Ok so I’m building a modest 6” diameter fusor and would like to add one or more ion guns in the future after I’m mastered it with out them. So how many and what size conflat ports should I put on it and in what arrangement.
Chris Franklin
Ports for ion guns
-
Richard Hull
- Moderator
- Posts: 15024
- Joined: Fri Jun 15, 2001 9:44 am
- Real name: Richard Hull
Re: Ports for ion guns
Two guns would be considered a minimalist configuration.
On an 8" fusor, 6 guns would be considered an absolute maximum! The thing would be lousy with screws and flanges and the shell body would be like swiss cheeze prior to welding with just a web metallic fabric representing its former self.
2.75" ports would be considered the norm on such a system for the gun ports, but alot would be dependent on the ion gun current rating. Micro whimpy guns could easily be fashioned on the 1.33" port design while a 10 ma or larger gun might demand a 3.38" port. This might require that the fusor shell swell to up to 16 inches in diameter when using 6 such guns. Naturally, the fusor would almost have to be differentially pumped and the chamber, itself, exhausted into the 10e-6 torr range to achieve a decent MFP.
Joe Zambelli is the only amateur that I ever knew who constructed his fusor with ion gun ports from the inception of the project! It is a work of art and for sale. ($10,000+ range with all vaucuum system gear and electronics included...see image in image dujour).
viewtopic.php?f=18&t=7737#p55696
He never put any guns on the device and his ports remain blanked off.
All this gun stuff demands that nasty vacuum and electrical issues be tackled. It throws the whole effort into a new class of hassle, expense and seed energy requirements.
Get out your checkbook, clean and oil the lathe and milling machine, batten down the hatches and hang on.
Richard Hull
On an 8" fusor, 6 guns would be considered an absolute maximum! The thing would be lousy with screws and flanges and the shell body would be like swiss cheeze prior to welding with just a web metallic fabric representing its former self.
2.75" ports would be considered the norm on such a system for the gun ports, but alot would be dependent on the ion gun current rating. Micro whimpy guns could easily be fashioned on the 1.33" port design while a 10 ma or larger gun might demand a 3.38" port. This might require that the fusor shell swell to up to 16 inches in diameter when using 6 such guns. Naturally, the fusor would almost have to be differentially pumped and the chamber, itself, exhausted into the 10e-6 torr range to achieve a decent MFP.
Joe Zambelli is the only amateur that I ever knew who constructed his fusor with ion gun ports from the inception of the project! It is a work of art and for sale. ($10,000+ range with all vaucuum system gear and electronics included...see image in image dujour).
viewtopic.php?f=18&t=7737#p55696
He never put any guns on the device and his ports remain blanked off.
All this gun stuff demands that nasty vacuum and electrical issues be tackled. It throws the whole effort into a new class of hassle, expense and seed energy requirements.
Get out your checkbook, clean and oil the lathe and milling machine, batten down the hatches and hang on.
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
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
Re: Ports for ion guns
Chris - Just a point or two to add to Richard's thorough picture of the basic issues and plumbing challenges of incorporating ion gun(s) into the fusor.
In my work, I have to deal with a sealed high voltage electron beam emitting tubes. These devices operate above 50 kV and are rather small - less than 6 inches long. The state of internal pressure is always of interest since withstand voltages depend on good vacuum. Internal pressures typically are in 10-8 or better torr... except when things are not right.
We measure internal pressure of the sealed tube making use of its ability to produce an ion current... of mixed species. Ion currents typically run in the nano to pico amp range for an electron emission current of some 100 plus micro Amps... so the relative ratio at high vacuum is about 10-5 to 10-8 of ion current per emission current. ...0.001%. or less.
However when the internal pressures are high. ( above 10-6 torr), ion production increases vastly, for the same emission current. I recently had a tube with a few uA of ion current for the same 100 uA of emission current... a few percent now.. about several thousand times more efficient at ion production, although not much good in that condition for high voltage standoff.
At higher pressures up to some limit... one gets even more ion current for the same emission current. Unfortunately, I don't yet know the functionality of the higher pressure regime. I do know that it is nonlinear with respect to emission (or electron) current. With more emission, you get disproportionately more ion current, up to the point where all ions possible have been made. Then ion current "saturates" with respect to further increases in emission.
The point of all this is that these particular devices are not intentional ion guns, and are extremely simple in design, and yet probably good enough to make ion guns for fusor work.
One needs an electron source that can handle pressures up to about 10 micron, say.. lower would be better. A lamp filament or even a simple fine tungsten wire will work. You also need an anode to collect the electrons emitted from the filament cathode. The anode needs to built with an opening in it so that ions created by impact from the electrons, can exit A rectangular or circular port will work fine. It should be fairly open.. perhaps a mm in diameter(width) .. so that the potentional from the ion accelerating electrode at high negative potential , can "reach in" and extract the ions made.
In the fusor configuration, the ion gun will be at one or more ports..I would suggest diametrically opposed... and the ion accelerating electrode, at high negative potential, will be on the standard inner grid.
I will suggest here for discussion, that we might NOT need any fancy gas regulation system. One would fill the fusor with D2 to a rather low pressure typically way below 1 micron (0.001 to 0.0001torr) and seal off the fusor. Then the filament is lighted and current drawn to the local anode at a modest voltage 100 V would the maximum needed. Electron current (emission current in my example above) is set to a number like 100 or more uA. The anode voltage (innergrid ) should be a fairly high level, 1 to 10 kV. The ion current should be measured and plotted as a function of emission current, which will then give the story of your device, at this pressure.
If two ion guns are used... then within the inner grid one should see the evidence of the poissor.
It should obvious, that there is a particular advantage here. Most of the fusor current will be in the form of ion currents from the ion guns. Since the ion gun somewhat aims the ion beam... then the likelihood of direct collision with the grid is dramatically lower, and one major problem .. grid heating goes away.
A second thing happens,,, efficiency should rise. By how much, and if so at all... are things to be shown experimentally. I am reasonably certain, that this can be shown easily with any fill gas whatever. So one doesn't need to worry about neutron measurements.. at first.
The last expected benefit, is that ion energies will now be mostly at the full grid potential, rather than a grand stew of every energy from about 15eV on upward. This should improve neutron yield significantly.
So there's a few things to consider...
I am speculating that filament structures from fluorescent lamps might be quite suitable for this... they have a simple straight helix shape.and they are very robust. The one possible spoiler is whether they are contaminated with the mercury used for the UV output.
This could be an nice experiment for those with bell jars to do. Just need to remember the usual precautions about letting electron beams impinge on the bell jar walls, if you are using a glass one.
Just some ideas I have been mulling over . This seemed to be a good place to lay them out.
Dave Cooper
In my work, I have to deal with a sealed high voltage electron beam emitting tubes. These devices operate above 50 kV and are rather small - less than 6 inches long. The state of internal pressure is always of interest since withstand voltages depend on good vacuum. Internal pressures typically are in 10-8 or better torr... except when things are not right.
We measure internal pressure of the sealed tube making use of its ability to produce an ion current... of mixed species. Ion currents typically run in the nano to pico amp range for an electron emission current of some 100 plus micro Amps... so the relative ratio at high vacuum is about 10-5 to 10-8 of ion current per emission current. ...0.001%. or less.
However when the internal pressures are high. ( above 10-6 torr), ion production increases vastly, for the same emission current. I recently had a tube with a few uA of ion current for the same 100 uA of emission current... a few percent now.. about several thousand times more efficient at ion production, although not much good in that condition for high voltage standoff.
At higher pressures up to some limit... one gets even more ion current for the same emission current. Unfortunately, I don't yet know the functionality of the higher pressure regime. I do know that it is nonlinear with respect to emission (or electron) current. With more emission, you get disproportionately more ion current, up to the point where all ions possible have been made. Then ion current "saturates" with respect to further increases in emission.
The point of all this is that these particular devices are not intentional ion guns, and are extremely simple in design, and yet probably good enough to make ion guns for fusor work.
One needs an electron source that can handle pressures up to about 10 micron, say.. lower would be better. A lamp filament or even a simple fine tungsten wire will work. You also need an anode to collect the electrons emitted from the filament cathode. The anode needs to built with an opening in it so that ions created by impact from the electrons, can exit A rectangular or circular port will work fine. It should be fairly open.. perhaps a mm in diameter(width) .. so that the potentional from the ion accelerating electrode at high negative potential , can "reach in" and extract the ions made.
In the fusor configuration, the ion gun will be at one or more ports..I would suggest diametrically opposed... and the ion accelerating electrode, at high negative potential, will be on the standard inner grid.
I will suggest here for discussion, that we might NOT need any fancy gas regulation system. One would fill the fusor with D2 to a rather low pressure typically way below 1 micron (0.001 to 0.0001torr) and seal off the fusor. Then the filament is lighted and current drawn to the local anode at a modest voltage 100 V would the maximum needed. Electron current (emission current in my example above) is set to a number like 100 or more uA. The anode voltage (innergrid ) should be a fairly high level, 1 to 10 kV. The ion current should be measured and plotted as a function of emission current, which will then give the story of your device, at this pressure.
If two ion guns are used... then within the inner grid one should see the evidence of the poissor.
It should obvious, that there is a particular advantage here. Most of the fusor current will be in the form of ion currents from the ion guns. Since the ion gun somewhat aims the ion beam... then the likelihood of direct collision with the grid is dramatically lower, and one major problem .. grid heating goes away.
A second thing happens,,, efficiency should rise. By how much, and if so at all... are things to be shown experimentally. I am reasonably certain, that this can be shown easily with any fill gas whatever. So one doesn't need to worry about neutron measurements.. at first.
The last expected benefit, is that ion energies will now be mostly at the full grid potential, rather than a grand stew of every energy from about 15eV on upward. This should improve neutron yield significantly.
So there's a few things to consider...
I am speculating that filament structures from fluorescent lamps might be quite suitable for this... they have a simple straight helix shape.and they are very robust. The one possible spoiler is whether they are contaminated with the mercury used for the UV output.
This could be an nice experiment for those with bell jars to do. Just need to remember the usual precautions about letting electron beams impinge on the bell jar walls, if you are using a glass one.
Just some ideas I have been mulling over . This seemed to be a good place to lay them out.
Dave Cooper
Re: Ports for ion guns
To add to Dave's
A 12v car bulb will also work as a filament - even a 100w headlamp and they are clean - I have not tried a halogen bulb because of the iodine contamination.
-
Richard Hull
- Moderator
- Posts: 15024
- Joined: Fri Jun 15, 2001 9:44 am
- Real name: Richard Hull
Re: Ports for ion guns
Dave has done a great job in explaining the losses in guns and the poor ion current output to seed energy input at lower pressures. Much of this has been discussed in detail in the earliest posts in this forum.
One of the finest systems would be the one that I hope to morph fusor IV into, and that is the classic Hirsch-Meeks design with a ring filament and ionizer grid. This is the logical next easiest step in looking at higher fusion numbers. Ion guns would logically follow this step....at least for me.
The ionizer grid might exist only a 1/4 inch off the shell and could be made of very coarse, 1/2" mesh screen wire. It would not need to be particularly transparent I would think.
It is important to realize that any custom ionization energy would have to be figured into the total energy consumption of the device along with the acceleration energy in any effeciency calcs.
Richard Hull
One of the finest systems would be the one that I hope to morph fusor IV into, and that is the classic Hirsch-Meeks design with a ring filament and ionizer grid. This is the logical next easiest step in looking at higher fusion numbers. Ion guns would logically follow this step....at least for me.
The ionizer grid might exist only a 1/4 inch off the shell and could be made of very coarse, 1/2" mesh screen wire. It would not need to be particularly transparent I would think.
It is important to realize that any custom ionization energy would have to be figured into the total energy consumption of the device along with the acceleration energy in any effeciency calcs.
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
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
Re: Ports for ion guns
Following up on RIchard's thought about the ring filament. This could be a tungsten wire ring around the Fusor's periphery, spaced out from the shell a small distance on ceramic standoffs.. Heated by a small current ( if the wire gage is small enough) . The local anode could be as simple as a flat strip with a narrow slot in it. The slot would need to interrupted in a few places so the strip had some mechanical strength... to hold its shape..
Biasing the filament ring a volt or two postive relative to the fusor shell ground would make the walls a repeller and generally help direct the electrons out to meet the gas molecules.
This should greatly reduce the basic Fusor current. The ion generating current is at low voltage, so the power would not be large. A few milliamps at 100 volts...
If the basic gas discharge heating is removed, the fusor heat should be almost nonexistent.
I will be interested to see how your design developes, Richard.
Dave Cooper
Biasing the filament ring a volt or two postive relative to the fusor shell ground would make the walls a repeller and generally help direct the electrons out to meet the gas molecules.
This should greatly reduce the basic Fusor current. The ion generating current is at low voltage, so the power would not be large. A few milliamps at 100 volts...
If the basic gas discharge heating is removed, the fusor heat should be almost nonexistent.
I will be interested to see how your design developes, Richard.
Dave Cooper
-
Richard Hull
- Moderator
- Posts: 15024
- Joined: Fri Jun 15, 2001 9:44 am
- Real name: Richard Hull
Re: Ports for ion guns
Actually, I plan to mime the original hirsch meeks patent save for the indirectly heated ring filament which the Farnsworth team had whipped up in there Pontiac street tube lab. See the patent.
The filament supplies electrons in the shell-ionizer grid zone. The grid was a variable positive voltage 0-200 volts relative to the grounded shell. It attracted the electrons which were repelled by the shell. The electrons that didn't ionize on their way to the ionizer grid and also didn't hit the ionizer grid, but instead, went through it, were slowed and turned by the increasing field gradient of the intensely negative inner grid and accelerated back towards the shell where they were again stopped, repelled and reattracted to the ionizer grid. In theory, there would be a cadre of serpentining electrons threading in and out of the ionizer grid region crashing into neutrals and ionizing them.
The ionized deuterons would be attracted out of the ionizer system towards the inner grid.
Many of the older losses still remain, of course, but at least you are warranted a bulk ionization near the shell and the opportunity for a larger number of fusion energy deuterons streaming into the inner grid to do fusion.
This was the Hirsch-Meeks 1970's patent idea. It worked brilliantly with D-T, giving isotropic numbers with these gases on the order of 10e10n/sec using a 50kv supply. With more elevated voltages near 80kv, Meeks recorded 10e12 on one occassion. An image of the device is seen on my website under the Farnsworth original team photos.
We see Hirsch at his desk at APTI in DC during one of my visits with him. On his desk, between the camera and Hirsch, is the actual Hirsch-Meeks device used in the 1968 "hotel linen cart fusor" demo before the AEC. The device can be seen on the cart also in this group of images with Gene Meeks and Steve Blasing in the image.
Richad Hull
The filament supplies electrons in the shell-ionizer grid zone. The grid was a variable positive voltage 0-200 volts relative to the grounded shell. It attracted the electrons which were repelled by the shell. The electrons that didn't ionize on their way to the ionizer grid and also didn't hit the ionizer grid, but instead, went through it, were slowed and turned by the increasing field gradient of the intensely negative inner grid and accelerated back towards the shell where they were again stopped, repelled and reattracted to the ionizer grid. In theory, there would be a cadre of serpentining electrons threading in and out of the ionizer grid region crashing into neutrals and ionizing them.
The ionized deuterons would be attracted out of the ionizer system towards the inner grid.
Many of the older losses still remain, of course, but at least you are warranted a bulk ionization near the shell and the opportunity for a larger number of fusion energy deuterons streaming into the inner grid to do fusion.
This was the Hirsch-Meeks 1970's patent idea. It worked brilliantly with D-T, giving isotropic numbers with these gases on the order of 10e10n/sec using a 50kv supply. With more elevated voltages near 80kv, Meeks recorded 10e12 on one occassion. An image of the device is seen on my website under the Farnsworth original team photos.
We see Hirsch at his desk at APTI in DC during one of my visits with him. On his desk, between the camera and Hirsch, is the actual Hirsch-Meeks device used in the 1968 "hotel linen cart fusor" demo before the AEC. The device can be seen on the cart also in this group of images with Gene Meeks and Steve Blasing in the image.
Richad 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
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
Re: Ports for ion guns
I remember the general construction approach of the Hirsch -Meeks fusor. A neutron flux of 10+12 /SEC is in the watt output level. Getting tantalizingly close to enough output to be useful.
From my own limited experience with dieliberately making ions, it seems desireable to have a fairly large electron current at modest voltages, in the ionizer. Also, the structure of the ionizing chamber needs to be as open as possible, so the central negative field of the inner negative grid can be felt by the ions. A good geometry to accomplish this is a transverse electron flux.. that is an electron flow that is at right angles to the radial field of the negative central grid. This suggests some type of windowed ion gun... so that the ions once formed are immediately drawn out and accelerated inwardly.
I am not sure how well electric fields penetrate a fine mesh. I need to get around to modeling that. But intuition says the mesh might screen pretty well, thus shielding the ions , taking longer for them to find their way out.
But with our low pressures, the differences might not be large, whatever configuration the ion gun takes.
Dave Cooper.
From my own limited experience with dieliberately making ions, it seems desireable to have a fairly large electron current at modest voltages, in the ionizer. Also, the structure of the ionizing chamber needs to be as open as possible, so the central negative field of the inner negative grid can be felt by the ions. A good geometry to accomplish this is a transverse electron flux.. that is an electron flow that is at right angles to the radial field of the negative central grid. This suggests some type of windowed ion gun... so that the ions once formed are immediately drawn out and accelerated inwardly.
I am not sure how well electric fields penetrate a fine mesh. I need to get around to modeling that. But intuition says the mesh might screen pretty well, thus shielding the ions , taking longer for them to find their way out.
But with our low pressures, the differences might not be large, whatever configuration the ion gun takes.
Dave Cooper.
-
Brian McDermott
- Posts: 682
- Joined: Wed May 28, 2003 6:28 pm
- Real name:
Re: Ports for ion guns
Joe Zambelli apparently tried to replicate the H-M device as well. His pictures are at:
http://myweb.wvnet.edu/~guf00478/iec/construc.html
I left some spare ports on my fusor so I could try that variant in the future, when time and resources allow.
http://myweb.wvnet.edu/~guf00478/iec/construc.html
I left some spare ports on my fusor so I could try that variant in the future, when time and resources allow.
-
Richard Hull
- Moderator
- Posts: 15024
- Joined: Fri Jun 15, 2001 9:44 am
- Real name: Richard Hull
Re: Ports for ion guns
I knew Joe tried it as we discussed it over the phone about 2 years ago. He found it difficult to control manually. This is why I haven't rushed into the effort. His numbers did go up by about 1 order of magnitude.
Dave, that was not 10e12 flux is was 10e12 isotropic. I know it is just words, but they mean a lot different power density. 10e12 "iso" is indeed a watt.
For those interested in the derivation of Daves calcs
2X10e12 (total fusions=2Xthe number of neuts) X ~3.5X10e6 (average ev/fusion) X 1.6X10e-19 (joules/ev) = 1.3 joules or ~1watt-second.
Richard Hull
Dave, that was not 10e12 flux is was 10e12 isotropic. I know it is just words, but they mean a lot different power density. 10e12 "iso" is indeed a watt.
For those interested in the derivation of Daves calcs
2X10e12 (total fusions=2Xthe number of neuts) X ~3.5X10e6 (average ev/fusion) X 1.6X10e-19 (joules/ev) = 1.3 joules or ~1watt-second.
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
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
-
sharks63128
- Posts: 20
- Joined: Thu Jan 27, 2005 1:10 pm
- Real name:
Re: Ports for ion guns
Thank you all very much for the help. Below is a sketch of my plan. Will the location of the ports work with future ion guns?
- Attachments
-
-
Richard Hull
- Moderator
- Posts: 15024
- Joined: Fri Jun 15, 2001 9:44 am
- Real name: Richard Hull
Re: Ports for ion guns
Yes, this looks oK if two guns are your goal.
Thanks to Brian for catching the error on the calcs. I was still locked in D-D mode.......As are we all.
D-T makes a big difference don't it!
Richard Hull
Thanks to Brian for catching the error on the calcs. I was still locked in D-D mode.......As are we all.
D-T makes a big difference don't it!
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
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
-
Brian McDermott
- Posts: 682
- Joined: Wed May 28, 2003 6:28 pm
- Real name:
Re: Ports for ion guns
Specifically, what trouble did he run into?
Perhaps this should be in a new thread?
Perhaps this should be in a new thread?
-
Richard Hull
- Moderator
- Posts: 15024
- Joined: Fri Jun 15, 2001 9:44 am
- Real name: Richard Hull
Re: Ports for ion guns
Joe found that the filament and ionizer system really performed too well and the main fusor acceleration current was seen to suddenly spike. The ionizer system current had a more far reaching effect on accelerator current to the point that the ionizer system had to be run at vey low levels that were tough to control. All of this made the pressure much more critical to control, as well. Lots of variables. perhaps too many for manual control. The only way to evaluate is to actually do it oneself, I figure.
I sort of wished Joe had reported on this personally, but he was not much for online interaction then.
Richard Hull
I sort of wished Joe had reported on this personally, but he was not much for online interaction then.
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
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
-
Brian McDermott
- Posts: 682
- Joined: Wed May 28, 2003 6:28 pm
- Real name:
Re: Ports for ion guns
Just to correct those calculations a bit:
I recall you saying that Hirsch's 10^12 machine ran on D-T fuel. If that is the case, every neutron signifies one fusion, and each fusion releases 17MeV. (The original calcs assumed D-D fusion.)
So the calculation ought to be:
10e12 (total fusions=the number of neutrons) X ~1.7X10e7 (average eV/fusion) X 1.6X10e-19 (joules/eV) = 27.2 joules or 27.2 watt-second
When he achieved these results, what were the voltage and current being used? Assuming the total input power was between 1 and 10 kilowatts, this is (relatively) much closer to breakeven than the typical 7-10 orders of magnitude.
I also posted this in the construction/operation forum, since this topic is no longer pertinent to the original thread.
I recall you saying that Hirsch's 10^12 machine ran on D-T fuel. If that is the case, every neutron signifies one fusion, and each fusion releases 17MeV. (The original calcs assumed D-D fusion.)
So the calculation ought to be:
10e12 (total fusions=the number of neutrons) X ~1.7X10e7 (average eV/fusion) X 1.6X10e-19 (joules/eV) = 27.2 joules or 27.2 watt-second
When he achieved these results, what were the voltage and current being used? Assuming the total input power was between 1 and 10 kilowatts, this is (relatively) much closer to breakeven than the typical 7-10 orders of magnitude.
I also posted this in the construction/operation forum, since this topic is no longer pertinent to the original thread.