Fusor2_DMS

For posts specifically relating to fusor design, construction, and operation.
User avatar
russssellcrow
Posts: 68
Joined: Fri Dec 27, 2019 12:40 am
Real name: Russell Crow
Contact:

Fusor2_DMS

Post by russssellcrow »

Greetings to all! At Dallas Makerspace we've been collecting hardware for the next iteration, Fusor2.

After slogging through building a Gieger counter, a Scintillator, a Fast Neutron counter, and a Spectrophotometer; we decided it is time to upgrade the Fusor to some real vacuum hardware so we will have something to measure, duh.

We went with a 6-way 8 inch with 4 DN40 ports as the main vessel. We've begun making the adapter to attach a Consolidated Vacuum Systems diffusion pump (old school, but bombproof) with our 2 stage mech pump.
Fusor2_baffleplate_polished.jpg
Diffusion_Pump.jpg
And here are the pics of the new Chamber:
275ich.jpeg
8inch_conflat.jpeg
User avatar
russssellcrow
Posts: 68
Joined: Fri Dec 27, 2019 12:40 am
Real name: Russell Crow
Contact:

Re: Fusor2_DMS

Post by russssellcrow »

Edit: should be "DN25 ports"
User avatar
russssellcrow
Posts: 68
Joined: Fri Dec 27, 2019 12:40 am
Real name: Russell Crow
Contact:

Re: Fusor2_DMS

Post by russssellcrow »

A little progress, down in the machine shop. I cut the zero length vacuum adapter on a CNC plasma cutter, then surfaced on a Colchester lathe, polished with a Ryobi sander using 220, 320, 600, 1000, and 1500 grit Wet or Dry. Next came the bolt holes on a vertical mill and hand tapping rig.
20211219_180630.jpg
We considered using this 19" network rack, but then opted to build and stain a wooden rolling table. Next maybe a Nixie tube display for the neutron counter. Yes, this is going for Steampunk appeal.
20211219_142816.jpg

Resized_20211219_205842.jpeg
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Fusor2_DMS

Post by Richard Hull »

Wow! You really did it the hard way, but saved money. Why spend it if you have access to CNC! Really enjoy seeing the big stuff brought to bear. A nice start on your fusor setup.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Fusor2_DMS

Post by Dennis P Brown »

Wheels are useful. Agree with Richard, a good cost saving measure but wood is, in some ways, superior to metal in any case.
User avatar
russssellcrow
Posts: 68
Joined: Fri Dec 27, 2019 12:40 am
Real name: Russell Crow
Contact:

Re: Fusor2_DMS

Post by russssellcrow »

I've been looking for HV feedthroughs rated above 60kV, they are expensive! So I figured, what the heck, give it a shot making one.

I figure a spark plug is good for over 100kV, and about 20 Barr of pressure. You never see exhaust leaked from the porcelain on plugs, even after 100,000 miles, so the seals must be incredible! The trick is to attach a high temp electrode to the plug.
20211230_213805.jpg
Tungsten is about as high temp as obtainable, and plenty high strength to hold our electrode (a tungsten carbide wedding band, nod to another poster on Fusor.net), and just happens to be the "T" in TiG welding. So the jig is designed to hold the plug, and the back of a TiG assembly in place (note the plastic test tube borrowed from Biology Dept to make the rod holder cylindrical and predictable). Lighting and sticking the tip in the right place is my job. PLEASE NOTE: the jig is drilled all the way through at the bottom, so the plug's electrode protrudes and touches the steel table, which is well grounded.

The 1st attempt was an abject failure, because I had used a Resistor Plug. They put resistors in modern spark plugs to quell RF interference caused by the ignition, so it doesn't jam the car's radio. The Miller TiG just kept dancing the ignition spark around the electrode, but never fired the Arc.

I then obtained a non-Resistor plug, and the procedure worked almost flawlessly.
20220101_171658.jpg
This still needs to be sealed into a 2.75" SS conflat. I will cut off the factory gasket, and use a Copper washer instead. Then JBWeld in the threaded hole and on the plug, and about 30 ft-lbs of torque should do the trick.

Now, if anyone knows how to punch a hole through a Tungsten Carbide wedding band, please chime in!
User avatar
Liam David
Posts: 518
Joined: Sat Jan 25, 2014 5:30 pm
Real name: Liam David
Location: PPPL

Re: Fusor2_DMS

Post by Liam David »

Good work on extending the electrode. Do be aware, however, that a spark plug will not withstand 100kV. You might get some 20kV before it arcs over. Also, pressure-tight is not the same as vacuum-tight (consider e.g. what difference a 0.0001 bar/s leak makes to 20 bar vs. 0.00001 bar). Spark plugs can leak between the porcelain and metal, solved using some JBWeld.

You might be able to cut the carbide ring with a diamond-tipped bit for a rotary tool. EDM machining would probably be the best method though.
User avatar
russssellcrow
Posts: 68
Joined: Fri Dec 27, 2019 12:40 am
Real name: Russell Crow
Contact:

Re: Fusor2_DMS

Post by russssellcrow »

For all the cool stuff we have at Dallas Maker Space, we do not have EDM (although there has been talk)

The Blacksmiths tell me you can heat TC in an inductive forge, then just hit it with a hammer and punch to make a hole. It's not a big investment, $13 for the ring. I think I will take the risk.
User avatar
russssellcrow
Posts: 68
Joined: Fri Dec 27, 2019 12:40 am
Real name: Russell Crow
Contact:

Re: Fusor2_DMS

Post by russssellcrow »

Do be aware that spark plugs withstand 100,000 volt charges, across the planet, in millions of cars, on a daily basis. This is part of what pushed the CAFE Standards from 20mpg to 27mpg overnight, these modern electronic ignitions.

Coil-over-Plug has become the standard, it delivers extremely hot sparks which wires could never achieve.

Try to keep up.
User avatar
Joe Gayo
Posts: 404
Joined: Sun Jan 06, 2019 9:34 pm
Real name: Joe Gayo
Location: USA

Re: Fusor2_DMS

Post by Joe Gayo »

Realize "russssellcrow" that you are the student and David and others here are the teachers. No one should "keep up" with your strategy if they desire a better than demo device.
User avatar
russssellcrow
Posts: 68
Joined: Fri Dec 27, 2019 12:40 am
Real name: Russell Crow
Contact:

Re: Fusor2_DMS

Post by russssellcrow »

If you have data which refutes what is known, please present it.

Otherwise, just keep it, nobody is interested!
User avatar
Joe Gayo
Posts: 404
Joined: Sun Jan 06, 2019 9:34 pm
Real name: Joe Gayo
Location: USA

Re: Fusor2_DMS

Post by Joe Gayo »

I find it interesting, that two separate people inform you of a flaw and your response is to deny and ignore it. I hope you aren't in charge of anything important.

There are many sources of data that can refute your claim, but consider why the plasma forms in the first place. Electric field breakdown is well understood. Most materials used for vacuum feedthroughs are designed to 220V/mil. If the ceramic on the spark plug is 0.125in thick that would give you 27.5kV of "punch-through" dielectric strength. However, breakdown over surfaces can be much less depending on their intersection with the electric field lines. Besides surface contamination, the electric field can promote electrons to migrate and cause a surface breakdown. This is a distinct possibility if one could possibly apply 100kV to that spark plug since the electric field would cause electron field emission. All of this is an over-simplified explanation and there are many other factors to consider (i.e. triple-points, sputter contamination, insulator charge-up, etc.).

Read -
https://cds.cern.ch/record/1693330/file ... 4.0952.pdf
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Fusor2_DMS

Post by Richard Hull »

I say let the man fail! Nothing at all teaches like abject failure. You do, you fail, you learn.

As all here know, I started with spark plugs and went on to very special spark plugs. I did my first fusion, 20k n/sec, with a special racing plug. We have lots of posts here on spark plugs recommended for demo fusors.

I found out real quick in my real fusion quest that going much over 20kv applied to any plug was asking for failure. By ditching the plug and going to a more formal vacuum rated technical grade feed-through insulator, that I could get up to 40kv or more without "air arcing". Air arcing is the classic fail for auto plugs. Internal arcing under vacuum is another spark plug fail mode.

I found it amazing the way spark plugs are built. They are designed to continuously take a pounding from below at huge compression ratios without leaking. They are assembled and sealed fast from below as that is where the beating occurs. However, they are not rated for keeping out 14.7 PSI of air at the molecular level from above against an outer-space vacuum level from below. This is not what they are rated for. I found that a little epoxy where the porcelain touches the metal plug body on top would seal well enough. However, it is the air arcing and internal weakness of the plug that limit its ability to handle high voltages above 30kv with hundreds of watts flowing through it. Automotive apps never have any real wattage demanded of them. A 20kv spark can jump about an inch in air, but in an auto plug it is shorted out in a .035-inch gap. The tip never gets to 20kv as the impedance of the average coil as a power source is so high that much of the voltage drop during the arc appears within the coil secondary. Naturally, the wires must be rated for the full voltage as short intervals of full voltage can appear until the plug arcs, (milliseconds), and then the coil secondary will have the bulk of the voltage across it.

Like Franklin said....Let the experiment be done. We are here to teach our experiences related to fusion and all facets of all technologies touched upon in the effort. Just about everyone of the old boys and electronics engineers here have tried everything imaginable. Most seemingly brilliant efforts fail and we report our results. We teach, all listen, but some just have to learn their own way. Thank goodness it is by the doing and by experiment. That is the way it is done in real life. Most of us rather learn from and be taught by someone who has "been there, done that" as opposed to reading in some book what someone claiming to know something is writing about in that book or a magazine. The real world intrudes on our reading and school based learning, even college level teaching. The old adage about "those who can't do, teach" finds that teaching motto smashed to pieces here, for "all that teach here have done and can do". Students, as always, can take it or leave it or test it.

Note: I did get that first real fusor, in its first iteration to 30kv with the special order racing plug by getting an oversized silicon boot and filling the inside with HV silicon putty and forcing it down over the wire, HV terminal and insulator. However at real fusion voltages over 30 kv the internal failure of the plug showed up, killing the auto plug from use in a real continuing fusion effort I attach images of fusor III, first iteration.....I did mill and machine and TIG weld 100% of FusorIII. FusorIV and V were built from all stock conflat fittings. (I learned my lesson here again after fusor III to use formal vacuum rated fittings), I still had to bore and TIG weld everything with only a tiny amount of machining. We live, "do", and in the doing know how to improve the "do" next time....repeat until it is the best ever....and perhaps repeat yet again.

Richard Hull
Attachments
Note: this plug was changed to a far more suitable special racing plug and got to 25-30kv before failing.  Iteration #2 had a formal vacuum rated insulator
Note: this plug was changed to a far more suitable special racing plug and got to 25-30kv before failing. Iteration #2 had a formal vacuum rated insulator
No knife edges on this baby!  I sealed the big flanges and window port using pure indium wire.  Worked great!
No knife edges on this baby! I sealed the big flanges and window port using pure indium wire. Worked great!
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Fusor2_DMS

Post by Dennis P Brown »

One thing I've learned here (and in work, as well) is that being polite even when correct but challenged is both respectful and appreciated by everyone. I have certainly been on both sides (wrong ... ok, extremely wrong, on an issue now and again or correct but told otherwise) and no matter always was glad afterwards when I didn't respond in a non-professional manner. Sometimes it can be both - some applications due to special design works fine but in more typical situations, does not. In which case we have to remember that as well.

I've taken my lumps now and again and moved on - we all make mistakes and shouldn't get concerned but are here for one real reason - enjoyment of sharing a fascinating subject with others both to learn and to teach.
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Fusor2_DMS

Post by Richard Hull »

Taking "lumps" are OK provided they are "learning lumps". After the pain of the lumps subsides, the smart man will be wiser. We sometimes take pride in having read this or that and feel we know something. However, in application, the real world and the universe of physics intrudes into our pridefulness, about what we think we know. We learn that, which is in books, even the best books, are tales of highly controlled and often isolated descriptions of idealized or mathematically derived experiments or setups. Pressing what we have learned from books or school into a mixed process situation will show us there are always "catch 22s" involved around what we have learned and think we know. Solving these real world issues around such mixed processes makes us truly knowledgeable and often a bit more humbled.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
User avatar
russssellcrow
Posts: 68
Joined: Fri Dec 27, 2019 12:40 am
Real name: Russell Crow
Contact:

Re: Fusor2_DMS

Post by russssellcrow »

I can get a little snarky when challenged with facts that conflict with what I've seen and tested, (and I've had a couple of shots), all apologies for any offense taken. "Try to keep up" is a motto I live Moore and Moore these days.

What Richard Hull holds forth is the PROCESS, and this is the crucible in which we purify our concepts. Materials, techniques, and technology race forward; harnessing them to do what we want is often a whole lot of beating one's head against a wall.

I am going to test the dielectric breakdown voltage of the NGK plug, since all I have left to do is the bottom side construction. I'll report back the data.

Best regards!
User avatar
russssellcrow
Posts: 68
Joined: Fri Dec 27, 2019 12:40 am
Real name: Russell Crow
Contact:

Re: Fusor2_DMS

Post by russssellcrow »

Fusor2 is coming along pretty well, still there are many things left to do (as always).

Our 1st Vacuum Test went pretty well using only our Roughing Pump, measured by our BOC Edwards Pirani gauge (P.No D02187000) sampled by our standard platform Heltec ESP32 with Wifi webpage interface.



Starting Vsignal = 9.71 Vdc
Starting webpage Value = 4095
Ending Vsignal = 5.26 Vdc
Ending webpage Value = 2325

Using the transfer chart from BOC Edwards, this puts the pressure somewhere between 10 E-1 and 10 E+0 Torr, so we will need to make a calibration curve. With a 2 stage mechanical pump the lowest we should expect is 5 x 10 E+0 Torr.

We have the Diffusion Pump with a room temp Chiller/Radiator loop charged and ready for the next test. It is awaiting only the Pneumatic Control Relay to close the main Roughing Valve. When we have hit the bottom of the Pirani range (2.0 Vdc), we will fire up the Bayard Alpert Gauge to see how low this system can go.

What is the trick to orienting pictures? I tried adding this pic, then I remade one with a 180' flip, either one still oriented upside down.
Attachments
Added_MFC_D2_Electrolyzer_+180.jpg
Added_MFC_D2_Electrolyzer_+180.jpg (71.63 KiB) Viewed 8440 times
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Fusor2_DMS

Post by Richard Hull »

Corrected image but is now a 72kb file. I rotated the image to vertical, took it into paint and without alteration saved it again. Then removed your old image and installed an upright version. Try this and then edit your larger image back into your post. Many times, cell phone images are reversed or upside down when uploaded.

A very nice looking setup, indeed.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Fusor2_DMS

Post by Dennis P Brown »

Agree with Richard, very nice setup.

A bit confused on your nomenclature: That is we generally use 10^ for a base 10 exponent. So 1 torr is 1*10^0 and half a torr would be 5* 10^-1. This makes what base we use crystal clear (i.e. not base 'e'.) Not terribly important but how we use powers here.

Assuming this is the case, then a good two stage pump should achieve 5*10^-3 torr or 5 microns. Your nomenclature is saying 5 torr for a typical two stage pump. Also, 10E-1 = 10 E+0 but you know that - it is easy to mistype exponents using that nomenclature - so another good reason not to use it. ;)
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Fusor2_DMS

Post by Richard Hull »

I have always preferred 10e as in torr only! 10e-3 torr is never used as it is 1 micron. Use microns to 1 micron and then switch to 10e-6 torr for a very deep vacuum. Most of us operate in the positive whole micron range. (mechanical pump and fusor operational vacuums) Our gauges, as found surplus, are micron TC gauges. Only those speaking and bragging or saddened by their best bottom pressures use torr....10e-4 torr should be used and not 0.1 micron. Euro folks can speak in pascals, bars, etc. and that is fine. However, most here can barely afford a TC gauge in microns. Microns are where we actually do fusion and operate.

I have always suggested torr for rotten vacuums....100 torr. Using Torr negative exponentials for really nice deeper vacuums.
For good "technical" and fusor vacuums, speak in microns.

Just sayin'

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
User avatar
russssellcrow
Posts: 68
Joined: Fri Dec 27, 2019 12:40 am
Real name: Russell Crow
Contact:

Re: Fusor2_DMS

Post by russssellcrow »

I'm gonna let the nomenclature conflict go (you can Paschal, Torr, or In-Hg as you wish), we found a giant leak on Sunday (June 5, 2022), with a really simple technique.

We turned on the 2-stage roughing pump with both valves closed, and hit a hard A/D of 1550 (about 2.5Vdc on the BOC graph, 10-E3 microns). So, we re-torqued all the system seals, then went looking for leaks.

We sprayed just Water (no fancy Helium Sniffer) onto just about every seal in the HVAC Loop, and it went off the scale when we sprayed the Deuterium Injector Port.

This works because if you have a vacuum leak, it sucks water droplets into the system, which instantly evaporate and raise system pressure immediately.

To fix, I'm torn between stir welding the connector, or just using JB Weld. JB Welding is reversible, stir welding is not.
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Fusor2_DMS

Post by Dennis P Brown »

Water!? NO! That is the worse substance to use. Use either alcohol (absolute is best) or better, "Dust Off". Getting any significant amounts of water in the system is counter productive.

Epoxy has issues - it with tend to release over time due to thermal expansion issues of most metals. Then leak. While I'm against vacuum grease in general, it is better then epoxy for test purposes to check for sealing. There are special high vac liquid sealants on the market that work well. One area where epoxy is good is with sealing against ceramics.
User avatar
russssellcrow
Posts: 68
Joined: Fri Dec 27, 2019 12:40 am
Real name: Russell Crow
Contact:

Re: Fusor2_DMS

Post by russssellcrow »

HaHa, water is "the worst"; I can think of a lot worse...Hydrazine or Nitric Acid come to mind.

Using an atomizer bottle of water, you can watch the vacuum gauge go up, and it comes right back down in seconds as the microdroplet of water evaporates and gets purged. Besides, we are running out of Helium.

So Dennis, you don't approve of JB Weld in a threaded joint? I'm thinking I'll smear the threads of the Swagelok 1/4" adapter and torque it into the Conflat of the Deuterium Port. The next day we'll repeat the vacuum test. If it still leaks (water spray test), I'll buff the inside with a grinder, and TIG stir weld the adapter to the Conflat using no filler rod.

Check this person out, they have no more heat sensing nerves in their hands, but a lot of skillz:

https://m.youtube.com/watch?v=990RYJXYTPY

Best,
Russell
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Fusor2_DMS

Post by Richard Hull »

I have TIG Welded in the vacuum world for years. Doing what is shown in the video requires long attained skill in current control, a very sharp point on the tungsten tip and rather rapid movement as seen following the "tack" welds. Watching the video, I noticed the skill as I saw one of the tack-welds did not fully connect and the welder noticed this and immediately re-tacked the tiny joint. It is a great joy to watch such artifice.

Machining advantageous weldment lips on the work is another way to avoid filler rod and make the work secure and look professional. I do not like or use filler rod in my TIG exploits. Mostly because extreme strength is never demanded in any of my SS vacuum system work. No need to lay down a heavy bead.

Machine to advantage first, and then TIG weld.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
User avatar
russssellcrow
Posts: 68
Joined: Fri Dec 27, 2019 12:40 am
Real name: Russell Crow
Contact:

Re: Fusor2_DMS

Post by russssellcrow »

Good advice Richard; in order to make a good "autogenous" weld (without a filler rod), the tolerance between the 2 pieces must be very tight (<0.010 inch is preferred).

In that video, the guy without gloves just bears onward in a straight bead, and it looks pretty good. He is prly welding a stainless exhaust, not a High Vacuum System.

If you are making a rocket (or a high vacuum), you prly want to "stir" the weld. What is a stir? When you strike a TIG arc, and you keep it close to the part, you will see that the ionized liquid metal is "attracted" to the arc. The arc starts on 1 side or the other, but to make it cross the gap, you have to "stir" it in circles. My friend, the Scottish Engineer, recommends that I stir the weld in Figure 8s, but I have found that simple circles work just fine.
Post Reply

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