All journeys

For the VA Tech fusion team to discuss and display their work

All journeys

Postby fusordoug » Sun Jan 30, 2011 8:16 pm

Begin with a single step, and sometimes the journey is the destination. Just getting the ball rolling here, this forum is intended for the VA Tech crew to use to ask questions, document results, and generally keep in touch. I might ask later to move some things from here to the main topic forums on the board, or link here from there when we get some super-nice stuff here.

Here's some pictures Lee Hall sent along to put up to get the ball rolling. They are just getting going now, so they can use our help -- ya'll other members chime in as seems appropriate.
And of course, the Tech crew -- you know more about these pix than I do, so comment!
Entire Setup.JPG
Entire setup

Gas System.JPG
Gas system

Grid.JPG
Grid

Power Supply.JPG
Power supply -- another Spellman here

Vacuum Pump.JPG
Pump


I'm going to encourage them to get a high vacuum pump to add to that last. This operating regime is right on the edge with just a mech pump, and having plenty makes all the rest a lot easier, particularly when you want to get to purity.

Go for it guys, we're rooting for ya :)
Why guess when you can know? Measure!
User avatar
fusordoug
Site Admin
 
Posts: 46
Joined: Wed Jul 14, 2010 1:59 pm

Re: All journeys

Postby Lee Hall » Tue Mar 01, 2011 11:27 pm

Over the past couple of weeks, our design team has been hard at work designing and installing a vacuum bypass loop and a new fill system with more sensitive flow control. We are consistently having a bad experience ordering parts from AirGas; needle valves instead of diaphram stop valves, shipping parts that weren't ordered, poor advice on some components, etc. At the recommendation from AirGas, we bought some brass compression fittings to connect our 1/4" copper tubing to our valves and Tees. After a long and exhausting journey with attempting to get these fittings to work, we have determined that it is obvious these fittings are not vacuum rated. We have ordered new fittings from a vacuum supply company that should be in over the next couple of days.

However, we have been successful in installing a 102.3 kOhm ballast circuit between our power supply and high voltage feedthrough. At the lowest vacuum we have been able to achieve recently (67 mTorr), we decided it was important to test out our electrical system while we waited for vacuum parts to arrive. We cranked our power supply up to -1 kV and noticed that it was reading around 5.6 mA of current as well. Amazed since we had not seen such a current to this point, we turned on our camera looking into the reactor and noticed a startling sight. For the first time, we have gotten a glow. It was arcing from about 3 different places on our spherical, spiral cathode without tripping our power supply. The glow started off as a purple tint, but then briefly changed to orange as the voltage was increased to around -2 kV. At this point, the current read out to around 14.6 mA, fluctuating slightly. The glow finally settled to a more 'white' glow. Noticing our cathode heating up (slightly orange), we decided to shut it down. We view this as big step toward successfully achieving fusion; we are confident that once our vacuum troubles are fixed it will only be a matter of time. Pictures are included for viewing pleasure (hopefully).
Attachments
Arc67mTorr.JPG
Arc67mTorr_2.JPG
Lee Hall
 
Posts: 24
Joined: Sat Jan 29, 2011 8:51 pm

Re: All journeys

Postby johnf » Wed Mar 02, 2011 12:49 am

Lee
very good --most excellent progress

now with your resistor a dead short at 1kV would be 10mA so the 5mA shows that the arc voltage is around 500 volts and your 2kV voltage figure also reflects this 1500 odd volts dropped on the resistor leaving around 500 volts on the arc.

now you need better vacuum to get the arc voltage up

it all coming together
johnf
 
Posts: 433
Joined: Sun Aug 08, 2010 3:51 pm
Location: Wellington New Zealand

Re: All journeys

Postby Doug Coulter » Wed Mar 02, 2011 10:25 am

Yes, it's a big deal to get from a pile of parts to a glow, as you guys now know!

Your pictures show the same thing mine do when the pressure is too high, and I also see 500-1000v on the grid proper under these conditions. We also see the little "bugle" formations here while pumping down from too-high pressure, and if they appear at random and move around, you've got a good uniform grid. If they're always at the same place - not so much. You'll rarely need to run over about 15 ma in real life with a fusor this size (and you'll need some fan cooling on the resistor to do that for long). Here we find our best output of neutrons when we set our supply (at full voltage) to limit around that current or a bit less, then adjust the gas pressure to stay under that limit, running about 10ma average in our setup for best output in our "normal/static" mode.

We've had no troubles due to grid heating here up to the point of melting a Ti one at about yellow heat. If anything, that cleans them up and gets rid of any gas they had ad or absorbed. You just need to be sure you're not hurting the end of your HV feed through with heat, which shouldn't be much problem -- it's fat and leaves the tank so it can take some heat out with it.

I've been able to use compression fittings fine, but most people don't understand at first just how perfectly clean and correctly formed it all has to be to work right. Heck, I've even used flare fittings, like the plumbing kind, but again -- have to do it just so. There is a flare fitting in my D2 supply (provided with the regulator) and it doesn't leak at all. I've had probably less trouble with Lesker than you've had with your vendor, but no one is perfect. The prices are all so high we get most of our stuff elsewhere anyway -- ebay and general scrounging (sometimes at the Va Tech auctions!)

I suspect you're going to need another stage of pumping -- diffusion or turbo. The situation is that if you could get the actual nameplate ratings of a typical 2 stage vane pump, you'd be fine, but that never happens even in a leak tight system, even after baking it out and pumping overnight. They rate these with a big degree of "specs-man-ship" to say the least. Any HV pump will then be far too much -- but that's what valves and bypasses are for. A good pumping system will cut down your outgassing time considerably as well. If you have trouble scrounging a HV pump on your budget, get in touch as we have some spares here. Right now, you're working on the lowest bottom of the Paschen curve -- you need to get over to the steep left side of it. At a too high pressure, you can indeed get to an arc, and the mean free path is such that you can't get any ions up to full speed due to collisions with neutrals -- and with too much gas, you can't ionize it all with any current under a lot of amps.

Strictly speaking, an arc is the lowest voltage and highest current mode, like a welder, and that's not what you have here, you simply have a normal glow discharge (which is good).
You can find some good information, along with correct terminology here, in the Phillips book. The first couple and the last chapter (with a neutron generator design) should help you there.


Congrats, you're on the way now!
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: All journeys

Postby Lee Hall » Wed Mar 02, 2011 3:11 pm

We ran another test today trying to get our LabVIEW program to control the power supply and changed some things around with the voltage and pressures. We got up to around 20 mA at one point with a 64 mTorr vaccum, which made some cool pictures. I have been thinking about what that little 'ball' in the center of the cathode is and my best explanation is that it is a collection of ionized air molecules. But our tests today showed that it was arcing from this ball, not the wire spiral.

Arc64mTorr.JPG
64 mTorr, ~20 mA


Once we decided to shut it down, we chose to allow the reactor to depressurize through the leaks we have to look at what happens to this glow. It looks like it gets bigger and less [dense?]. The current increased during this depressurization as well.

Arc64mTorr_2.JPG
~150 mTorr, ~7 mA
Lee Hall
 
Posts: 24
Joined: Sat Jan 29, 2011 8:51 pm

Re: All journeys

Postby Doug Coulter » Wed Mar 02, 2011 4:55 pm

My first good plasma looked a lot like yours. Here's a picture. This was done in some mixed gases I was playing with -- air, argon, neon, just to make it pretty, and be sure my gas gage wasn't reading off (most of them do on hydrogen -- they read high usually, sometimes very high).
plasma.jpg
First light here

This was at fairly high pressure so a 5kv power supply could light it up. This drug down the supply voltage to about 1kv at its current limit.

Until you can get the thing to just refuse to light at all at full power supply voltage, you still don't have good enough vacuum (and you've not tested the electrical hookup at full volts either). Remember, if you want say 99% purity, you've got to get roughly 100 times lower than the operating pressure first (and hold it there until all outgassing is over with -- which can take a week or more), or flow enough D through there for it to be 100 times what the leaks are.

Please don't call that an arc, it's not! It's a glow discharge. H Ions cannot really glow, by the way. All glow is from electrons dropping from higher to lower energy levels in an atom...an H ion has no electrons attached, or it's not an ion by definition if it's hydrogen (neglecting the rare chance that you can indeed have a negative H ion with an extra electron if you try hard enough) -- it can only lose one electron because it only has one to start with. Air ions can glow, because they can lose an electron but still have some left that might be in higher energy states to fall back into a lower orbit and release the energy as a photon. But in general, all visible glow is going to be from neutral atoms decaying after recapturing an electron. Most all the energy states available in other cases of high Z atoms give photons from UV to X ray energies.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: All journeys

Postby Doug Coulter » Sat Mar 05, 2011 4:16 pm

I just had my fusor torn down for various improvements, and I re-ran a pumpdown under voltage conditions. It appears your gage is telling some very large lies yo you (or is much closer to the pump with some restriction between it and the tank).

I checked the pressure range over which I see those cute blue bugle formations, and it is: 1.9e-1 millibar, many times higher than you are saying. Just below that, they dissapear and it starts to look more like what you saw here, with blurry at first, then a distinct focus showing up around 2.2e-2 mbar, and best fusor operation at about 1.6e-2 mbar.

So, you are a factor of ten at least from fusion on your vacuum system. Once you've seen this, it's almost better than a gage for a lot of things. My gage is a Pfeiffer pkr 251 and it's right on the tank, so it reads actual tank pressure better than one in the vacuum line would. It's known to read somewhat high on H or D, but....even with that, it's reading 10x or more what your gage is telling you at the same actual tank-gas voltage/current conditions.....your work is cut out for you.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: All journeys

Postby Lee Hall » Sat Mar 05, 2011 8:22 pm

I really don't like to hear that but I know we need to know! We have been fighting suppliers, hose connections, and various other problems for two straight weeks now and still haven't gotten back below 64 mTorr (according to our gauge). Shown in one of the original pictures on this thread, our sensor is located just downstream of the sphere (blue cap, horizontal cylinder on the left). Shown below is a picture of the gauge itself.
sensor readout.JPG
Pressure Sensor

Until now, I was fairly certain it was reading the correct pressure since it would read around 706 Torr when open to the atmosphere, which is close to the standard atmospheric pressure at 2000 ft (706.6 Torr). Also, we were aiming to test out our reactor when we are able to achieve pressures around 0.5 mTorr, so I'm not sure how much that helps us out with this new information. I'd certainly feel better if we had an additional pumping stage to reach lower pressures. But, we're on break as of now so it's not looking like we'll get much work done for about a week. Regardless, some fittings we need won't get here until March 15th at the earliest. Thank you for the heads up; this is certainly something we'll look at ASAP.
Lee Hall
 
Posts: 24
Joined: Sat Jan 29, 2011 8:51 pm

Re: All journeys

Postby Doug Coulter » Sat Mar 05, 2011 8:42 pm

What type of gage is it? Thermocouple?

At any rate, yeah, I don't like handing out bad news to anyone, much less you guys -- I'd rather see you kicking butt totally, frankly, and I think it might be possible. The intent is to make it so you get to fusion with less fooling around, is all that's motivating me, and while I was doing those tests today, saw some benchmark numbers on pressure vs what I saw. Figured I'd just pass that along. I have a guy coming down at some point with pro movie gear to do a documentary on us, and I'll have him do a split-screen off the tank view and our gage during a pumpdown that should actually be some useful data for all -- I don't have the stuff to do that just yet myself.

What you're probably going to find out is that once you have a couple of "touchstone" points, the fusor itself will be your best possible gage anyway -- being able to see what's going on in there is key, whether by camera or mirror (or that really neato chunk of thick lead glass like I have). Due to how steep that side of the Paschen curve is, most gages just aren't going to have enough resolution to show much in the range of interest, which is pretty tiny. For example, in my own fusor, at first the pressure I could run was about 2.2 e-2 mbar. Now with ion sources and all that other jazz, I can get down to 1.4 e-2 -- not exactly an enormous difference, and at say 9e-3, I can't get it to draw any current over microamps (plenty of X rays from field emission, but basically no fusion). At this point, I just use the volts of a current limited supply (that second grid) to tell me when I'm "there" and its far more accurate than any gage I've seen.

All the thermal type gages read way high on H (and isotopes) compared to air, sometimes almost as much as the ratio of atomic weights -- big number -- "air" is normally called Z=29. My opinion is that you don't really care that much as long as it's repeatable so you know when to turn on the power and so on. The hassle is, when you have leaks (I'm still hunting mine) you don't know the composition the gage is seeing, so that tosses another big variable into things. Ion gages (the PKR-251 has both) have their own weirdnesses with different gasses, depending more on how easy they ionize than how thermally conductive they are. Nothing's perfect. The pressures here are too high for a mass spectrometer to work right.

Getting a better pump before you get the leak isn't going to do a ton of good, from my experience. Due to what happens in the transition from viscous to molecular flow, even my super duper 530 l/s turbo can't take on even a ridiculously tiny leak -- a pinhole would be huge compared to what I have, yet it takes me from e-9 mbar to 2e-5 for example after an entire day of pumping down. Well, that doesn't tell the whole story. With no leak I get to e-6 mbar in a couple minutes, e-7 in an hour or so, e-8 overnight, and e-9 in a couple days with some baking here and there.

I'd bet you can find something surplussed at Tech -- we go to the auctions, but you guys get first crack at stuff like that, I believe (the auctions are frequented by resellers who bid everything to 1/2 to 2/3 retail, so we don't buy much at them). If not, well, you know who to call....we have some spare stuff here, some of which you've seen and we didn't pay much for it - we can let it go at cost. But it's not time for that yet. You should be able to get to fusion, albeit kind of lousy, with what you have, and it's more important to fix those leaks than have a killer pump for now.

Once you get there, and have the other issues hammered out, then....a good secondary pump is going to take you real far, but not until then.

Oh, for those of you reading along -- we're both being lazy here -- Lee is talking torr, I'm talking millibars. They're not all that different (760 torr == 1 bar), so it kinda works anyway as long as we don't start using things like microns (of which unit?) and like that. All our gages have more error than the conversion factor, depending on what gas is being measured!

We have a saying in stock trading -- "the ticker is the truth". Same goes here -- what you see in the tank is the truth, the gage is more of a guess or estimate unless you go real crazy and have one of those spinning levitated balls measuring friction or a manometer, both of which don't work real well down in this range anyway.

What you'd like to see is this -- the ability to completely turn off current drain by having a good vacuum. That will happen at not real far below the ideal pressure for a fusor. If you can get your gage down to 1/10 or 1/100 of that, you can start talking about having fairly pure fuel -- and it does matter, we've tested that. Up to about .5% impurity, you get one class of behavior, and better neutron output. Above that level, you get better stability of the fusor, but much less actual fusion. It seems a little impurity helps the thing stay "lit off" -- but makes it draw lots more current too, and a lot of that is wasted in ionizing the various components of air, which can really eat the power and heat things up -- probably because things like O, N, Ar, can be multiply ionized and transfer ionization to the D on collisions. The higher the Z of the impurity, the worse this effect is. They found at JET and ITER that just going from tungsten to carbon "beam stops" they reduced losses tremendously (big integer factor), even though in either case the amount of the high Z stuff that was in the plasma was a fraction of a percent...This isn't the same conditions, but the same effect does seem to occur with a fusor.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: All journeys

Postby Doug Coulter » Sun Mar 13, 2011 3:31 pm

I happened to be testing some things, and created some video that may be of use to anyone starting out on this path.
In the first one, I flooded the tank with way too much deuterium, and powered up with a decent current limit, 10 milliamps.
This shows what I saw, which you can relate to the pix from the Tech guys above.
http://www.youtube.com/watch?v=sY_jKJQUAlg


Note the smaller grid in the foreground is so poorly shaped it doesn't show bugles at any time -- so the tech guys are at least getting close. FWIW, that little guy does make a few neutrons if your test gear is sensitive enough, but it's way down there.

I'm not going to claim the gage numbers I read out here in millibars are absolutely correct, in fact, they're a little low if anything -- part of what I was doing today was recalibrating the gage.
All gages, this one included, have some sensitivity to gas type, and either read lower or higher depending on what they were calibrated for. My gage is a PKR-251 Pfeiffer, for which the manual is here:
PKR-251Manual.pdf
PRK-251 manual with gas type error numbers
(609.7 KiB) Downloaded 331 times

(read the appendix for the cal factors). Any thermal type gage will be similar in response to gas content.

Here I'm doing my trick to reduce the gas pressure in steps so you can see how the picture evolves, and when the neutrons start coming out. Note all the crappy reflections -- which is why I usually do these at night (and I shouldn't have had the radio on WVTF at the time, but hey, this is a Tech thread).
http://www.youtube.com/watch?v=Sb3x2yPFcB8
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Next

Return to VA Tech Fusion Project

Who is online

Users browsing this forum: No registered users and 1 guest

cron