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

   Home   Help Search  
Pages: [1]   Go Down
Author Topic: Homebrew variable capacitor materials  (Read 1631 times)

Posts: 2075

« on: April 08, 2008, 08:44:09 AM »

I've seen a couple of web pages (including at least one on E-ham) about making variable capacitors from scratch. Usually involves some form of sheet metal for the plates and washers/nuts/threaded rods for assembly and spacing. Looks perfectly reasonable.

Optimally, though, what's the ideal material for making the plates? Brass, copper, and aluminum are all readily available and I can work them with sheet metal tools pretty well. Brass seems to be the most durable, aluminum next, and copper last. But those are just my perceptions.

What options are there for plating copper or brass? Do they really need to be plated, anyway? I mean, except for the connections you want the plates to be insulated, right? I know I can just dunk them in molten solder - true silver plating (as opposed to the think poorly bonded film gotten with used photo fixer) seems quite a bit more difficult.

I'm guessing that most of the variable capacitors I have knocking around in the junkbox (age between 30 and 70 years old) are either aluminum, brass, or some sort of plated steel for the plates.

For insulators/frames, I really like working with G-10 and other glass fiberglass materials. Traditionally the frames and spacers etc. were ceramic. Is ceramic just traditional, or actually superior to fiberglass materials made to be electrical insulators?

Posts: 1044

« Reply #1 on: April 08, 2008, 09:25:00 AM »

This should tell you something: "I'm guessing that most of the variable capacitors I have knocking around in the junkbox (age between 30 and 70 years old) are either aluminum, brass, or some sort of plated steel for the plates."

I agree that brass is probably best, from the workability and durability standpoint. Silver plating using photo fixer is viable if you electroplate and not just "dunk plate."

If your capacitor is for a high-power tank circuit, you should strive for the best electrical contact between vanes and rotor and plates and frame (e.g. welding or brazing the parts together), and keep the plate edges smooth to avoid potential spark gaps.

Just my 2ยข worth. Tell us more about your project.

Posts: 245


« Reply #2 on: April 08, 2008, 09:53:52 AM »

Hi tim,

You don't mention what you would like to use the variable capacitor for. For high-power RF use there are some other considerations as well.

First is the voltage breakdown rating of the cap. The plates (and some other metal surfaces) need to be well polished and smooth or the breakdown voltage will be degraded a lot. Commercially, the individual metal parts used to be tumbled in walnut shells to smooth them out prior to assembly. Don't know how it's done today.

Second is the effective series resistance of the structure. This is important when the currents through the capacitor can be high. One place where this is difficult is the contact to the moving (rotor) portion of the capacitor. Butterfly capacitors eliminate this moving contact. Welding the pieces together usually reduces series-R quite well compared to compression contacts.

Third is the quality of the insulation (usually the frame). Good materals are available nowadays that have low absorption.

Fourth is the minimum tuning capacitance. New capacitors have a trapezoid-shaped end frame (rather than square or rectagular) to reduce the minimum capacitance.

Other issues include mechancial stability, temperature stablity, etc.

Others can chime in - but I think it should possible to achieve a Q near 1000 in the HF range. I do not know the impact of the surface plating or oxidation on the capacitor Q or absorption.

I'll bet someone reading this thread knows...

  -- Tom, N5EG

Posts: 5436

« Reply #3 on: April 08, 2008, 10:54:34 AM »

Your going to have to give us some expected frequency, power, and voltage info here.
Ceramic is the best for high power HF applications.
But for homebrew experimentation, lower power use, the simplest material would probably be PC board.
What are you hoping to build?


Posts: 20542

« Reply #4 on: April 08, 2008, 11:40:54 AM »

If you're talking about making conventional rotor/stator "meshed plate" variable capacitors, the material used for the plates isn't so important but the way it's cut and finished is.

Many are aluminum, but the trick is getting the edges cut cleanly enough that they have no points or burrs, or even sharp edges.  Most high quality capacitors use plates that have been tumble deburred to make them very smooth and eliminate any sharp edges or points anywhere.

When I worked in the RF instrumentation field many years ago, a common method was to tumble deburr using walnut shells (I think -- it was some kind of large nut, anyway), and it was a long process to make good plates for a variable capacitor.

Ceramic's likely a better choice than G-10/FR-4 etc.  Teflon, Delrin and some other materials are okay.  G-10 is probably okay for low voltage/low power stuff.


Posts: 2358

« Reply #5 on: April 08, 2008, 04:33:43 PM »

If you Know of some one that does reloading of rifle or pistol shells i bet the have a Tumbler with all the media you need to buff and smooth out the parts.

While i know i did not offer you much help or advice i did maybe give you some insight on how and where you could get the parts deburred and looking great for free at. Jeff

Posts: 9304


« Reply #6 on: April 08, 2008, 04:40:40 PM »

Variable cap plates are still tumbled in Walnut shells to polish them. Without that voltage rating is seriously reduced.

With any reasonable care at all capacitor Q should be in the thousands, even 5000 is not uncommon for modest reactance values with reasonable construction. If you weld plates and use a good form factor for the capacitor along with very solid connections, it's possible to get into the tens of thousands for Q.

By the time you make a good cap it is often much cheaper to just buy something.

73 Tom

Posts: 2075

« Reply #7 on: April 10, 2008, 12:53:40 PM »

I was thinking of using at least one butterfly capacitor for a balanced, link coupled HF antenna tuner. Nominal capacitance range would probably be 20-500pF and power levels would be 100W plus some margin. Impedance levels on the output side are likely high (several hundred ohms) so voltage levels would be several to ten times what you'd expect at 50 ohms.

Right now I've got no shortage of old variable capacitors, but none are as balanced as a true butterfly capacitor.

Loaded Q's of the input and output tank are not ridiculously high. Not sure if having a variable capacitor Q of 1000 vs 5000 makes any different when, loaded with the antenna, the circuit Q is 10 or 20.

Does this help narrow the design down any?

Posts: 31

« Reply #8 on: August 13, 2008, 05:22:27 AM »

I know back in the day when the Japanesse made 365 pf variable caps really small. They replaced the air with another dielectric. As we all know most insulators have a higher dielectric constant and higher breakdown voltage than air.

I'll drink to that by taking a beer bottle. A brown beer bottle has a dielectric constant of 7.5. If we go back to Benjamin Franklins' time they had leyden jar capacitors.

I wrap the outside of the bottle with aluminum foil and pour a conductive liquid like copper sulfate into the beer bottle. A beer bottle has a glass thicknes of .075 inch, and a breakdown voltage of 120,000 volts. And to meet OSHA regs I have to label the contents of this bottle and put a skull and cross bones on it.

Needless to say vary the amount of copper sulfate in the bottle and you have a variable capacitor. If you filled a 12oz brown beer bottle close to the brim you can expect to get 1nf of capacitance.


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!