Splitter/Combiner with a twist?

[hanlonmi06]

Does such a critter exist that is oriented to the scanner hobbyist that could efficiently allow something like a multi band (discone) and a single band (yagi) antenna configuration to one radio?

All I've found so far are Ham radio duplexer/triplexer products that have freq specs oriented around the VHF/UHF ham bands. The problem I am attempting to solve is having the wide band features of a discone for monitoring a variety of things, and a simulcast issue on an 800mhz P25 system. I am literally within view of one of the simulcast towers and when I have a yagi pointed at that tower, the system comes in great, virtually no simulcast drop outs. The discone either goes def or has such bad distortion its just miserable on 800mhz. If there was a duplexer that had a DC-512mhz port A and then a 760-860mhz port B spec it could perhaps be a possible option to try, but no one makes such a thing that I can tell. Or am I missing something obvious all together?

 

[prcguy]

Older versions of this amateur diplexer work ok for what you want and they show up on Ebay cheap here and there. Diamond® Antenna ~ MX37 Duplexer

Here is exactly what you need but they are expensive new. I've picked one up on Ebay cheap but its rare. Microlab 80-520 698-2700 MHz 50dB Diplexer Filter (BK-24N) from Solid Signal

There are other similar commercial units, all expensive and the trick is to find them used surplus for a good price.

 

[hanlonmi06]

Welps, just goes to show that with enough time and money, you can do just about anything...

As for the Diamond, how much of an impact does that port starting in at 900 have down at the 760mhz range? If its "negligible" or "not too shabby" I think I have a comet CF-413 laying around, which is what got me started on thinking this train of thought.

 

[Ubbe]

You can get a $35 filter from Mini-Circuits that will pass DC to 550MHz and attenuate 30dB at 750Mhz and even more higher up in frequency. Put that in line with the Discone and then connect the yagi in parallell between filter and scanner using a T-connedtor. The yagi will be so ineffecient below 500Mhz as it probably will give no interference at lower frequencies.

If you would like to optimize and be sure that the yagi are fully isolated at lower frequencies and doesn't attenuate the signal at all, get another $39 filter for the yagi and put both filters directly on the T-connector.

Low pass filter
High pass filter

/Ubbe

 

[JimD56]

I use a Comet CF-413B for exactly what you need. I use Port A 0-460 mhz on a Diamond Discone and Port B 840-1400 mhz Yagi and it works great. Both antennas are in the attic.

 

[hanlonmi06]

I guess I was thrown off by the specs a bit, I tend to overthink things sometimes. I'm going to put the 413 in line and see how it goes (I forgot that I even had it during this whole time I've been fighting trying to get this problem narrowed down until looking for something else last night). Thanks for the replies, the mini circuits site likes look a cool site that I need to spend some time on!

 

[prcguy]

I have an older Diamond MX37 and it passes down to the low 800s ok but I think the newer ones are tuned a little higher. Making a diplexer from a Minicircuits 550MHz low pass and 800MHz hi pass works but its not perfect. I did this in a system where the filters were used as a diplexer after a preamp so its not that critical but there was a few dB of ripple within each pass band. I had SMA connectors on the filters and used a tiny SMA T adapter to keep line lengths very short, but apparently its not short enough.

I guess I was thrown off by the specs a bit, I tend to overthink things sometimes. I'm going to put the 413 in line and see how it goes (I forgot that I even had it during this whole time I've been fighting trying to get this problem narrowed down until looking for something else last night). Thanks for the replies, the mini circuits site likes look a cool site that I need to spend some time on!

 

[Ubbe]

Mini-Circuit state 0-0,6dB loss for the low pass filter between 10MHz-520MHz and 0,3-0,4dB for the high pass between 700MHz-900MHz and you really need to connect them directly to the split without coax as the coax will work as a stub filter when a filter begins to stop the frequencies and go high impedance and could attenuate the signal up to 20dB depending of the coax lenght and frequency.

They have other types without connectors, and you may also pull out the circuit board from its enclosure, and fit in a box of your own to get short connections where the two filters connects together and then to an outside connector on your new metal box.

Mini-Circuits are a very generous company as I asked for some balun transformers and they sent me those free of charge and no shipping cost as "samples". The sale price where only a dollar or two a piece so no big cost for them but still very generous.

/Ubbe

 

[prcguy]

If you sweep an individual filter it will meet specs. Connect two together with a T adapter and it will react different due to crazy impedances presented to each filter.

Mini-Circuit state 0-0,6dB loss for the low pass filter between 10MHz-520MHz and 0,3-0,4dB for the high pass between 700MHz-900MHz and you really need to connect them directly to the split without coax as the coax will work as a stub filter when a filter begins to stop the frequencies and go high impedance and could attenuate the signal up to 20dB depending of the coax lenght and frequency.

They have other types without connectors, and you may also pull out the circuit board from its enclosure, and fit in a box of your own to get short connections where the two filters connects together and then to an outside connector on your new metal box.

Mini-Circuits are a very generous company as I asked for some balun transformers and they sent me those free of charge and no shipping cost as "samples". The sale price where only a dollar or two a piece so no big cost for them but still very generous.

/Ubbe

 

[hanlonmi06]

I got the 413 put in line tonight. Immediate results. I'm actually dumbfounded how well that worked!! I cant believe I had it sitting in a bin of odds and ends and forgot all about it!

As an aside, I am surprised that there doesn't seem to be something like this already in existence specifically targeted towards the scanner hobby.
Perhaps the economics of it aren't there if a Ham product can be repurposed.

 

[prcguy]

If someone would make and sell a good diplexer that covered 100 to 520Mhz on one port then 698 to maybe 1300MHz on the other port with low loss (under .5dB on the low side and under 1dB on the high side) and good isolation and under $100, I think it would sell very well.

In my opinion, for many scanner users, a Discone supplemented with a gain type 700/800 antenna is a perfect setup that makes the best use of the Discone and fixes its shortcomings of a lousy radiation pattern in the 800MHz range. It would also allow for options like a preamp for just the 800 side or both or other configurations.

I got the 413 put in line tonight. Immediate results. I'm actually dumbfounded how well that worked!! I cant believe I had it sitting in a bin of odds and ends and forgot all about it!

As an aside, I am surprised that there doesn't seem to be something like this already in existence specifically targeted towards the scanner hobby.
Perhaps the economics of it aren't there if a Ham product can be repurposed.

 

[Ubbe]

If you sweep an individual filter it will meet specs. Connect two together with a T adapter and it will react different due to crazy impedances presented to each filter.

That might be true if the interact in the cross over band. But the filters I suggest use a cap in series for the high pass and a coil in series with the connectors for the low pass and there's a "boarder" band at 600Mhz that doesn't get used. So no crazy impedancies with those filters, just going high impedance with a 20dB return loss so they interact nothing at all at the frequencies the OP wanted to use.

It's easy doing filters when there's a 520-750MHz band where you do not need to receive anything. The two filters and a T-connector are less than a $100 are are a perfect solution for most users that do not need to receive the terestial television band. You can also very easily put a low-noise $25 amplifier at the 800Mhz antenna to make it a $100 investment, being it a yagi or a vertical with gain to overcome any coax losses.

/Ubbe

 

[prcguy]

Back in July of last year I used a Minicircuits 550MHz low pass and 800MHz high pass to make a diplexer and I swept it and it has some unwanted ripple and loss as I mentioned. This is not speculation, its measured. Look at post #49 here and you can see in the lower left of the picture the filters are mounted right to an SMA T adapter with no cables. prcguy's mast mount filter/preamp/diplexer thingee

The only reason its acceptable in my case is the diplexer is placed after the preamps where the system gain and noise figure is set upstream. I would not use the same components in a passive system, I would use the excellent commercial diplexer I had pictured in post #1 in the same thread. Well, looking at how the tower top thing is wired up, when power is removed relays bypass some filters and the amplifiers, but the basic diplexer made with two Minicircuits filters are still there. Maybe I need to fit that big Microlab diplexer back in there somehow.

That might be true if the interact in the cross over band. But the filters I suggest use a cap in series for the high pass and a coil in series with the connectors for the low pass and there's a "boarder" band at 600Mhz that doesn't get used. So no crazy impedancies with those filters, just going high impedance with a 20dB return loss so they interact nothing at all at the frequencies the OP wanted to use.

It's easy doing filters when there's a 520-750MHz band where you do not need to receive anything. The two filters and a T-connector are less than a $100 are are a perfect solution for most users that do not need to receive the terestial television band. You can also very easily put a low-noise $25 amplifier at the 800Mhz antenna to make it a $100 investment, being it a yagi or a vertical with gain to overcome any coax losses.

/Ubbe

 

[Ubbe]

If you sweep an individual filter it will meet specs. Connect two together with a T adapter and it will react different due to crazy impedances presented to each filter.

Back in July of last year I used a Minicircuits 550MHz low pass and 800MHz high pass to make a diplexer and I swept it and it has some unwanted ripple and loss as I mentioned.

But you say it meet specification when sweeped, and that are 0,6dB loss at the most, so no problem with that low loss figure.
If you get other results when you connect them together you must have other issues not related to the filters but to the additional connections you have made. Or the filters you selected to use are not the best suited for your application if they interact at some frequencies you want to use.

/Ubbe

 

[prcguy]

I swept each filter separate and they meet spec. I connected the two filters to a very small SMA T adapter with one filter terminated while I swept the other and the two filters connected to the T adapter cause ripple and extra loss. This is with no other cables in the system except for the scalar analyzer cables that were calibrated out.

I think the setup is acting like an unterminated T adapter in the system because in the 100 to 550MHz range the port with the 800MHz high pass filter is a very high impedance and above 800MHz the 550Mhz low pass filter looks like a high impedance and in both cases the T adapter is a very short length of feedline and there are also some components in the filter that might add additional lengths.

You can get away with a small T adapter like this at HF but at 400, 500 or 800MHz its going to show up as a problem. For me, I don't like it but it may actually work better than the amateur grade diplexer rated at 460 and 900MHz.

But you say it meet specification when sweeped, and that are 0,6dB loss at the most, so no problem with that low loss figure.
If you get other results when you connect them together you must have other issues not related to the filters but to the additional connections you have made. Or the filters you selected to use are not the best suited for your application if they interact at some frequencies you want to use.

/Ubbe

 

[hanlonmi06]

So, without the test lab equipment to confirm system performance my derriere dyno tells me I gained 10 horsepower by adding stickers.

...by that I mean, what I was accustomed to hearing, and what I suddenly heard different and now better tells me that adding the duplexer in the antenna system was a net positive. I would love to have access to test equipment that could confirm the real numbers, but then again this is part of the joy of the hobby. Down the road, I may definitely experiment with some other components mentioned here and see where it leads. the replies have been super helpful!!

 

[prcguy]

In a few days I can sweep and post the results of a Diamond MX37 diplexer so you can see how it does above 460MHz and below 900MHz. It might be just fine but I suspect up to several dB loss at 512MHz and 700MHz. The 413 diplexer obviously working fine for you but its nice to know the details and if they are acceptable or you move on to something better.

So, without the test lab equipment to confirm system performance my derriere dyno tells me I gained 10 horsepower by adding stickers.

...by that I mean, what I was accustomed to hearing, and what I suddenly heard different and now better tells me that adding the duplexer in the antenna system was a net positive. I would love to have access to test equipment that could confirm the real numbers, but then again this is part of the joy of the hobby. Down the road, I may definitely experiment with some other components mentioned here and see where it leads. the replies have been super helpful!!

 

[Ubbe]

.... and in both cases the T adapter is a very short length of feedline and there are also some components in the filter that might add additional lengths.

If you look at the filters datasheet you'll see that the group delay, that will mimic the delay of a lenght of coax, are at it longest delay just at the filters frequency where it starts to attenuate. You probably had the filters interact too much by having their pass frequencies too close to each other. The filters I suggested do not have that problem but then there's no 550-700Mhz reception either, but that's only the televsion band that most scannerist doesn't want to receive and actually would like to surpress.

/Ubbe