Z-Match for 6 meters

Antenna Index page
Home page Z-Match for 6 meters
By John Haserick W1GPO

Background:

The Z-Match is an impedance matching network designed for use with any transmitter that is within the design frequencies of the device. The Z-Match allows the transmitter to operate at a maximum efficiency by matching the source impedance (transmitter output impedance) to the load impedance (antenna, duplexer, isolator, etc.).

Building One:

I have attempted to make many filters for 6 meters, but none have really worked without excessive loss. I know that the GE MastrII on a local 6 meter repeater required a commercially-built Z-Match between the output of the 100W PA and the Wacom Duplexer, so I thought maybe our 6 meter Micor repeater would also benefit from one. Mine (and probably the commercial version) is just a pi network with input and output variable caps to ground with a coil between the caps. I built it inside a 2"x2-1/4"x4" die cast aluminum box with two N-female chassis mount connectors, #12 enameled wire, and two variable caps, one 50pF and the other 100pF maximum. Click on any photo for a larger view.

The Z-Match coil itself is 5/8" OD, 4 turns spread about 1" long, with about 5/8" pigtails on each end. The #12 wire was close wound over a 1/2" tube then the turns were stretched outside the tube. With the N connectors disconnected, and the Z-Match tuned to 53 MHz, the coil grid dipped at 78 MHz. I would suggest two 75pF caps instead of the 50 and 100pF caps I used, because the 50pF cap ended up close to fully meshed, and while the 100pF cap was not too sharp in adjustment, a 75pF would be less, however I used 100pf to give the Z-Match more range of adjustment, if needed. I recommend a minimum of 500V rated caps for up to 150W and under 2:1 SWR, with 700V rating preferred for up to 250W. Also since the pictured caps are hard to find, 75pF variable caps with two mounting posts and a metal hole plug for shielding over the adjustment shaft slot hole in the aluminum box would be an alternative. Also Teflon insulated N connectors are preferred, otherwise the insulation may melt with soldering.

The tricky part was by trial and mostly error getting the correct inductance on the coil, and that required a sweep generator and return loss bridge. Originally I thought I'd just set the two caps at 35pF (to allow for tuning out different reactances and impedance ratios) and place in a coil that grid-dipped at 53 MHz. That produced a 6 dB loss and 2:1 SWR! I was about to give up, but I discovered the return loss bridge sweep looked good at about 27 MHz, so turns were removed and the coil spread until eventually I got 0 dB loss and 1.0:1 SWR at 53 MHz. For some strange reason the cap settings only changed a small amount. The position of the coil placed the bottom of the 4 turns (about 5/8" OD coil) about 2 mm from the inside of the aluminum case. I do not know if the SWR would have been higher if the coil was positioned more towards the center of the box rather than almost touching the case, but I thought it best to possibly mimic part of a 50 ohm cable between the N connectors.

Performance Results:

Connecting it between the 110W output of the PA and the coax to the duplexer actually increased the output of the duplexer from 89 to 92 watts, and dropped the reflected power to the PA from 2 watts to 0.5 watts, not a whole lot, but I expect it would have made more of a difference if there was a bigger mismatch. I believe the output of the PA also increased by a few watts without making any PA tuning adjustments.

I measured the difference in PA current with and without the Z-Match between the continuous duty Micor PA and duplexer. It turned out that the PA drew 14 more watts of input power with the Z-Match inserted to produce about 4 more watts from the PA, and for just 2-3 more watts out of the duplexer, so the Z-Match is affecting the PA tuning and efficiency. Retuning the output cap in the Micor PA made no difference, possibly because it was already at the minimum end of its adjustment range. I suspect the Z-Match is consuming 1-2 watts out of the 110 watts out of the PA, however reflected power from the duplexer to the PA was reduced from 1.8 watts to 0.5 watts by the Z-Match. The bottom line is that for the continuous-duty Micor PA, the repeater was better off without the Z-Match for our low SWR situation. This might not be true for a MastrII PA, or in a higher SWR situation.

Performance Update:

Bob WA1MIK suggested that I should try placing the Z-Match exactly 1/2 wavelength electrically from the Tx power sensor, about the equivalent of 5 inches in Teflon coax longer than the Tx BNC output connector. The Z-Match was relocated from the top of the repeater cabinet to the Tx port of the duplexer, and a new longer length of 1/2" Superflex was fabricated using the existing intra-cabinet coaxial cable and rack-mounted Bird wattmeter, plus a Teflon pigtail representing the coax inside the Tx, using a special set-up with the antenna analyzer. This set-up was a long random length of RG213 coax from the analyzer to an N tee, with a 50 ohm dummy load on one port and the 1/2 wavelength of coaxial cable with an open connector on the end opposite the tee connector end. The analyzer was swept to determine the frequency of 1.0:1 SWR and 0 Reactance, and the Superflex trimmed until the Tx output frequency was reached.

The results were much better with the new Z-Match placement and Superflex cable in that the reflected power could now be tuned out to zero reflected and the peak output from the duplexer followed the same Z-Match tuning for zero reflected, which was not the case with the original set-up. Also the Micor forward and reflected readings on the built-in power sensor tracked exactly with the rack-mounted wattmeter in the repeater cabinet. In addition, the Micor PA no longer lost efficiency when the Z-Match was inserted; in fact, about 2 more watts came out of the duplexer with the same PA input with the Z-Match in place (maybe that is where the original 1.8 watts of reflected power went.)

In conclusion, the Z-Match either needs to be placed right at the Tx output connector, or at multiples of 1/2 wavelength electrically away from it.

A Higher-Power Version:

The first version worked great up to 100 watts or so, but I needed something to handle up to 400 watts. Variable capacitors with a high-enough peak voltage spec were hard to find but eventually Bob WA1MIK pointed me to some on a popular auction site that I bought for $18 each. I made this version using a slightly larger enclosure.

Construction Details:

The inductor was made from #10 enameled copper wire, 9/16 inch OD, 5-1/2 turns. It measured 0.030uH out of the circuit. More details can be found above.

The variable capacitors are 100pF with a nylon extension shaft. These extensions were cut off and slots were cut in the brass portion of the shaft. The caps ended up about 40% meshed.

The box is a Hammond 1550D. Adjustment holes are 1/2 inch diameter for 1/2 inch metal hole plugs. With 1/2 inch holes, metal hole plugs should be used, especially if duplexing, and to keep out dust and bugs. It might work without much RF leakage just to drill the tuning rod hole just slightly larger than the nylon shaft.

The flat jumpers between the caps and coil were made from #5 braided Solder Wick. [From WA1MIK: A piece of braid from RG-400 or RG-214 coax would also have worked fine, as would a piece of copper flashing bent into an "omega" shape to allow flexing with temperature changes.]

I used lead-free solder. 60/40 solder would have looked a lot shinier. It's possible my lead-free solder has some silver in it.

One standard N-female and one standard N-male connectors were used. A couple of the N connector mounting nuts are smaller on purpose to create a larger space to the hot side of the adjacent capacitor, to reduce the chance of high-voltage arcing.

Tuning and Performance:

For some reason, shorting the metal tuning shaft to the aluminum cabinet changed the tuning slightly on one cap more than the other cap. An insulated tuning tool was used for tune-up; the alignment was sharp but not overly touchy.

The Z-Match had approximately 0.1 dB insertion loss, but with 105W from a Syntor X9000 that had its harmonic filter adjusted for 6 meters I got 108W out with the Z-Match inserted and adjusted at 53.5 MHz, which is the frequency the coil was adjusted for and caps tuned to, for 1.01:1 SWR. The antenna analyzer photo below shows the result.

The inductance of the coil is critical for the lowest loss into a 50 ohm load. You find the frequency of lowest SWR or best return loss first then spread or compress the coil to slide that original frequency to the desired frequency. The Z-Match is only good for less than ± 0.2 MHz before it needs to be re-peaked to the new frequency to keep the insertion loss to 0.1 dB. The photo below shows the insertion loss.

With the Z-Match tuned to about 52 MHz, and the N connectors disconnected, the coil resonates at about 68 MHz. (If I remember correctly, for some unknown reason, the other Z-Matches resonated closer to 80 MHz when disconnected.) The spectrum analyzer sweep below shows the return loss.

Here's another spectrum from 3 MHz to 203 MHz (the center is 103 MHz), showing the overall response of the Z-Match when tuned to 53 MHz.

Contact Information:

The author can be contacted at: jhaserick84 [ at ] comcast [ dot ] net.

Back to the top of the page
Back to Antenna Index page
Back to Home

This article created on Monday 09-Jan-2017

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.