting a different IF frequency when you program in a UHF frequency than what you get with a VHF frequency then you have a problem with the phase-lock-loop system or the programmable divider. It should produce the same IF frequency regardless of the band.

 

[kb2vxa]

"But if you are getting a different IF frequency when you program in a UHF frequency than what you get with a VHF frequency then you have a problem with the phase-lock-loop system or the programmable divider."

Hey, I didn't design the thing so it's not my problem. (;->) Now if you're familiar with the 101 then you should be familiar with the frequency step programming sequence using the front panel channel switches. Talk about primitive! Remember this was before CPU control, numeric keypad entry and display, I had a couple of ham rigs that were rather complicated dialing up frequencies too being the steps were selected by rotary switches. The problem wasn't in the PLL and VCO circuitry (that never changed) but rather the divider chips in use at the time and BTW still in use in CB rigs since all they have to do is come up with 80 frequencies (40 ch T/R pairs) in 10KHz steps with the Class C channels locked out.

If you can come up with the BC-101 manual you'll find the programming formula from which you can easily figure out which switch controlled which frequency step in the divider. At one point I had it down to a science because I was listening to federal frequencies not listed in the channel chart but being so long ago memory fails so please don't ask. (;->) Hmmm, too bad I disposed of all my old service manuals and data sheets when the archives became too much to carry around in a box. The original full size factory blueprints are probably a bit of a collector's item these days, oh well.

 

[jackj]

I have repaired a few of them, Warren. In fact, I used to own one until a few years ago when I gave it to my son (still worked but had a few bad memory locations). I'd have to look but I think I have the blueprints around here somewhere. Surprised me when I ordered a service manual and that's what I got. The 101 used a few proprietary IC's and I think the memory chip was one of them. But I don't remember the IF being offset by band.