Sunday, April 4, 2010

Conversations and updates on the Polywell


For those of you out there who have not been following the Polywell research recently, the group at the Navy have come out with their next generation design for a test reactor, WB 7.1. WB 7.1 looks to be a duplicate of WB 7 with some improved diagnostics. They have also put a nice picture of WB-7 in operation. MSNBC did another blog post on the machine. I sent out links to let people know about the developments. A good friend had some questions and comments about all this. Below are the answers to his questions from my own research and from M. Simon of Talk-Polywell. The answers are included here with links and the pictures for your enjoyment.



MSNBC article


1. What are the pieces that touch the vertices of the loops?

The metal loops are connected by a very strong insulator. The blocks are white, so I am guessing they are probably Beryllium Oxide. Beryllium is a very nasty thing to work with. It is transparent to X-rays, which is why we like it. The stuff is ~$105 per pound; those blocks in there might be ~300-400 dollars.

2. Why are there so many freaking holes in design 2?

The holes are for the vacuum pumps, electron guns, Ions guns, and Neutron counters. As long as we fuse DD or DT we are making neutrons. We count the neutrons to determine how well we are doing. Neutrons can also cook the reactor, destroying the rings and making them radioactive. The holy grail of fusion, however, is fusing PB11, a much hard thing to fuse. If that were to happen there would be low amounts of neutrons, low amounts of radioactivity.

The holes are for diagnostics. A production reactor would not need so many. pB11 would reduce neutrons by a factor of about 1E3 to 1E6. This is caused by side reactions. The Hydrogen used in the reactor would need to be purified as much as possible in order to eliminate Deuterium.

3. This is designed to heat water as a means of energy exchange?

I do not know about the DD & DT energy exchange. There is probably a steam cycle in this scheme. I believe you can catch neutrons in a blanket of Lithium (~$300 per pound). If we fuse PB11, there is a great way to get the energy out: direct conversion. PB11 fusion generates alpha particles or hydrogen nucleus, when these strike metal they can draw a current. The theoretical efficiencies on this process are very high.

D-D and D-T would probably use a steam cycle. However, alpha collection may be worthwhile even in those cases. It would be an engineering/economic question.

4. Does this thing still only cost $200k to build?

No. I think $200K was a guess at the final commercial price. It is still allot cheaper than NIF (~$4.2 billion) and ITER (~$1.23 billion for US). The beauty is in the size and the scale. As of right now, you do not need exotic technologies to get this to work. The e-gun is pre-WWII, a very mature technology. The modern ion gun is late 70's. The rings could be superconductors, which could create cooling issues.

$200K would be for a small pulsed experiment. My estimate for a production reactor is about $.50 to $1/watt. The experimental net power machine (100 MW) is estimated at $200 million. That would include a reactor building (for shielding).

5. 100MW.....holy crap.

Bussard showed that the fusion rate increases with the 7th of the radius. He showed that the net power out increases as the 5th of the radius. He also showed that increasing the rings increases confinement which makes it work better. EMC2 inc. is assuming he is right, and that is why they are calling the 20 ft x 20 ft x 20 ft device a 100 MW machine.

Power is a function of R^3 (volume) and B^4 (magnetic field). Net power will be less (to account for losses). The R^7 is for copper coils. For superconducting coils the scaling needs to be separated into its components due to limitations on maximum field of the super conductors.