This is a resend of my email to Nate last night. It looks like good old yahoo stripped my reply.
Hi Nate, A UHF pass band cavity for example will pass only a UHF frequency that it is tuned for. On frequency signals coming into it will see 50 ohms. Off frequency signals will see a short circuit and will be greatly attenuated. The input loop of the cavity (as well as the output loop) looks like a short circuit at all but the tuned frequency. So anything that happens to be in parallel with the loop will also see the short circuit if the frequency is not that to which the cavity is tuned to. So if you had a half wave length cable between the cavity and your T connector, then the short circuit at the cavity (off frequency short) would also look like a short circuit at the T connector. No problem for the UHF signal as that frequency sees 50 ohms at the T. but any other frequency sees a short circuit at the T and would be attenuated there. Now if that cable was a quarter wave length instead of a half wave length, the short circuit (off frequency short) would be transformed to an open circuit at the T connector. That would allow all other frequencies to be present with no attenuation at the T. If you used a random length of cable here, you may be ok and you may not be depending on how far away from a quarter wave length the cable happened to be. This is exactly how a duplexer works. The cables between the T and each cavity set is a quarter wave length at the opposite frequency for which the cavity is tuned to. The quarter wave length cable connected to the T always wants to see a short at the other end at the frequency that it does not want to pass, as the quarter wave length transforms the short to a open which does not load down the other side of the circuit.. With close spaced duplexers sometimes the two cables may be very close in length or the same, as the cable is not near as high a Q as the cavity is. Further to your current post, there is more of a chance of hitting a bad cable length than there is of not. Only a quarter wave length will transform short (low impedance) to a high impedance to the T. All other lengths will be something with lower impedance at the T, which you don't want. If everything is set up right you should be able to remove the "other" cable from the T and notice no difference on the remaining receiver. As far as passive splitters go rather than using the above cavity setup, it is always a good idea to use a splitter for isolation between receivers. The same thing can happen as described above (with no cavities) as the tuned input stage on the receiver will do the same thing to rob power from the other receiver depending on cable length between a T and the receivers. A splitter really does not rob any power from the system. If you think about it, each receiver is going to "pull" half of the power from the antenna line anyway. So just the fact that you connected the two together, the total antenna power is going to be shared between the two receivers if you are lucky enough to not have other losses due to impedance match problems sucking more than that out. So the 3 db loss that a splitter gives you is there anyway. But with the splitter everything sees the proper impedance and maximum power is transferred to each. UNUSED PORTS ON A RECEIVER MULTICOUPLER A receiver multicoupler by the way is basically just a multi way splitter. A TV splitter is built exactly the same. There are two different type of couplers used. One type the unused ports must be terminated in order to maintain 50 ohms at all other ports. The other type it doesn't matter if they are terminated or not as the port on this type is isolated by the amount of port to port isolation. Depends on the manufacturer. VSWR LOSSES ON RECEIVER ANTENNA LINE Yes you can have vswr losses on your receiver line going up the tower depending on the match that the line sees down at the BOTTOM of the tower. It works just the opposite of a transmitter line. The transmitter line requires a good match at the antenna end of the line in order to keep the vswr low on the line. This will keep lose to a minimum. But when the energy is originating at the antenna end of the line, then the match at the bottom of the line becomes important to minimize vswr loss on the line. A mismatch at the bottom of the line causes reflections (with part of the received power) to go back up the line and become attenuated by the regular line loss. At the top of the line that energy is re-reflected back down the line again and is further attenuated by the regular line loss. This re-reflected energy eventually adds to the signal (minus the amount lost). So the worse the match the more reflected energy there is going back and forth on the line and less ultimately reaches the receiver. The more original line loss there is the worse the problem becomes. This is another reason for paying attention to proper matching at the receive end. You end up with the same kind of loss that you get on a transmitter due to high vswr. 73 Gary K4FMX > -----Original Message----- > From: Repeater-Builder@yahoogroups.com [mailto:Repeater- > buil...@yahoogroups.com] On Behalf Of Nate Duehr > Sent: Wednesday, March 10, 2010 5:42 AM > To: Repeater-Builder@yahoogroups.com > Subject: Re: [Repeater-Builder] Re: Dual receivers on one antenna for RX > only site > > > On Mar 9, 2010, at 8:37 PM, n...@no6b.com wrote: > > > t most certainly does. Try random length cables from the cavities to > the > > T instead of 1/4 wavelength (like one local did several years ago) & > watch > > your sensitivity drop by over 20 dB if you're unlucky (as he was). > That > > mistake literally killed off a local radio club, as few of the members > were > > able to use the repeater following the addition of the T & wrong > cables. > > Thanks both Bob and Skipp for explaining that one odd-ball configuration > that would crush the receivers with random cable lengths that just > happen to hit the right "sweet spot" to do this. > > I suspect, that if someone saw a 20 dB loss while installing this setup, > they'd at least STOP and start asking questions -- maybe they wouldn't > "get it" that they'd hit this "perfect storm" combination -- maybe > they'd think they had some kind of receiver failure when it suddenly was > "really deaf" -- but I also doubt that *most* people would hit the > problem. > > Would you agree with that assessment? (Skipp's comment that if there's > a train wreck to be found, he'll be there... I know that feeling.) > > I guess what I'm saying here in a round-about way is... random cable > lengths really shouldn't be that much of an issue in a setup like this, > but yeah... agreed... once in a while it'll bite you like an alligator > (had to get that elephant/alligator theme in here, just one more time! > GRIN!)... > > I've seen lots of people get away with it. > > As far as the 3dB lost in a true broadband splitter -- also true, of > course, Bob -- at most of the sites where we have to share a receive > antenna with multiple rigs, the site measured noise-floor is so high the > 3dB doesn't have much of an impact... just keeping the local crud out of > the receivers is difficult enough -- sometimes that 3dB loss helps, > instead of hinders, so to speak. :-) > > I guess we should all probably also mention the evils of not terminating > all the unused ports on a multi-splitter with 50 Ohm loads, too... if > we're going to get this picky, right? ;-) > > -- > Nate Duehr, WY0X > n...@natetech.com > > facebook.com/denverpilot > twitter.com/denverpilot > > >
