What band are you dealing with and its frequency spread? The length of cable including the loop's effective length and the star's contribution should equal an electrical odd multiple of a quarter wavelength. Easiest way to verify the actual length required is to remove the loop from the can (or adjust it for minimal coupling) and put it in the star with a cable of approximately the correct length and use 2 of the other star ports to measure the resonant frequency of the loop and cable to see if it is in the mid of your frequency spread.
All the cables would then be the same length to the star if the loops are the same. The open stub would be close to the electrical 1/2 wavelength if required to make the output return loss symmetrical and better. I personally usually do not use a stub and try to make the cables work alone but then again, I am using an Agilent 8753ES network analyser that makes the job very easy. The stub is used if the frequency span is larger or the cables used are not the ideal lengths. The open circuit stub is trimmed to make the return loss symmetrical. Harold --- In [email protected], "skipp025" <[EMAIL PROTECTED]> wrote: > > Hi Gareth, > > > "Gareth Bennett" <garethb@> wrote: > > I am appealing to the wisdom of the group for information > > regarding Transmitter combining (Multicoupling). > > OK, we'll try to help. > > > I am interested in the maths and formula involved in cable > > lengths, coaxial stubs, etc for the "Star" style of TX > > combining. > > Yes there is math, but almost every combiner system is an as built > based to the overall system requirements and performance trade > offs. The cable lengths are based no relative odd 1/4 wave lenghts > with lots of things changing the actual length. > > > I have inherited a site that has been added to over the years > > with different equipment and filter styles. I need to re-build > > the TX (and RX) side to tidy it all up. > > The first think you'll need to do is make a spread sheet of your > tx and rx frequencies. > > > Each TX leg is like this... TX, Isolator, Cavity filter, > > 3/4 wavelength coax to star combiner... Nothing flash, > > just want to know the maths behind it and to fine tune > > the system.... > > There are rules of thumb for the combiner engineering, based > on a number of variables... first the frequecy list, second > the allowed mix problems are checked, third the materials > you have to construct the combiner, fourth the loss or performance > trades vs frequency spacing, fifth the actual equipment you'd > like to use. > > > Especially the Transmitter coaxial stub information that > > usually is found on one spare combiner port. > > Measure the existing stub for open or shorted end. It's most > often a harmonic trap/stub, 1/2 the length of the in band > quarter wave length. > > > Thanks for reading :-) > > No problama, it's hard to be specific without more information > from you. If you ask Telewave - Sinclair or the Andrew Decibel > RF Engineers about making a combiner, they'll want your frequency > list first. Sometimes you can make a single antenna do everything, > sometimes you split the antennas. In problematic frequency > combinations, you might need to have/include both additional tx > and rx antennas to avoid system gremlins. > > When a multi channel pre-configured tx combiner works right > out of the box, I would consider the owner and mfgr very > good and a bit lucky. Combiners are where I picked up the > "no free lunch" line. In close space combinations, you try > to keep the power loss to less than 60%. Also, your results > will probably vary... > > cheers, > skipp > Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/Repeater-Builder/ <*> To unsubscribe from this group, send an email to: [EMAIL PROTECTED] <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
