Andrew Seltzman

Compilations of the works of significant long term members of the site. Induction into this area is by long term contribution and by nomination only.
Post Reply
Andrew Seltzman
Posts: 811
Joined: Sun Feb 01, 2004 8:02 pm
Real name: Andrew Seltzman

Andrew Seltzman

Post by Andrew Seltzman »

I'll have to update this to a more detailed description and bio over the next few weeks or so but here is a start:
  • I was born and grew up in Raleigh, NC
  • I went to Georgia Tech for undergrad and received BS degrees in Physics and Electrical Engineering. I built three fusors in my dorm room, the latest of which I still have and have been upgrading since
  • I went to grad school at University of Wisconsin - Madison and received a MS in electrical engineering, and a PhD in Plasma Physics
  • I am currently (as of 2020) a postdoc at Massachusetts Institute of Technology, where I work on lower hybrid current drive in tokamaks
Advice to new fusioneers:
  • e-Bay is your friend, search for vacuum parts at a discounted price there
  • Avoid thermocouple gauges, they are temperamental and only go down to ~1mTorr, and then only at lower resolution and accuracy. Instead use a more modern multi sensor gauge. I recommend the MKS-901p, you can find them on e-bay for about $40. They are compact, have analog and digital rs-232 computer output, set-point relays, and gas type correction. They measure from atmospheric to 1e-5torr.
  • If you can afford it, invest in a dry turbo pump system early on. You won't regret it and will never go back to any other vacuum pumping method. This combination pairs an oil free (dry) diaphragm vacuum pump with a turbomolecular-drag hybrid turbopump. To run with a diaphragm pump (typically 1Torr vacuum) it MUST be a turbomolecular-drag hybrid. My recommended pump brand for turbos is Pfeiffer (I think they are a bit higher quality and more reliable). Turbomolecular-drag hybrids from pfeiffer are denoted by the TMH (ISO flange), TMU (conflat) prefix or TPD(molecular drag only; it's a lower pumping speed but more than sufficient for fusor use unless your vacuum chamber is huge), I would recommend buying a TMH-071P or TPD011 Pump (these will usually cost <$1000 used). Buy a pump that has the model TC100 or TC110 integrated controller.
  • Avoid pipe threads where ever possible, instead invest in NW or KF type hoses for your vacuum runs, particularly after your high vacuum pump
  • If you are serious about building a neutron producing fusor, start off with buying conflat flanges hardware for your vacuum chamber, and use them with copper o-rings where ever possible. Viton o-rings slowly permeate air and water vapor; large numbers of o-rings will affect gas purity, particularly at lower pumping rates and will decrease neutron numbers.
  • Radiation safety is important, particularly as you start to get to lower vacuum (<20mTorr) and higher voltage (>15kV). The hazard is x-rays not neutrons. Use a lead glass shielded viewport (I recommend the Lesker VPZL-275XLG, it's about $162. With a 0.24" thick lead glass insert it will screen out all x-rays in the typical armature fusor voltage operating range). This will provide x-ray hazard free viewing of your plasma. At 40kV, 18mA my fusor produces ~40R/h at the surface of a borosilicate glass viewport (this attenuates rapidly with distance due to inverse square law and air absorption of the lower energy soft x-rays) with the lesker lead glass viewport (in conjunction with a 1/8" thick stainless vacuum vessel), dose rate is reduced to <1mR/h
  • Use macor for the in-vacuum insulator. Alumina insulators tend to arc near the grid (but are generally a good starting place for most people before moving to more advanced ceramics), boron-nitride absorbs and water and releases it as it is heated, and quarts can be cracked by arcing at the lips if your power supply has large enough filter capacitors (most switching power supplies do). A field control cap over the top of the ceramic insulator just under the grid will reduce arcing. /list]

    Links to highlights my work:
    My personal website:-----------------------------
    My google scholar profile with publication list: ... by=pubdate
    Thread on my latest fusor:------------------------viewtopic.php?f=6&t=10294
    Thread on my ion source design:-----------------viewtopic.php?f=12&t=5011
    Thread on thermal imaging of the fusor grid:--viewtopic.php?f=6&t=8734&hilit=Mark3
Andrew Seltzman
User avatar
Dennis P Brown
Posts: 3272
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Andrew Seltzman

Post by Dennis P Brown »

While conflat are superior for high vac, I don't see any advantage for most fusors - 10^-6 torr isn't needed so, not clear what is superior for that type flange over standard, and cheaper (usually) KF adapters. I have had great results using KF's since going down to mid 10^-5 torr is more than enough (but my KF's work fine to mid 10^-6 torr but that, I wouldn't recommend.) They are cheap, a lot of availability and steel bellow tubes are easy to get for that type of connector (also, good deals on those, as well.)

I have both a turbo and water cooled diffusion pump and frankly, far prefer the DP over the turbo. Why? Simplicity and control - I have no fear of accidental gas overload on the DP but a turbo, that can be a bad day. While I agree for most, (due to water cooling) a DP isn't necessarily a easy install, but in cost, there is generally no comparison. While I got my turbo, cable & controller for $10, doubt others will ever get that lucky. My DP was $100 but came with a manual gate valve and adapter coupling. Unlike the turbo price, the DP assembly can be had for reasonable prices on ebay if one is careful and willing to wait and negotiate when available.

Still, your advice is excellent and these are just my experiences due to random ebay purchases.
User avatar
Richard Hull
Posts: 15111
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Andrew Seltzman

Post by Richard Hull »

Andrew, great bio and thanks for the background on your life to this point. Any cool photos of your youth or hobbies? I hope you will, indeed give us more on yourself in future.

There is no doubt that I prefer conflats over all other seals, but I have been amazed at the depths of vacuum I can reach with my turbo and the two KF 25 fittings in my hi-vac line to the chamber. I owe this to the key fact of using a thin coating of apiezon "H" grease on the sealing rings. As the chamber is 100% conflats, I have to look at the Hi-vac KF fittings as the weakest seals in the system, yet they serve well. In the end, the fusor doesn't require high vacuum during operation. Where heat exists, I prefer conflats. There is zero heat on my Hi-vac lines so they serve.

Andrew is in the High vacuum biz and his thoughts are sage. One could do worse than follow his advise. He is one of the "old boys" now. Newbies, pay attention to the old boys! We've been there, done that, stumbled, got up and did it right thereafter.

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
Post Reply

Return to “Compilations of long term technical contributors”