Up one level Antenna Systems Index
Back to Home
  Coaxial Cabling Lengths between the Duplexer and the Repeater versus a "Z-Matcher"
By Eric Lemmon WB6FLY
HTML'd and edited by Mike Morris WA6ILQ
  Print this Page

A question was asked on the repeater-builder mailing list:
"Anyone have any idea what length of coax I should run from the transmitter and receiver to the duplexer?"

A reply from Eric Lemmon WB6FLY:
It depends.   If your transmitter power amplifier is exactly 50 ohms source impedance (not likely) and your duplexer input cavity is exactly 50 ohms load impedance (not likely) and your interconnecting coax is exactly 50 ohms (pretty likely if it's good quality), then the length of the coax is irrelevant.   Ah, but we know that nothing in this world is exact, especially PA source impedance!

Perhaps the best course of action is to make up the interconnecting cable to neatly and conveniently fit the installation, and see how it works.   If the power to the antenna, measured with a thruline wattmeter at the antenna connector of the duplexer, is about what you'd expect after the loss in the duplexer and cables, then you probably don't need to do anything.

If you determine that the power to the antenna is significantly less than you think it should be, then you might try installing an impedance matcher ("Z-Matcher") right at the transmitter output connector, and see if you can tune the PA for the expected output.   If that makes a huge difference, you can either leave the Z-Matcher in line and get on with your life, or you can experiment with varying lengths of cable to find the length which transforms the impedance to a value similar to that created with the Z-Matcher (if you DO use a critical length, please label it as such for the next person that works on the repeater).   My response here is addressing only the cable from the transmitter to the duplexer, but Z-Matchers have been used on the receive side as well, although the effects are not as profound.

Both TESSCO and Talley Communications sell impedance matchers. The ones made by EMR (www.emrcorp.com) and by Telewave (www.telewave.com) are similar in construction, performance, and price (at the time of this writing (2004) about US$70).   I think they are fairly equal in quality and performance.   The photos below are of an EMR VHF matcher.   The UHF unit looks similar except for the coil.

73, Eric Lemmon WB6FLY

It's not obvious in the above photo but the brass disks are screw-off caps that expose the pistion trimmer cap adjustment screws.

Editors Note:
The impedance matching provided by a magic length of coaxial cable or by a Z-matcher is required since few solid-state PAs present a 50 ohm source. Some come very close, and the best are far more forgiving than the majority of aftermarket amplifiers. Tube-type PAs generally do not have this matching problem, because they have tuning adjustments to match the load.
Wacom discusses optimizing the cable length between transmitter and duplexer in their duplexer tuning writeups.   One of those is here:

The information on cable lengths starts on page 4.   Despite what they say, DO NOT use RG8 or RG9 - use something better.   And note that by using a Z-matcher you're not fixing the problem transmitter (you probably can't), you're just making the transmitter end of the coax look like a 50 ohm device to make your duplexer happy since the tuning curves of the duplexer are to some degree sensitive to the impedance it "sees" at its I/O ports.   So consider the Z-matcher as a new permanent part of the transmitter, and label it as such (something like "Do Not Remove Z-Matcher From Transmitter ‑ See NNN" (where NNN is the knowledgeable person's name).

Tuning the Z-Matcher: (more from WB6FLY)
Bearing in mind that the basic function of an impedance matcher is to add a tuned output circuit to the power amplifier, the first step is to tune the PA with the Z-matcher connected as close to the PA as possible. Use a good RF power meter, such as a Bird 43, and terminate its output at a good 50 ohm load. If the power climbs above the rated value as you tune the transmitter just reduce the drive level until you have achieved the best possible match (and efficiency) at full rated power. What you have done at this point is to optimize your PA for a purely resistive load.

For maximum benefit the impedance matcher should always be permanently attached as close to the transmitter PA output connector as possible. The insertion loss will be in the range of 0.15 dB.

Ah, but the "real world" is hardly ideal, and duplexers are no exception. Because duplexers are resonant devices, there will be some reactance seen by the transmitter when it is connected to the duplexer's TX input. One of the ways to minimize (it cannot be totaly eliminated) this reactance is to always have a 6dB attenuator on both cables of the network analyzer when tuning the cavities. The attenuators also prevent ringing in the cables when the cable length is harmonically related to the wavelength of the signal. This ensures that the tuning of the cavities is based on 50 ohm inputs and outputs. This is perhaps an oversimplification, but it's close!

The next step is to connect the PA to the duplexer, with the Z-matcher in place, using a non-magic length of 50 ohm cable. Then, attach your power meter and dummy load to the antenna connector of the duplexer. Key the transmitter, and fine-tune the Z-matcher for maximum forward power. If your antenna system is known to be a good match, you can connect it and relax.

It is interesting to watch the PA current while tuning the Z-matcher. While the current will usually be at minimum at the point of maximum efficiency, it doesn't always tell the full story. There are many ways to build a PA, and some of them are stable but not very efficient, and some are the opposite. To protect the PA, you want to limit the power dissipation to a safe level. Therefore, you must lower the drive level, which lowers the current, and also the power output. Moreover, always be ready to reduce the drive level when you are tuning a PA with an impedance matcher.

Very important! When tuning an impedance matcher the wattmeter is inserted between the duplexer and the antenna, not between the impedance matcher and the duplexer's transmitter input. I'm assuming here that your antenna and feedline are a good match to the duplexer, so the power output should not change when you remove the wattmeter. Actually, that's one way to confirm that your antenna and feedline are a good match: Use a Bird 43 or equivalent wattmeter connected at the duplexer's antenna connector, to measure the power going into a dummy load right after the wattmeter. Then connect the antenna feedline in place of the dummy load. Ideally the forward power will be nearly the same. If there's a lot of reflected power then you have an antenna problem to investigate.

Editor's Note:
The basic function of an impedance matcher (the "Z-Matcher") is to add a tuned output circuit - the one that the equipment designer/manufacturer left out - to the power amplifier final stage.
Start by tuning the transmitter (with the Z-matcher) into a dummy load, and optimize the output power and current draw (the overall transmtter efficiency). Then add the jumper cable to the duplexer and the pretuned-to-50-ohms duplexer. Put the thruline wattmeter AFTER the duplexer and before the dummy load. Verify the system. If everything is 50 ohms it should not need ANY tweaking. Now remove the jumper cable and dummy load and connect the feedline and antenna.
You do not put the thruline meter between the z-matcher and the duplexer because the change in electrical length (the presence then the absence of the wattmeter and its jumper cable) will change the tuning. Put the meter after the duplexer and before the dummy load. Note that changing the drive level of an amplifier stage changes it's input impedance, and the output impedance is affected by the output level.

The benefits of a Z-Matcher over precise cable lengths: (more from WB6FLY)
You can (effectively) transform one impedance to another by cutting the interconnecting cable to a specific length (or multiples thereof) through a process of cutting, reterminating, testing, cutting some more, reterminating, testing, etc.
Or, you can simply connect an impedance matcher and accomplish the same result in a very short time. I'll admit that sometimes I get lucky with the TX to duplexer cable, and the impedance matcher makes no worthwhile gain. In that rare case, I leave it out. I'll also admit that sometimes the impedance matcher doesn't reach an optimum match within the range of its adjustments; in that case, I probably need to put the PA on the bench and find out why it's not performing properly.

A case history:
My local radio club has a Packet Node with both VHF and UHF ports, and a Yaesu FT-1500M 2m transceiver is used for the VHF side. It feeds a Decibel Products DB-224E 4-bay dipole antenna that has a very good match at 50 ohms and a 1.05 VSWR at 145.030 MHz. We set the power to about 10 watts, thinking that the radio would "loaf" at that setting, and the power was appropriate for the hilltop location. One day, the radio stopped transmitting, although its receiver was fine. A bench check proved that the M67781L PA module had blown, so another module (US$80!) was purchased and installed. While I had the repaired radio on the bench, I tested the radio through a Bird 43 meter into a dummy load. Hello! At the "low" power setting, the radio was very inefficient, the module was very hot, and the source impedance of the M67781L module was far from 50 ohms. In other words, the old module was toast because we ran it at low power! I then connected an impedance matcher and tinkered with the tuning and the drive level to find a happy medium. I was able to find a sweet spot where the ratio of power output versus DC current was greatest (i.e. the best efficiency), and the PA module was running cool. In this particular application, cutting and experimenting with cable lengths to find a match is not very practical.

73, Eric Lemmon WB6FLY

Editors Note:
Tony King W4ZT has some information on the Z-Matcher that GE built into some MASTR II stations at the W4ZT Z-Matcher page.
The schematic to the GE Z-Matcher is here.
It's part of LBI-30201G.
DB Products also sells Z-matchers, and their tuneup info is here.

Back to the top of the page
Up one level Antenna Systems Index
Back to Home

Text by Eric Lemmon WB6FLY  4-Jan-2004 and 11-March-2005
HTML Copyright and Editing 3-18-2005  by Kevin K. Custer W3KKC

This page originally posted on 3-18-2005

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.