11
$\begingroup$

After about 01:10:10 in the 2022-02-10 SpaceX Starship Update part of Musks answer to Tim Dodd's question about Raptor 2 development:

The only remaining issue that we're aware of is melting the chamber, so... that thing really wants to melt. It's got like on the order of a gigawatt of heat, so it's pretty hot. Like a gigawatt is what a nuclear power plant produces, so it's really desperately trying to melt at any time.

Question: Is a Raptor 2's thermal output really comparable to that of a nuclear power plant?

$\endgroup$
8
  • 4
    $\begingroup$ Technically, any two numbers are comparable to each other. There's "just" a slight (or not so slight) scale factor... $\endgroup$
    – RonJohn
    Feb 12, 2022 at 23:27
  • 6
    $\begingroup$ They're in the same ballpark, but the comparison is rather uninteresting, it's more of a marketing quip (because it sounds cool if you don't think about it too deeply) than anything else. Sort of like saying that a Pop Tart has enough energy to power a PlayStation 5 for 70 minutes (200 kcal / 200 W = 1.16 h). I mean... it's true in some sense, but it doesn't really mean anything. $\endgroup$
    – Jason C
    Feb 13, 2022 at 6:37
  • 3
    $\begingroup$ @JasonC do you know how small a Raptor 2's combustion chamber is? The point is that it sustains a gigawatt of heat generation yet doesn't melt. The comparison is incredibly interesting. $\endgroup$
    – uhoh
    Feb 13, 2022 at 11:27
  • 3
    $\begingroup$ @uhoh Exactly! The point is that it sustains a gigawatt of heat generation but doesn't melt. :) The gigawatt is the interesting bit. The nuclear power plant comparison is the silly bit, and could be left out without losing information. Consider: Most of us (definitely me; probably you too?) don't actually have a real concept of the thermal output rate of a nuclear power plant, we just have a vague sense of "wow, very heat!". I mean, I bet if I told somebody "A nuclear plant has massive thermal output, in the ballpark of a Raptor 2 rocket engine!" they'd still say "wow". See... $\endgroup$
    – Jason C
    Feb 13, 2022 at 18:46
  • 2
    $\begingroup$ (If it still doesn't make sense, think about it this way then: "A strawberry Pop Tart has only 200 calories, while a brown sugar cinnamon Pop Tart contains enough energy to power a full-bore PS5 for nearly 70 minutes." Clearly the brown sugar cinnamon one sounds more intense, even though both flavors are ~200 cal (food cal = physics kcal, FYI), and regardless of whether or not the reader has a concept of how much power a PS5 uses. No real info was added to the latter, it just changed the perception.) $\endgroup$
    – Jason C
    Feb 13, 2022 at 19:06

5 Answers 5

12
$\begingroup$

Power is in units of "energy rate," joules per second, and thus it is relatively easy to estimate for a rocket engine: take the "energy rate" of the exhaust gases at the nozzle exit (a slight underestimate of chamber thermal power because of losses and because exhaust gases still have lots of thermal energy):

$$P = \frac{\dot{m}v_e^2}{2}$$

Using values from Wikipedia (presumably Raptor 1 approximate specs):

$$P = \frac{650 \frac{kg}{s} \cdot (3200 \frac{m}{s})^2}{2} \to P=3.3 GW$$

Reactors are often quoted by their electrical power output ($MW_e$) which can be around ~1GWe per reactor unit. The thermal power output ($MW_{th}$) can be ~3-4 times higher than the electrical output so the comparison is completely valid.

Another consideration however is the enormous scale difference between a rocket engine's combustion chamber and a reactor core (~bucket sized VS ~room sized):

CANDU Reactor core with humans for scale:

CANDU Core 1 CANDU Core 2

(Images from CANTEACH)

Raptor 1 Raptor 2

(From here and here , respectively)

$\endgroup$
6
  • 4
    $\begingroup$ Though it’s a mite unfair to compare just the combustion chamber section to the reaction chamber, which also has the fuel rods and all that. The rocket engine + fuel is just a tad bigger …. but an interesting comparison nonetheless $\endgroup$ Feb 12, 2022 at 7:57
  • 9
    $\begingroup$ @fyrepenguin also interesting to note that the rated power output of a typical nuclear reactor is a fraction of the power output it could actually operate at - nuclear reactors arent operated anywhere near their limits, while rocket engines are… $\endgroup$
    – Moo
    Feb 12, 2022 at 9:25
  • 4
    $\begingroup$ @Moo well, they can be run to their limits ... once $\endgroup$ Feb 12, 2022 at 10:57
  • $\begingroup$ It seems like a pity to not mention nuclear thermal rockets! $\endgroup$
    – ikrase
    Feb 12, 2022 at 11:35
  • 1
    $\begingroup$ @fyrepenguin Somewhat true, the CANDU uses unenriched uranium, so a better GW/m^3 is possible with enrichment, but the "fuel rods and all that" are the reason for the nuclear reaction, so excluding those from the size doesn't make sense $\endgroup$ Feb 12, 2022 at 14:00
13
$\begingroup$

On average it isn't wrong as a "close-enough" analogy.

All but one of Britain's 10 operating nuclear power stations are scheduled to close by 2023 and two of these are planned to close at the end of 2010.

Britain's existing 11 GWe nuclear fleet is planned to be replaced with modern PWR reactors, as existing AGR and Magnox stations reach the end of their 30-year operating lives.

Emphasis mine

Source https://www.imeche.org/policy-and-press/from-our-perspective/energy-theme/nuclear-power/about-nuclear-power/how-does-it-work/nuclear-power-stations

$\endgroup$
1
  • 2
    $\begingroup$ Note that to produce 1GWe takes about 3GW thermal power in a PWR or 2.5GW thermal power in a more efficient AGR. $\endgroup$ Feb 13, 2022 at 19:45
3
$\begingroup$

I can't say they aren't the same "order" but we're not quite there yet. In fact, the reactors used in nuclear power stations are considerably more powerful than 1GW.

The "headline" output is the electrical output. The common Pressurized Water Reactor technology is not very efficient.

The oldest reactor currently in operation in France has an electrical output of 910 MW, but its thermal output is 2785 MW (source https://www.world-nuclear.org/reactor/default.aspx/BUGEY-2), the newest will have a thermal output of 4300 MW.

$\endgroup$
6
  • 3
    $\begingroup$ 3 GW sounds a lot like "on the order of a gigawatt" to me $\endgroup$ Feb 12, 2022 at 18:15
  • $\begingroup$ @BrendanLuke15 - did I say it wasn't ? $\endgroup$
    – grahamj42
    Feb 12, 2022 at 18:20
  • 4
    $\begingroup$ The phrase "considerably more powerful" seems like you are arguing that the two are not comparable $\endgroup$ Feb 12, 2022 at 18:30
  • 1
    $\begingroup$ Thanks for your useful answer with helpful data from cited sources +1 I prefaced it with your comment just to alert readers you didn't "say it wasn't" Welcome to Space! :-) $\endgroup$
    – uhoh
    Feb 13, 2022 at 0:54
  • 1
    $\begingroup$ @uhoh - thanks for your edit. I was replying more to the OP's "comparable" than the original "order of". $\endgroup$
    – grahamj42
    Feb 13, 2022 at 21:35
2
$\begingroup$

If you calculate the power from fuel flow of 140kg of methane per second with a specific energy of 50MJ/kg, you end up with 7GJ/s or 7GW of power, which is even a bit more than the thermal output of the largest nuclear nuclear reactors delivering 1.6GW electric and 3 times that thermal. So I would say it is even a bit more powerful than the largest existing nuclear power station.

$\endgroup$
2
  • 1
    $\begingroup$ Great, and Welcome to Stack Exchange! If you can add a link or cite a source supporting "the thermal output of the largest nuclear nuclear reactors delivering 1.6GW electric and 3 times that thermal." that would improve your answer. Stack Exchange works a little different than most other sites, answer posts need to support their assertions. Thanks! $\endgroup$
    – uhoh
    Apr 24, 2023 at 11:25
  • $\begingroup$ Please give details of your calculation. Mr. Musk is not given to understatement and you are claiming that he is understating the engine power by a lot. $\endgroup$ Apr 25, 2023 at 2:05
1
$\begingroup$

I worked at a nuclear plant, both as they guy in the control room that directed the reactor operators, in charge of the plant when at night when the bosses were gone, and then as a trainer for 25 years. Among my specialized topics were reactor theory and operating kinetics, thermodynamics, fluid flow and heat transfer, basic physics and electrical theory. And as a simulator instructor on the control room sim for initial training and qualified crew drills; the simulator is very realistic. Is that good enough to support my assertions? :-)

The plant I was at, somewhat older and not the biggest, had about a 950 MWe electrical output from a core thermal power of almost 3,000 MWth. So 3ish GW. Definitely in the ballpark!

Of course it ran at that power 24/7/365 with only a month offline every two years for maintenance!!! Right about at 96% capacity factor. Not just a "measly" two minutes! lol

Oh, FYI, since many of you can understand: core size was about 14' high active fuel length, about 16' diameter; 177 fuel assemblies with 208 ~1/2 inch fuel tubes each. Pressurzed Water Reactor, so at about 2155 psig operating pressure the water didn't boil, it was about 80 degrees below the boiling point. Four 6 MWe reactor coolant pumps circulating around 180 million pounds per hour flowrate.

The heat is carried to two steam generators through inside it's similar sized tubes; 70 feet long and about 15,000 of them each. Water was boiled on the outside of the tubes in a separate loop at about 1000 psig, and comes out superheated by about 70 degrees above it's boiling temperature (a true gas, no entrained moisture,) approximately 11 million pounds per hour, to the one-million-ish horsepower main turbine. Which turns the 25,000 volt, 25,000 amp generator. The main transformer turned that into about two thousand amps at 345 KiloVolts to the grid.

The turbine exhaust steam containing the rest of the heat, about 2/3 of it, goes to the condenser, because you can't pump steam back to the steam generators, you have to condense it first. The two pumps that sent the water back to the steam generators were turbine driven themselves, at about 12,000 horsepower each. The condenser has two twin parallel passes of about a million gallons per minute of cooling water through it's 28 foot, 16,000 tubes. Another separate loop. That goes out to the cooling tower through 9' diameter pipes where it "rained" out of nozzles on a distribution header about 50 feet above the ground; the 500 foot tower had a natural draft created, about 20 degrees above ambient temperature. Most of the water returned via a 11 million gallon canal to be pumped again, but evaporative losses were approximately 10,000 gallons per minute. About a swimming pool of what is basically a cloud coming out.

If anyone wants info on all this superfluous info, I can get you lots of links.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.