Back to the Antenna Systems Index page Back to Home

Introductory Information about Transmitter Combiners… Compiled, written and HTML'd by Mike Morris WA6ILQ

Comments, critiques and suggestions for this page (or any page at this web site) are be appreciated.

There is some confusion over the naming and functionality of Duplexers, Diplexers and Combiners (even Wikipedia gets it wrong occasionally). A duplexer is an in-band device, a diplexer (or triplexer or quadplexer) is a multiband device, and both are covered in different sections on the Antenna Systems page.

What is a Combiner?

The term "combiner" is frequently used in the consumer world differently than in the Land Mobile Radio (LMR) world… Your author has seen it used in the consumer world to describe a situations where two roof-mounted television antennas (pointed in different directions) are "combined" to one feedline that runs into the house and feeds to one (or more) television receiver(s).

This article is concerned with transmitter combiners that allow two (or more) LMR transmitters on different frequencies into a single transmitting antenna without causing intermod or damage to the transmitters. The LMR combiner is a single band (also referred to as "in-band") device and are usually used in LMR with transmitters in the 5 to 500 Watt area. Each combiner consists of a system of cavities and isolators that allows multiple transmitters to share one transmitting antenna while preventing the output of one transmitter from feeding into the output of another. A combiner is not a duplexer and does not connect to a receiver, however I have seen a duplexer used as a two-transmitter combiner. A combiner provides isolation between the different transmitters while having a minimal insertion loss to the antenna.

The overall configuration of a two-port (two transmitters, one antenna) cavity-based combiner resembles a "T" (just like a duplexer) with the transmitters connected to the two sides and the center point is connected to the transmit antenna. Combiners with more than two ports resemble a star where the transmitters feed the legs and the center point of the combiner connects to the feedline and then to the transmit antenna. Note that the antenna has to be resonant to the entire range of transmtter frequencies. Some antenna designs are narrow, some are broad. The presence of a combimer will limit your choice of transmit antennas.

Picture a conventional transmitting system: a transmitter feeds an isolator (or circulator), then a pass cavity, then a feedline then the transmit antenna. The pass cavity after the isolator is NOT optional.

Then circumstances change and now require two (or more) transmitters to share that same single transmit antenna. The configuration now consists of multiple transmitters, each with an isolator and a pass cavity. The outputs of the pass cavities are coupled to a common point by precise length coax jumpers (the same sitaution as intercavity jumper cable length in a duplexer). This jumper length involves the V_f of the cable AND the frequency (which translates to wavelength, and that translates to the cable length). Any error in the length of that cable (due to the wrong V_f or wrong measurements) will result in a mismatch, and that is likely to create higher Insertion Loss and poor Return Loss.

This jumper cable length has bitten people in the rear when they move a duplexer or a combiner from 460 MHz (or higher) to 440 MHz. In some cases that difference requires longer intercavity jumpers for optimum performance. The page maintainer works a couple of days per week for a shop that maintains 30+ repeaters on 460 MHz and 40+ repeaters on 470 MHz that are part of a 8 site trunking system (from 10 to 19 repeaters per site). Several sites have combiners ranging from 2 to 8 ports. Plus a few sites ahve a UHF amateur system on one port of the combiners. The jumpers for 440, 460 and 470 are all different lengths.

You will want to use high quality cable for those combiner cables (please, no RG-8, RG-213 or LMR-type cable!). Mil-spec RG-400 and RG-214 or similar is very common in combiner cable harnesses.

As I said above, a conventional transmitting system has a transmitter feeding an isolator (or circulator), then a pass cavity, then a feedline then the transmit antenna.

In a combiner-based transmitting system the difference is the insertion loss of the combiner assembly. On a properly designed and implemented combiner it can be as low as 1.2 dB or as high as 3 dB. But at a crowded site that may be your only choice - having a repeater on the air through a combiner or not having a repeater at all.

Note that most repeater systems that use a combiner usually have a separate receiving antenna. I've seen shared VHF and UHF receive antennas that feed two or three repeaters and I've seen site master receiving antennas that are multicoupled to over 40 receivers.

I have equipment at several southern California sites that have such a master receive antenna system for UHF. The broadband (406-512 MHz) receive antenna is usually positioned on the top of the tower to maximize the receive performance and, by the way, maximise the receiver to transmitter isolation. The receive antenna feeds either 7/8 or inch-and-five-eights Heliax™ feedline which connects to a high-level AngleLinear preamplifier inside the building, then that feeds a high-level splitter and each splitter port feeds an eight cavity window filter. At most sites one window is at amateur UHF, another is for the 455-460 MHz commercial repeater receivers, similarly another is for the 465-470 MHz commercial receivers, a third at 473-476 MHz commercial, and the last one is for the 508-511 Mhz commercial receivers. Different geographical areas have different 3 MHz wide windows between 470 MHz and 524 MHz. The output of each window filter feeds an AngleLinear distribution amplifier and that feeds a number of system receivers. This situation is covered in more detail on the System Engineering page.

On VHF things are very different. The repeaters have inconsistent (almost random) offsets and are scattered throughout the the 132-144 MHz government band and the 148-174 MHz commercial band. Likewise in the 406-420 MHz government band.

Don't forget that the antenna that the combiner is feeding has a finite power limit. By the time you combine (for example) five transmitters, each with 100 watts out, through a combiner that may have as much as 3 dB loss, that's still 250 watts that's going up the feedline to the antenna, and (poof) goes the under-rated antenna. And remember that after the antenna went (poof) the SWR just skyrocketed and the combiner needs dummy loads that can handle the full transmitter power in continuous duty, and appropriate ventilation to get rid of the heat. There are combiners with more ports, and I have seen multiple installations that required an antenna rated at 500 watts. Not many manufacturers make those, and the ones that are available are usually EXPENSIVE and sometimes are a special order and can have extended delivery times (i.e. sometimes are made to order).

So why do people use combiners? Because a combiner system just may be the only way to accomplish what is needed. It may be that there is a limited amount of tower space, or maybe the tower is close to maximum on wind / ice load. If a combiner is the only way to do the job then you engineer the system so that the overall system RF power budget can handle the insertion loss and have the system still be usable. The most common UHF method is to use a community receive antenna that feeds the receivers on multiple systems then combine anywhere from two to twelve transmitters into one transmit antenna.

Let's go into specifics: At one site that I visit semi-regularly the site owner has community receive antennas at the top of the tower… one for 222 MHz, one for UHF (406-512 MHz) and one for 800 MHz. On my first visit (over 20 years ago) I noticed that the 222 MHz antenna was feeding four systems through an AngleLinear High Level amplifier and a 5-port splitter. On that visit I observed that the UHF receive antenna was feeding over 50 systems through a more complex distribution amplifier system. The site owner uses multiple UHF transmit antennas rated at 500 watts, and multiple five-port combiners each feeding a separate transmit antenna. The combiner ports are rated at 150 watts each, he conservatively limits transmitter power to 125 watts. Using this design five transmit antennas would support up to 25 transmitters. Eight transmit antennas would support up to forty transmitters.

There is some system engineering involved with combiners. Some of the concerns are obvious (like power input per port), some are not. There is a minimum frequency separation amount between transmitters on a single combiner, the actual minimum depends on the design of the combiner, typically 75-100 KHz at VHF and 100-350 KHz at UHF. An example of 100 KHz is here. Other UHF units have a 250 KHz KHz minimum spacing (like this one). There is a maximum as well. You need to check the manufacturers data sheets. Plus you have to consider the bandwith of the transmitting antenna.

You MUST NOT have the difference between any two two transmit frequencies in a combiner be equal to any repeater offset at that site! and if there are multiple buildings at a site you need to know what is in the other buildings! As I said above the repeater offsets are all over the place on the 148-174 MHz (commercial VHF), and the 132-144 MHz and 406-420 MHz government bands. Yes, a combiner with two transmitters (on any band) that are 600 KHz apart and transmitting simultaneously will nail any 600 KHz offset 2M repeater at the site… And there are several cities in the USA that have AM broadcast stations right on 600 KHz… In that situation all you need is one transmitter and a mix opportunity. And I know of another site that is poisoned to 220 MHz repeaters because there is a 1600 KHz AM broadcast station less than a quarter mile away. The same thing would happen if you had a combiner with two transmitters that were 1.6 MHz apart on any band. And yes, any two commercial transmitters 5 MHz apart (no matter what band) can nail every UHF repeater at a site. More on that situation below…

And some areas have commercial repeaters between 470 MHz and 524 MHz with a +3 MHz offset, so if you are in one of those areas you also have to watch for two transmitters exactly 3 MHz apart on a combiner.

And since the commercial tenants at a site pay the bills the amateur systems (if any) have to fix the situation (or live with it).

A "live with it" story…
Back in the 1980s the author was told a story about a situation in Texas where two FM broadcast transmitters on the same tower were exactly 5 MHz apart… This poisoned the tower for anything on UHF commercial LMR. As the story went Motorola went so far as to offer to fully fund a frequency change of either one of the FM stations. Both station owners said, essentially and politely, forget it. As a result there are UHF systems (both commercial and amateur) on that tower but they are NOT on a 5 MHz offset.
Supposedly that situation still exists today. If anyone has any details on that tower situation (such as the location, frequencies involved, etc) please let the page maintainer know and he will update this paragraph with the details.

Another "live with it" story… This one does not involve a combiner but it is still relevant about what happens with two transmitters and a mix opportunity.
Your author has commercial and amateur equipment at a major commercial site where many years ago two transmitters (in two different buildings) mixed, and the mix product killed every single repeater with a ±5 MHz offset in over a dozen buildings, on both the UHF commercial and amateur bands.
The story was complicated:
a) The ham repeater owner also didn't have all the information but would have forseen the problem.
b) The two meter coordinator didn't have all the information but would have forseen the problem.
c) The paging company was a long established tenant, and had a clean transmitter. Therefore the the ham system, being the new guy, had to fix the problem.

The rest of the story: The site has a dozen buildings across a ridgeline that is 3/4 of a mile long. The ham coordination group knew about most of what was up there, including the 152.24 MHz, 152.48 MHz, 158.70 MHz 162.6625 MHz and 163.25 MHz paging transmitters, but did not know about a paging system on 152.21 MHz. They coordinated a 2 meter repeater that followed the band plan and listened on 147.81 MHz and talked on 147.21 MHz. The ham system owner also had no idea that the 152.21 MHz transmitter existed (he had a background in commercial 2-way and would have been able to predict what was going to happen!).

His 147.21 MHz system was built, began testing and worked fine. There was only one issue: when it and the 152.21 MHz paging transmitter happened to be keyed up at the same moment. The resulting 5 MHz mix product itself mixed with every UHF repeater output and created products +5 MHz and -5 MHz. The physics didn't care if you were low-in/high-out or high-in/low-out.   EVERY single ±5 MHz UHF repeater in a dozen buildings up and down the ridgeline jammed itself and instantly became useless… but only for the duration of the simultaneous transmissions from both the 2 meter ham system and the paging transmitter. This was in the days when the users were a mix of older crystal controlled radios and newer synthesized radios. Hence swapping frequencies with another ham repeater was not practical (some users of both systems would have to buy a new seet of crystals).

The immediate (and final) solution? The ham system owner flipped his frequencies and used the problem side of his pair (147.21 MHz) as the input. That worked for many years and the user community grew to several hundred regular users.   In the 1980s cellphones became popular and decimated the paging industry. As a result a number of the paging channels went dark.   When the 152.21 MHz transmitter shut down the intermod situation went away. The repeater owner at that time left things alone for a number of years because there were still a lot of crystal controlled radios in use and everyone had crystals on the "upside down" pair. After synthesized 2 meter radios had been around long enough that the crystal radios had been retired the repeater was flipped "right side up". And finally, after 30+ years of operation the repeater matched the band plan.

Back to combiners:
When you get right down to it, a 1.2 to 3 dB loss in the combiner may not be that bad. In many cases the insertion loss of an isolator / circulator, the duplexer and the feedline on a conventional configuration repeater can be equal or greater than that (especially if the feedline is LMR-series cable (or even 1/2 inch Heliax™ instead of 7/8 inch).

As an example, let's look at one UHF system I am familiar with… it is at a different mountaintop commercial site (ham-owned) with separate master receive antennas for 220 MHz, 440 MHz, 902 MHz and 1200 MHz), each feeding multiple systems.
However we are discussing a UHF system:
1) The UHF receive antenna is a 10 dB UHF Super Stationmaster at the tower top, which is at 120 feet (36.5m) above the ground.
2) The receive feedline is 150-160 feet (45-60m) of inch-and-five-eights Heliax™.
3) Due to the high RF levels at the site (over a dozen buildings) there is a AngleLinear High Level Preamp between the Heliax™ and the ham band window filter.
4) The window filter is 8 cavities in series which make a several MHz wide "window".
5) The output of the window filter feeds an AngleLinear distribution amplifier that feeds over a dozen UHF amateur systems in the building.

The actual UHF repeater system:
1) The system receiver is a MICOR that was rebuilt from 460 MHz to match the factory 440-450 MHz design (this information is on the MICOR page at this web site), and that resulting receiver was peaked for the single frequency of the repeater.
2) The 330 watt transmitter is also a MICOR (retired paging transmtter PA deck). There is a backup 100 watt PA deck that can be switched in by a logic output on the repeater controller. That function switches a pair of Dow-Key RF relays plus an AC relay that switches the AC power to whichever amplifier power supply is active.
3) The RF relay on the amplifier output(s) feeds one port of a combiner. The insertion loss of the combiner is very close to the loss found in the previously used configuration (an isolator plus pass cavity plus duplexer plus about 100 feet of feedline). The combiner feeds another run of inch-and-five-eights Heliax™ which feeds the actual transmit antenna (which is an 8 or 9 dB fiberglass stick at about 80 feet (25m) on the tower. And the tower is at the peak of a 5500 foot mountain.

Combiners are expensive not only in RF energy but also in money (unless you have a source of good used / recycled cavities and isolators). A complete two port turnkey UHF combiner at new retail prices (these prices are from a 2009 quote) will set you back about US$4,600, a four port about US$8,400, an eight port about US$16,000, and that's plus shipping, installation, and final tweaking.
See these manufacturer web sites (and don't forget to ask - some manufacturers have amateur radio discounts, some do not):

• Bird Corp's "Combiner and Receiver Multicoupler Design" document   55 KB PDF (local copy)
• db Spectra transmitter combiner products (off page link to the db Spectra web site)
• Sinclair transmitter combiner products (off page link to the Sinclair web site)
• Sinclair trunking combiner products   254 KB PDF (local copy)
  Trunked radio systems frequently have anywhere from two to thirty transmitters close-spaced at one site. This Sinclair writeup is worth reading for the background information. One of the sites that the author visits semi-regularly has twelve trunked UHF channels (100 watts on each channel) inside a 1.4 MHz chunk of spectrum. Six 2-port combiners allow twelve channels on six transmit antennas. Another site has six trunked UHF channels (90 watts on each channel) in 350 KHz of spectrum and on three 2-port combiners. A third site has five 90 watt UHF transmitters combined onto a single antenna.

The antenna fed by the combiner also has a maximum frequency bandwidth (the point where the SWR becomes intolerable). Plus, you have to really look at the potential for intermodulation and mixes - and don't forget, you have multiple receivers to worry about! At the sites mentioned above, if they were to add one more transmitter to any of the combiners the site manager has to worry about over 50+ receivers at the site (and there are multiple tenants in his building) and just on UHF there are receivers from 406 MHz to 511 MHz. Remember the articles on receiver isolation at this web site? Go re-read them, but now realize that this site manager has to worry about 50+ UHF receivers! Plus the base frequency and the 3rd harmonic of every VHF system! Plus the low band systems (Red Cross on 47 MHz and another building down the road has California Highway Patrol all across 39 to 44 MHz).
And there are a dozen buildings on that mountaintop, and one building is operated by a TLA (three-letter-agency) that is very closed mouthed… And every building owner and site manager on that mountaintop has every other owner / manager in his phone book…

The most unusual transmitter combiner I am aware of is one on Mount Wilson, north of Los Angeles. That site is home to 90% of the FM and TV broadcast transmitters and has a "footprint" that covers 15 million people (around 37% of the population of the entire state of California… and remember that Los Angeles county has more people than 43 of the 50 states). The combiner was custom designed for the site and combines seven digital TV transmitters (channels 32, 44, 48, 49, 50, 51 and 56) into one custom designed antenna array that has to handle from 580 to 756 MHz. And that's not the only TV transmitter combiner on Mt. Wilson.

And there are a number of combiner articles on the page here.

Contact Information:

The author, Mike Morris WA6ILQ, can be contacted here.

Text, layout and hand coded HTML © Copyright 1995 and date of last update by Mike Morris WA6ILQ
This web page was split from the main Antenna System page on 12-Nov-2011.

This web page, this web site, the information presented in and on its pages and in these modifications and conversions is © Copyrighted 1995 and (date of last update) by Kevin Custer W3KKC and multiple originating authors. All Rights Reserved, including that of paper and web publication elsewhere.