New idea for making a grid

For posts specifically relating to fusor design, construction, and operation.
cooldayr
Posts: 3
Joined: Sun May 16, 2010 12:08 pm
Real name:

New idea for making a grid

Post by cooldayr » Sun May 16, 2010 11:49 pm

Hello
Recently I have come up of an idea for making a PERFECT grid
If you have ever heard of shapeways you would no they are a 3D printing company (and are cheap)
If you can design a grid and upload it you can have it printed in steel making a perfect grid
You would however need to design it in a 3d program

Tyler Christensen
Site Admin
Posts: 551
Joined: Tue Feb 24, 2009 2:08 am
Real name:

Re: New idea for making a grid

Post by Tyler Christensen » Sun May 16, 2010 11:55 pm

When I was just starting out about a year ago I thought the same thing and designed a model of a grid in 3DS and uploaded it, but it said steel was not an option for that geometry size and shape. Perhaps that has changed.

Although for higher power operation steel is a very poor grid, at a minimum stainless steel is required to avoid extreme out gassing that steel would exhibit, along with the need for higher temperature tolerance. It could work well for lower power fusors though, although for lower power there is little need for optimal grid geometry.

cooldayr
Posts: 3
Joined: Sun May 16, 2010 12:08 pm
Real name:

Re: New idea for making a grid

Post by cooldayr » Mon May 17, 2010 2:49 am

My bad it can be printed in stainless steel

Jerry Biehler
Posts: 902
Joined: Tue Nov 24, 2009 8:08 am
Real name:
Location: Beaverton, OR

Re: New idea for making a grid

Post by Jerry Biehler » Mon May 17, 2010 6:51 am

Its not true stainless steel its stainless powder mixed with a bronze that basically brazes the powder together. And the finish is not so hot too.

User avatar
Doug Coulter
Posts: 1312
Joined: Sun May 27, 2007 7:18 pm
Real name: Doug Coulter
Location: Floyd, VA, USA
Contact:

Re: New idea for making a grid

Post by Doug Coulter » Mon May 17, 2010 9:04 am

Materials aside, and even good SS isn't truly up to snuff, a "perfect" grid has to be geometrically possible, which of course depends somewhat on one's idea of perfect. You cannot for example, design a spherical grid that has equispacing of the wires/elements at all points, forgetting for the moment that you have to connect to it someplace. With a cylinder grid, my choice, you can at least have that in the middle, but not the ends so easily, and have to tolerate end-effects, which I suppose you could call "fringing" for a fairly loose definition. But at least it can be truly right in one area, and even the end effects may not all be bad in real operation.

There are also some other tradeoffs involved, like how much interception area the conductors create, which has a lot to do with losses due to grid-ion collisions (not to mention secondary electron emission). But too small and there are implied sharp corners to generate field electron emission and also be sputtered off by ion hits during operation, changing the dynamic of how it works over fairly short operational times. And making it so weak as to be hard to handle and less able to take heat.

Some of which can be countered using techniques similar to beam power electron tubes, catching things at low energy to shadow things they would hit at high energy, taking more than one grid to accomplish. Or you could make the elements/wires so thin that direct interception became less, but then you have mechanical strength issues, particularly at high you'd probably try to run it at anyway.

Only after that does the exact design and fab begin to really matter (and then, it does matter, a lot).

Precision is pretty important, so new fabrication tech that promises this is always welcome, but frankly doesn't even have the specs available in any old design beam power tube re accuracy and relative alignment over temperature and shock variations. These had spacings between elements sometimes in the microns range, and stabilities to match -- and managed that over some pretty wide temperature swings. This tech doesn't seem to have that if more than one part has to be used and the parts match to those levels.

When I can get the equivalent of CAD/3-d printing in the material of my choice, say tungsten/rhenium and it has precision specs like frame-grid vacuum tubes did I will get pretty excited and it will be a real step forward.
Why guess when you can know? Measure!

Hector
Posts: 158
Joined: Mon Jul 07, 2003 1:15 pm
Real name:
Contact:

Re: New idea for making a grid

Post by Hector » Tue May 18, 2010 11:42 am

I remember back in the 90's when I visited Dr. Miley at UI he and Robert Stubbers , (now Dr. Stubbers) explained to me how when they first started making Fusors thought that the grids had to be "perfect", so they would polish them for hours. After some experimentation they discovered that it did not make a difference.

User avatar
Doug Coulter
Posts: 1312
Joined: Sun May 27, 2007 7:18 pm
Real name: Doug Coulter
Location: Floyd, VA, USA
Contact:

Re: New idea for making a grid

Post by Doug Coulter » Tue May 18, 2010 1:19 pm

Of course, there could be a lot of other reasons for that observation -- like the grid field was so inaccurate at a macro scale that the surface finish of microns didn't matter, something else wasn't up to snuff, and so forth. If they were posting record Q's, the observation might be much more meaningful.

For what it's worth, we don't see huge differences due to surface finish here, and at any rate, sputtering tends to make the grid self polishing. We DO see enormous differences when wires are bent, unequally spaced, or other geometric errors exist in either the grid of the shape of the surrounding outer electrode. The main effect we see from surface roughness is that it becomes easier to start (light off, in the vernacular) with some roughness, probably because of some field electron emission helping with getting ionization started at the beginning. But of course that goes away quickly as sputtering makes everything smoother unless some gross error was made constructing the grid (an errant pointy wire end sticking out for example).

Many things dealing with charged particles only work a little, or very poorly, until *everything* is right.
Current Q's reported for fusors indicate that this is where we are now -- very poor operation vs power input, as many here have pointed out.

A look through the history of vacuum electron tube development would be most educational for most people here and elsewhere working with fusors. Very easy to make say, a triode with low mu and low transconductance -- with almost anything laying about the shop and sloppy hand work (I've done it myself, just for fun -- even built an oscillator with it, barely).

To get to more modern tube specifications, vastly better, it took serious precision and understanding about how charge behaves in various structures and fields, which the charge motion itself affects. We are using particles of the other polarity that are also heavier, but much of the same basic behavior (with appropriate scaling) obtains here.

Modern (relatively speaking) vacuum electron tubes use grid wire spacings down to the 100's/inch, spacing to cathodes in the sub mil range, and alignment between fine pitch grids that are marvels of mechanical design in something that changes temperature quite a bit, and has E fields strong enough to deflect weak structures. Often, fairly expensive and exotic materials were required to meet these specifications. And they didn't do all that for the fun of it in those profit-driven companies, it's how they got the performance up by factors in the 2-3 digit range from the old V or W filament and a bit of screen wire and a flat plate -- which is about where fusor development is *now*.

Or take the case of a modern quadrupole mass spectrometer, or Quistor ion trap - sure, you could hook up some rods or rings and caps to some AC and DC with arbitrary bad mechanical precision and guesses at the AC and DC fields, ratios, and AC frequency, and it would be pretty easy to say the concept was bunk, you never saw mass selectivity or very little...and that rod spacing accuracy or shape wasn't important, because you didn't have the other stuff right, and simply assumed it didn't matter. You'd be dead wrong, of course, as modern mass specs would prove, but it might seem reasonable as an observation.

So I believe the original post was relevant. Just wasn't the whole story.
Why guess when you can know? Measure!

bk8509a
Posts: 40
Joined: Sat Oct 31, 2009 8:47 pm
Real name:

Re: New idea for making a grid

Post by bk8509a » Wed May 19, 2010 5:31 pm

Just pointing this out, as I was about to hop in this bandwagon.

"DISCLAIMER:
Please note that the materials we use for manufacturing the models make the models suitable for decorative purposes and they are not suited for any other purpose. The models are not suitable as toys to be given to children. The models should not come in contact with electricity or food & drinks and should be kept away from heat."

User avatar
Rich Feldman
Posts: 1321
Joined: Mon Dec 21, 2009 11:59 pm
Real name: Rich Feldman
Location: Santa Clara County, CA, USA

Re: New idea for making a grid

Post by Rich Feldman » Wed May 19, 2010 6:32 pm

Brian Kelleher wrote:
> Just pointing this out, as I was about to hop in this bandwagon.
> "DISCLAIMER:
> Please note that the materials we use for manufacturing the models make the models suitable for decorative purposes and they are not suited for any other purpose. The models are not suitable as toys to be given to children. The models should not come in contact with electricity or food & drinks and should be kept away from heat."

If you'll all please excuse some chattering on a tangential topic...

Got to watch out for those disclaimers! Here's one of my favorites, from the TDK catalog of high-power ferrite cores. http://www.tdk.co.jp/tefe02/e16_2.pdf Right after the 2-kilogram UU core sets, they have core clamping straps with app notes that include:

• "Do not use the bands for purposes other than to secure ferrite cores. The bands are not edible. Keep them out of the reach of children."
Attachments
core_straps.JPG
core_straps.JPG (5.89 KiB) Viewed 2308 times
All models are wrong; some models are useful. -- George Box

User avatar
Chris Bradley
Posts: 2930
Joined: Fri May 02, 2008 11:05 am
Real name:

Re: New idea for making a grid

Post by Chris Bradley » Wed May 19, 2010 6:40 pm

reminds me of a disclaimer I saw in a microwave oven instruction book that read something along the lines of "do not use this oven as a projectile in a sling".

(I believe there was a phase when geeks thought microwave ovens were the coolest and most amusing projectile for their trebuchet projects!)

Post Reply