Alex Calburean - Fusor Update

Current images of fusor efforts, components, etc. Try to continuously update from your name, a current photo using edit function. Title post with your name once only. Change image and text as needed. See first posting for details.
Post Reply
Alexandru_Calburean
Posts: 50
Joined: Thu Dec 05, 2013 9:30 pm
Real name: Alexandru Calburean

Alex Calburean - Fusor Update

Post by Alexandru_Calburean »

Hello all,

It has been a while since my last update as I have been busy with school and some other activities as well. I am posting this to give an update on my fusor and note some interesting things going on. I have attached three pictures of what seems to be star mode. The two that look most similar are taken at the same time at an unknown pressure (vacuum line for gauge was not yet made) and at about ~15kv @ ~13mA and the thrid at ~18kv @ 11 mA but at a lower pressure. These runs were done with my old Bertan 40kv 5mA supply. Ever since the first posting, I have found one of the main leaks but still have a rather large leak somewhere (my guess is the electrical feed through I made at home or the viewport). After about an hour and a half pumpdown I got to 11 microns on my thermistor vacuum gauge but when valved off the chamber pressure rises with about 2 microns a second and then increases to around 10 microns/s and finally 50 microns/s as the pressure gets higher and higher. There is most likely some outgassing as well so that is something to note.

As for my power supply, the old Bertan 40kV @5mA supply I had was destroyed in an arc somewhere in the secondary (either the rectifying circuit or the windings themselves). I have been on the hunt for a supply and finally picked up Andrew Seltzman's X-Ray transformer that he was selling. The new supply is capable of -50kV @ 30mA with a short duty cycle but should run fine for 2-5 minute run intervals with current limiting to 10mAs. I figure this for two reasons 1. Andrew himself ran it at that current rating without any damage for extended periods of time and the transformer and oil system has a rather large thermal mass thereby allowing me to avoid thermally overloading the secondaries as fast. The supply has a phase angle controller (with a random fire solid state relay) for voltage modulation which uses a 0-5 volt reference where 0-5 volts is 0-full power. I understand that where there are short rise times coupled with large inductance, stability issues tend to ensue, but I also know that these controllers can be constructed in such a way that said issue ceases to become an issue, so if anything other's inputs in this area would be greatly appreciated. Now here is where things start to get interesting, and interesting most likely because of my inexperience with such devices. First, There is a resistive ballast being used in series with the primary of the X-Ray transformer, which is a 900 W hot plate. The reason this is particularly interesting is as follows: the plate has a thermostat which is presumably like a relay in that it will open and close the circuit based on the heat or lack there of measured, so it then follows that if the hot plate is not turned on the circuit remains open and no current can flow. When I hook this up to my transformer, however, the hot plate's indicator light remains off and there is no indication whatsoever (heat, light, or otherwise) that it is on, but the transformer is operational. I know Andrew successfully operated with a 750 W space heater, so I know the problem is not on the transformer's end but rather on my end. Maybe there is something that is going on that I am not aware of that some of the more experienced members can inform me about. Another thing I noticed in my current setup is that the high voltage probe that I have (fluke 80k-40) seems to arc somewhere next to the input when the voltage is ~20kV therefore making voltage monitoring at and beyond that level impossible. The probe has been retrofitted with a screw with the screw head sawed off where the pointed tip of the probe was before, my guess is that the sharp points from the cut is causing the corona discharge that I hear as arcing, but again, some input on my hunches would be greatly appreciated. The most interesting thing by far however is what happens when I turn up the voltage on my supply.

Today I pumped down to 21 microns (as low as my vacuum system will go without waiting an hour to asymptotically stabilize at ~11 microns) and connected the high voltage without voltage monitoring but did have current motoring. It should also be noted that the transformer had the resistive ballast on the primary (900 W hot plate connected), and was being controlled by a 0-5 volt reference circuit on the PAC. I then plugged everything in and slowly turned up the voltage, and by slowly I mean in increments of about ~ .1 Volts per each turn on the potentiometer, as I am deathly afraid of this transformer. The voltage seemed to climb as I heard the typical high voltage air crackling until I got to about ~20kV, at which point it tripped the breaker and shut down everything including my two pumps. Three things should be noted here, first, the resistive heater again showed no indication of being on, and second, the fusor started to light off only at about 20kV (this corresponds to 2 volts on the pot and I generally know how many degrees turned that is even though I had no HV probe connected), and third, the amperage spike for the brief second it drew current was only a spike to 10mA not 60 or so mA like on one of Richard Hull's HEAS run videos. At this point I decided to check my circuit breakers and identify one with no current being drawn so I could use the full 15A on standard AC mains. I proceeded to hook up the transformer and its ballast load to the circuit and cranked up the voltage again, yet in spite of my efforts the same thing happened. Now clearly I am doing something wrong because Andrew was able to operate without drawing enough amperage to overload every breaker in his work space. Some input here would be nice, as I am rather confused about the high current draw. It is also curious that the hot plate doesn't seem to be functioning by any means (no heat, no light, etc) yet seems to allow the transformer to operate and keeps the brief amperage spike to 10mA so I am at a loss for words in trying to explain what happened. As a small note on the amperage spike: I did notice it took Richard's fusor a few seconds to get up to that 60mA peak spike where as my transformer only had about a second before power was cut, so that may explain that. By far though I am most curious about why it is taking a 2.75in conflat cross fusor 20 or so kV to even light off at ~21 microns, and how it is possible that a presumably current limited transformer with a hot plate as a resistive ballast is tripping a 15A breaker as soon as there is even a small 10mA amperage spike in the secondary, which is presumably the fusor trying to light off.

Other than those few issues my fusor seems to be coming along well, and I am planning to write about the relationship between pressure and Bremsstrahlung emission in the fusor for a rather large paper we have been given in school (for those familliar with the IB program: I am referring to the Extended Essay). After I get this paper out of the way I will focus on a deuterium system, and once that is complete I plan to talk to some contacts at some local universities for neutron detectors to go for neutrons. As many here have proven time and time again, slow and steady does indeed win the race and seeing as how the fusor will most likely be a life long hobby of mine, I plan to continue working on my fusor to hopefully do some new and useful research regarding the fusor's performance (not that I expect high Q out of the fusor). I'll try to contribute more frequently but this is a "as time permits" ordeal, so given that summer vacation is coming up, I am looking forward to being far more active on the forums.

All the best,
~Alex Calburean
Attachments
Camera flash off
Camera flash off
Camera flash on
Camera flash on
Lower Pressure Star or Jet mode
Lower Pressure Star or Jet mode
User avatar
Dennis P Brown
Posts: 3160
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Alex Calburean - Fusor Update

Post by Dennis P Brown »

First off, best of luck with the leak detecting! I know what a pain that is!

Your post about " I am planning to write about the relationship between pressure and Bremsstrahlung emission in the fusor for a rather large paper we have been given in school " is something I really look forward to reading! Best of luck on that research. If you can give some details on how you propose to measure the Bremsstrahlung radiation I'd be very interested to know about that.

By the way, you might want to post your question about tripping a 15 amp breaker in the high voltage section (with the details) so someone reads it (not as likely here!) and then answers that question in the correct section. Also, your issue about current draw @ 21 microns is very obvious and needs to be posted there,too.

In a similar manner - post your vacuum questions in that section, too. I have far, far too much experience in finding vacuum leaks (trying to reach/maintain low range of 10^-6 torr in a home made system is not fun!) -with leaks that are both large and small as well as determining what is a leak and what is out gassing. Photo's of all your system/connections and pump(s) in the vacuum forum with detailed questions would make helping you to solve this problem easier for you and us! Your failure to get down to a few microns and your rise rate is troubling considering the design of your chamber - CF flanges and you have those issues is very surprising.

You say you are using a hot plate to control the output of a high voltage end of an X-ray power supply!? That sounds very unsafe. First off, 10 Kilo-volts or higher across a 110 v relay and hot plate housing isn't really proper. Worse, the entire hot plate will most likely be at that high voltage and unless behind a shield, a rather great electrocution danger, I'd think. What is wrong with a high wattage resistor? I mean saving money is good but a hot plate as a HV resistor isn't what I'd think is proper nor saving money. Maybe I read the post wrong ... .
Alexandru_Calburean
Posts: 50
Joined: Thu Dec 05, 2013 9:30 pm
Real name: Alexandru Calburean

Re: Alex Calburean - Fusor Update

Post by Alexandru_Calburean »

Dennis,

Thank you for the fast response. I will move each of my concerns to their related sections, I just wanted to have everything in one place so as to provide an overview (however detailed) of my current situation in my update.

As for the Bremsstrahlung radiation I am using a small GM counter from school, which if I remember correctly has something like ~2% quantum efficiency but it was enough to start registering counts. I believe the highest output the I was able to achieve was something like 500 pulses in a second. I am doing this because I figure Bremsstrahlung if a direct result of the immediate deceleration of particles, causing them to emit a photon, then If we increase the pressure we decrease mean free path which will not allow for a very fast final velocity as acceleration time is decreased. Because of this reduced mean free path, particles in the reactor will lose energy before high velocity due to elastic nature of collisions. This will also lower the amount of particles emitting Bremsstrahlung radiation because smaller mean free path will reduce acceleration time for majority of particles therefore decreasing their total overall kinetic energy before a potentially Bremsstrahlung emitting collision. We shall see if the hypothesis ends up being correct or not.

Like I said, I agree with you in that I should move my worries to their appropriate forums but I just wanted to address your concerns regarding the ballasting. The transformer is being controlled via a phase angle controller but is being ballasted via a 900 W hot plate. This ballasting is done via a ballasting adapter (like is shown in this video: https://www.youtube.com/watch?v=AsdZJUF1guU) Since this is adapter is on the primary not the secondary it does not see high voltage so no worries there, unless there is something I am missing. I chose the hotplate because it requires no modification to the system Andrew already created; it is literally just plug and play. Like I said, I don't mind using a high wattage resistor but there is already a proven means of ballasting in place, and I am usually of the philosophy: "If it ain't broke, don't fix it." Thanks for the note about safety concerns as well; like I said, I am deathly afraid of this transformer, so short of dressing up in a thick silicone or rubber suit when I operate it, I am willing to do anything to protect myself.

All the best,
~Alex Calburean
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Alex Calburean - Fusor Update

Post by Richard Hull »

I just posted a FAQ on transformer ballasting in the fusor power supply forum.

viewtopic.php?f=29&t=9383

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
Alexandru_Calburean
Posts: 50
Joined: Thu Dec 05, 2013 9:30 pm
Real name: Alexandru Calburean

Re: Alex Calburean - Fusor Update

Post by Alexandru_Calburean »

Hi Richard,
Thank you for posting that, I've read it over a few times and will go the resistor route on the secondary if I cannot operate my supply like Andrew did when he was using it. Again, I very much appreciate the faq about ballasting as the searches did not yield definitive answers for "whether I should have a ballast in series with the primary or the secondary" as I saw people advocating for both means and even some bits about inductive ballasting. I will also type up a proper post in the HV and Vacuum sections when I get a long stretch of free time. Until then I'll also throw in this video of my setup that I took tonight in the spirit of images du jour and showing ones progress. Here is the link to that; feel free to comment: https://www.youtube.com/watch?v=lfs_UA0vhKk

All the best,
~Alex Calburean
Andrew Seltzman
Posts: 815
Joined: Sun Feb 01, 2004 8:02 pm
Real name: Andrew Seltzman
Contact:

Re: Alex Calburean - Fusor Update

Post by Andrew Seltzman »

Hi Alex,

Breaker problem:
No clue yet why this is happening a 750w ballast on the primary should definitely prevent the breakers from tripping unless they have some arc detection system. One thought, the phase angle controller might draw higher peak current for any given output then would be used with a variac (it fires part way through the AC cycle).

Arc problem:
I'd say the problem is certainly in the feedthrough, as that size and thickness of ceramic will not hold off the high voltage. The feedthrough probably arced through the ceramic insulator. There is probably a conductive channel burnt in the ceramic from the high voltage center conductor to the fusor shell and it has been getting worse(more conductive) over time. This is probably why it is now tripping the breaker / arcing at a lower voltage, rather then anything with the ballast.

HV probe:
As far as the probe setup, definitely connect the ground clip on the probe to a ground if the banana jacks are not plugged in or plugged in to a multimeter or the ends will float at a high voltage (if it's plugged into the terminal block on top of the transformer it will be grounded). Also the wire connecting the tip of the HV probe looks like it won't stand up to the voltage and the electrical tape on the tip most certainly won't. I would recommend sitting the tip in one of the HV output connections of the transformer like I had in the youtube videos on my website (for reference to fusor.net members on the xray transformer system: http://www.rtftechnologies.org/physics/ ... former.htm)

Vacuum system:
Don't use the hot melt glue, it won't do anything and will make things harder to service. The teflon tape will make a reasonably good seal seal and can give a good vacuum:(http://www.rtftechnologies.org/physics/linac.htm). You might coat the teflon tape with some vacuum pump oil which may help. Also don't use JB weld or glue to seal vacuum connections, it will probably outgass a lot and not seal very well. It may also be causing virtual leaks(trapped gas leaking through pinhole leaks into the cross)

Ballasting:
Richard and others have had good success with ballasting the secondary which also works well.
I put the ballast on the primary since:
(1) HV power resistors are hard to find
(2) Connecting an external HV resistor would have required 2 more HV connections to insulate and I didn't want to put the ballast resistor in the oil tank for serviceability reasons.
(3) There is no capacitive filtering in parallel with the output. If you have a capacitor in parallel with the output, the primary ballast won't do anything to protect the system from stored energy in the capacitor; you would need a secondary ballast between the capacitor and the fusor.
(4) It's simple to ballast the primary and I had the parts at the time and was on a deadline to finish the experiments for my senior thesis

As far as the large arc you saw, is it possible that most of what appeared to be the arc was actually reflected light from a smaller arc?

Andrew
Andrew Seltzman
www.rtftechnologies.org
Andrew Seltzman
Posts: 815
Joined: Sun Feb 01, 2004 8:02 pm
Real name: Andrew Seltzman
Contact:

Re: Alex Calburean - Fusor Update

Post by Andrew Seltzman »

Focus on building a better feedthrough, this will solve a lot of your problems. Using a 3/4 OD 1/2" ID high alumina or quartz or glass tube with a second 1/2" OD 1/4" ID insert as the insulator will work a lot better. Remember to take the insulator almost all the way to the grid on the inside since an arc can still form on the inside. Look into getting some swagelok / cajon
Look at the construction of the feedthrough here:
http://www.rtftechnologies.org/physics/ ... d-grid.htm

This might help if bored through to 3/4 ID on a lathe
http://www.ebay.com/itm/MDC-Hybrid-3-4- ... 3a662d14fa
if you replace the stainless ferrules with teflon ferrules to seal to a glass or ceramic tube.
http://www.ebay.com/itm/Lot-Swagelok-3- ... 338d4f0d66
with a cap on the other end
http://www.ebay.com/itm/Swagelok-SS-Cap ... 417fd09db9
or
http://www.ebay.com/itm/LOT-OF-2-NEW-SW ... 2c8042edf4
Andrew Seltzman
www.rtftechnologies.org
Alexandru_Calburean
Posts: 50
Joined: Thu Dec 05, 2013 9:30 pm
Real name: Alexandru Calburean

Re: Alex Calburean - Fusor Update

Post by Alexandru_Calburean »

Andrew,

Wow! Thank you for all the advice and help once again. I will take care of all of that but it is great to hear that it is something as minor as the feed-through. You are correct in saying that my feedthrough arcing is likely the problem. It too was constructed with JB weld and seeing as how jb weld has a metal in its composition it is likely to conduct easily. Also makes sense in terms of that arc channel you just described. The ballast is still a mystery but the arc may have been just the light from a smaller arc which is more sensible and probable. I will now work on that feed-through but I cannot say how relieved I am that it is not something more major. Thank you for the guide and parts list Andrew. Also, thank you everyone else for the comments. Well time to get to work.

All the best,
~Alex Calburean
User avatar
Dennis P Brown
Posts: 3160
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Alex Calburean - Fusor Update

Post by Dennis P Brown »

I see a major issue with your idea to measure the Bremsstrahlung radiation using a GM.

When you say
As for the Bremsstrahlung radiation I am using a small GM counter from school, which if I remember correctly has something like ~2% quantum efficiency but it was enough to start registering counts. I believe the highest output the I was able to achieve was something like 500 pulses in a second. I am doing this because I figure Bremsstrahlung if a direct result of the immediate deceleration of particles, causing them to emit a photon ...
The trouble is, the few ionized deuterium atoms that are producing Bremsstrahlung will be swamped out by the many orders of magnitude of other events also emitting photons - x-rays are produced by electrons colliding with gas molecules, the electrodes and the walls of the chamber. Worse still, de-acceleration of air molecules producing such radiation (including hydrogen!) will all produce photons with energy very close to the area you might be interested in (esp. the hydrogen atoms which will be emitted from the walls - i.e. water molecules that are fully ionized. Getting this trace contaminate out of a system is very difficult.)

A simple GM counter will detect all these x-ray photons. The photons you are interested in will be of a very specific and narrow energy range (i.e. wavelength) and I believe you will need a very narrow band pass filter to get a signal that will mostly be sensitive to just this specific wavelength of light. Filtering out the lower energy photons will be easy (and of course, this will also reduce your desired energy photons as well) but the higher energy photons might be a problem (maybe not - others here might know more.) I believe a specific crystal (with a specific atomic lattice) viewing the plasma region at a specific angle could do the job as a narrow band pass filter (but the required window might be an issue - glass thick enough to withstand the pressure differential might stop the desired x-ray/photon signal.)

You should research how this process has been done in the pass by others. Also, others here may have ideas or already have done this.

aside: If your GM tube is giving 500 counts a second and if this is due to x-rays (maybe not), you might need to concern yourself about shielding of your fusor - at high voltages (above 25 KV, this can be a problem.
Alexandru_Calburean
Posts: 50
Joined: Thu Dec 05, 2013 9:30 pm
Real name: Alexandru Calburean

Re: Alex Calburean - Fusor Update

Post by Alexandru_Calburean »

Dennis,


Sorry for the late reply as school and driver's education has kept me from thoroughly reading and properly responding to address your points. I'll start with the simplest of all; the radiation concerns, I am building a lead box to cover the 2.75 CF cross, but I am leaving the viewport open. I plan on running in the ~40kV range so x rays coming out of the shell will be shielded, the the fusor will be operated from a distance, and finally the run times will be kept short, thereby keeping my exposure to a minimum. The viewport will have a usb webcam and will be pointed away from anything it could harm, so no worries there.
As for the Bremsstrahlung emission, I did not expect the wavelengths of other high energy photons emitted to be so close to my area of interest, in fact I expected them to be either much much shorter or much much longer, which is ignorance on my part so no excuses. Plain and simple I will have to look into what others have done in the past to filter specific wavelengths as per your suggestion, and study ionizing radiation a bit more. Again thank you all for the replies and Dennis for the great heads up, finally, sorry for the late response Dennis.

All the best,
~Alex Calburean
User avatar
Dennis P Brown
Posts: 3160
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Alex Calburean - Fusor Update

Post by Dennis P Brown »

No problem on the response.

As for a lead box - two issues. First, lead is toxic so all exposed surfaces must be painted. Wash hands well after touching lead. Second, the high voltage of a fusor can be deadly - having metal panels surrounding a fusor that is excited with 40 kV could be very dangerous (all the lead pieces should each have its own earth ground if they are not behind a plastic shield.) So keep this issue in mind.

I use non-lead shielding that is also non-conducting and non-toxic. It is extremely cheap, as well. That is, I use slate from Home Depot - any abrasive blade cuts it and I use "Liquid Nails" to hold layers together. The stuff comes in large squares, has 1/4 the density of lead (so is easy to determine proper thickness.) Also, no disposal issues!

For an X-ray line filter you should look into silicon wafers (again, rather cheap) and these come with specific crystal orientations (just ask) - like 100, 001, and even higher orders. Set at the correct angle (will be fairly shallow), specific and exact X-ray lines can be reflected into a shielded detector through a pin hole. Look up the topic of Bragg x-ray reflection. Again, a window material may be the biggest issue - that is not a minor issue and be careful to address it before getting any crystals (i.e. calculate the required x-ray energy photon you need and the possible intensity you will need.)
Alexandru_Calburean
Posts: 50
Joined: Thu Dec 05, 2013 9:30 pm
Real name: Alexandru Calburean

Re: Alex Calburean - Fusor Update

Post by Alexandru_Calburean »

Dennis,

Again sorry with the delay in response time, unfortunately it seems that the delay will exist until the school year ends. As for the lead, thanks for the tips, I knew about its toxicity but figured that I am going to sandwich it between steel plates. As for this radiation shielding assembly coming into contact with high voltage, I plan to create a HDPE shield around my feedthrough to prevent it arcing to the shielding. I am concerned because yesterday I managed to hit about 6000 counts per second with the GM probe against the viewport so lead shielding will be a must on my system. I also plan on getting a proper high voltage feedthrough, but that will have to come later as I do not have much to spend. The one I built, when covered with shrink wrap and a spark plug cover, can go to about 25kv before its starts arcing. As for the high voltage system I have some news on that as well, I got a 20 amp variac as a means of controlling the transformer and am now able to run at about 25kv max and at about 2-3ma ballasted and can therefore also confirm that the ballast is working properly. Its very odd to me, however, that the fusor is only drawing ~75 watts at 25kv, 3ma, and ~40 microns, I would have expected it to draw 20 or so mA as per Doug Coulter's Q plot data but our chambers, grid ratios, and gas compositions are different so it may be a combination of all of those factors. Also, thanks for the tips on where to get some of the slate you mentioned that should help the safety process along.

As for an X-ray line filter, thank you so much for the information, I have looked into that a bit here and there but I do not have much free time now a days. I'll look some more into it this weekend but I might also have access to a single channel amplifier at a local university. I'll be taking data at a university because my school and the program that I am in (The International Baccalaureate Organization) do not feel comfortable letting me submit a paper that has data taken from any radiation producing devices without radiation safety controls. As frustrating as it is having my safety controls not count as proper safety controls for whatever reason, it is just the way the cookie crumbles I suppose, but like I said it might give me access to some nifty NIM based measurement and discrimination systems. If not I will look into buying some silicon wafers like you said but will look further into Bragg x-ray detection and window material specifics. Again, thanks for all the help!

All the best,
Alex calburean
Post Reply

Return to “Images du Jour”