Acrylic coilformer now finished. Comment. / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Acrylic coilformer now finished. Comment.

1 2 next

Move Thread

LAN_403

ragnar

Wed Mar 29 2006, 11:46AM

Registered Member #63 Joined: Thu Feb 09 2006, 06:18AM
Location:
Posts: 1425

As you may have seen me aspiring previously to high frequencies with high powers (fingers crossed with high reliability, HA!), I thought I'd make the first step in the right direction with a quality resonator.

Couple of problems with small, high-frequency resonators:

- the resonant frequency encourages short, thin resonators wound with thin wire.
- the thin wire will represent a relatively high resistance (skin effect) and gets hot,
- the former may catch fire
- the power throughput is limited

So I've designed a coilformer that will:

- significantly minimise the inter-turn capacitance,
- facilitate the use of my own 19-strand x 0.15mm litz wire to give a nice low resistance for higher powers,
- not be in intimate contact with the windings, and allows forced/passive convection around the windings,
- offers a better (longer) tracking/flashover path, and dirt on the former won't be so much of an issue,
- more great stuff that I'm hoping people will add to this thread... :)

A few potential issues with the cross-section I've drawn, cut, and assembled are:
- the sharp corners might present an early breakdown point, perhaps I need a hexagonal or octagonal shape rather than square,
- more issues I'm hoping people will add to this thread... :D

The former is made from lasercut 3mm acrylic sheet, slotted so the two can connect as seen below.

This was just a proof-of-concept, very rough. My planned dimensions are 20cm high, 7.5cm diameter, 100 turns (or an average of a turn every 2mm).

The final profile of the former's edge will be sawtooth not squarewave, which will encourage the litz wire to bunch up rather than flatten out.

I'm hoping to pump lots and lots of power through this lil fella at ~5MHz.

enjoy! =)

Criticism and suggestions welcome.

Pre-empting questions:
No, the 40N60 is only a gauge for size, it's got nothing to do with the former or the driver.
Yes, that's a 7KG roll of 0.15mm copper wire.

WaveRider

Wed Mar 29 2006, 12:02PM

Registered Member #29 Joined: Fri Feb 03 2006, 09:00AM
Location: Hasselt, Belgium
Posts: 500

Hi BP,
At high frequencies, using thick winding wire will reduce winding losses. Litz wire may help, but remember that helical resonators with Q factors in the 100s can be constructed for freqs into the low microwave bands. These use fat windings. The losses are proportional to the product of effective wire diameter and the winding length whereas EM energy storage depends on the resonator volume. Since Q is proportional to Energy Storage/power loss per cycle, the best resonator performance is attained by maximising the ratio b/w effective resonator volume and wire length.

Of course there are tradeoffs when you want to maximise other things...like the length of your arcs.. Good luck!

Cheers!

ragnar

Wed Mar 29 2006, 12:31PM

Registered Member #63 Joined: Thu Feb 09 2006, 06:18AM
Location:
Posts: 1425

Dr Bill =)

wrote ...
...At high frequencies, using thick winding wire will reduce winding losses. Litz wire may help, but remember that helical resonators with Q factors in the 100s can be constructed for freqs into the low microwave bands. These use fat windings...

I found this article (referred by SM) worthwhile consideration, too:

interesting thinking about current density uniformity across the 19 windings and capacitance between the strands.

wrote ...
...maximising the ratio b/w effective resonator volume and wire length...

In the case of the resonator, the best volume:wirelength ratio is where the height of the resonator is the diameter. I determined this empirically (and later found out on the web this is standard and expected) with a past project.

For a totally unrelated project, I made a C++ program; you specify a target inductance and the diameter of wire you're using, and it will calculate the ideal number of turns and the ideal diameter to achieve that inductance using the shortest length of wire... it suddenly occurred to me that the ratio between number of turns and coil diameter was a constant, and proportional to the thickness of wire! Silly me.

I find it very rewarding to be able to cross-fertilize knowledge/experience(/realisation of my own ignorance/stupidity) gained in one project with another.

In that case, I should redesign my former for the height to equal the diameter. Thanks very much indeed!

(click to enlarge)

1143635494 63 FT6036 Weightratioasymtote

Marko

Fri Mar 31 2006, 06:30PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

I vote for single thick wire, it will be simpler to wind at least

With these frequencies you will end with more skin surface, silvered wire may also help if you can find some.

But now, turns are so spaced that your LC resonant frequency tends to be ultimately high, far away from 1/4wl frequency of helical resonator like this.
So we are back to what-is-a-tesla-coil again...

Sulaiman

Fri Mar 31 2006, 10:07PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

From the little I've read
Tesla was very keen on optimising the Q of his resonators
but he wanted high voltage/charge operation
An arcing TC has such a low Q that close-wound is normally acceptable.

BTW, I think maximum Q for a single-layer solenoid is approx.
when radius=height, not diameter.

Steve Conner

Sat Apr 01 2006, 10:03AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

It looks nice, but I bet it breaks out off the corners. If you can figure out a way to scale it up to make a cheap coil former that's like 7ft tall by 2ft diameter (without sharp corners) I give you 100 cookies and a big hug

ragnar

Sat Apr 01 2006, 10:46AM

Registered Member #63 Joined: Thu Feb 09 2006, 06:18AM
Location:
Posts: 1425

'evening Steve,
for lack of a good description, I'll rectify the sharp-corner-issue before cutting the next former, attached is a pic.

Even though the laser only has a bed of 600mm x 500mm, scaling up is just simple acrylic fabrication - make notches and joiners where you want to connect multiple segments.

As for the roundness issue, I'll use more perpendicular tiers to give a rounder cross-section.

Thanks for emphasising that potential issue before I make the thing =-)

Desmogod

Sat Apr 01 2006, 01:49PM

Registered Member #139 Joined: Sat Feb 11 2006, 11:01AM
Location: Perth, Western Australia
Posts: 358

After our conersation the other day, If you can make it, I reckon a hexagonal or octagonal former may be a better option for you.
I want one of those laser cutters!

ragnar

Sat Apr 01 2006, 02:22PM

Registered Member #63 Joined: Thu Feb 09 2006, 06:18AM
Location:
Posts: 1425

And when you do get your lasercutter, then you'll have to learn how to use it. When things go wrong, you'll have to learn how to fix it. If the mirrors go out of alignment, you need to get in there and tweak it. When stuff catches fire, you have to extinguish it. When it needs regassing, you'll need to pay through the nose.

it's never as easy as it looks...

...but it's still my baby... *stroke*... *stroke*...

Just trying to calculate how many perpendicular iterations of this cross-section I should do.

5 looks about right... but still sharp corners... hrrm

Marko

Sat Apr 01 2006, 02:34PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Hi,
I think you can get the same using very thick wire with thickest possible insulation, and wind it around standard PVC tube.

For my 10Mhz resoantor (that what I still dont have driver for) I used telephone wire with thick insulaton, wire thickness is 0,5mm and af insulation about 0.4mm, so I have each two truns spaced for around 0,8mm total (or maybe even more since turns aren't absolutely tight).

But now you should absolutely use ths nice former, for try at least.

cheers...

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob