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: Electronic capacitor to replace a vacuum variable  (Read 3862 times)

Posts: 118


« on: April 08, 2008, 06:54:48 PM »


A transmitting loop antenna requires a vacuum variable to handle the high voltages/energy in the circuit and offers low loss resistance. Unfortunately, vacuum variables are expensive, need to be controlled with an add-on motor which can be an issue outdoors, and any damage to the capacitor will result in large expenditure to replace it.

As an alternative, I was wondering how well would an *electronic* capacitor work? The idea is to bunch some fixed high voltage capacitors in parallel and activate each using something like a FET. Since I know the approximate values of capacitance required for 80m, 40m and 20m for the loop I built, I can use three fixed capacitors around those values and have a floating set of, say, 10 1pf capacitors for fine tuning.

Even if it works, one issue is finding a FET to handle the high voltages. Any alternatives for that?

Thoughts, comments, flames?


- Siddhartha

Posts: 3333

« Reply #1 on: April 08, 2008, 07:25:55 PM »

I have built many mag loops (some quite large, others not so big) and faced the same issues you mention.  I suppose you could get some kind of industrial-strength PIN diodes to work, but the mechanical engineering needed to fit and cool them, to say nothing of the resulting uncharacterized operation, would be most daunting.  And the PIN diodes you might use would cost far more than a simple stepper motor and surplus vacuum variable would.

Much cheaper and easier to make would be a screwdriver motor running a trombone capacitor.

Posts: 4546

« Reply #2 on: April 08, 2008, 07:33:20 PM »

There are plenty of devices built for horizontal sweep service in TV's and CRT style video monitors that can handle hellacious voltage spikes, but they tend to be bipolar silicon and would need to be decoupled from the control line.

Then I wonder if you'd be ahead of the game to think in terms of relays to switch doorknob caps in and out of a stack, but the difference in cost & complexity might not be much of an advantage over a vacuum variable. Then there's the matter of avoiding a hot switch lest the contacts see too much arcing. Which means you'll be tuning by trial & error until you get the bandspread figured out.

Maybe someone here can speak from experience on this, but I'm thinking a vacuum variable or a nicely weatherproofed conventional cap is probably the path of least resistance...


Posts: 338


« Reply #3 on: April 08, 2008, 08:39:57 PM »

Hi Siddhartha,

In small transmitting loops, the radiation resistance is very, very low. Thus the currents in the loop tuning capacitor are huge, and the voltage across that capacitor is also huge. Capacitors for transmitting loops need to have extremely low series resistance or the loop will not perform. In fact, welded butterfly capacitors are used to keep the resistive losses as low as possible. Even at 100 watts, the capacitors need a breakdown voltage that can approach the range of 5 to 15 kV (depending on loop size, and band of operation).

Even if you could to accurately predict the values of the 3 needed capacitors (one for each band), each one would have to be of very hi Q construction, and the switching element for it would need to have resistance lower than is achievable electronically. Even a wire has too much resistance; straps or metal frames are needed.

The fine-tuning capacitors would each have lower current if they have small capacitance compared to one of the 3 fixed capacitors since the current divides between the fixed and the tuning capacitors. Then the problem would be breakdown voltage across those small capacitors and the associated switching element.

The capacitor is a tough problem part with small loops. The loop is usually welded to keep the resistance low. These types of antennas are working at the far edge of circuit values and have corresponding difficulties.

  -- Tom, N5EG


Posts: 9748


« Reply #4 on: April 09, 2008, 01:41:35 AM »

You cannot use any type of pressure contact or semicondutor in this application.

The reason is even with fairly low power  power any loop with good Q and high efficiency (say over 20% efficiency) will have circulating currents in the dozens of amperes and voltages in the thousands of volts.

This is why good loop designs use very wide area welded or solidly bolted connections and smooth wide conductors, and why capacitors must have very high voltage ratings.

Of course if you use a lossy loop design or poor construction the voltage and current will decrease, but for systems with any reasonable value of efficiency the capacitor is very critical.

I would expect you can find a suitable vacuum capacitor for less than $150. The motor might be another $20-50 for a low speed high torque gear drive motor.

There are many bad designs with poor efficiency around that still make people happy. There are sliding printed circuit board capacitors, there are loops made with coax shield. Poorer design loops where efficiency is in the single digits or less are much less critical for capacitor ratings because most of the energy is distributed in heat losses. You just have to decide how important efficiency is and how much you are willing to pay for the system.

73 Tom

Posts: 6900

« Reply #5 on: April 09, 2008, 04:42:16 AM »

I don't know what a "*electronic* capacitor" is.
I do know that a vacuum capacitor is used because of the extremely high voltages involved, 10+ Kv.  At this level, even moisture in the air can cause breakdown.  They are the most cost effective solution.
I don't know of any FETs that can handle that type of voltages, most relays can't either.


Posts: 1289

« Reply #6 on: April 09, 2008, 10:25:19 AM »

Even in low voltage situations using a FET to switch a capacitor usually is futile, the FET can have a horrendous amount of drain to source capacitance, it just is unreal how high that value can be at low drain to source voltages... bipolars are a lot better for that. Probably the screwdriver and tubular capacitor is a great idea for homebrewing something. I priced high quality motor driven surplus capacitors recently and they were expensive...

Novice 1958, 20WPM Extra now... (and get off my lawn)

Posts: 9748


« Reply #7 on: April 09, 2008, 03:38:44 PM »

A trombone capacitor has high ESR and limited Q.

Keep that in mind.

The reason ist has low Q is the series path through the capacitor is very long. This adds series inductance that greatly increases the capacitor ESR. It is actually very unlikely to have higher Q than a traditional air variable with wiping contacts.

The most effective capacitor is a butterfly capacitor of almost square form factor. The STATOR plates are welded , the rotor plates do not need to have any special treatment at all. A good butterfly cap can have very low series inductance and very low ESR. It is possible to have Q's in the tens of thousands with a proper butterfly cap.

A trombone cap would not only have much less Q, probably in the low thousands at best, but it also has limited tuning range. This is because series inductance increases as capacacitance decreases, and this lowers the reactance range.

There are really two very important parameters for efficiency. One is minmum series resistance, the other is minimum series inductance. Both are equally important.

The capacitor really needs to be the smallest "box" with the smallest path length possible.

73 Tom

Posts: 8916


« Reply #8 on: April 09, 2008, 05:17:37 PM »

"The capacitor really needs to be the smallest "box" with the smallest path length possible. "

Yeah, and a butterfly is the way to go if you're particularly concerned about mechanical ruggedness.

Honestly, though, I think a ceramic vacuum variable would survive quite well as long as you don't drop the thing off your roof.  But they do come, it seems, at a higher price.

I'd imagine just falling over onto a poorly placed rock while you had the magloop resting on something out in the yard to grab a tool could shatter a glass-enclosure vac variable.  

But stick with those choices.  Having built a magnetic loop around a (fortunately free, mildly unfortunately glass) vacuum variable, I can say it's an awfully useful antenna, and I'd be happy with it if I were somehow stuck with it as my only one.  But there's a reason you don't see a magloop in every restricted ham's backyard...

Capacitor Sticker Shock.

One little component for a homebrew antenna that costs $100?  $150? $200? $250?  Then you have to motor drive it?


But I think it's a mistake to head down a road that eschews vacuum variables or expensive butterfly capacitors just because of the initial shock. Too many magloop projects are scrapped because of the initial capacitor price, and I think it's a mistake.  I know I never built one because of that (I've always wanted to try one but never *needed* to try one)

Low minimum capacitance, high voltage ratings, and very low series resistance make butterfly or vacuum variables the correct choice, hands down.  Motor drive can be cobbled together from hamfest parts.  It's pretty easy to set up a simple limit switch arrangement using 1/4-20" stainless threaded rod and a couple of microswitches and reversing diodes.  You don't need any real indication of tuning except the sound in your receiver.

So I'd say two things:

1) Max Gain Systems ( has some of the best prices I've seen on the web for vacuum variables.  I'm not affiliated with them, nor have I scoured the depths of the web looking for the absolute best deal, but they do beat a couple of other places hands down, price-wise.  Make sure you check the flange situation though.  A lot of manufacturers use silver-plated flanges with large surface area to make mechanical connections to the capacitor.  That's OK, and I just soldered those flanges to my loop with wide solid straps and that worked well, but you don't want to cobble together something, so make sure you get the right flanges.

2) Think of it in these terms.  In the end, you get a true multiband antenna that you built that is one of the most efficient antennas for its size.  How many el-cheapo "limited space" wunderantennas is that worth? Your basic miserably inefficient wunderwhip costs maybe $99.99.  How many hams have tried, and been ultimately disappointed in, two or three of those, and none of those are remote tunable! In those terms, $150 or $200 for a capacitor and some more cash outlay for the motor to drive it is kind of a bargain.

Small transmitting loops are not free, but I think the investment is likely to be worth it.  An EZNEC model of my 4 foot octagon magloop of 3/4" copper tubing, including a guess at 30 milliohms of flange contact resistance, shows an efficiency of 58% on 10.1MHz (not counting earth losses).  G4FGQ's magloop progam says 55%.    I doubt many other four foot antennas can match that, and certainly you could waste plenty of money on commercial offerings to have built a good small transmitting loop.




Monkey/silicon cyborg, beeping at rocks since 1995.
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!