scaleablity of a reactor?

[JimmyT]

Rematog,

I’m reluctantly forced to agree with you about the probability of regulation. It is interesting to note, however, that all the safety issues discussed do not involve the reactor. But rather the high voltage transmission equipment. And this equipment is present at all power facilities, be they coal powered, hydroelectric, etc.

[Aeronaut]

Jimmy- I didn’t mention reactor safety since each was assumed to be shielded to below background radiation level. Now I’ve stated that design point. Physical and electrical safety were the main thrust of my last post. Should the site encounter trespassers, I’d bet on them being kids 9.9 times out of 10, so I believe that illustration is perfectly valid.

Rematog, I hadn’t figured the cost of capital at all, as you pointed out. Could I get you to calculate a 10M$ construction and initial operating bond over 10, 20, and 30 years, please?

The 6 to 10 reactor figure I mentioned provides at least one reactor to cover the local network’s scheduled maintenance requirements. Planning for 10 reactors on 5 sites allows for 5 of them operating at as much as 98% rated load availability to pay the system’s way, as I noted in the previous post, albeit not quite as plainly. Given the potential load variations that I noted above, the main dedicated local reactor could also run at 9x% full rated available power and sell most of its output to said futility.

My previous post was designed to leverage the existing infrastructure to the fullest possible extent, including electrical and safety considerations. Michigan is one of something like 28 states that is required by law to buy a certain percentage of its power from wind farmers and fart burners. The big utilities are obligated to shadow every kW of this power with a fossil fuel-fired plant to fill in the gaps. FF provides the most predictable way to do that without building new plants, which will begin operation grossly underloaded.

The plan I’ve laid out offers the utility industry a graceful way out of burning fossil fuels without stranding partially depreciated assets, while expanding in 5 to 10 MW distributed chunks, on somebody else’s dime.

[Rematog]

10 million @6%

30 yr amortization = $719K/yr
20 yr = $860K/yr
10 yr = $1,332K/yr

10 million @8%
30 yr = $881K/yr
20 yr = $1,004K/yr
10 yr = $1,456K/yr

The above does not take into account taxes, depreciation, profits(?).

I’d also point out that the quick cost of power I did earlier today did’nt include any overhead for administration, insurance, profit, etc.

Rematog

[Aeronaut]

Rematog wrote: 10 million @6%

30 yr amortization = $719K/yr
20 yr = $860K/yr
10 yr = $1,332K/yr

10 million @8%
30 yr = $881K/yr
20 yr = $1,004K/yr
10 yr = $1,456K/yr

The above does not take into account taxes, depreciation, profits(?).

I’d also point out that the quick cost of power I did earlier today did’nt include any overhead for administration, insurance, profit, etc.

Rematog

Thanx for the numbers, Rematog.

From the bottom up, the utility partner would bill admin, insurance, profit, and other reasonable service amounts and deduct them from my township’s power bill. Like the way I reversed a common term that everybody thinks they know the meaning of? As this unfolds, I’m thinking of this more and more as one of those fabled Public/Private Partnerships which is coming into style for funding and managing large projects. In this case we would have 3 parties- the township voters, township government, and the utility company. I don’t think we’d have to look very hard to locate energy grants from the feds and possibly the state. Who knows- maybe they could provide enough loans to do this without, or with a much smaller bond issue.

Something to keep in is that Michigan’s entire economy is too closely tied to Detroit’s fortunes. Governor Granholm’s recovery strategy has centered around turning Michigan into the world’s renewable energy capitol by luring solar panel and windmill factories to locate here. She’s a lame duck, so a FF plant could give her the triumphant exit she’s got to be looking for. Not just a feather in her cap, but a full war bonnet when she goes to Washington.

Local politics aside, how many reactors do I need to reverse my $1,500/yr property taxes after all bills are paid? Not eliminate my property taxes, reverse the cash flow, please.

[Brian H]

Rematog wrote: Aeronaut,

Very interesting analysis. Now your getting a feel for “obligation to serve”, which is what a regulated utility has to it’s customers (the utility takes on this duty in exchange for the regualated monopoly status).

Not mentioned in this is the added cost per house to “own” their own power supply. If the FF module, installed on site, costs $1,000K, and the distribution system to these 100 homes costs $500K to install, that adds $15k to the construciton cost of each of these 100 homes.

I went back to a cost of power I’d work up in an old post, and maintenance for a FF module (1 day every 4 weeks + 2 weeks/year, I’d be glad to post if you’d like) @$119,240/year. Now, as this is being done in field, rather than a centralized plant site, multiply this by 25% (time to drive to site each day, cost of vehiles, and crafts being less well supervised) = $149,000 per year / 100 homes = $1,490/year. Plus, you have to have a back-up power supply (the Grid?) for those 24 days per year. (12 of the days would only have about 8 hrs down time so it works out to about 96% availablility, not a bad number at all).

With a cost of money of 6%, the capital investment costs + maintenance costs: $15K@6%= $900 + $1,500 maint = $2,400/yr or $200 Month.

While not a huge electric bill, this does not seem to me to be the “cheap power” the board is assuming.

WHY?

Because, the 5MW module is being sized for peak load which was done so it could be DEDICATED to a DISTRUBUTED location. The grid allows the central station to keep it’s units much more evenly loaded, so they actually generate a much larger percentage of their rated capacity.

So, Aeronaut, your example, to me, provides another arguement against distributed generation. (and I’ll not even discuss the “keep the kids from jumping the fence” safety issue.)

Rematog

The big number in your calcs is the maintenance ($1500), but your frequency number is way high, IMO. The maintenance cycle I’ve seen referred to is 2X/yr, 2 days downtime each. (9 hrs. cooling off, refuel, replace any degraded parts/items, restart). Your capital costs are also VERY ‘generous’, possibly double, IMO. Roughing it, that would cut the cost per household in half. I think you also overstate the margin necessary for peak for 100 homes; that’s a sufficiently large ‘sample’ that usage patterns are likely to track the overall grid fairly closely, with only a few % points wider peaks and valleys. Which would further dilute the cost/household. Your peak allowance of 50kw/household is probably triple actual max load. It takes quite a bit of hypothesizing to see 100 homes simultaneously drawing 15kw or more.

IAC, such a “cluster” would probably have some linkage to the overall grid, if only to be able to feed excess power into the system, which would be most of the time, maybe even virtually all the time.

IOW, I think you’re pushing your conservative ‘worst case’ assumptions way too far.

P.S.
There’s another category of cost reduction I haven’t taken into account, as it’s rather tricky to figure. It’s the reduction of ongoing and prospective alternative expenditure on conventional (=existing) power system construction and maintenance. That would vary hugely from situation to situation, I think.

[Rematog]

Brian,

I’d be glad to know from someone (Dr. Lerner?) what the expected maintenance cycle is, but in my posts from May, 08, no one challanged this. In that post (for a central station repowering, I also used 10 shifts for duration of annual overhaul. In that analysis, I assume day & night shifts. In this case, I assumed only day, as having night shifts in a remote location might even yield negative productivity. Note the costs above are for well paid ($25-$40/hr) skill craftsman. These costs also include money for estimated parts and materials. My judgements on the costs, wage rates, durations, are based on >25 years in heavy industrial maintenace and engineering experience. Yes, I know, “appeal to authority”….. but jeez, I have been doing this work since Reagan’s first term…

Cost. The 300K to 500K figure put out by the FF developers, is for a module, FOB the factory dock. It does not include foundations, cooling, controls, fences, etc. So I’ve consistantly used $1,000K. This is dirt cheap. FF at $200/kw capital and no fuel cost, vs new coal plant at about $1,600/kw to $1,800/kw, with large fuel cost. Fission now has construction cost of $2,358/kw, as estimated by the Congressional Budget Office in May 2008. My personal belief is that that is low and would be more in the range of $3,000 to $5,000/kw. So, $1,000,000 for 5 MW ($200/kw) is cheap!

[Aeronaut]

Rematog wrote: Brian,

I’d be glad to know from someone (Dr. Lerner?) what the expected maintenance cycle is, but in my posts from May, 08, no one challanged this. In that post (for a central station repowering, I also used 10 shifts for duration of annual overhaul. In that analysis, I assume day & night shifts. In this case, I assumed only day, as having night shifts in a remote location might even yield negative productivity. Note the costs above are for well paid ($25-$40/hr) skill craftsman. These costs also include money for estimated parts and materials. My judgements on the costs, wage rates, durations, are based on >25 years in heavy industrial maintenace and engineering experience. Yes, I know, “appeal to authority”….. but jeez, I have been doing this work since Reagan’s first term…

Cost. The 300K to 500K figure put out by the FF developers, is for a module, FOB the factory dock. It does not include foundations, cooling, controls, fences, etc. So I’ve consistantly used $1,000K. This is dirt cheap. FF at $200/kw capital and no fuel cost, vs new coal plant at about $1,600/kw to $1,800/kw, with large fuel cost. Fission now has construction cost of $2,358/kw, as estimated by the Congressional Budget Office in May 2008. My personal belief is that that is low and would be more in the range of $3,000 to $5,000/kw. So, $1,000,000 for 5 MW ($200/kw) is cheap!

Rematog, the only numbers I’d whittle away at are how long (and the assumed where) of the annual overhaul. I’d like to think that 9 hours or more after taking the reactor off-line the crew could do a core swap in only a few hours on-site. That would be a function of how the shielding, core, and access are designed into the reactors, along with task-specific jigs, dollies, etc. The core would be rebuilt on a properly instrumented rebuild bench and very likely verified on a burn-in rig, similar to aerospace procedures. This goes almost without saying that the utility provides these services in their overall service contract. Who else can all parties trust?

[Rematog]

Aeronaut,

You may be right that I’m on the high side on duration, but again, I could be on the low side of parts costs…..? I cetainly don’t have hard data. But, we’re nit picking the numbers, and the point was the order of magnitude. For exampe, if maintenance is only 60% of what I guesstimated, then cost per household would go from $200/month to $150/month. A nice savings, but not revolutionary.

And, I can see FF as revolutionary… but that it does not REQUIRE distributed installation to acheive this revolution. Which is good, as I think the regulators and/or public would be NIMBY to any nuclear reactor….and fusion is a nuclear reaction.

On another note, I don’t see utilites as providing core rebuilds. If this was done in a shop (nice thought there Aeronaut!), I’d guess it to be more likely factory offered or specialy technical companies. Utilites are regulated public business’s, with very high barriers to doing work in the un-regulated arena.

Rematog

[Aeronaut]

Rematog wrote: Aeronaut,

You may be right that I’m on the high side on duration, but again, I could be on the low side of parts costs…..? I cetainly don’t have hard data. But, we’re nit picking the numbers, and the point was the order of magnitude. For exampe, if maintenance is only 60% of what I guesstimated, then cost per household would go from $200/month to $150/month. A nice savings, but not revolutionary.

And, I can see FF as revolutionary… but that it does not REQUIRE distributed installation to acheive this revolution. Which is good, as I think the regulators and/or public would be NIMBY to any nuclear reactor….and fusion is a nuclear reaction.

On another note, I don’t see utilites as providing core rebuilds. If this was done in a shop (nice thought there Aeronaut!), I’d guess it to be more likely factory offered or specialy technical companies. Utilites are regulated public business’s, with very high barriers to doing work in the un-regulated arena.

Rematog

Rematog,

I, too see FF as a revolutionary nuclear technology that could easily inspire NIMBY and/or dismissal as pie-in-the-sky utopians if improperly presented. That’s why I described my system in the order I did in today’s first post. I wanted a safe pastoral undertone, lol. None of us can accurately guess the numbers, but we’ve accomplished a lot today and yesterday examining how both business models can evolve simultaneously. Neither is right for all situations, nor does either preclude the other. Factory service just might work with all parties concerned, but I’d sure like to see the existing credibility of the utility ease the negotiations and transition.

[Brian H]

Rematog wrote: Aeronaut,

You may be right that I’m on the high side on duration, but again, I could be on the low side of parts costs…..? I cetainly don’t have hard data. But, we’re nit picking the numbers, and the point was the order of magnitude. For exampe, if maintenance is only 60% of what I guesstimated, then cost per household would go from $200/month to $150/month. A nice savings, but not revolutionary.

And, I can see FF as revolutionary… but that it does not REQUIRE distributed installation to acheive this revolution. Which is good, as I think the regulators and/or public would be NIMBY to any nuclear reactor….and fusion is a nuclear reaction.

On another note, I don’t see utilites as providing core rebuilds. If this was done in a shop (nice thought there Aeronaut!), I’d guess it to be more likely factory offered or specialy technical companies. Utilites are regulated public business’s, with very high barriers to doing work in the un-regulated arena.

Rematog

Eric’s estimate of cost FOB factory door under full mass-production was MAX $250,000, and he gave examples of large cost ratios between hand/small volume and large volume items (e.g., the beryllium window being 14x cheaper per in² in even low-volume standard size vs. custom hand-made). There is also something that feels wrong about your allocating all site prep and connection costs to the new FF installations; there is a re-use/replacement issue here.
Anyhow, transmission lines are in place and comparatively inexpensive compared to the savings from siting on cheap/easily serviced, or otherwise economically attractive land. The options are considerably greater for FF clusters than coal or nuke plants, e.g.

[JimmyT]

Rematog

The big number in your calcs is the maintenance ($1500), but your frequency number is way high, IMO. The maintenance cycle I’ve seen referred to is 2X/yr, 2 days downtime each. (9 hrs. cooling off, refuel, replace any degraded parts/items, restart). Your capital costs are also VERY ‘generous’, possibly double, IMO. Roughing it, that would cut the cost per household in half. I think you also overstate the margin necessary for peak for 100 homes; that’s a sufficiently large ‘sample’ that usage patterns are likely to track the overall grid fairly closely, with only a few % points wider peaks and valleys. Which would further dilute the cost/household. Your peak allowance of 50kw/household is probably triple actual max load. It takes quite a bit of hypothesizing to see 100 homes simultaneously drawing 15kw or more.

IAC, such a “cluster” would probably have some linkage to the overall grid, if only to be able to feed excess power into the system, which would be most of the time, maybe even virtually all the time.

IOW, I think you’re pushing your conservative ‘worst case’ assumptions way too far.

P.S.
There’s another category of cost reduction I haven’t taken into account, as it’s rather tricky to figure. It’s the reduction of ongoing and prospective alternative expenditure on conventional (=existing) power system construction and maintenance. That would vary hugely from situation to situation, I think.

Fifteen Kw/household does seem like a big load….now. But maybe not if this undercuts natural gas for heating. And maybe a sizable fraction of households overnight recharge their electric vehicles ….and God only knows what else. Recharging their light sabers maybe?

[Aeronaut]

Brian, yours are the numbers I see in larger production as a tinker who’s worked in manufacturing for many years, in many market niches, including aerospace. The basic reactor and solenoid should be a breeze to mass produce. Ditto for capacitor plates, possibly for entire cap banks. I can see every last one of these modules having an aftermarket similar to engine parts and PC components, which would drive the cost of parts WAY down. Now throw in cheap energy…..

The priciest component will be the X-ray harvester, with its thousand or more foils of differing thicknesses. The obvious solution would be to stamp them out on presses and distribute copies of those thousand dies to as many as a thousand stamping plants. (I like this! All these shuttered factories can go back to work again!). Talk about a job from Hell. Just designing the dies, let alone making them, is going to cost dearly in time and capital. Parts this thin are lousy candidates for the molding process because even if the metal flows, it will cool before reaching the end of the mold. Now I’m thinking Chemical Vapor Deposition (CVD) which the semiconductor industry uses to stack thin layers of metals and semiconductors to make CPUs, GPUs, memories, etc. This eliminates all of the die and press problems and invites somebody to mass produce those machines to “cash in on the lucrative fusion aftermarket”, further reducing that part’s price, in time.

The downside of distributed generation will be the price of the step-up transformer and the high current switching and wave shaping circuitry needed to distribute even the 13.2kV I used in yesterday’s example. High current parts go for mil-spec prices, as do high voltage parts. Compact designs optimizing for high current AND high voltage are where Rematog’s numbers will average up into his pricing ball park. Hopefully that, too, will change with mass production.

The single most important part of Rematog’s guesstimates are not the actual numbers, but the categories that they’re listed in. Labor costs suggest union labor. “Nuke plant” site prep will be priced at government prevailing rate labor, so a simple 25 by 25 foot steel building is no longer priced the way it could be if we could light up a few million high school and college age people with “The Wonder of Fusion”, thus obliterating the Nuclear Mystique and turning these cool little buildings into neighborhood status symbols. Heh, heh, heh.

Rematog’s also pointed out that consumption is inversely proportional to price. An all electric car with a range of even 200 miles would be very appealing to me. I’d either keep my ’05 gas burner for road trips or rent a hybrid for the rare occasions I really would need more than 100 miles in a day. Now we’re talking about charging two or more electric cars most nights. Throw in some light sabers for the road warriors, 30A to 50A “cheap resistance heating” furnaces, water heaters, clothes dryers, and Thanksgiving dinners, and the 15kW average requires a LOT of instantly available reserve capacity.

The utility industry’s problem is how to plan and execute for anticipated growth in demand despite a lag time of something like ten years or more per plant, which is likely to strand capital IF fusion ever leaves the labs. Sucks to burn fossil fuels, but what else is proven (reliable) until thousands of townships, villages, and regions rise up “out of nowhere” and demand new legislation permitting them to partner with the utilities to own their own, extremely reliable, local fusion plants. All it takes is one township to light the fuse…

I therefor believe it is in the utility industry’s best interests to put systems into place that would welcome and fast-track the integration of these local systems. The PR benefits alone would be stupendous, and the transition out of burning fossil fuels could easily be scheduled along a thirty to fifty year plan.

[Brian H]

Aeronaut wrote: Brian, yours are the numbers I see in larger production as a tinker who’s worked in manufacturing for many years, in many market niches, including aerospace. The basic reactor and solenoid should be a breeze to mass produce. Ditto for capacitor plates, possibly for entire cap banks. I can see every last one of these modules having an aftermarket similar to engine parts and PC components, which would drive the cost of parts WAY down. Now throw in cheap energy…..

The priciest component will be the X-ray harvester, with its thousand or more foils of differing thicknesses. The obvious solution would be to stamp them out on presses and distribute copies of those thousand dies to as many as a thousand stamping plants. (I like this! All these shuttered factories can go back to work again!). Talk about a job from Hell. Just designing the dies, let alone making them, is going to cost dearly in time and capital. Parts this thin are lousy candidates for the molding process because even if the metal flows, it will cool before reaching the end of the mold. Now I’m thinking Chemical Vapor Deposition (CVD) which the semiconductor industry uses to stack thin layers of metals and semiconductors to make CPUs, GPUs, memories, etc. This eliminates all of the die and press problems and invites somebody to mass produce those machines to “cash in on the lucrative fusion aftermarket”, further reducing that part’s price, in time.

The downside of distributed generation will be the price of the step-up transformer and the high current switching and wave shaping circuitry needed to distribute even the 13.2kV I used in yesterday’s example. High current parts go for mil-spec prices, as do high voltage parts. Compact designs optimizing for high current AND high voltage are where Rematog’s numbers will average up into his pricing ball park. Hopefully that, too, will change with mass production.

The single most important part of Rematog’s guesstimates are not the actual numbers, but the categories that they’re listed in. Labor costs suggest union labor. “Nuke plant” site prep will be priced at government prevailing rate labor, so a simple 25 by 25 foot steel building is no longer priced the way it could be if we could light up a few million high school and college age people with “The Wonder of Fusion”, thus obliterating the Nuclear Mystique and turning these cool little buildings into neighborhood status symbols. Heh, heh, heh.

Rematog’s also pointed out that consumption is inversely proportional to price. An all electric car with a range of even 200 miles would be very appealing to me. I’d either keep my ’05 gas burner for road trips or rent a hybrid for the rare occasions I really would need more than 100 miles in a day. Now we’re talking about charging two or more electric cars most nights. Throw in some light sabers for the road warriors, 30A to 50A “cheap resistance heating” furnaces, water heaters, clothes dryers, and Thanksgiving dinners, and the 15kW average requires a LOT of instantly available reserve capacity.

The utility industry’s problem is how to plan and execute for anticipated growth in demand despite a lag time of something like ten years or more per plant, which is likely to strand capital IF fusion ever leaves the labs. Sucks to burn fossil fuels, but what else is proven (reliable) until thousands of townships, villages, and regions rise up “out of nowhere” and demand new legislation permitting them to partner with the utilities to own their own, extremely reliable, local fusion plants. All it takes is one township to light the fuse…

I therefor believe it is in the utility industry’s best interests to put systems into place that would welcome and fast-track the integration of these local systems. The PR benefits alone would be stupendous, and the transition out of burning fossil fuels could easily be scheduled along a thirty to fifty year plan.

Good analysis and comments! As for the load, I was suggesting that 15kW would be a peak, not an average. Heating would be significant in the northern winters, cooling significant in the summers, but even so 15kW is pretty big. And that’s STILL only 1.5MW demand / 100 homes.

Note that the housing/building is part of the factory unit. I assume that various shell alternatives would be offered over time, and prettification is always a local option, I suppose. As for the foils, I suspect the difficulty is less than you describe; I’d like to see what analysis Eric has on hand about this. Perhaps the patent filing would provide some clues.

The 200 mi BEV exists, if you want a $100K roadster soon, or a $50K sports sedan in 2 years or so: http://www.teslamotors.com/ , http://tinyurl.com/TeslaConvoy . The 200 miles costs about $4 recharge at current prices, a fraction of that with FF. The $50,000 sports sedan would cost the same to lease and run as a $30,000 ICE gas-burner.

Your lead-time comment points out another contrast: it would be slashed by about 90% for FF when the factories are up and running. This will almost make power grid planning a JIT operation, or at least far more responsive.

[Aeronaut]

Brian H wrote:
Good analysis and comments! As for the load, I was suggesting that 15kW would be a peak, not an average. Heating would be significant in the northern winters, cooling significant in the summers, but even so 15kW is pretty big. And that’s STILL only 1.5GW demand / 100 homes.

Note that the housing/building is part of the factory unit. I assume that various shell alternatives would be offered over time, and prettification is always a local option, I suppose. As for the foils, I suspect the difficulty is less than you describe; I’d like to see what analysis Eric has on hand about this. Perhaps the patent filing would provide some clues.

The 200 mi BEV exists, if you want a $100K roadster soon, or a $50K sports sedan in 2 years or so: http://www.teslamotors.com/ , http://tinyurl.com/TeslaConvoy . The 200 miles costs about $4 recharge at current prices, a fraction of that with FF. The $50,000 sports sedan would cost the same to lease and run as a $30,000 ICE gas-burner.

Your lead-time comment points out another contrast: it would be slashed by about 90% for FF when the factories are up and running. This will almost make power grid planning a JIT operation, or at least far more responsive.

Oh, yeah, I drooled on my keyboard the first time I visited those links, lol. Now to get a FF license so I can afford one of them.

15kW in my opinion is only on the low end of first shift hours. I think it would be higher after 5:00 local time. My range, furnace, and hot water are all gas-fired, which would change if the township eliminated or even reversed at least my electric and gas bill.

Yep, grid expansion will come down to the time required to finance and permit each installation. That’s a very cool thing because once somebody breaks the ice all bets are off regarding how fast the demand for FF installations and electricity is going to take off.

[Brian H]

Aeronaut wrote:

Oh, yeah, I drooled on my keyboard the first time I visited those links, lol. Now to get a FF license so I can afford one of them.

15kW in my opinion is only on the low end of first shift hours. I think it would be higher after 5:00 local time. My range, furnace, and hot water are all gas-fired, which would change if the township eliminated or even reversed at least my electric and gas bill.

Yep, grid expansion will come down to the time required to finance and permit each installation. That’s a very cool thing because once somebody breaks the ice all bets are off regarding how fast the demand for FF installations and electricity is going to take off.

Here’s a sample study of the kW usage impact of upgrading AC units in Florida: http://www.fsec.ucf.edu/en/publications/html/FSEC-PF-379-04/ . Note the overall usage mentioned — about half of all annual power goes to heating and cooling. The peak cooling demand even in the summer is ~3kW, much less with modern Variable Speed Air Handler units. I conclude from this that a peak demand per household would be <<10kW. Therefore, 1 MW / 100 homes. Total power used during the year was around 17MWh, or about 2kW average demand, with daytime peaks in the 7 kW range.

Note the para in this site that refers to peak demand usually in the 3-6 kW range, but sometimes “as high as 8 kW”. So I think my figures above are very reasonable.

Budgeting a 15kW peak demand is at about a 2:1 safety margin. At that level, a single 5MW FF generator would be able to safely cope with 5000/15 = 333 residences. Commercial and industrial users are of course a completely separate issue and market.

[Author]

Posts