THIS PAGE UPDATED APRIL 20, 2006 WITH MAJOR DESIGN REVISIONS!
SARA Coverage map Visit the Official SARA Web Site at www.n4nw.org
Please be sure to read this brief page on VHF Propogation Basics
Please read our Standard Disclaimer regarding content of this site.
Most of the equipment can be had for relatively cheap, with exception of a duplexer. So I began exploring the possibilities of building my own duplexer from scratch. I located a wealth of information on the web, and began to investigate the pro's and con's of each approach.
Personally, the most desriable situation would allow me to place a duplexer inside the rack with the repeater itsself. Only one of the DIY projects would allow this approach. It was design developed by Sam Kennedy. I printed his web page with the duplexer construction details to a PDF and it can be viewed by clicking the following link -> KT4QW Helical Duplexer <- These files can also be viewed by going to Sam's web site. I read the description of how this design is constructed, and decided that I would pursue the heliax duplexer described on several other sites.
The first design I was able to locate was the WB5WPA Heliax Duplexer.pdf. I decided this was the way to go due to the simplicity of the design. I have also been flooded with requests for the information after mentioning my project on the repeater builder list server. I posted this pdf because Jim's page apparently comes and goes due to local provider issues. If possible, I would visit Jim's web site first. Allowing access to the various on-line calculators and the like. If Jim's site is unavailable, you can get enough info from the above pdf to get started.
I also followed another link to the following page maintained by Glen, WA7X... describing how he used the heliax notches to boost the performance of his more traditional "can" type setup. Glen's page is located here. Glen provides some good construction advice. For more information, Glen provides the following email address mail@wa7x.
So far, I have completed three of the reqiured 8 filters needed for my repeater project.
A plot of the "first-cut" can be viewed here.. try1.pdf.
A plot of the first filter with the appropriate capicator added to reduce insertion loss can be viewed here 1finished.pdf.
And finally, a plot of three completed filters in series can be viewed here... 3in series.pdf.
As time allows, I will post photos of the construction techniques that were employed by myself. But, as this is still a "work in progress", the photos will have to wait for completion of the project.
As promised... a couple of my construction variations...
The first step after determining the resonant length is to add 1 1/2"
before cutting the heliax...
Then drill clearance holes for the BNC connectors one inch beyond the resonant length. Next short the inner conductor to the outer conductor on the bottom end by soldering 1/2" copper strap across the bottom
The top of the Heliax is cut using a Dremel tool ... I was
able to find a bit that resembles a tiny saw blade on a shaft. Set the depth so
that the case of the Dremel is used as a "depth gauge". Insert the
bit inside the inner conductor, and clip the inner conductor like in the
picture. I set the depth to be 1 1/2". Then use a razor knife inserted
through the slit ( also cut with the Dremel ) to cut the dielectric loose so it
can be removed with the cut portion of the center conductor. Discard the cut
away items. Insert the "gimmick cap" and tack the braid to the center
conductor ( slide up or down for rough tuning )
This photo shows the method for attaching the "fine-tuning" tab. Use
a Dremel tool and cut a slit in the side to insert a piece of 1/2" copper
strap, and solder in place with a "big-honkin'-iron".
The trim cap or inductor.. depending on TX or RX notch, is added last.
I also plan to use plastic "cap-plugs" on either end to keep "critters" out of the finished assembly.
Nice clean package... no need to buy double sided PC board and solder a box together.
MAJOR DESIGN UPDATES - May 2006
After much research into the mechanical instability problem with the previous design, I began to search for a different way to construct the "gimmick" capacitor. I have come to believe that this capacitor in conjunction with the "trim-tab" are the source of tuning instability that has plagued so many of this projects builders.
I was forced to tune the filters in the cabinet after transporting them home. I found that just the ride (in the back of my Blazer from my office to my site) was enough to put most of the eight filters tuning so far out-of-whack... that they were un-usable as a duplexer! Physically it looked just the same, but some of the notches had moved drastically. I constructed a mount for the filters that would fit entirely within the repeater cabinet, and set about the task of re-tuning the filters.
I presently have the design in service on my own repeater, and have noticed no major excursions of tuning since it was put on the air.
This is how I was able to fit the filters into the same cabinet with the repeater and associated other gear...
In the interest of coming up with a design that could be shipped by "The Brown-Shirt Brigade" also known as "United Parcel Smashers" I made mention of my project, and desire to "stabilize" the design. Another Ham on the Repeater-Builder list server mentioned that he knew somone who had constructed one of the beasties as well. That is how I made the acquaintance of Brian - G4UTM, a fellow Ham Radio Operator in the UK. Brian had constructed a duplexer for service by a local 6m repeater owner. Brian had his own interesting twists on the original design...
I very much liked the cross-bar tuning precision, as well as the elimination of the troublesome "trim-tab". Very significant to performance is the lack of inter-filter cable connectors. This drastically reduces insertion loss!
The proud Papa (Brian - G4UTM) poses with his completed beastie!! Reports have it in service and functioning quite well!
Anyway, after many emails back and forth discussing the subject, and possible plans of attack. Brian and I had discussed using "piston-style" capacitors to accomplish the feat... but I had been unable to locate the correct values for our project. Then one day I received an email from my friend across the pond containing the following photos....
And the exploded-view version....
.
Brian had taken the bull by the horns, so to speak, and constructed this beta capacitor for testing... his original attempt used a different dielectric, which prompty melted when exposed to the temps produced when soldered into place.. so he ginned-up this little beauty using a Teflon dielectric instead.
I was extremely impressed with his work, and asked if he would manufacture one for me to test, as Brian didn't have access to any 1-5/8" heliax at the moment... and he might not get any for several months. Brian obliged my request, and promptly sent a small packet by international air-mail. It arrived about a week later. I had to wait several days for an oppurtinuty to construct yet another filter for testing. I have a pretty good supply of heliax, and the wait was almost unbearable!! I was very excited at the prospect of putting the cap through the wringer, and hopeful it would prove to be a viable option. Unfortunately, I am not allowed to bring the "big-gun" test equipment home on a whim. It is usually on a job-site so I had to wait for my chance.
The following weekend I toted the toys home, and sneaked a few hours away from the spouse to try the cap out.
It was an astounding success!!!!
The following plot shows it to be a very viable replacement for the "gimmick cap" as described in previous designs. I tossed it around like a rag-doll trying to disturb the tuning, and found it to be a very stable design. The following plot is from a HP8753E Network Analyzer, and was plotted after the "abuse" session.
The full series of test plots can be downloaded by clicking on the plot above.
In fact, the test was so successful that I have ordered more Teflon rod for machining to create a full set of 8 caps for a full-blown test of this design.. in which I will use 1/2" heliax for the interconnecting cables with no connectors. I hope to have the teflon rods shipped to Brian shortly. In fact, I just received a tracking number for the shipment. I got so excited about testing Brian's cap, I didn't get any pictures taken before it went into service. I do promise a complete article regarding construction and tuning when we are both satisfied no further advancements on this design can be made.
May 8, 2006 - New Information
A new set of 8 "gimmick" capacitors is on it's way here from
across the pond. I have a few revised pictures and a working drawing to post at
this time... but no new test data until the new caps arrive.
The first item is a photo of the revised design using 0.5" hex bar. The
overall length has been increased ever so slightly to accomidate the ability to
tune through the desired setting, rather than being able to "just get
there".. this should allow a little more margin for error.
The idea for the use of a larger outer diameter brass rod was to allow the
constructed cap to be inserted and soldered into place with a minimum of
hassle. Using the 0.5" hex stock, it is a snug fit into the inner
conductor. This should also make it much easier to position and solder into
place.
This picture illustrates how the cap mounts inside of the inner helix material.
While it may appear to be slightly off-center, it is due to the corrugations in
the inner helix material. The cap "self-centers" allowing the builder
to have both hands free for the soldering exercise to follow. As always,
Brians' work is first rate! It generally takes around 4 to 5 days to send a
packet over, I am hoping to get some construction time in this weekend.
Allowing time for the 6m Sprint. After all, contesting vs. building a new
duplexer... the Sprint wins!
This is Brians' working drawing... I hope to post a nice CAD drawn PDF file
once I complete the project. But for now, if you want to get a jump on the
project... this will have to do. I expect the project to be a success with all
the previous testing...
But you never know how things will go. Hope for the best, more reports to
follow!!
His assistance and expertise is greatly appreciated!
Until then, wish us both good luch & much good DX!!
73's for now..
Mike
About Me | Contact Info Index | News Index | My Hobbies Index
This document maintained by k5jmp@arrl.net.
Material Copyright © 2006 K5JMP