Tuned splitter as cheap bandpass filter

[nanZor]

I've read the sticky threads about combining or splitting feeds, but it got me thinking about just using a tuned-port TV splitter as a simple bandpass filter by not using one of the ports, or just terminating it.

I happen to have a Radio Shack #15-2586 handy doing nothing so I thought I'd test it out. It has a separate VHF and UHF output, and one input. It is NOT just the simple wideband 2-port splitter. I'm too embarrased to show you what it looks like attached to a handheld.

From the best I can determine from other A/V forum threads, it probably has about 4 db loss on each of the ports, and I can't determine the isolation - however I'm not concerned too much since I'm terminating one of the unused ports anyway and just utilizing the bandpass function.

(and 4db was just enough to get rid of some heterodynes on my favorite tower channels that I had to formerly lock out - rather than go overboard with the built-in 20db scanner's attenuator - so I might live with this 4db loss everywhere since those towers are so important to me)

My guess from reading is that this typical unit has a 5-216 mhz bandpass for VHF, and a 470-1100 mhz bandpass for UHF. I'm only guessing since I can't find specs nor have the equipment to test it out. No amateurs are currently active, so maybe it will work at 440 - I'll try it during the daytime.

There is a definite split:
480-860 can be heard only on the UHF port.
42-165 can be heard only on the VHF port.

I haven't listened long enough to determine how bad (or good) the bandpass slopes are.

So I guess in a pinch, instead of using really custom notch/bandpass filters, if you have one of these laying around and are getting hammered from a cell-tower when listening to a vhf-dedicated scanner, or maybe suffering from some sort of vhf paging / FM broadcast interfering with your dedicated UHF scanning setup, this might do the job.

You might even be able to pop the cheaper #15-1296 into service for a vhf bandpass. I got a feeling you get what you pay for, but sometimes you have to use what you have!

Obviously a dedicated high-quality notch or bandpass filter custom tuned for specific needs is the real ticket, but I found this to be a fun test.

 

[nanZor]

Update:

The VHF port can pass up to the amateur 220 mhz band.
The UHF port passes down to the amateur 440 mhz band.

As suspected, terminating the unused input port is very important. So is not chaining a bunch of barrel adapters. F-F male is just as bad as any slide-on connector when it comes to leaking 800mhz especially.

I also tried using this as a tuned combiner for two different antennas - a general purpose vhf and the RS 800mhz duck on some short 3-foot jumpers. Even tuned, it had the problems mentioned in the other sticky forum threads.

So essentially I'll use this when needed as either a high-pass or low-pass filter hinging somewhere around 300mhz center with high-quality direct connections and the unused tuned port terminated.

 

[kb2vxa]

"As suspected, terminating the unused input port is very important."
Yes it is, terminating unused lines/ports brings the SWR back into acceptable values which reduces or eliminates signal leakage via common mode currents.

"So is not chaining a bunch of barrel adapters."
I sure wish we had a Snopes for this sort of urban legend, nobody listens when this fallacy is explained.

"F-F male is just as bad as any slide-on connector when it comes to leaking 800mhz especially."
I talk to the wind.

"So essentially I'll use this when needed as either a high-pass or low-pass filter hinging somewhere around 300mhz center with high-quality direct connections and the unused tuned port terminated."
Sounds like a plan.
Now if you're resourceful you can make your own terminations out of F connectors, 1/8W metal film resistors and pot them with epoxy making neat, waterproof, and professional looking little units.

 

[prcguy]

Modern F connectors are not as bad as you think. Canare has been producing very high quality F connectors for many years and they are the standard of the satellite broadcast industry. Compression type F connectors used by most cable TV companies and satellite dish installers test as good as the Canare units and have very respectable VSWR and insertion loss to 2GHz or more. Most male to male F adapters are not so great but the 3GHz rated female to female F adapters are just fine. These will usually have a blue dielectric and there is another color that designates 3GHz rating but I forget which.
prcguy

 

[Uplink]

PRC and Hertz, you guys are RF GURU's, all your posts on RR are what make this forum kick a$%^&*!!! HAVE A HAPPY 4th, DON'T BLOW ANY OF YOUR FINGERS OFF!!!

 

[prcguy]

I will never be a Guru, I refuse to wear that little red dot on my forehead.
prcguy

PRC and Hertz, you guys are RF GURU's, all your posts on RR are what make this forum kick a$%^&*!!! HAVE A HAPPY 4th, DON'T BLOW ANY OF YOUR FINGERS OFF!!!

 

[nanZor]

No guru here either - just a confirmed tinkerer!

The good news about the RS splitter is that there doesn't seem to be such a drastic attenuation factor of 4db or so. With no real equipment, I can't verify it with real specs, but once again the insanely weak VOR station was used for testing in and out of circuit.

This VOR station is so weak that I have to have the squelch open, and my ear pressed to the speaker to hear it with my eyes closed. I can barely make out the "LHS" morse ID. I can almost make it go away just by thinking about it.

With the makeshift bandpass filter inline on the VHF port, and with good direct connections, I can still just barely hear it.

Putting TV-quality gear into place for scanning always makes me cringe, but this might do in a pinch.

Re the adapters: I was actually happy to witness firsthand just how poor they were when testing this thing out. It would be hard to really tell with extremely strong signals, and in fact the wide-open preamp on the Pro-164 (mislabeled "ATT" button) on 800 mhz trunking actually helped prove it. I'm saving this for another thread....

 

[krazybob]

There is another approach that you might consider.

At SBN - The Scannerbuff Network - we use a Diamond MX2000 and MX3000 triplexer reversed. This gives tuned ports on 30-100, 100-175, and 200-1000. We use a separate antenna for low band (CHP), VHF (CALFIRE and others) and UHF or 800mhz. This is obviously not for a handheld installation But for a base it kicks the snot out of a "scanner" antenna. Because these are tuned ports images and birdies tend to disappear. But because we are using gain antennas tuned for a specific band we cannot run an amplifier and sometimes have to add 3-20dB of attenuation using F barrel attenuators. The net result is still better than a discone or other all-band antenna.

There is nothing wrong with using F connectors. They are just BNC connectors in a different form. I use compression connectors by Ideal and have jars of connectors, coax strippers, and the compression tool. They give a solid connection that is waterproof. For serious feeds I use LMR-400 and UHF or N connectors, but I also use RG6 Quad shielded 75 ohm cable. Up through 512mHz its loss isn't bad. Climbing towards 800 it becomes questionable except for strong signals. I also use CATV splitters good though 3GHz and feed 8 radios. The scanners automatically choose the best antenna (path of least resistance.)

 

[nanZor]

Those Diamond MX2000 / 3000's look interesting. Nice to hear that they work well in reverse. Great - another thing to add to my list to look at when visiting the candy store. That's right up my alley since I like building my own antennas specific to each band...

Here I go again - for a guy that doesn't like hanging out in the TV section of Radio Shack, I seem to be spending a lot of time there.....

Found a RS #15-579A "TV Interference Filter" which is just a 54-900 mhz bandpass filter and has a nicely crimped f-connector on a pigtail. Does it work? Seems to. It does attenuate the 42mhz low band systems that I monitor, and probably does so to some extent on the amateur 6-meter band. However I have no way of giving figures to going from full-scale on the 164's s-meter down to 2-bars at 42 mhz. But I guess if one is getting hammered from an HF or low-band VHF station, this works to provide at least some attenuation down there. It *might* help if you have a long run of feedline acting as an HF antenna in common-mode that isn't choked with ferrites, etc.

But how is the insertion loss? Well, now at least with the tuned splitter AND this "tv interference filter" attached to the input of the splitter, I'm detecting some insertion loss. The good news is that after about 2 minutes of listening to the originally weak vor station, it is still there - but just barely. Took awhile to convince myself that I am actually hearing the morse-id, and not just imagining it. So it stays inline. My high-vhf operations still seem fine.

Essentially with these two units ganged together, I have a 54-250 mhz bandpass filter on the vhf port without too much insertion loss to affect my operations. So it will stay inline for the time being.

Again, without any test equipment, this is all by ear and no specs to go by.

Fun!

 

[smason]

Putting TV-quality gear into place for scanning always makes me cringe, but this might do in a pinch.

QUOTE]

I don't have the specifics handy, but the output of modern satellite LNBFs that feed the receiver are up in the gigahertz range.
Good quality, properly installed F connectors and splitters (in some rare cases, usually multiswitches are required) seem to do the job quite well in that application so I have no qualms using them from 30-950 MHZ.

 

[nanZor]

Yessir - all those F - F male adapters have now been replaced with 1-foot f-f jumpers. A little bit more cable management, but the results are worth it.