Nuclear Reactor Simulator | Hacker News Reader

Nuclear Reactor Simulator

dalton-nrs.manchester.ac.uk

481 pointsloopion125 comments

How bizarre! Me and a colleague ( who's still a good friend ) wrote this in like, 2012-2013! Well, more accurately we ported this from a older Delphi/Pascal program that was using raw OpenGL calls ( a legitimately very impressive bit of code for what it was written in ) into WebGL ( a relatively new piece of technology at the time )

The primary reason for porting this to a modern platform was ( as far as I understood ) so that it could be run on interactive events for schools and such. It's definitely more optimised for fun than realism. I remember a few good memories of various people who were actually nuclear engineers complaining that realistically the demand would never change that fast on any commercial nuclear power plant!

I think the idea is that this is sped up significantly.

Also, in real life, there's not a human twisting the dials for "more steam" or "more reactor", that's undoubtedly handled with PIC controllers and software. Humans are just keeping an eye on things and running checklists when the software doesn't respond properly.

Also most nuclear reactors are run at near full capacity and load variance is dealt with by other energy types. The reason is that the cost base isn't really fuel but the construction and running which is all fixed so regardless of energy pricing the reactor is just run at its peak stable performance all the time.

This is a process very few nuclear power stations do, they get started and then they stay running for enormous lengths of time stopping only for fuel replacements or other maintenance.

Yes, thus the term base load power.

However, while a 1GW nuclear power plant is only ~6,000$ an hour in fuel costs that still adds up. So, curtailing nuclear in favor of solar/wind with zero fuel costs is still a net gain assuming it’s not going to cause other issues. Further the massive ramp off of these sources is becoming a serious issue for the nuclear industry.

Solar/wind don't have fuel costs, but they do have maintenance costs, and they tend to get replaced fairly often relative to nuclear installations, and while none of the above lasts forever. Like a modern wind turbine is rated at 20 years, but most installations are operating at half that in favor of installing larger turbines instead of using existing turbines that produce less power. Gridscale solar is much in the same situation, where, the panels themselves are rated for 25-ish years, but, are replaced early in favor of something newer and more productive.

The reasons for the early replacement is simple really, land costs alot of money, and not all land is suitable for solar/wind, so, you can't expand your solar/wind farm past a certain point, so it makes sense to get into newer, better, generation devices as cost/subsidy availability allows.

Hence why you really have to look at total cost over time instead of just static operating costs, or post-subsidy costs, or just cost/MWH

What you'll find is that solar/wind are indeed pretty cheap, but, they're also only pretty available, so you need base load, and if your goal is clean power, you need clean base load, for that job there's nothing even close to nuclear.

Really what we should be doing is yanking 100% of the future oil/gas/coal subsidies and shoving them at standardized reactor/plant designs and uranium mining. Then allow that standardized design to bypass existing nuclear regulators, and spin up new regulators and regulations to fast track those standardized designs being rolled out. That's more or less what France did, and it's served them well so far.

We can probably run with that model until fusion power is a reality, or we find a way to make solar panels way more efficient (we're running at like 20% currently, but we can get to ~40% in a lab, theoretical maximum is 80%, so room left to grow there), or some other major breakthrough.

> so you need base load

First base load isn’t a benefit it’s a downside. The ideal generation has low fixed cost per kWh and flexible generation right now that’s hydro where you get a limited number of kWh per month but have a lot of flexibility when in the month you’re producing power. Natural gas turbines fill the same niche at higher cost per kWh which opens the door for “base load” generation as long as it’s cheap enough.

What’s really scaring nuclear is battery backed solar as ~5 year construction timelines + 50 year lifespan means they need to complete not just with todays low prices but solar and battery prices from 5 and 55 years from now. Battery backed solar is lower risk and similar ROI today, but projecting forward things keep getting worse for nuclear.

> Gridscale solar is much in the same situation, where, the panels themselves are rated for 25-ish years, but, are replaced early in favor of something newer and more productive.

That never really happens for grid scale installations. Land costs are seriously negligible unlike roof space which is more limited.

1 acre of solar farm in a decent location for solar generates ~500,000 kWh per year. Ex: https://en.wikipedia.org/wiki/Springbok_Solar_Farm At even extreme land costs of say 1,000$/acre per year you’re only adding 0.02c/kWh as in 2 / 100th of 1 cent per kWh.

Old solar farms are almost pure profit. There’s literally 30 year old solar panels still in operation and little reason to replace them any time soon. Sure, old panels are lower efficiency but not low enough to really matter here.

I mean, if you have have gigawatts of base load online inside of 5 years, that sounds like a good model to me. We're more than 5 years away from viable gridscale batteries for the sole reason that you have to manufacture them and currently we aren't even close to having enough capacity to do so. Current deployments of batteries are like 80% efficient, and they're not designed to run indefinitely, usually only being used for peak loads, and really never more than 4 hours at a time. If you have 12+ hours of darkness, well, you're going to need alot more batteries than you think.

Also, Springbok is in a highly ideal site in the western mojave, and it's capacity factor is only 31%, so, it's not really generating it's nameplate power very often. It's also 1400acres of land used, which isn't trivial. Granted it's mostly worthless desert land in this case, that won't be the case everywhere. Sure the up-front cost/mwh of PV is much lower than nuclear ($40/MWh vs $82/MWh), you also need 12+ hours worth of batteries, which can easily place your total costs at roughly double what nuclear is, which is why you see exactly zero installations operating like that. In your scenario with no base load but batteries, they'd have to. Batteries also piss away roughly 20% of the energy you put into them, so, you need 20% higher capacity and 20% more generation.

It seems that nuclear is mostly curtailed by regulation. That's somewhat a scale problem because every plant is bespoke, were they standardized, there would likely be less need for regulation. That drives costs and lead times down significantly. Realistically, nuclear is $6/MWh, which is very, very cheap, comparable to pure PV, less than PV with batteries, the problem is building the plant which is incredibly expensive, and, with a high degree of uncertainty as to how expensive exactly. You could very well build a plant, and then have to rebuild it because some regulator found something they didn't like that was inherent to your design (that they'd looked at for years by that point). That's a thing that's actually happened.

What you need to remember is that the basic numbers you're given for things are often wrong for the things they appear to be useful for. Most of what gets quoted are hybridized numbers that attempt to account for a variety of externalities. When you look into what those externalities are, you'll find that they're accounting for costs that are a big deal in a small set of unusual circumstances, and failing to account for things that are a big deal in a large set of common circumstances. The body politic et al has substantial influence on this, as do corporate interests. Once you filter through all of that, you more or less arrive at the conclusion I've come to. Nuclear base load of at least 50%, PV/wind for the rest (mostly PV), build 4 hours of battery capacity, and your grid's stable and cheap for the foreseeable future. Whenever fusion power gets cheap, you can probably move to that, which at this rate will probably be around the time the reactors are reaching EOL given that we out and out refuse to fund it. (15.6B for renewables, 763M for fusion most of which is slated for ICF with no plans to generate power).

> 4 hours at a time.

This is a misunderstanding on your part. A 1 GW at 31% capacity factor is producing 7.44 GWh per day having “24h” of batteries would therefore be 24GWh and thus 3 days output.

“4h” * 1GW actually represents over half of daily output from a 1GW solar farm, just what you want to provide power for 1/2 the day when the sun isn’t shining. Except even better because batteries add flexibility to better follow the demand curve.

> Batteries also piss away roughly 20% of the energy you put into them

No that 80% efficiency number assumes grid>battery>grid so AC>DC>AC conversion. However PV is DC as are batteries, so doing PV>Grid involves 1 DC>AC conversion but PV > Battery > Grid also only needs a single DC > AC conversion allowing for effective round trip efficiency in the 90-95% range depending on battery chemistry etc.

Further, if you store ~50% of the output and it costs 5% of that in losses then you need 5% more solar not 50%. More importantly nuclear only has a 70-90% capacity factor not 100%. Thus roughly 2.3 - 2.9 GW of battery backed nuclear = 1GW of nuclear but solar scales much better due to flexibility.

> Realistically, nuclear is $6/MWh

Unsubsidized nuclear is nowhere close to $6/MWh today. A 1 GW reactor produces 1GW * 24h * 365 Days * 50 years * ~70-90% capacity factor. If you’re lucky that’s ~400,000 GWh over it’s lifetime at 6$/MWh that’s only 400,000 * 1,000 * 6 = 2.4 billion dollars not enough to even cover construction costs. If you were thinking 60$ / GWh that might cover construction cost + interest assuming it’s not a disaster like but not a 500 person workforce, insurance, fuel, maintenance, decommissioning, etc.

PS: 5 years is seriously unrealistic for nuclear. The US’s most recent reactor didn’t take 7 years to build it was 7 years late and that’s for an extension to an existing nuclear power plant. Even China’s nuclear power commonly taking 7+ years. Ex: Changjiang (2008 - 2015), Fangchenggang 3 (2015-2022), Fangchenggang 4 ( 2016- not finished) etc

Also further down this thread, there's a NE talking about how modern plants basically self regulate based on load, so, the reactor itself isn't throttled manually at all, but instead is moderated by cooling water temperature in a natural equilibrium.

Do you mean PID controllers?

Programable Industrial Controllers. That can include PID, or it can be far smarter or dumber.

The idea behind them is, you have purpose built, redundant, single purpose controllers for everything, so, swapping one out is no big deal, and doesn't require any knowhow. You load the config, plug it in, done. Usually they're very simple, doing jobs like "read this sensor(s) every X miliseconds, make adjustments to this output signal according to Y math formula".

A cursory search for Programable Industrial Controller gave me no hits apart from PLCs. Those would typically contain PID controllers. Could a link be provided to a PIC as meant here?

Did you google it without quotes?

https://www.google.com/search?q=%22Programable+Industrial+Co...

In am effort to relate to their userbase big G has made their search engine just search for whatever the fuck instead of what it was asked for, sometimes some combination of quotion marks and verbatim mode will trick it into being useful.

Why are you telling him what you think the idea is when he is the one who wrote it?

Verifying that your idea is understood by the audience the way you had planned is important for almost any author.

Decades ago, I attended a nuclear engineering course at Hinkley (or perhaps Oldbury - both power stations with gas-cooled magnox reactors.) I recall being told that the reactor had a slightly positive thermal reactivity coefficient. It had an automatic control system that did not work well, so they operated it manually. There was a simulator for training, run at 10X real time, and we all got to try our hand. It was not difficult, even at the sped-up rate.

There were multiple automatic safety systems ready to stop anything bad happening if the operators somehow lost control of it.

Back in the late 70s/early 80s there was a nuclear reactor simulator for the Atari 400/800 personal computers called SCRAM. I played it on my 400 when I was in high school.

https://en.wikipedia.org/wiki/Scram_(video_game)

Thank you for doing this.

This reminds me of a game idea I have.

I want to build a game where players assemble various components into engines or systems. Building the required system is itself a bit of a puzzle, but then players must also demonstrate that they can control the system using sensors and switches, etc, despite failures of various components. Can you build a reactor? If something breaks, can you figure out what broke using the sensors you placed? Can you fix the system with the switches you placed?

My favorite thing about flight simulators was always the simulated avionics. I get to click simulated buttons and watch simulated gauges, I love it.

Injured Engine is a similar game where you have to maintain a car engine and repair any part that breaks due to wear (but don't need to put it together): https://en.wikipedia.org/wiki/Injured_Engine

When I studied "computer engineering" (a very long time ago) one of our classes involved hardware debugging on PDP-8 minicomputers (which were old even at the time).

The lecturer would use a craft knife to make a tiny cut in a PCB trace somewhere in the machine. Then we would write debugging code, use oscilloscope and multimeter etc. to isolate the failure. A blob of solder repaired it.

That was such great fun and some powerful learning. It would be great if there was an online game/logic simulator that could do something similar.

Wow that must've been really fun!

It's not really similar to your idea, but your idea unlocked a memory of playing Gizmos and Gadgets as a kid.

Edit: So I decided to go and play it on an online archive site. It's just as fun as I remember.

I mean that's kind of what some security CTFs are built this way. They build a working system/application with some flaws, and you find them and break them. Sometimes in the more advanced CTFs like NetWars, you have to also fix the flaws and defend against your competitors.

The very first nuclear plant simulator I ever played with was a primitive text-based program, running on a DEC 11/780. To win the game you had to average a certain number of MWh over each turn (days). It seemed easy, just turn it on and let it run, but there was a catch. The game simulated normal fatigue, meaning that the longer it ran, the more likely something would go wrong. Also, the higher power you ran, the faster things would fatigue. The game required you to shut the reactor down for maintenance periodically. The longer you ran it, the longer the maintenance cycle. Success involved balancing uptime and power generation with required maintenance downtime.

As an aside, I learned about how nuclear reactors generate power when I was pretty young, so I was surprised to learn, just this past year, that there were people who didn't know that nuclear reactors just heat up water to make steam, same as any other power plant. There's people out there who (quite reasonably, IMNSHO) have a model of nuclear reactors directly generating electricity from the reaction.

Might it have been Oakflat Nuclear Power Plant Simulator[1] you're thinking of? I'm not aware of a version that ran on the DEC but it might very well have existed and it sounds quite similar. I can certainly credit it for developing my unhealthy fascination with nuclear power. Also perhaps my unhealthy tendency to push buttons and see what happens!

[1] https://www.mobygames.com/game/62490/the-oakflat-nuclear-pow...

Nope, much more primitive than that. I played it on paper terminal in the late 70s/early 80s.

I never realized this was the one I had on both C64 and DOS! When I played C64 I was too young (4-5yo) to actually play it. And the DOS version came on one of those disks comprised of half freeware games, half bootleg games. (Back when you could still buy those at legit retailers). Anyway that's germane because I seem to remember that once the reactor powered up it would immediately meltdown. I realized at the time - I think this was the developer's copy protection. Which struck me as much more entertaining and clever than just not working if it was a bootleg copy.

I didn't understand how nuclear reactors work until I was in grad school and even then I was aghast. You'd think if we could split an atom, we'd have a better way to make electricity than to make steam to turn a turbine. Of course, there are RTGs, and https://en.wikipedia.org/wiki/Betavoltaic_device

It's an incentives problem. Nearly all of the world's power is generated by spinning a turbine with steam, so, we have a direct, actionable, incentive to make that process as efficient as possible, and we have indeed done this and continue to do this. Modern steam turbines are wildly efficient.

That doesn't mean that they are the only thing that can be efficient, just that they're the current best option, and, any other option is going to have to show immense promise in order to get the funding to catch up to where we are now with steam.

The only thing on my radar that bypasses steam entirely is what Helion Energy is doing up in washington with their experimental fusion devices. The basic idea being that you're using a pulsed fusion reaction and intentionally not containing it, instead using the energy produced to push back on the magnetic containment and generate power. No idea if it'll work out to be viable, but it at least makes sense on paper.

You're correct. The design of nuclear reactors was intended from the beginning to replace steam generation, with the goal of not needed to redesign the entire generating plant and take advantage of the known tech as much as was reasonable. There were ideas around for just replacing existing coal/gas/whatever with reactors, to get the most value out of the capital expended on the plant. There are still studies around retrofitting retired coal generation with nuclear reactors.

Just like steam tractors were designed to replace horses.

Which is a fascinating factoid really, because the farm tooling was designed to run at a certain speed (maybe not intentionally, but, over time, that's the design criteria regardless), so the tractors were designed to run about as fast as a horse so all existing tooling could be retained.

It wasn't until much later that tooling was redesigned to work faster.

Former US Navy submarine nuclear reactor operator here.

Adjusting the steam output was kind of strange. On a submarine, the steam used to propel the submarine dwarfs all the other steam loads. As a result, there's a throttleman who controls that.

Even though this simulation is simplified, it's not too bad. It does hide some of the really interesting aspects of a water cooled/moderated nuclear reactor. The most interesting thing is that water makes the reactor self-regulating because of its negative temperature coefficient of reactivity. I'll explain.

When a uranium-235 atom absorbs a stray neutron, it becomes unstable and splits. This releases more neutrons. Very few of these neutrons will be absorbed by surrounding uranium-235 atoms. This is a good thing. Most will escape the fuel, and some will bounce around in the surrounding water. This slows the neutrons down, and some of them will bounce back into the fuel to be absorbed for more fission reactions.

Let's say 1,000 fission reactions occur. If the result is that 800 neutrons from those fission reactions are absorbed by other uranium-235 atoms, you'll have 800 more fission reactions. The reactor is sub-critical as the reaction will not be self-sustaining.

If 1,000 fissions cause 1,200 neutrons to be absorbed and react, you'll have 1,200 resulting fission reactions. The reactor is super-critical as the number of fissions will increase.

If 1,000 fissions occur and the result is that 1,000 neutrons are absorbed and cause 1,000 more fission reactions, the reactor is critical. "The reactor is critical" means the number of fission reactions is self-sustaining and neither increasing nor decreasing.

How can we affect how many neutrons bounce back into the fuel? We can change the density of the water. It makes sense if you thing about it. The denser the water, the more likely neutrons will hit a water molecule and head back into the fuel.

How can we change the density of the water? We change the temperature of the water. If the water is colder, it is denser and the more likely neutrons will bounce back into the fuel.

How do we change the temperature of the water? We pull more/less heat of out it by using more/less steam.

Putting this all together, as steam demand goes up, more heat is pulled out of the water. This causes colder water to enter the reactor. Colder water will reflect more neutrons. More neutrons means more fission. More fission means more heat. More heat means warmer water and this will attenuate the increase in fission until an equilibrium is reached.

If you're creating too much power, the coolant temperature will increase and the power output will lower. If you're creating too little power, the coolant temperature will decrease and the power output will rise. That's why water is a great coolant/moderator: its negative temperature coefficient of reactivity.

Thank you so much for sharing that! Beautiful / elegant and simple!

It's been decades since I looked at any of the details involved in any of the various types of reactors that have been designed. When I did, in the past, I hadn't even encountered concepts like "control theory" or spent any time with the subject matter of "systems engineering" or even "chemical engineering". I.e., areas where you start thinking about how to combine all of the different simple "laws"* and properties and such of energy and matter to create "robust" (ideally) or even just practical "systems".

Although I had read about the Chernobyl disaster, and "run-away" that occurred - the massive volumes of water being pumped in, partly as a result of such levels, at near boiling ... the steam voids, etc. I'm not entirely sure whether I really encountered the point about temperature and density, but, certainly, it didn't 'click' quite the way it did now when I read your description.

I love this kind of stuff - the "how it all fits together" from what can otherwise be these seemingly dry / 'dead' "laws" and such that can seem too simple / narrow / etc. to do much of use with - even if your teachers spend as much time as possible giving you homework questions etc. that certainly seem practice-oriented - but who gives a rat's-keister about whether comparing the weight of a duck to a putative witch might establish flammability and hence witchcraft when they're 15, right? ;)

* Simplified models describing various types of matter and physical processes - models that are valid (for some definition of ... as the mathematicians &/ Humpty-Dumpty [Alice in Wonderland / Lewis Carroll] might say) given certain assumptions / pre-conditions (on scale, frame of reference, etc.)

The Chernobyl disaster was partly because the design was graphite moderated, which does not have the safety that water does since it’s not self regulating due to the GPs explanation about that above. When the reactor started to go supercritical, it was reinforced by the moderator working better to create more neutrons, the opposite of what you’d want.

Excellent point - you are 100% correct, parent comment etc. were specifically about water-moderated types.

I was too grabbed by some of the later description and just connected it somewhat haphazardly to not very organized or accurate info rattling around in my head from years ago.

Thanks for pointing that out!

Additionally, they were ordered to disable all safety mechanisms and run the reactor in a known unsafe condition, causing a massive long-term disaster in... the Ukraine.

The way I attempt to explain the difference between the negative and positive coefficient of reactivity is it's like one car accelerates by pressing the gas pedal and the other car has an engine running WFO and all you ever do is press the brake pedal. It isn't a perfect analogy, but I think it gets the general concept across.

As an aside, "the Ukraine" has mildly offensive connotations. https://en.wikipedia.org/wiki/Name_of_Ukraine#English_defini...

It's a tad more complicated. Light water is both a good neutron moderator _and_ a good neutron absorber.

If you vaporize the water, thus reducing its density, it reduces both the neutron absorption, and it reduces the moderation efficiency. But crucially, the moderating efficiency matters much more in regular reactors, so the overall reactor power will drop.

In a graphite-moderated reactor, water's moderating efficiency might not matter much. So if you vaporize the water, there's going to be less neutron absorption, but there's still going to be plenty of graphite moderator to help neutrons to slow down. So the reactor power will _increase_ unless compensated by other means, and this can result in a self-reinforcing loop (see: Chernobyl).

BTW, the neutron absorption is the reason it's very hard (though not impossible) to make water-cooled breeder reactors that produce more nuclear fuel than they consume.

After the Chernobyl disaster, several remaining RBMK reactors were made safer by enlarging the cooling channels. This increased the amount of water present in the core, thus increasing the dependency on water's moderating effect, greatly reducing the positive void coefficient. It couldn't be completely eliminated, but it was reduced to a level where it can't result in prompt criticality anymore.

> supercritical

All reactors are technically supercritical. Chernobyl reactor became _prompt_ _critical_.

Normally, a small amount (just around 0.2%) of fission neutrons are emitted within a 1-3 seconds after a fission event. They are called "delayed neutrons", and this small percent of delayed neutrons is what pushes a reactor over the criticality threshold.

Since these neutrons are delayed, it gives enough time for control systems and natural feedback mechanisms to keep the reaction rate steady.

If you push your reactor past the delayed neutrons so that there are enough of prompt neutrons to sustain the criticality, you're screwed. The reaction rate can double within microseconds, far too fast for anything macroscopic to react. So within less than a millisecond your reactor can overheat, until the nuclear fuel becomes too hot to fission because its atoms move too fast (usually somewhere around 1000C).

And then it'll be followed by some extreme thermodynamics and chemistry: steam explosion, water-zirconium reaction, graphite moderator fire, etc.

Here's a research reactor that does prompt criticality excursions in controlled conditions: https://www.youtube.com/watch?v=pa0Fmcv83nw (with a countdown!)

> If the water is colder, it is denser and the more likely neutrons will bounce back into the fuel.

My memory is that the denser water thermalizes the neutrons in a shorter period of time and this is why reactivity is increased.

It's kind of the same idea, right? The more stuff there is to bounce off of, the faster it will slow down, the smaller the net distance it will travel, the increased chance it will thermalize, and less likely it is to escape. I could be missing a lot of nuance there as it's been almost 20 years since I went to naval nuclear power school. I'm definitely not the one to ask about the specifics!

I know land vs sea is different, but after decades of nuclear submarines working beautifully it's just so sad to me we don't have abundant SMRs by now.

Agree... although I do believe part of the problem is that a lot of US Naval reactors run on weapons grade uranium. Someone here probably knows more about this.

The fuel in US Naval nuclear reactors is enriched to a much higher percentage than civilian reactors due to size and longevity considerations. It has to fit the ship and refuels take months/years. A ship undergoing a refuel isn't a ship you can use.

In a civilian plant, you can have multiple reactors and refuel them on a rotating schedule to avoid downtime, having a larger reactor vessel isn't a problem, and all of that is also going to be less expensive - which is a huge factor.

I think the difference i >90% for military use and 3-5% for civilian use.

Some new SMRs are planning on using >5%.

With that grade fuel, can a nuclear submarine denote like a bomb?!

I am of the understanding that part of this is because there are different profit margins in mind with a civilian reactor generating power to be sold and a military reactor powering a vessel.

When that is combined with deregulation (or an anti-regulatory mindset) where things like insulation on water intake is deferred or ignored because it impacts the economics of the power plant, then building one becomes difficult.

Large nuclear reactors have better economics (per energy produced) than small reactors due to economy of scale benefits.

But SMRs can still be useful for small towns, remote communities, district heating, process heat for nearby industry and so on.

Dope explanation!

I have a question about the control rods: are they normally removed completely when you want the reactor to run? If they are partially inserted, that would seem to mean that the reactor fuel would burn unevenly, with the pellets at the bottom used up sooner than the top. Is that true, and is it a problem that has mitigations?

The rods are not completely removed. Control rods ravenously gobble up free neutrons. As they're pulled up, more neutrons get to the uranium. You are correct in that fuel at the bottom is used up sooner. As more fuel is used, the rods will have to be pulled up higher than they were before for the same effect. The design takes this into account.

Regardless of control rod use, there is non-uniform burnup. Fuel manufacturers use different enrichment levels in fuel pellets throughout the length of the rod to partially compensate for the non-uniformity.

The majority of PWR fuel assemblies have similar axial-burnup shapes – relatively flat in the axial mid-section (with peak burnup from 1.1 to 1.2 times the assembly average burnup) and significantly under-burned fuel at the ends (with burnup of 50 to 60% of the assembly average). Figure 1 shows a representative PWR axial burnup distribution. As is typical, the burnup is slightly higher at the bottom of the assembly than at the top. This variation is due to a difference in the moderator density. The cooler (higher density) water at the assembly inlet results in higher reactivity (which subsequently results in higher burnup) than the warmer moderator at the assembly outlet.

Quoted from "ORNL/TM-1999/246: Review of Axial Burnup Distribution Considerations for Burnup Credit Calculations"

https://www.osti.gov/servlets/purl/763169

Former submarine nuke with a masters in NucE here (it's fun to see us come out of the woodwork for this).

Rods are always in the core. To start a reactor that is shut down (with the rods are all the way on the bottom), you withdraw them slowly until the reactor is self-sustaining. From there, you increase power by increasing steam demand (as described in the parent comment above) and continue raising rods to increase or maintain temperature.

When the reactor is operating at power, the control rods are used primarily to 1) control steady state coolant temperature and 2) provide a safe and reliable way to shut the reactor down quickly (by dropping them to the bottom of the core -- this is called a reactor scram). If you have a short-duration power transient for any reason, you can "shim" the rods in to prevent a power spike that might cause a protective action to occur (you shouldn't really ever have to do this except for during emergency drills).

If the rods were drawn outside of the fuel region at power, they wouldn't be able to absorb any neutrons and wouldn't give you any way to control temperature or power. During some specific maintenance when the reactor is shut down, you sometimes might pull one rod further out for testing.

Your question on uneven burning of fuel is insightful. That can happen, and it's caused by an uneven neutron flux (# of neutrons traveling through a unit surface area per unit time) distribution. The core designers take rod positioning into account when determining how to distribute fuel throughout the core in order to maintain a "flat" flux profile.

thank you so much for the response. one more q if you dont mind: on the "how its made" show they show how the fuel comes from ore, to yellow cake, to pellets in zircon rods, to collections of rods in an assembly.

This is completely safe (compared to spent fuel), but how do you get the reaction started? do you have to "light" it with a neutron source when you're ready to use the fuel for the first time? or do you "light" it with radioactivity from existing fuel? or a neutron reflector?

In How-it's-made they didn't say anything like "the fuel assemblies are shipped to power plants with graphite moderators to prevent unwanted reactions during transit", so obviously there's no danger of an unwanted reaction outside of a reactor. So what kicks it off?

Criticality is a result of geometry. If you modify the geometry (by placing the rods in a reactor, by removing control rods, etc), you can vary the system from subcritical, to critical, to super-critical. No external neutron source necessary.

Fresh fuel pellets are “safe” in that they’re not going to kill you immediately, but they’re still fairly radioactive, not just from alpha decay, but from spontaneous fissioning, which produces neutrons. Pile em up and they’ll start a chain reaction all on their own. There’s even geological evidence of natural chain reactions in some uranium ore seams: https://en.wikipedia.org/wiki/Natural_nuclear_fission_reacto...

Sorry for the slow reply -- didn't realize there weren't notifications on HN.

U-235 (the fuel used in naval reactors) does undergo spontaneous fission, but not at a rate high enough so reach criticality. Like one of the other posters mentioned, you can make it easier to achieve and maintain criticality by changing the shape of the core (so that fewer neutrons leak out) but in general you do need a neutron source inside the core that is just always spitting out enough neutrons to help the reactor achieve criticality as the control rods are withdrawn.

Once the core has operated at power for long enough, some core materials become "active" (from irradiation) and may help contribute to the neutron source.

Straight U-235 is fairly safe (iirc), but even then I don't think you'd want to ship the fuel assemblies with any moderator as moderated neutrons are what make fission more likely.

You don't pull all of the control rods at the same time. You can fully withdraw some number of them, then control the reaction with a few more rods.

There are several strategies to provide additional control so it's not all at the bottom. For instance, in a PWR, reactivity decreases with temperature, so additional coolant can be injected where additional reactivity is needed. In the RBMK, a small number of rods were inserted from the bottom to provide more axial control.

After a civilian nuclear reactor is refueled mtc is positive. It makes for a fun few weeks.

I must seem completely seem out of place nowadays, but I actually learned that in school in Germany. Not to the nitty gritty detail, of course, but we spent a double lesson on every reactor type. What I've taken from it and remember to this day is that you usually can determine the reactor type from the shape of the building.

Here's the BBC's revision guide for 16 year olds in England and Wales. Only one reactor type:

https://www.bbc.co.uk/bitesize/guides/zyqnrwx/revision/2

Funny that it specifies a graphite moderator. Almost every country uses water as the moderator except Britain, which uses graphite moderated gas cooled reactors.

I was on AQA and I learned about water as the moderator. Could this be an Edexcel specific thing?

This is cool and takes me back to one of the first cool sim games I ever played as a child. Muse software's Three Mile Island for the Apple II. Mastery of this sim was an amazing feeling at the time.

Manual https://archive.org/details/ThreeMileIslandAppleIIManual

Game (Note view 7 save/reset state seems to be broken - avoid)

https://archive.org/details/a2_Three_Mile_Island_Special_Ver...

Was that the one where you had to perform maintenance on the heat exchanger and coolant loops to prevent fouling?

Yes, planning maintenance out into the future was a huge part of mastering the sim.

What an awesome game that was - even more awesome to know there are people that remember it. :-D

Here’s a link to a video of a ZX81 version, in a similar vein.

https://youtu.be/tB6CC8UbJLU

12 year old me went in with a single purpose: make it go bang.

Damn safety systems are messing up my fun.

Same

We need an RBMK simulator with a A3-5 button front and center.

And dosimeters with a maximum reading that equates to "Not Great. Not Terrible"

Was the first thing I tried to do, was so excited for a meltdown.

My favorite classic NPP simulator is SIMULA-C by Ralph Reuhl. I'm only finding this report [1] (from my comment history, hah), but both the C source code as well as MS-DOS executables were available. Back when I knew what the URL is, the Wayback Machine could fetch the .exe files but not the source archive.

Anyone remember this one? A screenshot can be found in the PDF at the end.

[1] https://inis.iaea.org/search/search.aspx?orig_q=RN:29043408

This bring back memories?

https://web.archive.org/web/20020928111017/http://www.simula...

Yep! The German version contains a tutorial and a manual, that's why it's larger; unfortunately the Wayback Machine did not go back in time (since I last saw this page) to capture the source code from 2002. :)

I emailed the author to see if I could get the source code to update it for a modern platform. Given the amount of link rot since that time it is unlikely that he replies but hope is eternal.

It's neat. What exactly am I controlling with the steam generator lever, the pump at the bottom? With the control rods it's more obvious since they move on the diagram.

You're controlling the flow of water through the steam generator.

https://store.steampowered.com/app/1428420/Nucleares/

Another nuclear plant simulator, haven't played too much but diving into later today with a friend.

Wow, they got their own steam-themed domain name and everything!

:-) there's also https://store.steampowered.com/app/2381620/Steam_Engine_Simu...

This one is super fun.

If I start the primary coolant pumps and "increase steam output", I get nontrivial output power for a while until a scram happens. This indicates to me that something is very off with the underlying model: there should be no power generated in the reactor in that situation.

And how is it these reactors are operating with half of the bioshield missing within a foot of my face and yet I feel no ill effects of acute radiation sickness?

The only destined purpose of a nuclear reactor simulator is to gradually let a massive meltdown happen, with gauges slowly increasing in tandem with thrill in anticipation of the upcoming drama and fireworks until you discover a bit too late that... it's too late.

The reactor keeps tripping for me before I'm out of the first "setup".

I got Pi as a score (314) on my first run-through. I've played simulators like this in the past, this is the first one that required interacting between multiple parameters interactively like this, well done!

Reminds me of this android gqme: https://play.google.com/store/apps/details?id=ru.DmitryLomak...

Just played this game and honestly I'm impressed how it's made. Interesting to see how it works even if it's simplified.

I tried, but I see only a loading page, it never starts.

I was hoping I'd have to source the fuel, too, but they didn't get into that.

https://www.satisfactorygame.com/

This reminds me that I have a disc somewhere of a ton of different Java applets, one of which was a basic nuclear reactor simulator. There was a dining philosophers one too.

The Music is awesome.

reminds me of

https://www.youtube.com/watch?v=rEgrn7vopZU

Hah, next step: Add a simple PLC the players can program to try to control it (or modbus to hook up an actual PLC) ;-)

This made me feel like a regular Homer Simpson.

Yeees, bring on the simulation tools.

why am i not able to cause a meltdown

My first thought on the Title "Play and learn how a nuclear plant works." this is was lead to this:

https://en.wikipedia.org/wiki/Windscale_fire

Seems impossible to make it meltdown and explode. Shutting off the coolant just makes it automatically SCRAM. No incoming tsunamis threaten to swamp your diesel generators, no xenon pits to slowly climb out of. Booooring. ;-)

My biggest beef with nuclear (and I've always said this) is its just so unsexy. No mushroom shapes, no passionate explosions followed by restful oblivion at the end. Its just reliable and faithful and keeps on pluggin' away. I like to live a little dangerously, and you should too!

I'm not so sure. There was a news article today about a leaky storage site for nuclear waste, coincidentally not too far from Manchester. [0] There are more ways than explosions to cause ecological devastation.

[0] https://www.theguardian.com/business/2023/dec/05/sellafield-...

Nuclear isn't just reliable and faithful and just keeps on plugging away- the modern reactor designs require extraordinary engineering to be reliable and require continuous maintanence and monitoring.

That too, I don't normally indulge in high-maintenance ;)

The foreplay also takes like 10-15 years, c'mon?!

There are some user-submitted levels in The Powder Toy that cover a subset of your requirements: https://powdertoy.co.uk/Browse/View.html?ID=5170

They should have a "Chernobyl mode" where it CAN happen. Add graphite tips to the control rods (or whatever else was the problem)

I am one of the programmers that worked on this port back in 2013, I can assure you that we (the programmers) also wanted to make a scenario where the plant would explode, however given it was designed to promote nuclear technology to school children (generally speaking), the people funding the project were not so keen to have nuclear technology shown off in that way!

> however given it was designed to promote nuclear technology to school children

... and that was not a problem for you ethically?

What ethical problem would you have here?

Yeah at least let us put graphite tips on the control rods.

No time to check this out now, but I'll be interested in reactions from my fellow (ex) Navy nukes ....

I gotta be both the reactor operator and the throttleman?!?

I have a friend who was a 'Navy nuke' but was discharged for wiring up the superbowl for the crew while on duty. You don't see THAT in the simulators!

Reply on news.ycombinator.com