First Bubbles!

For posts specifically relating to fusor design, construction, and operation.
Post Reply
Caspar12
Posts: 9
Joined: Thu Jan 20, 2011 6:08 pm
Real name:

First Bubbles!

Post by Caspar12 »

Today we finally received our long awaited bubble detector from BTI that Nick Depasquale was kind enough to forward to us. Since we started looking for one last December we of course immediately fired up the pumps to try it out!

The first 7 hours we spend to prepare everything for the first run (of course everything started to leak so we had to reassemble the chamber all over again) the detector didn't show any bubbles. So we measured no background radiation over a period of 7 hours.

Around 10 pm we were ready for the first HV testrun, only to find out one of the rectifier diode strings of the x-ray transformer failed. Because we had no more spare diodes we decided to go ahead and run the setup at half power. We can do this since it is a center tapped transformer thus by removing one of the rectifiers we only utilize one coil and get a half wave rectified output (you could call it a 50% duty cycle).

In the beginning we had a lot of trouble to control and stabilize the pressure while maintaining a deuterium flow of approximately 0.1 milliliter/s in a range that we could run the HV as high as possible. After 2 hours of tuning the D2 flow (with a needle valve) and diffusion pump flow (by squeezing a flexible tube between the chamber and diffusion pump) we finally had it fairly stable at 25kV.

We then started the first run averaging (due to variations in pressure) at 25kV peak at approximately 250W input power. This first run totaled to about 3 minutes at 25kV peak. And we were extremely pleased to discover the first bubble in the bubble detector.

The second run was at same voltage and power and lasted approximately 4 minutes at 25kV peak, this yielded no extra bubbles.

The third run at same voltages and power lasted approximately 5 minutes at 25kV peak and yielded another bubble!

Since we had no bubbles in the preceding 7 hours and received two bubbles exactly during the high voltage/power runs we are fairly certain we have finally achieved fusion! :-D

Our setup for the fusor is as follows:
- Chamber is a salvaged cylindrical steel chamber measuring 13.5cm in diameter
- The (steel) inner grid has a diameter of 3cm
- The pumps consist of an Edwards ES50 rotary vane pump for backing and an Edwards E02 diffusion pump (with DC704 oil)
- For the high voltage we use a 50kV (rms) 350W philips dental x-ray transformer housed in a custom oil container. (Since we don't use the center tap we run it at max 25kV rms)
- Deuterium supply consists of a electrolysis setup currently filled with 3ml of D20. Using platina electrodes at 15V, 0.1A and some baking soda we have a fairly efficient electrolysis setup. We dry the deuterium with silica gel and use a needle valve to regulate the deuterium flow.
- For pressure monitoring we use 2 Robinair thermistor gauges in a custom panel. Unfortunately these are not correctly calibrated but we estimate a pressure of 5mTorr during the runs. We want to add an ion gauge to get better information about the pressure.
- For hv voltage/signal analysis we use a fluke nixie multimeter and an analog hameg oscilloscope.
- To control the x-ray transformer voltage we use an 0-230V 5A variac.
- For radiation monitoring we currently use a DOM 410 geiger counter with a large glass gm tube.
- In the picture you can also see in the rack a pulse counter, heinzinger 10kV 10mA dc supply (currently not used), and our pc and I/O module in progress.

Hopefully we will be able to operate at full power (400W) next week when we have replacement diodes. We will try to get a higher neutron flux and try to operate it for longer periods since we had some difficulty to keep the voltage/power stable. Our future plans are to build a polywell in our new 28.5cm stainless steel vacuum chamber. (still under construction)

We will gladly answer any questions.

Bubbles from The Netherlands,

Martijn Rambonnet
Bent de Lange


Picture 1: The 2 bubbles in the BTI-pnd 30bbs/mrem on the left.
Picture 2: Clear starmode with D2, unfortunately the camera doesn't show the colours very well.
Picture 3: The entire setup.
Picture 4: More starmode
Picture 5: The chamber
Attachments
pic4.jpg
pic3.jpg
pic2.jpg
pic1.jpg
IMG_20110514_233152.jpg
User avatar
Steven Sesselmann
Posts: 2127
Joined: Wed Aug 10, 2005 9:50 pm
Real name: Steven Sesselmann
Location: Sydney - Australia
Contact:

Re: First Bubbles!

Post by Steven Sesselmann »

Bent,

Congratulations on first fusion, your neutron numbers will improve as you learn how to stabilize your system.

In the future could you report your experiment in full, so we can get a nice picture of how well it worked.


Deuterium supply..............:
Detector type....................:
Detector efficiency............:

Cathode Voltage Potential:
Current draw.....................:
Distance to detector..........:
Run time...........................:

From this we can estimate the neutron count

Well done, and hopefully Richard accepts you into the neutron club with two bubbles.


Steven
http://www.gammaspectacular.com - Gamma Spectrometry Systems
https://www.researchgate.net/profile/Steven_Sesselmann - Various papers and patents on RG
User avatar
Martijn Rambonnet
Posts: 18
Joined: Thu Mar 17, 2011 7:34 am
Real name:
Contact:

Re: First Bubbles!

Post by Martijn Rambonnet »

Hi James nice to hear you are still progressing with your experiments!

We will try to give some more detail on the deuterium supply soon but our electrolysis setup is extremely basic. We have modified a glass syringe for D2O (ebay) storage and we use the needle end of the syringe as a feedthrough for the platina electrode (platina wire from ebay). Experiments with other electrodes like graphite, stainless steel, steel, etc all failed. Over the electrode we have placed a smaller glass syringe upside down to which the deuterium line is connected and the outer electrode is connected from above. Our setup is quite similar to that of Andrew Seltzman (http://www.rtftechnologies.org/physics/ ... olysis.htm) but with a much lower budget (and nowhere near as nice as his setup). We borrowed his idea to use baking soda (NaHCO, ebay) as an electrolyte and we are very pleased with the results (opposed to for example kitchen salt).
To dry the deuterium we use silica gel (ebay, changes color from yellow to purple when is absorbs water damp) in a long piece of plastic tubing. Our whole electrolysis system is thus at approximately 1atm and we use a needle valve (CO2 aquarium needle valve, new ~15 euro) to regulate the flow into the chamber.

As I said we will try to give some more detail/drawings/photos in the future. We also have some other experiences from constructing the fusor on a very limited budget we will try to post soon.

Steven, for ease of reading a reformatted report of the experiment:
Deuterium supply..............: D2O electrolysis @ 1milliliter/second at 1atm
Detector type....................: BD-PND bubble detector
Detector efficiency............: 30 bub/mrem
Detector measurement.....: 2 bubbles

Cathode Voltage Potential: 25kV peak, 10kV average (half wave rectified)
Current draw.....................: 25mA at 10kV average (250W)
Distance to detector..........: 8cm
Run time...........................: 3 + 4 + 5 = 12 minutes

Using the your fusion calculator (http://www.beejewel.com.au/) this gives a fusion rate of 4e3 fusion/s and a Q in the order of 1e-11.
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: First Bubbles!

Post by Richard Hull »

Great work guys!

This is the kind of stuff we like to see. Lots of adapting with less than ideal materials. Lots of fiddliing to make it go and, yet, success! Key here is that you are about the second or third electrolysis fusor. It is nice to see this workaround alternative to bottled gas turning out successful fusion.

Your detailed description of the effort and its results, along with a full rundown on your gear was enough to tell me you did fusion. Sounds about like my first blast with fusor III back in the late 90's. Lots of effort for a little fusion recorded on the Bicron Scintillation system. The BTI beats all other detection systems in the hands of "first timers" even at this lower level.

I have added you both to the neutron club. Keep us informed on future runs with higher voltages.

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
Carl Willis
Posts: 2841
Joined: Thu Jul 26, 2001 7:33 pm
Real name: Carl Willis
Location: Albuquerque, New Mexico, USA
Contact:

Re: First Bubbles!

Post by Carl Willis »

Congrats, guys, on an operational and neutron-generating system. I know it can be more difficult for non-US fusioneers to round up their equipment, so kudos on overcoming those hurdles. I look forward to more reports on your efforts.
Carl Willis
http://carlwillis.wordpress.com/
TEL: +1-505-412-3277
Caspar12
Posts: 9
Joined: Thu Jan 20, 2011 6:08 pm
Real name:

Re: First Bubbles!

Post by Caspar12 »

Thanks all for the kind responses! We are excited to have it working, and very surprised that we were able to produce some neutrons even on our first attempt. I can't say it all worked flawless, but apparently good enough for some fusion.

It has taken us about 8 months of experimenting with all kinds of chambers, HV-supply's, grids, seals, feedtroughs, windows, pumps and all the other equipment to get to this point. The electrolysis was probably one of the easiest parts of the project.
Part of the equipment we salvaged from the waste bin at one of the faculty's that were either defective or to old to be used. Other equipment was bought from a Dutch equivalent to Craigslist, of course ebay and things like the bubble detector with help from here. That were some interesting projects on their own.

We will make a more extensive report (with pics) on all the steps we have taken to get it to work. It might be interesting for other fusioneers who are on a low budget.

The diodes have arrived, so we'll be able to make some new runs at full power shortly. After that we'll try to get the photomultiplier/scintillation detector working. But mostly we'll be concentrating on our next project, a superconducting polywell.

Attached a picture of our first plasma from some months ago, with a tacky-tape HV-feedtrough.
Attachments
First plasma.jpg
User avatar
Martijn Rambonnet
Posts: 18
Joined: Thu Mar 17, 2011 7:34 am
Real name:
Contact:

Re: First Bubbles!

Post by Martijn Rambonnet »

Hello all,

We received the diodes this week and we did a run at full power. However the diode string quickly failed again (although we made it 40kV, 1A instead of 35kV, 1A). We tried again with a second diode string creating a 40kV, 2A diode, which was able to withstand the continuous full power of the transformer. However after a short time running the transformer setup began to produce some smoke from within the oil so we had to stop the experiment. We have a suspicion it is arcing at the high voltage output of one of the coils where two wires are very close together. We are going to try improve the isolation this week.

The short run, a few minutes at ~30kV peak, full wave rectified, resulted in one bubble. We also did a background reading over three days but this yielded no bubbles.

We also attempted and endurance run with the Heinzinger operating continuously at 10kV and 10mA which resulted in a burned fuse. The third thing to fail was the electrolysis power supply (max 30V, we run it at 15V for electrolysis) which decided to go berserk and suddenly output 45V for no apparent reason.

Also a few hours were invested into the initial construction of an HV control panel (19 inch 1U unit) which incorporates an emergency stop button, two key switches as HV interlock (one key for each operator), and some computer/externally controlled relais to switch HV sources (x-ray transformer, photomultiplier power supply and in the future ion guns, electron guns, etc).


Current fusor construction history and details
As requested a short overview of our experiences from constructing a fusor on a minimal budget.

Vacuum chamber
Our first vacuum chamber was our diffusion pump. When we got the pump the diffusion pump was a mess and we had taken it apart for cleaning. After some initial tests with the first chamber and some major leaks we did a quick run with a Plexiglas lid on top of the diffusion pump and an electrode through it. Crazy idea of course but hey it worked and we were very happy to see our first plasma!

For the real chamber we use 2 cylindrical steel pieces with flanges we salvaged from the scrap metal bin of a workshop. This had 3 major problems. Viewing port, seals between flanges and material (see photo 1).

Since it is made from steel we probably have a somewhat larger outgassing rate. Leak/outgassing is usually about 0.1-0.5mTorr per second (@50mTorr) when it has been open and it is currently 0.006mTorr per second (averaged over 0-700mTorr). Either way this is well within the range of the pump near the required pressure around 5-10mTorr for fusion. It just uses a little bit more deuterium but that is of no concern to us.
The other problems with steel is that it corrodes quite easily and for aesthetics we have coated the outside with a little bit of oil to reduce corrosion. On the inside we have not seen any corrosion but since we keep the chamber under vacuum and very dry this is logical.

The second problem was sealing two flanges. Depending on which flange there is only a few millimeters of space for the seal. Our first attempt was to use tacky tape to make a vacuum seal. Normally this is used for vacuum bagging of composite parts for vacuum injection and can probably be best compared to bubble gum and clay. It is reasonably sticky but can be easily removed, any small residue can be dissolved with white spirit. We use a very thin ring of tacky tape on the outer flanges to make a good connection (where there is only 3mm of space). At the inner flange we use a custom o-ring. It is made from o-ring cord (nitrile rubber I think) which is cut to size and glued together at the ends with a cyanoacrylate type glue The major problem with this is that the o-ring is not fixed in a groove so it can easily pop out between the flanges when it is not carefully placed. other o-rings we use are for the inlets and outlets which consist of normal plumbing o-rings. For the window we have also used cyanoacrylate glue to seal some screws.

Since our vacuum chamber is not compatible with standard vacuum connections and due to obvious cost reasons we have made our own viewports. The initial viewports consisted of plexiglas plates which were bolted against the outside flanges (see photo 1). We discovered that Plexiglas was very prone to outgassing when it was being hit by beams from the plasma and this very much interfered with the pressure and stability of the plasma at higher voltages (>5kV). The other problem was that Plexiglas shields almost no radiation (I am pretty sure we would measure near 1000cps at 25kV with our glass beta/gamma gm tube). The current windows consist of a metal end plate with a sheet of glass clamped against it with an o-ring in between. The front window is a small Plexiglas window with a glass plate in front of it (n the vacuum chamber). This has greatly reduced the radiation and solved any outgassing problems.

High voltage
The first step in our high voltage carreer was to use a (multiple) car ignition coil(s) which were switched by several cooled 2n3055's at high frequency (basic 555 driver). This was enough for some nice plasma's but had very limited power. The second step was using a neon sign transformer (8kV, 25mA center tapped). We were unable to succesfully remove the center tap and could go no further than 4kV but this was again a step in the right direction. After investigating possible other power supplies (spellman, glassman etc very nice but way to expensive) we found an x-ray 50kV, 7mA transformer on ebay for 60 euro's. We created a container where the transformer could be submerged in oil (we use aviation mineral oil). For recitifcation we have tried to work with microwave diodes but they did not survive long. Currently we are using two strips of 35 1N4007 diodes (1kV, 1A, 0.02 euro a piece). For voltage monitoring we have a made a voltage divider from a special 10MOhm 30kV resistor from an old HV probe.

As an high voltage cable we use 20kV cable from ebay (few euro's, it is probably also possible to use an HV cable from a crt monitor but we haven't tried it yet) shielded by 2 layers of silicone (vacuum) tubing.

For the high voltage feedthrough we started of with a simple piece of metal through a plate of plexiglass. Due to heating of the electrode at high power this quickly failed. We now use an used 18mm aircraft spark plug. After some small modifications this worked great up to 10-15kV but at high voltage is still sparked from the electrode to the outside of the spark plug. We tried to prevent this by using a plastic tube (high outgassing with plasma's), glass tube (broke/burned almost immediately with a plasma), teflon (sparked through, danger this can produce extremely toxic gasses!) and finally we found a porcelain tube that we were able to fit around the electrode.

Vacuum
The vacuum backing pump and diffusion pump were obtained from the faculty who were throwing them away as scrap metal. The backing pump was in reasonable condition but the diffusion pump was pretty much broken. The heater had burned through, the oil had become a solid layer at the bottom, it was covered in tacky tape and several small parts were missing. We were able to obtain a new heater by chance from ebay (for only something like 20 euro's) and with a lot of effort we were able to clean the inside of the diffusion pump. With some improvisation were able to replace any missing parts and seals. The next problems was to find a small quantity of diffusion oil for a reasonable price. After a lot of calls and searching we were able to buy 25ml of DC704eu for ~40 euro's (still expensive, but other sellers only wanted to sell large quantities).

For vacuum connections we mainly used transparant vacuum tubing. To mimize outgassing we keep these lengths as small as possible and we use hose clamps to make sure they are thightly connected. For the chamber-pump connection we use a short large diameter reinforced piece of tubing with a clamp in the middle which we use to reduce the pump flow at low pressures.

For pressure monitoring we use 2 Robinair analog thermistor vacuum gauges with a range of 1mTorr-760Torr. Although they are currently not really correctly calibrated they are still extremely useful for reference values in tuning the vacuum for fusion.

As you probably have guessed we had our fair shares of leaks. Usually we have a suspicion were the leak is and we either use tacky tape to close the leak (and confirm the location) or we drench the connection in acetone (which gives a large jump in the vacuum measurement, warning: this will require some gas ballasting to get the acetone out of the backing pump oil, also this is not compatible with tacky tape seals). We have played around with an old ultrasonic leak detector but without any results.

Electrolysis
See our previous post for some details on our electrolysis system. Bent is currently working on a rendering and schematic of the exact details which will be posted in the near future.

Radiation detection
For x-ray monitoring we use a DOM 410 geiger counter with a large (~15cm long) glass geiger muller tube. It is sensitive to gamma and (hard) beta radiation. By chance we were able to buy a cheap CDV-700 with a 5979 front-end mica window probe which is also sensitive to (hard, not the low energy) alpha radiation. This proved to be very sensitive to x-rays but the probe unfortunately started leaking (if someone in the US wants a 5979 Richard Hull has a lot for sale).

For neutron detection we found no better alternative than a BTI bubble detector. We want a computer readable and non-expiring neutron detector so we are now working on a photomultiplier setup with the neutron scintillation foil from geohot.

Future plans
We are currently working on a larger stainless steel vacuum chamber (same design as Jeroen Vriesman's). We had taken a chance and bought a micro-ion gauge from ebay with an rs-485 interface but it seems to be broken so we will be looking around for an other ion gauge which will be necessary for the polywell.

Other efforts are focused on the design of the oil cooled (not yet super conducting) polywell design. We will also begin experimenting with electron and ion guns.

Photo 1: Third iteration of the reactor with large Plexiglas viewports
Attachments
IMAG0029.jpg
Jeroen Vriesman
Posts: 276
Joined: Mon Feb 07, 2011 4:08 pm
Real name: Jeroen Vriesman
Location: Netherlands
Contact:

Re: First Bubbles!

Post by Jeroen Vriesman »

fantastic!

you are planning to make a superconducting polywell wow.
User avatar
Martijn Rambonnet
Posts: 18
Joined: Thu Mar 17, 2011 7:34 am
Real name:
Contact:

Re: First Bubbles!

Post by Martijn Rambonnet »

The idea of a superconducting polywell is based on the fact that Bussard encountered a lot of problems with running his coils continuously (he did not use superconducting magnets).

This does however pose 2 major problems. The first is getting a superconductor; unless we can find some funding buying superconducting wire is too expensive, but it is also possible to make your own super conductor (YBCO).

The second problem is cooling the superconductor. Superconductors like YBCO require LN2 which ofcourse is a hassle to obtain/store but not impossible (a small cryocooler would be ideal but even diy ones are expensive).

Thus we are going to start with cooled restive magnets and if the results are interesting enough we hope to continue with superconducting magnets
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: First Bubbles!

Post by Richard Hull »

Very laudable effort on the future polywell work. It is good you are doing a simple fusor first. You are seeing just how tough it is to do simple stuff, or stuff that seemed simple. It is a big learning curve and you must master the simple fusor, vacuum, HV, gas (dueterium) handling, etc., before proceeding heavily into the polywell project.

Neutron detection is always a big issue too, but the BTI bubble detector doesn't lie.

Nice overall report on the physical setup and sorry you are experiencing numerous issues, system wide, but they will be solved one by one allowing much better operation.

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
David Geer
Posts: 136
Joined: Wed Nov 24, 2010 7:51 am
Real name:
Location: Colorado Springs, CO

Re: First Bubbles!

Post by David Geer »

There is a lead glass and lead acrylate/acrylic you can use for the view ports specifically for radiation shielding (the stuff found in x-ray rooms at hospitals and dentists). Not too expensive for the sizes you are looking at and can be prefab shaped to custom sizes and diameters, even with pre-drilled mount holes.

As for the superconducting electromagnets... I've been trying to find those for over a year now and the closest thing is a supplier through globalspec or alibaba for wholesale suppliers. If I find a cheap AND good source of Rare Earth magnets in custom shapes/sizes, I'll let you know.

SEG enhanced vacuum effects, hand-held cutting lasers for beam-on-target, polywell... these are all ideas I've been tossing around for a mk. 2 fusor once my truck is up and running and I can start collecting components for my first fusor.

I wish you good luck and we all hope to hear and see more of your next unit.

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

Re: First Bubbles!

Post by Doug Coulter »

Good Show! Here we found that those first bubbles were the hardest ones, and it wasn't so bad to move on from there -- we got improvements quick once we got everything reliable and realized what was important for this.

I note you're using an outer screen to try and sphere-icalize a cylinder tank. We got much better results just matching grid geometry to tank geometry here. Your outer grid might need more density to get a really round field as is.

http://www.coultersmithing.com/forums/v ... f=25&t=246 has some pix of this effect, trying various grids in various tank geometry.

What we found is it's more important to match than anything else.
Why guess when you can know? Measure!
Caspar12
Posts: 9
Joined: Thu Jan 20, 2011 6:08 pm
Real name:

Re: First Bubbles!

Post by Caspar12 »

Thanks for the input, that is a very interesting thought!

We have indeed made the outer grid to sphere-icalize the field around the inner grid, but didn't see much difference with our without the outer grid. I read on you website that making the (inner) grid as symmetrical as possible yielded a big improvement in reaction rate. We definitely got room for improvements there, seeing that our grid is far from perfectly spherical.

Unfortunately our DIY HV powerfully was pushed a little over it's limits on our last fusion attempt and is probably unrepairable. (smoke coming from the core of the x-ray transformer). So we have to find a replacement for that first, but after that we definitely have to give the 'grid with chamber geometry approach' a try.
David Geer
Posts: 136
Joined: Wed Nov 24, 2010 7:51 am
Real name:
Location: Colorado Springs, CO

Re: First Bubbles!

Post by David Geer »

The grids basically form an electromagnetic field around the core and having the varied polarities from outer and inner forces everything to the center where it tries to contain the particles. The emphasis on symmetry and quality manufacture of the grid yields a more even field cage that allows for better containment.

I'd like to see a rare earth electromagnet sonic welded/laser welded into a sphere cage and used as the inner grid. Similar to the Rodin coils or torus electromagnets. But I'm veering off the subject. Bottomline is this: if you don't give a grade A effort on the grids then you won't be getting a grade A result out of them. Quality = Results.
- David Geer
User avatar
Doug Coulter
Posts: 1312
Joined: Sun May 27, 2007 3:18 pm
Real name: Doug Coulter
Location: Floyd, VA, USA
Contact:

Re: First Bubbles!

Post by Doug Coulter »

You're probably zapping diodes that *seem* overrated because of arcs and the following inductive kickback that takes the voltage way over nominal. A small filter cap on the output plus the usual ballast may not be enough to protect them in some setups - we've run into that here too. You wind up either way-way overrating them (eg several to one), or taking a lot more care about preventing kickback surges from the transformer however you do that. We find the smallest practical ballast impedance is that which limits peak current into a short to a little under an amp -- and going to half an amp peak is better if you can stand the heat in your ballast resistor. EG, if you've got 40kv, you want at least 40k ohms in your ballast, and more is better sometimes, especially on a fresh grid that tends to arc more frequently.

We went with cylinder grids because we have a cylinder tank -- and matching is important.

A fusor is most definitely NOT a confinement device, however -- the ion beams come right out on the other side, and that glow you see in there at focus is neutrals that failed to react and got stopped for an instant -- then they too fly right back out when they get ionized again, being attracted away from the center and towards the grid wires at that point. It's a beam device, and the grid openings are acting like electrostatic lens elements. Monatomic hydrogen ions cannot make light -- it takes electrons dropping from one energy level to another to make light.

ChrisB would probably argue, correctly I think, that many of the fusion reaction are due to ions hitting stationary or slow moving neutrals. Our work would bear that out in most sets of conditions (but not all).

Given that, and with a cylinder chamber, it's one heck of a lot easier to make a good "lens" as all that takes is straight rods, evenly spaced (we use either 8 or 6) - uniform slots are far easier than tiling a sphere with eqispaced holes for the particles to be lensed through. In our case we used small diameter (.040") TIG welding rods, which come very straight, so it's easier to accomplish than wire bending. We used drilled graphite endcaps as one possible way to hold the rods in good alignment. Our second best grid is Ti wire with Ti washer endplates, spot welded, and the main reason it's #2 is we couldn't make it as precisely. All three materials mentioned work well for us. Carbon seems to make fewer energy wasting secondary electrons when hit, but to tell the truth, the effect is swamped by making the grid precisely -- which means it gets fewer hits in the first place, in our experience.
We are using about a 2::1 length to diameter ratio, anything around there seems to work well, but we've not explored real short or real long ones so far.

That aside, with the same speeds and feeds, we've found we get about the same amount of fusion as anyone else here does -- within experimental errors (or, lab cross calibration errors), maybe a bit more, but it's not tons more if it is.
(we can get more, and higher Q, but not in DC-stable modes - and I suspect that would be true with spheres too, but it seems no one has tried that with them yet)

It's just been easier to work with cylinders for us, so that's what we do for a conventional fusor. The HV stalk plugs one end (as it does on a sphere -- so not truly better symmetry either way) and we leave a hole in the other end so it doesn't get so hot as it does without one -- the ions can get in or out via that end hole too rather than just strike a flat end, wasting energy and heating the grid.
Why guess when you can know? Measure!
User avatar
Martijn Rambonnet
Posts: 18
Joined: Thu Mar 17, 2011 7:34 am
Real name:
Contact:

Re: First Bubbles!

Post by Martijn Rambonnet »

Thank for the valuable information! When we find a new x-ray transformer we will certainly install the ballast resistors. The disadvantage of the extra heat is minor if it prevents zapping the diodes.

We have not measured the (AC) signal kickback before the rectifying diodes but we did plot the output signal (DC) on an oscilloscope. We observed that when there was a plasma in the chamber there was a large kickback (positive) signal. We were wondering where this comes from and if we could solve this with an extra diode (array) in opposite direction to the ground to filter this out.

Also this cylindrical grid sounds intriguing and we'll produce one to compare with the current grid. We will try to design a stiff and precise design. I wonder what a toroid chamber with toroid grid would do...

Since we currently do not have an HV source we are working on the new chamber, improved DIY hv feedthroughs (so we don't have any arcing anymore), some initial polywell construction tests and we also have a very nice photomultiplier (from Richard Hull) to implement.
User avatar
Doug Coulter
Posts: 1312
Joined: Sun May 27, 2007 3:18 pm
Real name: Doug Coulter
Location: Floyd, VA, USA
Contact:

Re: First Bubbles!

Post by Doug Coulter »

I also sometimes wonder what toroids would do. So far, even the longest cylinder grid I've made doesn't have multiple kernels where rays come out. They all come out in the center of the length, even though you do see a pencil lead-like rod at the focus for the full length of the cylinder.

In my own opinion, the DC static/equilbrium fusor is where it is -- we've stopped seeing huge improvements all the time with it as we make changes based on observations, and it's just, well, where it is. So at my lab we're moving on to better systems to do basic study on, which probably won't make as many neutrons, but which will be good for learning -- and who knows where we wind up? Not us, or we'd just go straight there and skip the research step. But it seems we and others here have found what is at least the major "sweet spot" in the curves - there might be another one far from where we are running, but there's no hint yet where it would be, and plenty of people, us in particular, have been looking pretty hard for it.

One thing we've noticed here over and over -- anything that kicks our fusors (we have a few) off that static stability -- makes more neutrons for a bit till it finds that "attractor" again. But the fusor is so drawn to that stable place it makes it hard to quantify this, so we're looking into simpler systems.

We've done fusors with inductive ballast and gotten them to oscillate -- better.
We've tried warping the field with extra electrodes and time varying drive - better.
We've tried magnetic pulses -- better.
We've tried gating the ion source we use when we run extra low pressures to pulse it -- better.
(actually, lower pressures + ion source is just plain better -- but not original)

All these things make output better, but only while they last. As soon as equilibrium is re-established, it's back to roughly 1 mil neuts/second for a few hundred watts input. (at the 40-50kv voltages, Q does go up with voltage quite a bit)

Yeah, a fusor looks simple but... it's very much not the case. Between emergent behavior and the fact that people love to armchair theorize rather than build, and then build some hard to do metering gear and make real observations that are repeatable, it's not making much progress these days.

Everyone building a "simple" fusor seems to wind up at a certain neutron output and Q given similar speeds and feeds, and that appears to be "it" for that class of device so far. My main effort is on Q, not sheer output, but simply pushing a fusor up to "11" -- or even 20, gets you not much better of either, and in fact, Q tends to go down. When I put 2kw into mine, it's not all that much better than with half a kw -- but of course then it's a lot harder to get rid of the heat and X rays, which DO scale.

A ballast alone won't solve kickback, but it will prevent blowing pits in your grids and HV feedthroughs, and make the thing one heck of a lot easier to run. Kick back is what happens if your fusor suddenly stops drawing current, and the inductance of your supply has stored some energy -- voltage rises more or less without limit until something draws current again -- which in many cases is going to be your diodes breaking down -- or something else next time.
Why guess when you can know? Measure!
Post Reply

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