Tuned cavity / tuned interdigital filters like this normally start well above $500. At UHF frequencies, an insertion loss of 0.6 dB and stop-band attenuation like the filter I've demonstrated is unlikely. You would be paying well above $1k for that. I'm done discussing this. Have fun.
BTT SELECT-78 770/850 MHz Public Safety Filter
- Thread starter btt
- Start date
- Status
Again, can you show me plots for a passive commercial 770 band filter better than what I have demonstrated? Look at the stop band. With 60-70 dB stop band attenuation, your example of 100kW transmitters will be reduced to an effective 10-100mW. I can already see that there isn't a market for this, but damn, this is a tough crowd. Lots of talk. No plots.
Ok here is a 6MHz window filter I pulled out of my junk box that I had retuned from the 800 range centered on 902MHz. Its got 1dB insertion loss and 5MHz away from the 6MHz window edge its 36dB down and 10MHz from the window edge its over 60dB down. This cost $30 surplus. I have many more by Celwave that are cell phone industry surplus and dirt cheap that are a little better at about .6dB insertion loss and those live on a couple of mountain tops in front of my 900MHz repeaters before the preamp. The BTT filter plot showing 30MHz to 1.2GHZ is hiding the skirts so I can't really see how the skirts compare between the two. But this is not bad.
Yes these are expensive new but the bottom line is 4dB insertion loss is way too much. As a past designer of repeater front end and master receive systems on some of the most RF congested sites in the country, 2dB insertion loss is about the industry standard for max and even 2dB loss is hard to swallow. In my opinion 4dB is completely unacceptable even for hobby use. You take an instant hit of 4dB in the system noise figure and some weak signals will be unusable.
Yes these are expensive new but the bottom line is 4dB insertion loss is way too much. As a past designer of repeater front end and master receive systems on some of the most RF congested sites in the country, 2dB insertion loss is about the industry standard for max and even 2dB loss is hard to swallow. In my opinion 4dB is completely unacceptable even for hobby use. You take an instant hit of 4dB in the system noise figure and some weak signals will be unusable.
FreqNout
Member
I have several wideband tuned interdigital filters like that for 902-928 MHz as well. They are nice for ISM applications. They cost around $450. Not really the same thing. Not even close.
I’m not a filter designer and if I needed a filter for a commercial purpose it would be very expensive new and I usually try and repurpose a surplus unit. But I have had to buy some new ones and had to pay a lot like the last window filter we got from Telewave with eight 5” cavities. The point I’m making is 4dB insertion loss is way too much even for hobby use. With 4dB loss you instantly loose 60% of your signal and you can never get it back.
If you are willing to buy a filter with those specs and actually understand the implications then have at it, maybe you can get a qty discount then put them in series and use it as an attenuator.
If you are willing to buy a filter with those specs and actually understand the implications then have at it, maybe you can get a qty discount then put them in series and use it as an attenuator.
Last edited:
FreqNout
Member
Are you saying the $1750 passive public safety filter mentioned earlier in the thread is junk? It worked well for the person who mentioned it. The specs for that commercial filter are 4-5 dB insertion loss.
That public service filter does not interest me in the slightest. I was curious at first then I saw the specs. I think the disconnect here is what we are each willing to compromise in performance. If I must have something better and it costs more so be it and I’ll pay. If I can repurpose a surplus unit that meets my specs that’s ok too. But my experience dictates what insertion loss I can live with in a receiver front end and the filter in question here is not something that could ever work for me. Sorry if some of my posts got off track.
Last edited:
Moving forward.. It may not be obvious to some, but you can combine the active filter at the antenna end of the feedline, followed by the passive 770 band filter for an improved sensitivity with slightly worse IP3 specs, or you could use the passive filter only for improved IP3 specs with slight loss of sensitivity. We will likely see some real-world examples of real-world issues being resolved by these filters soon.
I’m intend to post a fuller set of information and pictures, but here is the summary, the 700/800 filter works well for my p25rx on 700.
Probably works well my other SDR’s as well, just haven’t tested as much of thIs. Also 800 interference is limited here so testing may not be as meaningful for 800.
Using the filter with my P25RX is like turning the hands of time back several years, before my home and office became grossly overloaded from broadbandend signals terminating at about 768mhz. The sites I could previously receive (including with a airspy) I can receive again.
The LWIN site in Lafayette (north) is a good test case for me because it has a 769mhz Ctrl channel and a lower signal level overall at my residence/QTH/main test point, then it would in the middle of city or my office; my home is somewhat on the edge of its service countour.
In my driveway, it goes from totally unusable w/o to fully usable with filter. Driving to the base of the offending cell tower made no difference (I was only about 50 feet from the base Of tower). On the closer to myself Scott LWIN site, I also got lots of wavering signals and some dropouts before the filter, but none afterward. Sitting side by side to an xts5000 scanning the site conventionally, we received the same stuff.
I will now be able to deploy the P5RX mobile again and I believe it will be enough at my office as well which is even worse b/c of two nearby cell towers.
I have a few screenshots from an airspy and SDRsharp showing the great improvement in noise floor for the 769 band and The improved SNR for a given control channel as a result. Briefly using it at my office, it helped my SDS100 lock onto an LWIN site, which it definitely could not do.before.
THE NET RESULT IS I CAN HERE THINGS AGAIN I HAD LOST due to the strong signals 768 and below, thanks to this filter.
Again more pictures/screenshots coming.
DISCLAIMAER: I was provided a filter to test, with the offer that I could purchase at a discount if I wished. I have no other business relationship with BTT/bluetailtech, other than being a happy customer of a P25rx
I absolutely will purchase this filter, and would purchase another one or two if I could.
Probably works well my other SDR’s as well, just haven’t tested as much of thIs. Also 800 interference is limited here so testing may not be as meaningful for 800.
Using the filter with my P25RX is like turning the hands of time back several years, before my home and office became grossly overloaded from broadbandend signals terminating at about 768mhz. The sites I could previously receive (including with a airspy) I can receive again.
The LWIN site in Lafayette (north) is a good test case for me because it has a 769mhz Ctrl channel and a lower signal level overall at my residence/QTH/main test point, then it would in the middle of city or my office; my home is somewhat on the edge of its service countour.
In my driveway, it goes from totally unusable w/o to fully usable with filter. Driving to the base of the offending cell tower made no difference (I was only about 50 feet from the base Of tower). On the closer to myself Scott LWIN site, I also got lots of wavering signals and some dropouts before the filter, but none afterward. Sitting side by side to an xts5000 scanning the site conventionally, we received the same stuff.
I will now be able to deploy the P5RX mobile again and I believe it will be enough at my office as well which is even worse b/c of two nearby cell towers.
I have a few screenshots from an airspy and SDRsharp showing the great improvement in noise floor for the 769 band and The improved SNR for a given control channel as a result. Briefly using it at my office, it helped my SDS100 lock onto an LWIN site, which it definitely could not do.before.
THE NET RESULT IS I CAN HERE THINGS AGAIN I HAD LOST due to the strong signals 768 and below, thanks to this filter.
Again more pictures/screenshots coming.
DISCLAIMAER: I was provided a filter to test, with the offer that I could purchase at a discount if I wished. I have no other business relationship with BTT/bluetailtech, other than being a happy customer of a P25rx
I absolutely will purchase this filter, and would purchase another one or two if I could.
Great review, thank you for your observations and opinion on it! This absolutely sounds like something that could help my situation as well as I live in a challenging RF environment.
Brief update, from my office (3rd floor/top floor) with 2 cell towers nearby and clearly visible, non of my unidens work at all on LWIN Lafayette or Scott sites (the two closest). The P25 without filter does not work either without filter.
With filter on p25rx, the Scott site is usable without obvious difficulty, no particular "sweet spot" for antenna needed. The Lafayette site works with filter, if I find the "sweet spot" (a few inches back and forth on my desk). So again, a noticeable improvement in practical use conditions.
With filter on p25rx, the Scott site is usable without obvious difficulty, no particular "sweet spot" for antenna needed. The Lafayette site works with filter, if I find the "sweet spot" (a few inches back and forth on my desk). So again, a noticeable improvement in practical use conditions.
freqseeker
Member
That's good to hear. Can you try the Unidens at your office with the filter? Just curious.
Please see the two following screencaps. I am running an airspy on a Sirio discone mounted on my tower at home, tuned to the LWIN Lafayette site. The filter provided a far better noise floor and resulted in a much better and ultimately usable signal to noise ratio. Note, I did not try to tweak decimation or gain, I left settings the same with and without filter. So it's a basic test, but it illustrates a real world test case.
Attachments
Glad to see that it is working for you. The second (passive) 770 filter is on the way to you. That filter will give better stop-band attenuation, but you will lose some sensitivity unless you place it after the SELECT-78 filter/preamp. Either way, it looks like both solutions will work for you. I think the "final" version of the passive filter will end up being ~3 dB of insertion loss. That is not a bad trade-off if it gets your receivers working. Even though there appears to be very little interest in the filters now (possibly due to the off-topic posts), I understand that demand could grow. I'm not going to invest in more parts until we see how the new MicroP25RX-II design goes. That will determine the future of BTT.
- Status