Call Search

New to Ham Radio?
My Profile

Friends Remembered
Survey Question

DX Cluster Spots

Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement

donate to eham
   Home   Help Search  
Pages: [1]   Go Down
Author Topic: Matching network coils, capacitors & enclosure  (Read 2319 times)

Posts: 52

« on: October 05, 2008, 06:53:19 PM »

I'm planning to build several broadband matching networks to install at the base of my antenna (one per band).

I've spent a bunch of time on the web researching good home brew construction techniques, but haven't found too much useful information.

I'm looking for ideas on materials and techniques to wind good quality high-Q coils and to make reliable high-voltage capacitors.  Suggestions for good enclosures and component spacing, etc, would also be appreciated.

Regarding coil design, I have some software that knows about solid copper wire, but I understand that using copper tubing can provide higher Q.  If the tubing is thicker than the skin depth at the frequency of operation, is the inductance per unit length of tubing the same as for solid wire of the same diameter?

Posts: 17352

« Reply #1 on: October 05, 2008, 09:40:30 PM »

I don't think that tubing coils will have a higher Q than the same diameter
of wire, but they may be better if you are buying the copper by weight
(since the tubing would be lighter.)

ON4AA's coil calculator includes some notes on coil construction at

and W8JI has some good info on his site:

One simple way to make the inductors is to drill holes for the windings
in a thin piece of plastic, then wind the wire around a slightly smaller
form and screw the coil in through the holes by twisting it.  This gives
a minimum amount of plastic inside the coil which reduces the self
capacitance.   You can also put notches in the edges of the plastic
to hold the turns.

There was an article in one of the ARRL Antenna Compendiums where
someone made a capacitor using a sliding plate and a piece of
glass for the insulator.  He recommended the thin glass used in picture
frames:  windows use a thicker glass and it is harder to get enough
capacitance.  That design used a motor drive with a screw thread to
move one plate back and forth relative to the other.

For fixed capacitors you may be able to use short lengths of coax cable
or pieces of double-sided printed circuit board if the voltage and current
requirements are reasonable.  I've seen circuit board capacitors built
with multiple plates of different sizes on one side so they can be
jumpered in different combinations for fine tuning.

Posts: 625

« Reply #2 on: October 06, 2008, 09:01:49 AM »

I've used plastic electrical boxes to design L networks for matching networks for 1/2 wave base fed vertical antennas.  The main need is to make them water proof.  I use fat short wires to hook the components up to minimize parasitics.  You can also use circuit board materials for good low inductance ground connections.  There isn't a need for shielding because the matching network is near the antenna anyway.

Low loss coax stubs are very cheap high voltage caps.  You cut them long and tweek them by clipping the length.  I usually use an air variable and an antenna impedance meter to set up the network and capacitor value.  Then I measure the cap and cut a length of coax to replace it.  I usually figure about 22pf per foot.

Stiff Cu electrical wire is fine to make coils with.  The ARRL had some info on homebrewing them on their web site (members only).

You really need to make sure you have modeled the effect of antenna impedance change with frequency on your "broad band" matching network.  I studied up on this a little bit a few years back.  There are well known ways to make broad band matching networks if you have a constant load impedance with frequency but an antenna is different.  There have been many attempts to produce matching and compensating networks for antennas that would essentially match the antenna and extend the bandwidth (matching BW).  Its really hard to do with lumped components because the antenna is a distributed impedance and has a different variation of impedance with frequency than a lumped network would have.  Theory then says you will need a great many lumped elements so simulate the required matching and compensating network.

So...what I am saying is you might want to first consider making the antenna inherently broad band.  Make sure losses are minimized in the antenna as well, before I would imbark on an ultimate low loss inductor/capacitor quest.  You might just be fiddling with a part of the antenna system that will pay back the least dividends.

Posts: 28

« Reply #3 on: October 06, 2008, 10:20:32 AM »

A lot depends on how much power and what impedances. I tried making a padding capacitor from aluminium sheet and thin glass. The antenna feed impedance was pretty close to 4000 ohms and at 400 watts, the results were spectacular! Beautiful flashes and holes punched in the glass round the edges of the aluninium, and the glass cracked.

Posts: 8911


« Reply #4 on: October 06, 2008, 03:00:11 PM »

"Beautiful flashes and holes punched in the glass round the edges of the aluninium, and the glass cracked"

You do need to be careful of

1) sharp edges
2) unknown dielectrics

If you make your own solid-dielectric capacitors, I'd probably recommend teflon.  Air, of course, is an option.

I wind my own matching network coils for the higher bands as self-supporting using #10 wire.  I wouldn't put it up on a tower that way and expect everything to be good for 20 years, but it's easy and quick if you just need a few microhenries.

I use lathed PVC coil forms for my bigger coils.  Not very helpful ;-)  Someday I'm working up a method to do the same thing with a dremel tool and a homemade fixture but haven't done anything there... think two pipe mounting discs with a coarse, commercial threaded rod at the center and a couple nuts at either end of a fixed frame that also sturdily holds a rotary tool with a burr in it.

The drill-holes-in-a-strip method seems to work for people, but is just too time consuming to be something I've tried, especially since I do have access to a lathe ;-)  

The strips probably do have higher Q over time because there's less leakage path.  The latest coil I built I sprayed down with electrical varnish, though, just because I figured it would be less lossy several years from now after it got dirty.  

I'd show you pictures, but my website appears to be down at the moment...

I haven't built many capacitors.  I did build a few copper foil and overhead transparency material capacitors that worked OK in moderate impedance networks at the 100W level, but mostly I've been picking up random air variables at hamfests for the years I've been a ham and finally have a job for them to do ;-)

Some of the 160m ops I know have sunk smaller air variables in a small box full of transformer oil to increase the capacitance, and voltage and current ratings a bit.  

I do use a pair of 10m/15m traps with homemade copper-foil and plumbers teflon tape traps and it seems to work OK, but I haven't tried to measure the properties well.

If I were going to make high-power-handling networks from scratch, I'd probably go with air or teflon dielectric capacitors and polish the plates well.  I hear (from W8JI I think) that high voltage air variables have been polished by tumbling the loose plates in walnut shells.




Monkey/silicon cyborg, beeping at rocks since 1995.

Posts: 680

« Reply #5 on: October 07, 2008, 07:18:16 PM »

HDPE (high density polyethylene)(white cutting board material) is your friend. It's tough, it's machinable, it's cheap, it's a good dielectric.

For making comb style supports for a coil, cut a strip, drill holes in a row down the middle of the strip that are slightly larger than the wire/tubing size, then cut off one edge of the strip, leaving a bit more than half the hole, so the wire/tubing snaps in.

Don't forget that nothing said you have to have round coils. Square and hexagonal works too.

For capacitors, unless you need adjustable, you can probably buy cheaper high current, high voltage caps. For instance, LowESR extended foil polypropylene caps (e.g. Cornell-Dubilier 942 series) can handle 10s of RMSamps at HF with 2kV voltage ratings.

Or, if you want to make smaller plate style caps, use multiple layers of thinner dielectric, rather than one thick layer. Keep in mind that if the voltage on the plates gets over about 1kV, you're above the corona inception voltage, resulting in inevitable dielectric degradation, and you're better off combining multiple caps in series.

Don't even think about oil immersing unless you like leaks. If you do use oil, Shell Diala AX or ExxonMobil Univolt are the two to use.
Pages: [1]   Go Up
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!