Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
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
About eHam.net

donate to eham
   Home   Help Search  
Pages: [1]   Go Down
  Print  
Author Topic: DIY Capacitors in Low Loss (High Power) Filter  (Read 1914 times)
AE7TE
Member

Posts: 53




Ignore
« on: October 11, 2017, 12:34:29 AM »

I've been investigating techniques for building Tesla Coils. Not because I want to make one, but because those parts have to be inherently low loss so they don't break down under the heavy RF fields of something that generates large electrical sparks. The inductors are made of heavy copper and are wound nowhere near lossy materials. The capacitors are made either with dozens of series-connected capacitors to make one large unit, or with something more exotic. Some builders make Leyden jars while others do something like a sandwich capacitor.

If I were to plan a set of harmonic filters for different HF bands, I'd have the following options...

- Parallel a lot of ceramic (NPO) capacitors together to get the values I needed. This would be only as reliable as the caps and not likely effective at higher powers.
- Buy a ton of different 500v silver mica capacitors. This seems to be what most people do. The problem is that buying so many different values is cumbersome and expensive.
- Craft high-voltage, low loss capacitors out of foil and film.

I want to go with the third option. One problem that presents is that rolled capacitors seem to have a relatively high inductance (being a long thin strip of foil with the lead wire at only one end). It's also possible that flat stack of foil pieces on stacks of wax paper or polystyrene film would form a better RF capacitor, because every plate would have its own connection to the terminal. One way I can think to solve the rolled capacitor problem would be to add several lead wires along the length of the cap, and then tie them all together once rolled. This should break up the roll length and result in lower inductance. I would need to dab some epoxy in the ends of the roll to stabilize the lead wires and prevent shorts between the plates. I've also considered soaking in a drying oil, to impregnate the capacitor and remove air voids that could result in shorter capacitor life.

It's also possible the thin foil used in kitchen rolls could not have current handling capabilities to be low-loss, especially at higher powers where the current through the capacitor exceeds an amp. That could be alleviated by using thicker aluminum, or foil made of copper or brass. I think I'd resort to that anyway if I started getting good at rolling caps. The lower loss would be essential when high currents flowed through a point. For example, the coupling of a tube plate to the Pi network should have a hefty capacitor suited for considerable RF current as well as low DC leakage. My main goal right now is to design fixed capacitors that will go in filters or tuners. As I've maintained, QRO parts are ideal for QRP because of their inherent low loss. I'd need to make these caps strong enough to be used in a Tesla coil even though they'll never see that kind of rough treatment.

Ed AE7TE
Logged
G4LNA
Member

Posts: 144




Ignore
« Reply #1 on: October 11, 2017, 02:42:41 AM »

I make my own capacitors out of brass shim and 0.25mm PTFE or I think you call it Teflon sheet in the US. Using on-line calculators to work out the size of the plates I can usually knock one up in an hour or so. I trim the capacitance up with my analyser by cutting a bit off the shim a bit at a time till I get the capacitance I want, I find it isn't far off what I calculated anyway.

They will take quite a high voltage if you look at the spec of the sheet.

I have got a PDF document of how I make them, but I'm at work at the moment, so I can't access my Dropbox where the file is located, IT have blocked it. I you don't mind waiting till later when I get I'll post the link if you are interested?
Logged

AE7TE
Member

Posts: 53




Ignore
« Reply #2 on: October 11, 2017, 03:36:57 AM »

I'd like to see the document you have. That would probably be pretty helpful.

I'll end up changing to brass or copper when I have the basic technique down. This will be for their superior conductivity over aluminum - I want the plates to have low losses because there will be large circulating currents in them. My thinking is that, with some practice, I'll be able to make some fixed harmonic filters capable of rejecting spurious emissions for all of the HF bands, and I'll be able to do it more easily than by ordering a handful of SM caps that might (or might not) be the correct values.

If I use something like stiff paper to make the capacitors, I'll have to soak them in some kind of oil which will permeate the paper fibers, then harden. Maybe I'll dip them in paraffin or beeswax to seal them.

I'm sure there are going to be people out there who scoff and say I'm re-inventing the wheel. I have an all-mode HF transceiver and tons of tech toys. These people are the same ones who moan about "no-code hams" and how the hobby is dumbed down. Let them scoff. I'm not going to be listening.

A primary goal is to learn how to make capacitors that will do well in AC-only, RF circuits. A secondary goal might be to make capacitors that do well for blocking DC as well as transferring RF.

Maybe if I get really good at this, I'll start making capacitors for filtering power at SMPS frequencies.

Ed AE7TE
Logged
AF6LJ
Member

Posts: 364




Ignore
« Reply #3 on: October 11, 2017, 05:26:41 AM »

A few thoughts;
A better material for your dielectric may be Teflon sheet which you can buy from places like McMaster Carr.
On the low bands aluminum will be good enough as far as a conductor goes. (160 and 80)
When you make your plates polish them and make sure the edges are broken and polished

Paper; even wax or oil soaked paper doesn't make a good dielectric at radio frequencies.
Logged

Take Care
Sue,
AF6LJ
KC4ZGP
Member

Posts: 1637




Ignore
« Reply #4 on: October 11, 2017, 06:02:28 AM »


Fashion capacitors from printed circuit board blanks.

Or two sizes of copper tubing in a 'U' configuration.

Kraus



Logged
G4LNA
Member

Posts: 144




Ignore
« Reply #5 on: October 11, 2017, 08:24:15 AM »

As promised, this is the link to the PDF in my Dropbox account.

https://www.dropbox.com/s/7xjt68300a55wlp/Bulding%20a%20capacitor.pdf?dl=0

I have used this technique to build the capacitors in a remote matching unit I built, also I'm not the only ham to use this idea, so it is a tried and tested way to do it.

Logged

G3RZP
Member

Posts: 8123




Ignore
« Reply #6 on: October 11, 2017, 02:55:47 PM »

I've done filters with Rogers Duroid Teflon PCB material. A lot depends on what frequency you operate at. A good rule of thumb is that for a capacitor made of PCB materials, you want a copper thickness of 5 skin depths. That gets the in depth current about 80 dB below the surface current. For copper, one skin depth is 2.6/sq.root. FMHz, which is about 0.002 inches at 2 MHz. So for 160, you would want 10mil copper each side of the dielectric. You would probably get away with about 4 or 5 mil - one skin depth is where the current has fallen to 1/e of the surface current, where e=2.728. (Note: I'm using 'mil' in the US sense of 0.001 inch - which we call a 'thou' for 'thousandth of an inch'-  not in the common European sense of 1millimetre).

People have used FR4 fibre glass PCB, but it's certainly lossy compared with Duroid, and it needs far more copper thickness to get the Q.

Another problem is if you have sharp edges or corners. That can produce interestingly explosive results, as I found out from some experiments with HV high current padding capacitors using window gals and aluminium siding....
Logged
KA9UCN
Member

Posts: 92




Ignore
« Reply #7 on: October 11, 2017, 08:38:14 PM »

Mylar has a high dielectric strength to thickens ratio. Is cheep. Light and is usually available.
Joe KA9UCN
Logged
G3RZP
Member

Posts: 8123




Ignore
« Reply #8 on: October 11, 2017, 11:18:23 PM »

What is the loss tangent for Mylar at HF? That is an important consideration.
Logged
G4LNA
Member

Posts: 144




Ignore
« Reply #9 on: October 11, 2017, 11:20:42 PM »

Another cheap source, if you want to go cheap, are freezer bags, they are cheap but you need to use three or four layers to achieve the dielectric strength which complicates building the capacitor.
Logged

Pages: [1]   Go Up
  Print  
 
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!