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

   Home   Help Search  
Pages: [1]   Go Down
  Print  
Author Topic: Current flow in magloop butterfly capacitor  (Read 3468 times)
JAHAM2BE
Member

Posts: 286


WWW

Ignore
« on: November 05, 2012, 08:41:25 PM »

I'd like to ask for advice on a specific small loop antenna geometry and specific butterfly capacitor geometry I am considering. I would like to know if the current flows might be lossy: the flow from the loop element to the capacitor, and flow within the capacitor.

First, some computer modeled images of my design. Images are very roughly to scale. A copper strip 360mm wide and 0.2mm thick is wrapped around a wooden octagonal frame whose left-to-right span (viewed straight on) is 1.06m (each octagonal segment 44cm long). Desired frequencies of operation will be 7 MHz - 28 MHz (though I doubt I will be able to reach 28 MHz with a butterfly capacitor design due to stray capacitance).

Images of loop geometry:
http://i50.tinypic.com/13yqb9v.png
http://i49.tinypic.com/33b3ll4.png

Close-up of the proposed capacitor geometry:
http://i45.tinypic.com/wmf3bn.png

The homebrew butterfly capacitor would use strips, not round rods, as the mounting posts on which the stator plates are mounted. The strip-shaped capacitor mounting posts would be soldered, at their top ends, to the main loop's strip-shaped conductor in the middle of the width of the loop's strip. The bottom of the mounting posts would be bolted to a polyethylene slab (i.e., the good old kitchen cutting board). The reason for using strips instead of round rods as the capacitor support posts would be for ease of construction, allowing soldering individual (copper) plates to the flat support strip, without necessitating the drilling of holes in the plates.

I haven't done detailed design work on the capacitor dimensions yet, but the desired capacitor size is approximately as shown in the drawing. (For example, I would avoid making an extremely long and narrow capacitor that extends halfway down the diameter of the loop; the capacitor dimensions would instead be confined to the smallest possible cubical region, that provides sufficient capacitance, located at the top of the loop.)

The support strips within the butterfly capacitor would be approximately 2-3cm wide (maybe 1/15th the width of the main loop conductor) and 0.2mm thick.
 
My concerns are that the butterfly capacitor strips, that conduct the current from the main loop to the stator plates, are small-width and lossier than the main loop conductor. Also, I wonder about the method of connecting the capacitor strips to the main loop strip. Is there some better way to interface the wide strip loop conductor with the butterfly capacitor (tapering connections, or some such)?

I don't have a feel for how critical these two parameters are (a. conductivity of conductors within the butterfly capacitor, and b. method of transitioning from the loop conductor to the butterfly capacitor conductors), or if they will be overly lossy in this particular design geometry.

Any comments appreciated.
Logged

N3OX
Member

Posts: 8847


WWW

Ignore
« Reply #1 on: November 05, 2012, 09:39:07 PM »

I don't know that it would make a meaningful difference, but I think it would be better to turn the whole capacitor so that the shaft was coming in and out of the plane of the loop.  Probably less convenient mechanically but then each stator plate can be practically directly connected to the loop conductor on one edge.

Again, this is just intuition and probably doesn't make a substantial difference... but I think that would minimize the stray inductance and added resistance.
Logged

73,
Dan
http://www.n3ox.net

Monkey/silicon cyborg, beeping at rocks since 1995.
JAHAM2BE
Member

Posts: 286


WWW

Ignore
« Reply #2 on: November 05, 2012, 11:51:45 PM »

turn the whole capacitor so that the shaft was coming in and out of the plane of the loop.  Probably less convenient mechanically

Interesting suggestion. OK, so something like this, for the stator plates:
http://i45.tinypic.com/bexcns.jpg

Now how to mount the rotor, in a precision manner such that the rotor plates do not touch the stator plates during shaft rotation, requires some more thought. Maybe I could solder on some copper strips as support brackets to hold a polyethylene block for the front and back insulating faces of the capacitor, then support the rotor shaft through holes in the polyethylene blocks. Also, the exact alignment of the main loop's ends becomes critical, as any slight misalignment will cause the stator plates to brush against the rotor plates.

Another issue with orienting the shaft normal to the plane of the loop is how to run the motor wires without upsetting the capacitor's electric field. I might just hope for the best, attaching the motor directly to the shaft and letting the control wires run straight down from the capacitor, though I've read that that such wires in the capacitor vicinity can cause problems (http://www.eham.net/ehamforum/smf/index.php?topic=74781.0).

Making beautiful-looking structures in the computer is easy, but mechanically realizing them is another matter... Smiley
Logged

JAHAM2BE
Member

Posts: 286


WWW

Ignore
« Reply #3 on: November 06, 2012, 02:09:27 AM »

After some more 3D sketching, I think I have figured out a feasible way to build a capacitor with the shaft perpendicular to the plane of the loop.

Build a support frame with thick copper bars and a bolted-on front and back plate:
http://i45.tinypic.com/10sc3h3.jpg

Solder in the stator plates:
http://i45.tinypic.com/spg1va.jpg

Insert the rotor assembly:
http://i45.tinypic.com/rhnx41.jpg

Test for capacitance, and when satisfied, solder the whole assembly (along the two top support bars) in its final position across the gap in the strip loop.
Logged

JAHAM2BE
Member

Posts: 286


WWW

Ignore
« Reply #4 on: November 07, 2012, 03:37:43 PM »

Judging by the low traffic to this thread, there doesn't seem to be much interest in this particular topic. But just in case someone in the future happens upon this thread looking for information, I thought I'd add a relevant link (as this sort of information seems very hard to come by):

http://www.eham.net/articles/9527

Search on that web page for the phrase "ANY small area contact". That post by W8JI contains advice relevant to capacitor design. In particular, connecting capacitor plates only at their corners, with small-area solder connections, seems like a bad idea. I think it would be better to rearrange the geometry such that instead of being soldered only at the corners, the capacitor plates would be soldered all along one edge (ensuring a wide solder ridge) to a wide supporting plate (not just two corner-mounted support posts).
Logged

N3OX
Member

Posts: 8847


WWW

Ignore
« Reply #5 on: November 07, 2012, 03:52:13 PM »

Test for capacitance, and when satisfied, solder the whole assembly (along the two top support bars) in its final position across the gap in the strip loop.

Sorry I didn't respond but I think this plan is a good one.

Quote
In particular, connecting capacitor plates only at their corners, with small-area solder connections, seems like a bad idea

Your sketches don't have small area solder connections, IMO.  I don't exactly know the scale but it looks like you're planning maybe 1.5-2cm of length where the cap is soldered to the connecting bars?

If you have really big plates you're going to get some stray capacitance issues.
Logged

73,
Dan
http://www.n3ox.net

Monkey/silicon cyborg, beeping at rocks since 1995.
JAHAM2BE
Member

Posts: 286


WWW

Ignore
« Reply #6 on: November 07, 2012, 08:45:04 PM »

Test for capacitance, and when satisfied, solder the whole assembly (along the two top support bars) in its final position across the gap in the strip loop.

Sorry I didn't respond but I think this plan is a good one.

I will admit that I may be obsessing over an insignificant detail, but re-reading W8JI's above-linked post about current flow at the edges and tapering connections has me worried again about the last-proposed capacitor geometry, with the shaft normal to the loop plane. Consider the current flow along the 36cm-wide strip loop: it is maximum at the edges; it is reduced over the central portion of the strip. Now, consider, with the last-proposed capacitor mounting, how current will flow from the strip into the capacitor plates. The plates will be mounted such that their normal vectors lie along the axis of the loop - in other words, some plates will be closer to the high-current edges, and most plates will be closer to the low-current central area of the strip. I wonder if this will create additional resistance or current hot spots.

Alternatively, trying to keep in line with W8JI's comments, I'm considering the following arrangement. (I'm away from my 3D modeling program right now, so a verbal description must suffice.) Mount the capacitor vertically with its shaft running down the central axis of the loop (similar to my initially-proposed mounting in my first post). Instead of mounting each set of stator plates on two vertical bars, mount each set of stator plates on a single, vertical strip of copper. So we have a wide (5-10cm?) vertical strip on the left that holds a set of stator plates, facing an opposing vertical strip on the right side holding the opposing set of stator plates. (The large opposing mounting strips will increase stray capacitance between the two sides of the capacitor, but if it reduces losses, I'm willing to accept that.) Then, taper the main loop's strip so that it goes down linearly in width from the full 36cm width down to the approximately 10cm width of the stator-mounting, vertically-oriented copper strip. This way - it would seem - the maximal current flow on the edges of the main loop strip will continue to flow on the edges of the tapering section and continue to flow on the edges of the vertical stator-mounting strip, thereby (assuming loop current is constant around the loop circumference and along the strip length) allowing every stator an equal amount of contact with the maximum-current-flow edges, in contrast with the shaft-out-of-the-loop mounting where only some stators are nearer the edges.

Opinions? Obsession over an unimportant detail, or possibly an improvement in capacitor ESR?

I have a butterfly capacitor modeled in FDTD electromagnetic field simulation software, so maybe on the weekend I can try to see if I can enhance the model to include a portion of the main loop, then see if I can detect any differences in current density with the different mountings.

Quote
In particular, connecting capacitor plates only at their corners, with small-area solder connections, seems like a bad idea
Your sketches don't have small area solder connections, IMO.  I don't exactly know the scale but it looks like you're planning maybe 1.5-2cm of length where the cap is soldered to the connecting bars?

Yes, that's about what I'm planning. I don't have a feel for how small "too small" is in terms of surface area for a very low loss solder connection in this particular demanding application (magloop).

If you have really big plates you're going to get some stray capacitance issues.

Right. I think a squeezing-plate capacitor would be the only way to have a very-wide-range homebrew capacitor, but after making several sketches I'm thinking the mechanics of making it work are too much trouble, so I'm back to the butterfly capacitor idea again.
Logged

WX7G
Member

Posts: 6328




Ignore
« Reply #7 on: November 08, 2012, 03:03:41 AM »

For maximum breakdown voltage the capacitor plates should be rounded rather than pointed.
Logged
K4PP
Member

Posts: 67




Ignore
« Reply #8 on: November 09, 2012, 11:27:11 AM »

"Opinions? Obsession over an unimportant detail, or possibly an improvement in capacitor ESR?"

I have a loop that uses clamped connections and small soldered areas. It doesn't use a butterfly capacitor but considering how well it works with the connections its not supposed to have, I believe any of your designs that you've shown will work exceptionally well. The biggest challenge will be the construction process, but it looks like you are well on your way.

K4PP
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!