| View previous topic :: View next topic |
| Author |
Message |
ravingdave
Joined: 27 Jun 2007
Posts: 645
|
 Posted: Sun Feb 01, 2009 11:24 pm Post subject: Why does it cost so much ? |
 |
|
One of the most beautiful things about the Polywell reactor concept is that it's simple. It should likewise be relatively inexpensive. Why does it need 200 million dollars to build it ? Perhaps i'm horribly naive, but I'm thinking the thing could be built for less than a million, perhaps even as little as a hundred thousand.
If we had someone with a notion like Harry Broderick...
I bet this thing would get built.
http://www.youtube.com/watch?v=eSsqsbv0FWg
What's the biggest expense ? Vacum Chamber and Pumps ? Power Supply ? Superconducting Wire ?
I think a dedicated amatuer could cobble all this stuff together rather inexpensively. Maybe not for a 100K, but surely for a million.
Too bad John Carmack already found something to work on.
While Politics and Religion are fun, I've been wanting to start moving back to the whole purpose of this Website, so i'm just thinking out loud.
David |
|
| Back to top |
|
 |
MSimon
Joined: 16 Jul 2007
Posts: 8780
Location: Rockford, Illinois
|
| Posted: Mon Feb 02, 2009 1:58 am Post subject: |
|
|
Yeah, a 5 to 20 MW 100KV HV low ripple DC supply for only $1 million. Got a broker?
_________________
Engineering is the art of making what you want from what you can get at a profit. |
|
| Back to top |
|
|
zbarlici
Joined: 17 Jul 2007
Posts: 201
Location: winnipeg, canada
|
| Posted: Mon Feb 02, 2009 5:50 am Post subject: |
|
|
HUMOR RELIEF TIME!!!!
Q. What`s the difference between a stock broker and a pigeon?
A. The pigeon can still make a deposit on a porsche. |
|
| Back to top |
|
|
windmill
Joined: 19 Jul 2007
Posts: 20
Location: Toronto, Canada
|
| Posted: Mon Feb 02, 2009 5:58 pm Post subject: |
|
|
Diversified Tech has some extremely interesting power supply designs using IGBTs that look like they would be right up our alley for Polywell open source development. Somewhere in one of their technical papers they mentioned a cost of $100/kw for production quantities.
http://www.divtecs.com/ |
|
| Back to top |
|
|
MSimon
Joined: 16 Jul 2007
Posts: 8780
Location: Rockford, Illinois
|
| Posted: Mon Feb 02, 2009 6:37 pm Post subject: |
|
|
| windmill wrote: | Diversified Tech has some extremely interesting power supply designs using IGBTs that look like they would be right up our alley for Polywell open source development. Somewhere in one of their technical papers they mentioned a cost of $100/kw for production quantities.
http://www.divtecs.com/ |
It is the price of the first one that is the current killer.
==
After looking at the design I'm impressed. It all depends on what the final requirements are. 1% regulation and .1% ripple is not bad. However the switching frequency (5 kHz) is low.
My design of 50 or 100 phases in series run at 30 khz would give tighter regulation. (.5 to .1%) and lower ripple (.05% to .01%) plus it would give an equivalent ripple output frequency of 50X to 100X i.e. 1.5 to 3 MHz. Very easy to filter even at very high voltages. Plus by adjusting the individual phases up or down with 14 bit control it would provide a final output control of better than 1 volt. If all phases were adjusted at the same time 10 V steps would be easy in a 100,000 VDC nominal supply.
I think the numbers my design would provide are better for the pB11 resonance. Of course if we are desperate for any results (esp if we start with D-D which has no resonance peak) 1% regulation and .1% ripple is fine. With 2 or 4 of the buck converters phased we might get regulation down to .25% which might be adequate. (i.e. +/- 250V). Of course your costs get multiplied by 2X or 4X. And with 30kHz conversion frequency supply closed loop response would be in the 1 to 3 kHz range. A 5 kHz switching frequency would give a control loop response of 200 to 500 Hz.
And I think my design might only be necessary for an experimental version. Once we find out what works we can relax the specs.
For a front end I was thinking of a 500VDC PFC off the shelf traction motor supply feeding the converter modules. Power quality is important.
_________________
Engineering is the art of making what you want from what you can get at a profit. |
|
| Back to top |
|
|
ravingdave
Joined: 27 Jun 2007
Posts: 645
|
| Posted: Mon Feb 02, 2009 9:07 pm Post subject: |
|
|
| MSimon wrote: | | Yeah, a 5 to 20 MW 100KV HV low ripple DC supply for only $1 million. Got a broker? |
To quote Sgt. Odball... (Kelly's Heroes)
Your specs might be necessary for a 100 MW power producing system, but something which could prove whether break even is possible or not, needn 't require so much energy, or low ripple. If you will recall, you and I and a few other people discussed the idea of a 60 hz pulsed reactor, and if the idea is workable, we might be able to get by with something like this.
http://www.publicsurplus.com/sms/all,ca/auction/view?auc=312042
If we absolutely HAVE to have DC, we can rectify it if necessary, and we could even use a voltage doubler circuit to push the voltage up to around 195K and come out with 120 hz ripple.(Per Phase. 360hz for all three) At that point we could LC filter it and build a voltage regulator circuit to clean up most of what was left.
(I haven't run the numbers on the capacitors required, but that might be the biggest hang up to this approach. )
A wise man once said ...
| Quote: | | Science is hard. Engineering is harder. Engineering is the art of making what you want from what you can get at a profit. |
In any case, it's fun speculating on various ideas that MIGHT work, while we're waiting for the Scientists and the Government to make some progress.
David |
|
| Back to top |
|
|
MSimon
Joined: 16 Jul 2007
Posts: 8780
Location: Rockford, Illinois
|
| Posted: Mon Feb 02, 2009 9:19 pm Post subject: |
|
|
Big supplies are required for continuous operation. And the power supples may be fairly device size insensitive. Graphs I have seen show device current going up with reduced size. Not a Polywell. But we don't know anything.
Pulsed operation (even 60 Hz) tells you nothing about lower frequency instabilities.
_________________
Engineering is the art of making what you want from what you can get at a profit. |
|
| Back to top |
|
|
MSimon
Joined: 16 Jul 2007
Posts: 8780
Location: Rockford, Illinois
|
| Posted: Mon Feb 02, 2009 11:37 pm Post subject: |
|
|
Voltage doublers don't regulate well.
_________________
Engineering is the art of making what you want from what you can get at a profit. |
|
| Back to top |
|
|
ravingdave
Joined: 27 Jun 2007
Posts: 645
|
| Posted: Mon Feb 09, 2009 11:46 pm Post subject: |
|
|
| MSimon wrote: | | Voltage doublers don't regulate well. |
This will regulate better.
This Idea uses that 40 MVA transformer from Anheim that I've previously linked. Cost of moving it is about $ 6,000.00 BOE. (That is if there is no government interference.) Might be able to find something cheaper and closer. I Know some people at AEP that might be able to put me on to something.
As for the rectifiers, there might be something out there that can handle those voltages and currents, but if there isn't I suppose a person could make their own custom Ignitrons. The caps can be home made using teflon sheeting, and the inductors can also be home made.
I haven't calculated the ripple yet, it took me a couple of hours just to draw the schematic. In any case, I think it's possible to do active regulation. For coarse voltage regulation saturatable reactors could be used in the transformer primarys, and those could be home made as well. (if something can't be picked up surplus. )
Just some ideas.
David |
|
| Back to top |
|
|
MSimon
Joined: 16 Jul 2007
Posts: 8780
Location: Rockford, Illinois
|
| Posted: Mon Feb 09, 2009 11:57 pm Post subject: |
|
|
Fine Dave,
Filament rectifiers. Are you kidding? No one uses those any more - if they can help it.
And it is not just the transformer. You need to regulate the voltage down to around 5 KV or so.
BTW such a system has 6% ripple before filtering.
With a power supply running at $.25 a watt a 5 MW power supply does not seem so daunting.
_________________
Engineering is the art of making what you want from what you can get at a profit. |
|
| Back to top |
|
|
Roger
Joined: 06 Jul 2007
Posts: 635
Location: Metro NY
|
| Posted: Mon Feb 09, 2009 11:58 pm Post subject: |
|
|
That was Sgt Oddball speaking to his tank driver, played by the same guy who played Capt Stubin on the Love Boat.
Said to Clint Eastwood:
..with so many positive vibes how could we lose?
Possibly the finest movie of its kind.
_________________
I like the p-B11 resonance peak at 50 KV acceleration. In2 years we'll know. |
|
| Back to top |
|
|
TallDave
Joined: 25 Jul 2007
Posts: 2524
|
| Posted: Tue Feb 10, 2009 1:01 am Post subject: |
|
|
| Quote: | | One of the most beautiful things about the Polywell reactor concept is that it's simple. It should likewise be relatively inexpensive. Why does it need 200 million dollars to build it ? Perhaps i'm horribly naive, but I'm thinking the thing could be built for less than a million, perhaps even as little as a hundred thousand. |
No, not possible. Concentric cooling is going to be very expensive, in addition to the above.
Also, too many people seem to think they can just build this, turn it on, and presto! -- cheap clean fusion power. Nuh-uh. It is going to require a lot of very expensive diagnostics and fine-tuning to get there, assuming it can be done at all, which is most definitely not a given. It certainly can't be done cheaply or quickly.
It's only cheap and simple and quick when compared to a $20B tokamak. |
|
| Back to top |
|
|
ravingdave
Joined: 27 Jun 2007
Posts: 645
|
| Posted: Tue Feb 10, 2009 2:01 am Post subject: |
|
|
| TallDave wrote: | | Quote: | | One of the most beautiful things about the Polywell reactor concept is that it's simple. It should likewise be relatively inexpensive. Why does it need 200 million dollars to build it ? Perhaps i'm horribly naive, but I'm thinking the thing could be built for less than a million, perhaps even as little as a hundred thousand. |
No, not possible. Concentric cooling is going to be very expensive, in addition to the above.
Also, too many people seem to think they can just build this, turn it on, and presto! -- cheap clean fusion power. Nuh-uh. It is going to require a lot of very expensive diagnostics and fine-tuning to get there, assuming it can be done at all, which is most definitely not a given. It certainly can't be done cheaply or quickly.
It's only cheap and simple and quick when compared to a $20B tokamak. |
I think i'm being misunderstood, and it's probably because I haven't emphasized one point enough.
I am not saying you can build a working power plant for $ 1 million. I am saying you can build something that will ignite a fusion reaction sufficient to prove the concept. Yeah, heat loading is going to be a B*tch. (if the idea works. )
What I am talking about, is that it ought to be possible to build something for $ 1 million dollars that can be fired up long enough to produce a 100 mega watts of fusion long enough for instruments to record it. ( of course if it achieves fusion, the heat will blow the superconductors, and the wiffleball will "explode", possibly damaging the machine. )
Once an experiment proves the idea to skeptical scientists (Hello Art ! ) then I have little doubt that the money floodgates will open and drown everything associated with polywell fusion in a cash stream.
In any case, I think speculating along this vein is fun. What else is there to do until someone pays for a bigger reactor ?
I guess we could talk about politics and religion.
David |
|
| Back to top |
|
|
ravingdave
Joined: 27 Jun 2007
Posts: 645
|
| Posted: Tue Feb 10, 2009 2:10 am Post subject: |
|
|
| MSimon wrote: | Fine Dave,
Filament rectifiers. Are you kidding? No one uses those any more - if they can help it.
And it is not just the transformer. You need to regulate the voltage down to around 5 KV or so.
BTW such a system has 6% ripple before filtering.
With a power supply running at $.25 a watt a 5 MW power supply does not seem so daunting. |
6% = 6kv of ripple. I've seen broadcast transmitters with 15Kv Plate voltage, so regulating 6kv of ripple is no big deal. As for the filaments, i'm just giving an example of what an amateur could do. I don't know if there are rectifiers out there that can handle 150 Kv at 200 amps, but if necessary, some could be made. Of course the filaments wouldn't be the cathode, they would just heat the cathode.
David |
|
| Back to top |
|
|
MSimon
Joined: 16 Jul 2007
Posts: 8780
Location: Rockford, Illinois
|
| Posted: Tue Feb 10, 2009 2:17 am Post subject: |
|
|
The supply needs to be able to go from 5KV to 100KV. That is the regulation problem I'm talking about.
_________________
Engineering is the art of making what you want from what you can get at a profit. |
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
