Archived - Fusor II

[myID]

Hi-

finally I made my way back to Germany...
Had a little time to set up my new toys- and the pictures show what the setup looks like now.
I call it Fusor II because it is much more then just a small change:

(first picture from left to right)
- new HVPS: Bertan 105, 30kV 33mA, changable polarity
- NIM BIN completed (cabling not finished)
- new pumps: Trivac D4B and Turbovac 50
- new HV Feedthrough (sort of shielded- at least electrocution safe)
- ESP2 with (different) small He3 tube in paintcan

Results (perhaps 1hr run time...):
- no ultimate Vacuum tests so far- how ever- after 40 min P< 10e-5 mbar so at least no big leak
- getting a stable plasma U>25kV is a bit hard...
- fusion: piece of cake with this setup: just introduced some D2 and the ESP started ticking...

Things to do/ come:
- change Oil in the Trivac
- condition the chamber
- Argon purge gas artifice (for glow cleaning and refilling the chamber after use)
- MFC for D2 to be set up
- connect Nim Bin to big He3 Tube (still no SHV cables- btw.: thanks for the efforts Bill and Doug...)
- higher Voltage (connected my homebrew bridge/ driver to my 2 HF HV tanks- NICE arcs so far but still lots of open questions due to strange waveforms.....)
- way to throttle Turbovac (at the moment throttled with inlet filter to decrease pumping speed- also somewhat protects the pump, pump is flanged directly to the chamber on the backside)
- complete shield (lots of lead sheet and paraffin (50kg...) already obtained) (will also act as moderator for activation experiments)
- change Trivac from three phase to one phase

Remarks:
- SHILED YOUR VIEWPORTS: in my setup at U > 25kV LOADS of Xrays....
- also without visible plasma still loads of Xrays
- get a proper power supply- AWESOME investment- worth every cent...
- the pump deal with the Trivac and the Turbovac was GREAT...
- inlet filter is contaminated with the phosphors from TV tube production: nice lightshow

More to come when I have more time because I am super busy with loads of other things still...

And: YES: I spent a fortune- now I am broke and will have to sell some Fusor I stuff

Kind regards
Roman

[Doug Coulter]

Lookin' good, Roman!

I'd second all those observations myself.
I also shield my viewport from the inside with a clipped in piece of pyrex which is turning brown -- better than than my nice main window/door! We (actually BillF) were very lucky to obtain a nice *thick* lead glass window from a radiology dept and boy do I like that thing for the outside! According to my survey meter, nothing gets through worth worrying about -- it stays very near background level. Due to the higher voltages I'm running, I also wrapped the entire tank in lead -- that one one heck of a job to do, but again, worth it. Safe to stand near it while running now.

People (even CliffS, I think) thought I was nuts to just go buy a really good power supply.
I happen to agree with you -- worth every cent. I love the safety factor and ease of use of a really good one. Just slap it on or off, all nicely current and voltage controlled, interlocked and gee, it just plain works. If there's trouble, it's elsewhere. I'm just having great luck with Spellman supplies, from the new SL2KW (50kv at 40ma but it beats that spec) and used ptv ones too.

I too run into "interesting stability" from time to time and am researching that, as the transition gives me huge bursts of neutrons. Not enough data to make a definitive report yet, but tell ya what -- let a little air or water vapor in there (a percent or so) and it gets much more stable -- which doesn't help neutron output one bit but makes it easier to run. I am also finding effects depending on whether the grid has any sharp edges or not -- seems a little field emission kick starts it a little easier....not necessarily good, but it's definately there.

Gas control is just one tough one. Need good control on both pump speed and gas inlet. Most mass flow controllers don't go low enough to not be wasting a lot of gas. I've made countless tweaks to that part of the system and everything that improves control improves the system as a whole proportionally more than that. It's a really big deal. I'm using needle valves and orifice tubes on inlet and both spinning the pump slower and using a conductance valve on it, and am *almost* there. I've now managed to run half hours once in awhile without touching anything at all....but that's the exception.

Sad you're broke, but hey -- who needs money if you've got the right toys?

[Richard Hester]

The only thing I really worry about is if the tubo pump will really be happy running in the horizontal position - that may be worth an email to Leybold. I always like to hang mine vertical.

[Linda Haile]

Turbos are fine running horizontal Richard. They will even run upside down if required.

A diffusion pump is another matter entirely, though.

[Chris Bradley]

Lyn Haile wrote:
> Turbos are fine running horizontal Richard. They will even run upside down if required.

Such a blanket comment ("Turbos are fine") is cobblers. Depends on the pump.

These Turbovacs are greased mechanical bearings which I presume are therefore supposedly meant to be 'OK' in horizontal operation but I'd not want to run one unless vertical. I run a Varian with ceramic bearings that handles inverted operation, but horizontal operation is still to be avoided even with those, due to the asymmetric radial loading.

I am sure Roman will take the advice from Richard who has a demonstrated experience of running turbos, rather than from yourself who has none, if he's not already asked the question for himself.

Nice work, Roman. That's quite a few hours of work to make everything look so neat.

[Linda Haile]

I beg to differ Chris. I've no experience with mag-lev's, though.

[myID]

Hi and

thank you for your advise- the manual of the Turbovac says: "come as you are" like Nirvana did- so horizontal mounting is fine. I did not find any information on optimal pressure rise curve when flooding the system....
Further more (I flood REALLY slow) I do not like the sound the pump makes at some RPM (guess hitting some resonance....).
Sooooo.... Flood faster to make the pump stop faster?

Btw.: I will build a "shadow cone shield" like Richard suggested once, and I use a capilarry and needle valve at the moment to control the D2 flow.

Greets

[DaveC]

Roman -

My compliments!.... Very nice setup.

I will second Richard and Chris' caution wrt to turbo orientation. Better safe than sorry.


Dave Cooper

[Chris Bradley]

Roman, what do you mean, exactly, by "flood the system faster"? I'm not sure I like the sound of that, either!

[Linda Haile]

The reason I wouldn't run a mag-lev horizontally is because I'm not sure they are designed to 'crash' in this orientation. Mag-levs have to be scrapped after two crash landings anyway, even in the vertical position.

(Never buy a mag-lev off Ebay!)

It is common practice to run conventional turbo pumps horizontally, as Roman's manual states.

[John Futter]

The Turbovac50's can run in any position
they are not maglevs, they run Barden SiN balled greased bearings.
I know, I have had lots of them apart fixing / cleaning them (all ex Ebay).

Anyway nice setup Roman

[Mike Beauford]

Hi Roman,

I will add my appreciation to all the others regarding your setup. Are you going to add a MCA into the mix on your setup?

Mike Beauford

[Doug Coulter]

Nooooooo!

Don't flood fast, the rotor tips will burn up! That's at least one source of the evil noise -- sonic booms off the rotor tips.

I have some experience with this one. I have two pfeiffer pumps obtained new. If just let spin down in a good vacuum (which takes a long time) then they may have hit a minor acoustic resonance for a bit while doing that. After just one too-quick flooding, they are noisier due to imbalances created by uneven rotor tip wear -- an inrush accident is about the worst thing you can do to one -- and the resulting moan/howl is something you never want to hear even the first time -- and after once, never again.

If it's easy to hear how fast its slowing down -- you're doing it too fast. Takes most of my systems a few minutes to go from programmed speed to 25% of that (standby mode) at the fastest rate I dare do it. Takes maybe half an hour or more if the vacuum is good. These have some amazing bearings, and a heck of a lot of stored mechanical energy.

There is an astonishing amount of foot pounds of energy stored in a turbo rotor, even a little one, about like a car crash in a big one. Enough to be mechanically dangerous, tear out the mounting bolts etc on an inrush accident in some designs-- Pfeiffer literature is full of warnings about this. You can even wring off a bellows you use to keep vibrations away from the main tank -- easy. (I use one on the big turbo here and it makes things a lot happier re noise and vibration).

So in slowdown or shutdown -- do let in a bit of gas to save time, but you're already over the top if the pump is noisy at more than one resonant speed -- you're burning off rotor tips with frictional heat. Say stay below e-1 millibar for that process more or less. Even the normal running pressure of a fusor (esp if you turn off the backing pump) is plenty to get one stopped in reasonable time.

If I'm just going to go to bed anyway, I just let it happen in high vacuum, it gets done.

The very best place to vent is behind the turbo, assuming you have a turbo-drag pump. This lets the more robust drag part have nearly all the friction and heat, and it can take it much better. Even a pure turbo -- better to wear the rotor tips that matter least.

[Richard Hull]

Very, very nice Roman. It's fusing and that is the bottom line.

Richard Hull

[Linda Haile]

I think you may have mis-understood me John.

I said these turbovacs will run in any orientation, but I wouldn't use a mag-lev in any position other than vertical.

I wouldn't buy a mag-lev from Ebay because there is no way of knowing if it has 'crash landed'.

BTW, Do you know if it is possible to rebuild a maglev that has crashed?

[myID]

Dear all,

I finished my shadow cone shield (20cm Paraffin Block with channels for samples, 5cm highly borated Paraffin, 10mm Lead) and felt better about doing some tests.

The 2 He 3 tubes were both in a distance of 30cm to the center of the grid. The smaller one uses the paintcan setup, the bigger one was connected to my homebrew preamp and my NimBin and inserted in the Paraffin Block.
D2 was regulated with a needle valve, Turbo and forepump were running at full speed, Power came from the 1kW Bertan supply.
Chamber size is about 13cm x 13cm x 13cm, Grid dia 2.5cm.

The values for U and I are about +-1; vacuum values- perhaps in the ballpark..?
Runs were pretty stable for few minutes, I just had to regulate the gas flow a little some times.

I attached the table with the values. The NimBin failed due to missing shielding (I guess), the "Paintcan" performed admirably well!

Intersting observations:
Even at low accelerating voltage significant count rates!
No star mode evident- could be expected due to non symetric field in cubical chamber (?)
The (what I think) hot plasma zone looks more like a glowing disc- the videos I recorded are empty... Shoot...

I would advise everybody to use the "paintcan" approach- can not be more simple than this, Noise immunity is good, Background is low, reliable due to few parts:
Paraffin filled paintcan, He3 (B10, BF3) tube in the center, a MHV plug (SHV) in the lid, a cable to the meter... Every meter with adjustable (high enough) sensitivity and adjustanble(high enough) Bias should do.

Kind regards
Roman

[Steven Sesselmann]

Roman,

Nice tidy experimental data, well done..

Do you have any idea what the TIER was during the runs?

Steven

[Andreas Welzmüller]

Hello Roman,
Iam also a new guy who works on a fusor-project in Germany.
Is it possible to cantact you?
best regards
Andreas