Advantages of the Dual Hemisphere?

For posts specifically relating to fusor design, construction, and operation.
Post Reply
bk8509a
Posts: 40
Joined: Sat Oct 31, 2009 4:47 pm
Real name:

Advantages of the Dual Hemisphere?

Post by bk8509a »

Here's a question I've been wondering about:
What are the advantages of using the dual hemisphere conflat design? Its the predominate chamber that everyone uses. From what I can see, it allows you to use a one grid system because you can use the outer grid as the chamber itself, but its price limits you to a tiny chamber and a usually tiny viewports.

I've been discussing using a 6 way cross system that has 6 6'' CF ports. I would then put in two grids on the inside and ground the outside one. The design is in the picture below.

I also have an alternative design that attaches both grids together using a ceramic stand off and just a grounding wire to ground the outer grid.

Although I've seen other designs such as the cylindrical system, the designer still uses the outer chamber as a ground and puts a cylinder grid on the inside. Can anyone enlighten me?

-BK
Attachments
Cross Design.jpg
cross.jpg
cross.jpg (16.09 KiB) Viewed 3999 times
Cross Design2.jpg
User avatar
Chris Bradley
Posts: 2930
Joined: Fri May 02, 2008 7:05 am
Real name:

Re: Advantages of the Dual Hemisphere?

Post by Chris Bradley »

Brian Kelleher wrote:
> From what I can see, it allows you to use a one grid system because you can use the outer grid as the chamber itself, but its price limits you to a tiny chamber and a usually tiny viewports.

What's your point? The question must then be; "how much do you want to see", and "how big a chamber can your pumps efficiently cope with/how much D can you afford to stream through it"?
bk8509a
Posts: 40
Joined: Sat Oct 31, 2009 4:47 pm
Real name:

Re: Advantages of the Dual Hemisphere?

Post by bk8509a »

The main question is Why is the dual hemisphere design predominate?
User avatar
Doug Coulter
Posts: 1312
Joined: Sun May 27, 2007 3:18 pm
Real name: Doug Coulter
Location: Floyd, VA, USA
Contact:

Re: Advantages of the Dual Hemisphere?

Post by Doug Coulter »

Hi, I'll toss in a thought or two. I'm the cylinder guy, by the way, and I like them -- no way you can "comb the hair on a sphere" and get actual spherical symmetry anyway -- ask any topologist. With the cylinder, at least the part away from end effects can be perfect (in theory). Heck, if you can't get it in theory, then why assume you can in practice? There's a chance, always, but it's kind of a slim one.

I do run the tank as the outer electrode for now, though I've done a little bit with two grids as well, but not enough to have interesting/good things to report yet, and never did simply ground the outer grid anyway -- I was using it more like a control grid in an inside out vacuum tube design, and letting ions hit the outer grid at lower energy input vs hitting the high voltage one with more wasted energy on them in those attempts, which are ongoing, but the two grid thing is much harder to do mechanically with any decent precision in this set of tests -- and metal sputtering onto any inter-grid insulating supports makes a lot of troubles unless a major design effort is put into that issue.

In any case, even with the outer grid charged to some kv negative, things do get past it and hit the chamber walls anyway, and that's pure loss. If your outer grid is really a solid sphere, you've just built a bowl in a bowl, no gain unless you are getting some other advantage that way, and indeed you may.
For example, testing with various outer to inner grid size ratios would be done this way perhaps.

We are doing one of our cylinder things in a big Tee (6" pipe, about 14" long). The vacuum system connects on the middle arm, and we just put a SS screen over that to keep the field shaped right -- it works fine for us.

Our other cylinder thing is in a side arm of a much larger tank, and it works great, some might say impossibly well, but we have good numbers and much replication to back up our claims.

Frank S posted some pix of a two grid system with the outer grid positive, and all that inside pyrex bowls to keep the ions in and to prevent making an inside out fusor between the outer grid and the tank walls. As far as I know, he never posted any actual neutron outputs from this test though. The pictures were nice anyway.

Personally, I like the big cross-type thing -- if you want a spherical E field in there (neglecting other sources of error for the moment) you can put screens over the port holes to maintain the field spherical, or make cylinders, whatever. I think the dual hemisphere is mainly due a "religion of the sphere" and some copy-catting going on, rather than any real science justification. I do know that it's pretty important for the shapes to match. We tried a nearly exact duplicate of Richard's grid here in a cylinder and the results were terrible, so bad we didn't go to the work of making a spherical outer conductor to test that way, as we were already doing better with cylinders than anyone is doing with spheres.

I like large view ports and a lot of ports to stick in diagnostic sensors myself. I can't see how anyone learns much from just making a copy of someone else's fusor, other than "things well known in the art" already. But that's me, I think there's probably some other opinions here on that.

Kind of had a chuckle over the $$$Newark in your art. Yup, good place to overpay at. I use Digikey a lot, but they are also on the high side -- but the service is great and they have the stuff, kind of like the McMaster of electronics.
Why guess when you can know? Measure!
adrian.f.h
Posts: 39
Joined: Tue Feb 17, 2009 1:04 pm
Real name:

Re: Advantages of the Dual Hemisphere?

Post by adrian.f.h »

Hi

Well I'm not actually fusing anything yet but I don't think my approach is that far away from it. My chamber is a small cross and it works fine so far but it's much too small. It's difficult to get in the inner grid and cleaning is a nightmare. So take care to get a chamber, that has somewhat bigger ports than those KF-25 on my Fusor. But even this should work.

Regardless of this, for me most of those readily build crosses are too expensive. Buying a sphere, the ports and getting this stuff welded together professionally doesn't seem to be a bad idea. Just use at least one large port for the hv-feedthrough.

Adrian
User avatar
Chris Bradley
Posts: 2930
Joined: Fri May 02, 2008 7:05 am
Real name:

Re: Advantages of the Dual Hemisphere?

Post by Chris Bradley »

Brian Kelleher wrote:
> The main question is Why is the dual hemisphere design predominate?

But your question implies it may not be the best solution after you've just pointed out benefits of cost and ease of use. What else would be the best solution?

...like asking why cars don't have 5 wheels.... I'm sure someone would build you a car with 5 wheels if you like.... what's your problem with the 4 wheeled type? Same here - why do you want to make the view bigger and need bigger vac pumps?

If visibility is what you want, do like me!!: viewtopic.php?f=15&t=7222#p49094
bk8509a
Posts: 40
Joined: Sat Oct 31, 2009 4:47 pm
Real name:

Re: Advantages of the Dual Hemisphere?

Post by bk8509a »

The reason I've brought this up is that I've found a 4 way cross just lying around. The OD of the flanges is 6", so I would need some reducers.

Adrian, How small is your chamber?

-BK
Attachments
cross.jpg
bk8509a
Posts: 40
Joined: Sat Oct 31, 2009 4:47 pm
Real name:

Re: Advantages of the Dual Hemisphere?

Post by bk8509a »

Doug to the rescue.

Always great advice. Really thought provoking. Although I'm super attracted to the aesthetically pleasing and interesting design of the spherical chamber I think you convinced me to use this cross that I found lying around in the department. The knife edges are fine. All it needs is some heavy duty methanol/acetone cleaning. I think I'll get a better view, save money, and save time, but I wont have that cool looking spherical design.

Still I need to think about it a bit more.

Thanks again Doug, you keep giving me prompt and valuable advice.

Doug Coulter wrote:
> Personally, I like the big cross-type thing -- if you want a spherical E field in there (neglecting other sources of error for the moment) you can put screens over the port holes to maintain the field spherical, or make cylinders, whatever. I think the dual hemisphere is mainly due a "religion of the sphere" and some copy-catting going on, rather than any real science justification. I do know that it's pretty important for the shapes to match. We tried a nearly exact duplicate of Richard's grid here in a cylinder and the results were terrible, so bad we didn't go to the work of making a spherical outer conductor to test that way, as we were already doing better with cylinders than anyone is doing with spheres.
User avatar
Carl Willis
Posts: 2841
Joined: Thu Jul 26, 2001 7:33 pm
Real name: Carl Willis
Location: Albuquerque, New Mexico, USA
Contact:

Re: Advantages of the Dual Hemisphere?

Post by Carl Willis »

Brian,

The appeal of the hemisphere approach to most do-it-yourself users is probably cost. I have built two fusors that way. The hemispheres are sold as architectural ornaments for a few dozen bucks and they have a nearly ideal fit with standard CF bored flanges. A single equatorial flange is all that is necessary to seal a spherical chamber, whereas with pipe stock and blank flanges the complement of premade flange hardware required would double. If money is an object, the tried-and-true RB Wagner flagpole ornament approach has a lot in its favor.

Also in favor of the spherical geometry: good symmetry of the radial electric fields (may or may not be of interest to a particular builder).

Since you have a CF cross, there's no particular reason to avoid using it.

-Carl
Carl Willis
http://carlwillis.wordpress.com/
TEL: +1-505-412-3277
adrian.f.h
Posts: 39
Joined: Tue Feb 17, 2009 1:04 pm
Real name:

Re: Advantages of the Dual Hemisphere?

Post by adrian.f.h »

I second Carl's answer. For me the most important reason why not using a cross like this is, that buying one and adding the reducers is that expensive.

May be there are better chamber designs regarding efficiency and such.

I don't see any reason why not using the one you got. At least as long as you don't have something better.
However you should at least check the prizes for all the stuff you still need to make this thing work and then you can still decide if building a new one is the better thing.

That's what I would do.

My chamber has an outer diameter of 7 cm (2,76 inch). That's small.....
I will beginn scrounging the parts for a bigger shell as soon as my current fusor works, although I have another chamber but this one is more suitable as a vacuum testbed.

I'm confident that I get a refill for my wallet soon.

Adrian
User avatar
Doug Coulter
Posts: 1312
Joined: Sun May 27, 2007 3:18 pm
Real name: Doug Coulter
Location: Floyd, VA, USA
Contact:

Re: Advantages of the Dual Hemisphere?

Post by Doug Coulter »

I forgot to add a couple things (and I put up too many words here anyway sometimes).

One is the big ports are great for getting in and out of the tank for various purposes -- but! Try pricing those gaskets! Owwww. This is a reason we use a lot of 2.75" stuff, and weld in nipples for it when we don't have enough -- great for gages, feedthroughs both low and high voltage. That smaller size is fine for all sorts of things, and not only are those gaskets a lot cheaper, but with a little "touchy-feely" on the bolts, you can reuse them about 3-4 times. You just only tighten to the point of no leaks, more or less; once you have once gotten your system leak tight and have some confidence in it, you can get away with that -- this is "post graduate" trickery. This does not seem to work out as well for the larger sizes for whatever reason. TIG welding these smaller things into any chamber isn't that hard at all, it's actually easier to do than most other welding is -- making the holes and jigging is, as usual, most of the work of this or any other welding.

For my big tank, I finally gave in and just bought a hinged door/window. Yes, it hurt the wallet, but I'm glad of having it every time I use it, which is a lot. Compared to the new turbo and the mass spectrometer that money was kind of in the noise anyway.

FWIW, I would advise anyone starting out on this to very definately NOT make a design that's either hard or expensive to get into and out of for testing things. That will limit what you test and what you learn. If every time you have an idea some part of your brain goes "ouch, that's going to cost me $20 in gaskets and an hour or two of just taking it apart and getting it back together on top of whatever it takes to try my idea" you won't try many things.

For BillF's smaller tank (which is a Tee, 6" ID), we (probably Bill) found an adapter flange for the 8" OD CF on one end. I then bored a bigger hole in this and added an O ring groove. A Viton O ring and a piece of borosilicate glass from McMaster give about the same effect as the fancy door, but at about $1000 cheaper, which suits us all fine. For the other end, where the HV comes in, we made a flange adapter out of a piece of billet aluminum that also holds an O ring in a non-conventional seal that works great. We only use 5 finger tight bolts on that, the vacuum does the rest of the work to make it seal.

Since it would be far too much for this forum to handle, I will be putting some details up on my own website real_soon_now for a bunch of things like this -- the picture selection and typing is "in progress" every day for that.
I will post to that effect when that's a done deal. The content will include detailed fabrication information on all the trick stuff.

For the moment, the first pic is the initial window implementation when we were first building Bill's system. The spring loaded window retaining washers have since been replaced with a somewhat improved set of finger clips for convenience. They are only needed to keep the window from falling off when there is STP air in the tank. Pressure drop when the system is under vacuum is what actually provides the sealing forces for the O ring compression. Very straight forward. You can see the SS screen we put into the tank to keep the E field uniform, it seems to work fine for that, and is one more layer of protection against dropping things into the pump stack. You can use this trick to get any internal tank shape you want as regards E field. Or as Carl mentioned, get some of that ornamental stuff and use that. I welded a 2.75" cross on the top for things like gages, ion source inputs, faraday probe outputs, gas inlet and so forth. I've since welded in another nipple at the bottom to support a high current feedthrough for evap and sputtering uses. Heck, "ports is ports" and it's hard to have too many.

The second picture shows the quick and cheap (well, fairly) flange we use on the other end. This end of the Tee had a problem with the knife edge that was going to be real hard to repair, so we took a piece of 6061-t6, made some jigs, and made what you see here. The bolt holes you can see don't go all the way through, they held the jig so I could easily turn this in my lathe. There is another set on the other side that also don't go all the way through for mounting the HV feedthrough. The jig in this case was simply a metal plate welded on a 1" diameter piece of steel rod with those 4 holes drilled in it, turned flat and straight before using it. The bolts handled the torsion load on the lathe while the 1" steel rod located the center on the hole we'd drilled in the Al plate first.

This has a step (not real obvious in the picture) that holds the viton O ring in the right place to work against the part of the CF on the tee just inside the knife edge. It seals great, and we just finger tighten the 5 bolts (which get sloppy loose under vacuum, but it just doesn't matter).

Cheezy as this looks, it's trouble free, no leaks at all other than the usual water permeation through the viton. This system is way overpumped with the 3" Varian diff pump, and we are in the process of replacing it with a nicely smaller air cooled one from a surplus leak detector to make it easier to run low gas flow rates.
Attachments
ChpWnd.JPG
QckFlg.JPG
Why guess when you can know? Measure!
Post Reply

Return to “Fusor Construction & Operation (& FAQs)”