Who controls reactor power?
I was standing in front of Queen La Chiefa for my Engine Room Lower Level qualification interview when he asked me that question. This was before pulling into port during the last underway, and this was big time for me. I was being grilled for my first “steaming watch,” meaning the reactor is critical while standing that watch. I didn’t even qualify for a steaming watch on Submarine NR-1. Nope, they only let me qualify as Shutdown Roving Watch.
Yet, the Engine Room Lower Level watch, even though a steaming watch, has very little to do with the reactor. Still, Chief Queen asked me a solid question in this. I knew exactly why he asked it despite this being a lowly Lower Level interview. You see, if you can answer that question—and you can explain the why’s and how’s behind it—then you have quite the strong grasp on how the submarine’s nuclear propulsion system works, including the role played in Lower Level.
Now if you, the reader, can actually already answer the question of who controls reactor power, then you also already know what a truly remarkable engineering feat this reactor plant was. Remarkable? Now that’s seriously one a hell of an understatement! There’s something very beautiful in the way it is all balanced. It’s downright elegant the way it all works together. Maybe you know this. You can skip this chapter then. If you don’t, or just want to go along for the ride since you bought the ticket, then read on!
Back in Naval Nuclear Power School, the second and hardest of the Peepayleenay schools, there was a statement summarizing the remarkable, beautiful, elegant engineering found in our naval nuclear propulsion units. They beat us over the head with the statement. We had to understand it. We had to think it automatically. Automatically in the sense that it was always in back of our mind as if this was the part of our brain which controlled breathing. Queen La Chiefa was making sure it was still there.
Yes, those highly intelligent Naval Reactor engineers condensed a fundamental element of reactor behavior into one serious five-word statement for us to memorize. Then the senior enlisted and officers like Chief Queen followed that serious engineering statement to its logical conclusion in order to turn it into a question to see if we understood it in a practical application. And finally, of course we junior enlisted swabbies took the correct answer to that serious question and used it to relentlessly bust the balls of any uptight, diggity twidget we encountered.
I’m sure those serious Naval Reactors engineers would be pleased to know that we twisted their important lesson for the purposes of skylarking. So, what was their reactor behavior summary statement? No, not yet. Before revealing that, let’s first work on answering the question Queen La Chiefa asked:
Who controls reactor power?
Well, it wasn’t me. Way down in Engine Room Lower Level? No, it certainly wasn’t me. I would control the systems that ultimately cooled the reactor, but like I said, this watch station I was attempting to qualify for had the least to do with anything nuclear. Standing Lower Level watch, I was basically the coner of the nuke world. Yeah, you gotta go climb up a few ladders to find the guy who controls reactor power. (I meant that literally. You literally need to climb ladders to Upper Level to find the guy.)
In earlier chapters, I mentioned that Maneuvering is the control center of the Engine Room, so you might possibly realize there’s a good chance you’ll find the guy who controls reactor power in there. I never did stand a watch in Maneuvering. Nor would I want to. You see, as a Machinist’s Mate, my watch stations were the largest areas on the boat presided by a single watch-stander. I had the room to roam.
To me, space was life in a submarine. I’d say it was key to maintaining my sanity. But those poor bastards in Maneuvering? After they got out of their coffin racks, took a shower in a little phone booth, and ate with a bunch of guys squeezed into a small table section, they then had to sit at cramped little control panels shoulder to shoulder with people next to them for their entire six-hour watch. Yucky!
Inside Maneuvering was a mixture of three enlisted swabbies and a junior officer. On the port side was a junior Electrician’s Mate like Hash Brown manning the Throttleman station at the Steam Plant Control Panel. On the starboard side, there is a senior Electrician’s Mate like Seaman Sample manning the Electrical Operator station at the Electric Plant Control Panel. In the center is a senior Electronics Technician like superhuman strength Christie manning the Reactor Operator station at the Reactor Plant Control Panel. Behind them sat the Engineering Officer of the Watch at, uh… the chair? I don’t know what it’s called, but it’s just a chair up on a platform. Yeah, I guess he just manned the chair.
So, four guys stuck in a little tiny box for six hours? It’s a damn certainty at least one of them would have gas. No thank you. I’ll be down in Lower Level if you need me. But back on point: we know that the answer to the question is one of those three enlisted swabbies inside Maneuvering. (We don’t let officers touch the controls; they aren’t good with their hands.) But which swabbie was it? The Throttleman, the Electrical Operator or the Reactor Operator?
Logic would dictate that the Reactor Operator controls reactor power. And yes, that is initially true. He is the guy who raised the control rods to the Estimated Critical Position (plus eventually a little more), which in turn raised reactor power from the source range through the intermediate range and finally into the power range. At least if his math was correct.
Note that if his math was incorrect and on the overly conservative side, to reach criticality safely, the method of handling rods was bestowed with a name not too dissimilar to a submariner’s love life while submerged. It’s officially called the “pull and wait” method. I kid you not. It’s called pull and wait, as I think it was somewhere around pulling the rods for three seconds and then waiting for signs of criticality for 57 seconds before pulling them out for three seconds again. And repeat as needed.
So, I meant once we reach criticality and make our way up into the power range, who controls reactor power? Yeah, you know this is a trick question. You know that someone is getting their balls busted over it, so clearly the answer isn’t the Reactor Operator. No, it’s not that senior-most watch station of the Electronics Technicians sitting there at the Reactor Plant Control Panel. Nope. It isn’t him. In your heart, you already know the answer is going to be the junior-most guy in Maneuvering.
That’s right. It is the lowly position of Throttleman at the Steam Plant Control Panel who controls reactor power. He just takes the reactor and its operator along for the ride. Short of drastic, deranged control rod movement, the only thing that the Reactor Operator does by raising and lowering his rods when we’re in the power range is adjust the reactor coolant temperature. So yeah, you’ll often hear some variation that the lowdown dirty Throttleman is the real Reactor Operator. Because it’s true!
So, we answered the question. Hash Brown controls reactor power! No I mean, the Throttleman controls reactor power. (There were two other guys hot-racking with Hash Brown who filled in when he was off watch.) Now then, what was that statement summarizing the remarkable, beautiful, elegant engineering which was drilled into our heads at Naval Nuclear Power School and made that torture of the uptight, diggity twidget Reactor Operator possible? I’m not sure if you’re quite ready for it just yet, but what the hell. Here it is:
Reactor power follows steam demand.
That was a letdown, wasn’t it?
Where the fuck is the beauty and elegance!?! I was promised beauty and elegance!
Well, it might seem underwhelming to you now, but this statement—reactor power follows steam demand—is a summary of the most amazing work of engineering I’ve ever encountered. Once you understand what that statement means, I would hope you too can see the elegance and simplicity in the engineers’ solution to controlling the beastly power of atomic energy. Their solution was a masterful, yet delicate and beautiful balance.
Note: There’s going to be a bit of reading to get to the point where you can make the determination of whether or not this is an amazing work of engineering. Brace yourself. This chapter is where we get into the reactor plant design portion of the book. So, if you didn’t like that nuclear physics chapter “Dubious Flirtation,” you might want to skip to the part with the motorcycles and strippers in the next chapter. Or just throw the entire book into a fire pit. (Hopefully you’re not reading an electronic copy if you are going to do that.)
Okay, here we go: Back in chapter four, I mentioned that one mole of Uranium-235, which is the size of a stack of quarters worth $3.75, contains the equivalent energy required to drive a car with reasonable gas mileage over four million miles. That’s true, yet completely misleading. For one, that small amount of uranium is insufficient to sustain a nuclear chain reaction. The minimum amount of fissile material required to sustain this chain reaction is called critical mass. Below critical mass, too many neutrons are able escape the mass of fuel prior to being absorbed by it.
Second, we do not want the nuclear fuel to explode like an atomic bomb. Rather, we want a gradual, controllable release of energy. For this to happen, criticality needs to be based on a mixture of both prompt neutrons and delayed neutrons. (Remember the term prompt-critical? It’s bad unless you intended to build a bomb.) Prompt neutrons are those emitted instantaneously by the fission event.
Delayed neutrons arrive fashionably late, at least 0.0001 nanoseconds after said fission event due to decay of the fission products, but more practically as many as several seconds later. In order for us to be able to control the reactor, we want the prompt neutrons to take us right to the edge of criticality, and then the delayed neutrons to get us over the finish line. This is mostly dictated by the geometry of the reactor core.
Third, we know that U-235 is a thermal fuel, meaning it requires low energy thermal neutrons in order to absorb them and cause fission events. Yet we also know that among the fission products of U-235 are two or three high energy fast neutrons. There is nothing in a solid mass of uranium to slow down—or moderate—the fast neutrons to the thermal energy region so that another U-235 atom could absorb these newly created neutrons. So, we need a moderator.
Wait. Fast neutrons and prompt neutrons? Which one is quicker? No, no, no. Two completely different distinctions. A fast neutron can be prompt or delayed. You see, prompt vs delayed neutrons refer to their time of appearance, while fast vs thermal neutrons refer to their energy state. Hope that makes sense. Now let’s get back to my list of reactor ingredients.
Fourth, we know that the final product of fission is heat. Let’s say we have a critical mass of nuclear fuel, proper core geometry and a neutron moderator. Absent some sort of coolant, this hunk of uranium will simply melt into a pile of deadly goo with the resultant fission products released contaminating anything they come into contact with for hundreds of years. As a bonus, not only does having a reactor coolant prevent a nuclear meltdown, but it also conveniently provides a means to transfer this heat to a medium which can produce useful work from it. (This “bonus” part is vitally important and will be covered downstream a few paragraphs.)
Fifth, even if the fuel doesn’t melt down, we don’t want the highly radioactive fission products carried away by the coolant. Who knows where they would end up? Some of them are actually a gas—like xenon—and most would eventually be released into the environment. So, we need cladding around the fuel for our first level of containment to keep the fission products inside the core even if there is a coolant leak or inadvertent discharge. This cladding needs to be corrosion resistant and transparent to neutrons.
Finally, we need a means to stop the nuclear chain reaction. We need the ability to insert non-fissile nuclei with large, microscopic cross sections for the absorption of neutrons. In short, we need a neutron thief. Steal those neutrons before the uranium can absorb them and continue the chain reaction. We do this with devices called control rods. Two notable elements that are good neutron thieves are boron and hafnium. An example of an undesirable neutron thief is one particular isotope of that pesky fission product gas xenon. (These neutron thieves, desirable or not, are formally known as poisons.)
This list of ingredients which included critical mass, reactor core geometry, a moderator, a coolant, a cladding, and some control rod riding neutron thieves were all baked together into the USS San Francisco’s S6G reactor plant, simply a masterpiece of engineering. I will now discuss the construction of her reactor plant to the extent that tactically clad men won’t slide down on ropes from black helicopters, come crashing through my windows, and escort me out in my skivvy shorts and handcuffs. (I really need to construct a writing bunker.)
Note that anything written here about naval reactors (and the secretive submarines carrying them) has been verified by me to be scattered about in the public realm either in published books or on the internet. I’m actually pretty amazed at the extent of information which is available out there on the subject. Particularly that which was written by former naval officers!
Wow you guys! Really pushing the envelope! Watch out for those helicopters though!
I suppose I should add a disclaimer that the accuracy of anything written here pilfered from those various works in the public domain can neither be confirmed nor denied. Maybe I couldn’t find what I remember the correct specs to be online, but needed them to carry the story forward… well seriously, how the hell would you know the difference unless you already knew the difference? In which case, I don’t need to tell you. (Just smile if you see a discrepancy, all you in-the-knowers.)
Now, for all you not-in-the-knowers out there, maybe I just settled on feeding some incorrect, yet public and widely accepted bull sheet info to you in order to maintain the story momentum, perhaps even put a smile on your face, and most importantly, to keep me out of jail. So, stand down black helicopter men. (Or at least wait until I have some pants on.)
Now we get to specifics. The USS San Francisco and all her earlier “Flight I” Los Angeles (688) class attack boat siblings were built around the S6G reactor plant with a D1G reactor core rated at 148 megawatts. This is double the power of the earlier Sturgeon (637) class attack boats, which had an S5W reactor plant with a S3G reactor core rated at 75 megawatts. Eventually the San Fran would be refueled with the D2W core used in later build “Flight II” and “Flight III/improved” LA boats, bumping her S6G plant to 165 megawatts. (At least that’s what it says on the internet. Is this info correct? Don’t you remember!?! I can neither confirm nor deny any of this sheet! It’s just what I found on the internet, so let’s move on…)
Anyone familiar with World War II naval aviation may recognize the US Navy’s alphanumeric reactor model numbers as it’s the same convention as the older airplane model numbers. The first letter is the type (A = aircraft carrier, C = cruiser, D = destroyer, S = submarine), the number is the generation, and the last letter is the manufacturer (B = Bechtel, C = Combustion Engineering, G = General Electric, W = Westinghouse).
The difference between the “plant” and the “core” is that the plant is everything built around the core including the reactor vessel, instrumentation, piping, pumps, steam generators, pressurizer and auxiliary systems. The core is just the core and has the same designation as the plant for which it was originally designed.
So, the USS San Francisco had General Electric’s 6th generation submarine reactor plant with GE’s 1st generation destroyer reactor core. That core already had the power output the Navy desired, so GE just built a new plant around it suitable for the Los Angeles (688) class submarine. A submarine class that was considerably faster than the preceding Sturgeon (637) class. How much faster? It was, wait… shh I think I hear a helicopter hovering nearby. We best be moving along now.
The S6G plant with either the D1G or D2W core is classified as a Pressurized Water Reactor (PWR) using highly enriched uranium as fuel, light water as a coolant and a moderator, zirconium alloy as cladding, hafnium for control rods and boron for strategic poisoning. Those are our specific ingredients!
Let’s start with the fuel. Due to the quantities of U-238 in natural uranium, it cannot be used as-is in thermal fuel reactors. The uranium intended to support a fission chain reaction needs to have a higher concentration of U-235 than the 0.72% found in natural uranium. Anything above that concentration is called enriched uranium. Anything below that concentration is called depleted uranium.
Low-enriched uranium (LEU) contains less than 20% Uranium-235. An example of an LEU is reactor grade uranium, which is in the 3-5% range. Highly enriched uranium (HEU) contains equal to or greater than 20% Uranium-235. An example of an HEU is weapons grade uranium, which is in the 80-90% range.
French submarine reactors use LEU enriched to 4%. Modern third generation Russian submarine reactors use HEU enriched to about 40%. The “Little Boy” atomic bomb dropped on Hiroshima used HEU enriched to 80%. What about the “Fat Man” atomic bomb dropped on Nagasaki? Fuck if I know. That sucker was made from Plutonium, not Uranium. So, let’s get back to Uranium enriched devices. Like the reactor on my boat.
Okay, now my reactor? Yeah, so with French reactors enriched to 4%, the Russian reactors enriched to 40% and atomic bombs enriched to 80%, well the reactor onboard the USS San Francisco was enriched to, oh say, uh… yeah 97%! Ninety-fucking-seven percent! Yeah, so my lil ole reactor was actually enriched beyond bomb grade.
[Note to helicopter men: Internet numbers. Internet numbers!]
Why in the hell would the United States Navy enrich a reactor beyond what an atomic bomb requires? Why would France, a country that gets a higher percentage of its electrical power from nuclear energy than any other country, go the opposite route with a LEU fuel? And why would the Soviets (and later the Russians) go somewhere in the middle?
Because of the skyrockets.
You see, a higher enrichment results in many desirable qualities: high power, compact core size, long core life and resistance to neutron poisons. But as enrichment increases, the cost skyrockets! It really all comes down to money. Deficit spending be damned, we’re rich! Bring on the skyrockets! So, we opted for bling. And what did we get for all this unfunded taxpayers’ skyrocketing money?
A thing of beauty.
Our rather expensive D1G reactor core could fit in the bed of a pickup truck yet powered the 6,900 ton USS San Francisco for 18 years before being replaced with a newer possibly more expensive D2W core. Eighteen years of immense power from a cube roughly the size of two refrigerators. At the time of writing, the latest plant is the S9G of the Virginia (774) class, and they are designed to last 33 years without refueling. French submarine reactors in contrast would have to be refueled more than ten times in that span.
Due to its unrefueled longevity, the S6G reactor does not have individually replaceable fuel rod assemblies filled with low-enriched uranium dioxide pellets like a commercial reactor. Instead, it has fuel plates made of 97% enriched U-235 inside a zirconium alloy cladding also containing strategically placed boron. These plates are layered together with multiple small vertical rectangular coolant channels and a large vertical H-shaped control rod channel. This tall rectangular vertically slotted assembly of fuel plates is called a fuel cell. Multiple fuel cells are then assembled into one box-shaped reactor core.
Theoretically, the most neutron-efficient reactor shape is a sphere. A sphere has the least surface area for a given volume, giving the neutrons less area to escape the mass of fuel. This shape will achieve critical mass with the least amount of fissile material. A sphere is not a practical reactor core shape, however. So… we have a box. And this box has far more U-235 than what is required for critical mass. The geometry of the core and spacing of the uranium and boron in the plates is such that delayed neutrons are required to achieve criticality.
Side note here: A relevant fact to this reactor geometry discussion—frequently emphasized in Naval Nuclear Power School—is that a nuclear reactor cannot explode like an atomic bomb. Sure, reactors can explode, but these are thermal explosions (from steam) or chemical explosions (from hydrogen), not atomic explosions. Semantics? No. Take Chernobyl for example, the biggest reactor incident the world has ever witnessed. Reactor 4 exploded on April 26, 1986, yet only one person was killed directly by the explosion. I’m not talking about dying from radiation. I know many more died from the incident. Yet only one of those poor souls was killed directly by the explosion. That is very much unlike the 150,000 people incinerated in Hiroshima by the Little Boy atomic bomb on August 6, 1945.
Also consider the fact that they actually operated reactors 1, 2 and 3 at Chernobyl for many years afterwards reactor 4 exploded. In fact, reactor 3 ran for over fourteen years after reactor 4 exploded. Yes, it’s true. The Chernobyl nuclear power plant was run all the way into December of 2000. Why is the fact that reactor 3 continued to operate significant? Because reactor plants 3 and 4 were in the same building. Do you think that if you set off a Little Boy atomic bomb in Chernobyl-4’s reactor hall that anything would be left of Chernobyl-3? Or the town of Pripyat for that matter?
Yes, reactors have more than enough fuel for critical mass, many times more than atomic bombs, just not the geometry for all of the fissile material to reach criticality at once like in those atomic bombs. So let me repeat this one more time and move on: A nuclear reactor cannot explode like an atomic bomb. (I don’t care what the ignorant television show and movie line readers say.)
Now in order for a nuclear reactor to achieve controlled criticality, we need to reduce the energy of those slippery little fast neutrons to thermal energy levels with our moderator. Fast neutrons are just too hard for U-235 to catch! The S6G is a Pressurized Water Reactor (PWR), and this pressurized water does double duty of moderating and cooling the reactor. Perhaps the best way to picture how neutrons are moderated is to think of slam-dancing. The reactor is a big old mosh pit.
All these subatomic particles just bouncing off of each other. I think you probably already know that you don’t want a moderator to gobble up neutrons, so it’s easy to understand that we want this moderator to have a low microscopic cross section for absorption reactions. What we want is a high microscopic cross section for scattering reactions.
Okay. So now pretend those frenetic slam-dancing party people you visualized earlier are actually their balls on a table. (Their pool balls on a table that is.) The cue ball is a neutron. The solid and striped pool balls are the moderator. If a high-speed cue ball hits a pool ball, a significant portion of the cue ball’s energy is imparted to that pool ball. You just thermalized the cue ball! All those solid and striped balls are good cue ball moderators. Now let’s say you cleared the table of the solid and striped pool balls and replaced them with bowling balls. You can probably see where this is going. When that cue ball strikes a bowling ball, it ricochets off with little loss of energy. The bowling ball is therefore not an effective cue ball moderator.
That little analogy is to help visualize the need for a moderator made out of particles with roughly same mass as a neutron. And what also has roughly the same mass as a neutron? Don’t say, another neutron! You can’t exactly go on down to the neutron store and pick up a bag of unbound neutrons to add to the reactor. I hope you are thinking, a proton! Because if you are, you would be correct. And if that wasn’t what you were thinking because you don’t know of any proton stores either, well I’m being slightly misleading.
You’re not exactly going to be finding many protons all by their lonesome that you can scoop up and add to the reactor. But you can find hydrogen. Hydrogen is simply a proton with an itsy-bitsy electron orbiting it. While you can find plenty of hydrogen atoms in diatomic hydrogen molecules (i.e. H2 – hydrogen gas), it is not dense enough to be an effective moderator. Plus, hydrogen gas is very dangerous to keep inside our little underwater capsule due to its explosive nature! No, a much better place to find hydrogen atoms is in good old fashion H2O – water! It’s safe, plentiful, and inexpensive. And it doesn’t absorb too, too many neutrons. So that’s why water is used as a moderator.
Now we need a coolant. The S6G, being a pressurized water reactor, uses light water. “Light” water means the hydrogen atoms in the water are H-1 (protium) isotopes. In other words, just plain old regular water. Well maybe not just plain old regular water. Tap water is too dirty to use. Reactor coolant is ultra, ultra, ultra-pure water. Ultra-pure light water.
Some reactor designs require heavy water, which means the hydrogen atoms in the water are H-2 (deuterium) isotopes. This apparently allows for the use of unenriched uranium as fuel. Heavy water reactors are common in Canada, ‘eh. Other reactor designs use gasses for coolant such as carbon dioxide or helium, while still other reactor designs use liquid metal such as sodium or lead-bismuth alloy.
Wait. Liquid metal coolant? That sounds wicked! Okay, let’s pause on the S6G for a minute to talk about those Terminator reactors. We, as in the US Navy, built one of them. Our second nuclear submarine, the USS Seawolf (SSN-575), used a liquid sodium cooled S2G reactor. It was not very successful due to poor reliability and difficulty maintaining it, so the Seawolf’s S2G reactor was eventually replaced with the S2W reactor. The S2W was our trusty dependable PWR type reactor found inside our first ever nuclear submarine, the USS Nautilus (SSN-571).
We bult one, but the Soviets constructed eight liquid metal cooled submarines in total. These subs were equipped with one of three different naval reactors that had liquefied lead-bismuth alloy as the coolant. The VT-1 reactor was used in the single experimental November class submarine K-27. (All other Novembers used the PWR type VM-A reactor.) The entire Alpha class of really, really, really fast attack submarines used the BM-40A and OK-550 liquid metal reactors.
I don’t know about you, but to me, there’s something about liquid metal cooled reactors that just seems so spooky. Like they should have their own sinister theme music. But the Soviets didn’t use them simply to be sinister. No, these liquid metal reactors are very compact, power dense creations—something great for a submarine.
But what’s not great for a submarine is the problem of keeping that metal coolant molten after the reactor shuts down. Once it solidifies, you can imagine how difficult it is to heat that metallic coolant back up to being able to flow through the piping! That plus the liquid sodium coolant in the Seawolf had properties which would make maintenance a wee bit difficult. Yeah, that particular liquid metal coolant had this tendency to explode on contact with air.
Uh, it’s your turn to vent off that reactor coolant header over there, okay Bob? Thanks a bunch.
Since those trusty dependable PWRs use non-explosive inexpensive light water for the coolant, a coolant that can be made by purifying ocean water inside the submarine, all naval reactors built now are of that type. But there is also a danger in using light water as a reactor coolant. You see, water in its liquid phase is a fairly good thermal conductor. Water in its gaseous phase—aka steam—is not a good conductor of heat at all. In fact, it’s a good insulator.
Inside the reactor vessel of a PWR, steam is the number one enemy. If you draw a steam bubble in the core of a PWR, you’re done. Pack it up and go home. Good job, asshole. You just melted down your reactor. Who raised you? Well, that is exactly what was done at Three Mile Island. TMI-2 was a Babcock & Wilcox model LLP pressurized water reactor. Emphasis on was.
Now how exactly does one draw a steam bubble in a reactor core? If I were to ask a few people at what temperature water boils, most people would probably respond “212*F” or “100*C.” And yet PWRs operate at much higher temperatures than that. I think those very same people realize that too. Hmm. Maybe my question should be, how exactly does one not draw a steam bubble in the core?
Well you see, there is a relationship between the temperature and pressure at which water boils, and it is a dynamic one. At atmospheric pressure, the temperature at which water boils—known as the saturation temperature—is indeed 212*F/100*C. But saturation temperature will be higher at higher pressures and lower at lower pressures. If you were to plot those points on a graph, this would be called a saturation curve.
Let’s look at a saturation temperature for a lower pressure. One example can be found in the USS San Francisco’s fresh water making distilling plant. We can boil the seawater at about 150*F (83*C) in that sucker. Fancy magic trick? Not really. We reduced the pressure of the system using air ejectors to create a vacuum of about 22.3 inHg (256 mbar) below atmospheric pressure. Reducing the operating pressure reduces the saturation temperature. This is advantageous as at around 165*F (92*C), biological life transforms into some pretty fucking hard boiler scale. (As if us swabbies needed yet another thing to clean.)
Now you want an example of saturation temperature for a higher pressure? Why don’t we look at the PWR? You know, the pressurized water reactor? But no, we can’t talk about the USS San Francisco’s pressurized water reactor’s operating specifications. I couldn’t find any declassified reactor coolant operating parameters on the S6G reactor plant, so let’s use figures from the most successful commercial pressurized water reactor.
With 55 of them built in the US alone, that would the Westinghouse model PWR. Yes, it’s a PWR model pressurized water reactor. I know, I know. Very original model name. Anyway, the Westinghouse model PWR reactor plant is very similar to both the General Electric S6G naval reactor plant and the Babcock & Wilcox model LLP commercial reactor plant found at Three Mile Island.
According to some US Nuclear Regulatory Commission documents I found online, the Westinghouse model PWR reactor coolant temperature is 547*F with a pressure of 2235 psig in hot standby conditions. The saturation temperature for an operating pressure of 2235 psig is 652.7*F. Therefore, the reactor coolant temperature is more than a hundred degrees below the temperature at which steam would be produced. That’s a comfortable safety margin. But we can also produce steam by reducing pressure. The saturation pressure for an operating temperature of 547*F is 1004.8 psig. Therefore, the reactor coolant pressure is greater than twelve hundred pounds per square inch above the pressure at which steam would be produced. Another comfortable safety margin.
That’s exactly how one does not draw a bubble in the core. You mind the saturation curve. Keep your temperature below the saturation temperature for a given pressure, and your pressure above the saturation pressure for a given temperature. Therefore, it stands to reason that if you wanted to draw that steam bubble in the core, there are two ways to accomplish this work of the devil: increase the temperature to saturation temperature for a given operating pressure or decrease the pressure to saturation pressure for a given operating temperature. (Basically, overheat it or vent it off.)
So, let’s get back to the Three Mile Island disaster where they killed their reactor dead like bug. While TMI reactor 2 was around full load, some sort of corrective maintenance deemed rather necessary and proper was performed. This unfortunately resulted in tripping the turbine generator set offline. That in turn caused a reactor scram. (A scram is when the control rods are rapidly and fully inserted into the core in order to stop the nuclear chain reaction.) Yet even with an immediate termination of the fission process, reactors create substantial residual heat from the decay products following a shutdown, particularly after running at high power for sustained periods.
For whatever reason, the backup cooling systems at TMI-2 were taken out of service. Regulations prohibited reactor operation with these systems out of service, but that didn’t stop them. They had to get the power to the people! However, without these functioning cooling systems, the decay heat in the reactor just continued to increase the coolant temperature unabated, which in turn increased the coolant pressure to worrisome levels. The pressure rose high enough to actuate an electrically operated pressure relief valve, a device used to prevent a coolant system rupture.
Once pressure was restored to below the setpoint, power was automatically cut off to the relief valve. Unfortunately, the valve was then mechanically stuck open. The only thing killing the power actually did was turn off the open indicating lamp on the control panel. Things looked normal to the operators, yet reactor coolant kept being dumped into the storage tank.
The reactor coolant pressure continued to drop, and it eventually reached the saturation pressure for the coolant temperature in the core. Then a steam bubble formed inside the reactor. There’s that pressurized water reactor’s number one enemy! Steam does not conduct heat well, so once the core was “uncovered” as they say in this business, it rapidly overheated and melted down. Billions of dollars right down the toilet.
How did the operators allow this? Well, since the core is always supposed to be covered, there are no reactor coolant level indicators on the reactor vessel itself to see an interface of steam and water. Instead, the coolant level is monitored in the device that maintains the reactor coolant pressure. This device is conveniently called the pressurizer, which is where the interface between the steam bubble and liquid coolant is supposed to be. Ordinarily, if the level in the pressurizer is satisfactory, the core is considered covered.
At first, the pressurizer level at TMI-2 was actually increasing after the scram due to the thermal expansion of the reactor coolant. Once the pressure relief valve opened, the coolant level slowly began to drop. Since the relief valve is at the top of the pressurizer, the coolant released there is steam. You don’t lose as much coolant when relieving pressure by discharging steam as opposed to discharging liquid coolant. So, the level didn’t decrease precipitously enough to alert the operators of the loss of reactor coolant casualty occurring.
When the pressure dropped to saturation, steam began to form at the heat source. This source of course was the reactor with its significant decay heat. Once the core bubble formed, the expanding volume of steam in the reactor displaced liquid coolant back into the pressurizer, and the indicated level began to rise again. The operators mistook this level rise and that extinguished open indicating lamp for the relief valve as proof that the relief valve did actually shut after the pressure excursion.
For the life of me, I have no idea how the operators either didn’t look at the coolant pressure indication or if they did, didn’t realize that it was at saturation pressure. This blows my mind. My training in the Navy nearly twenty years after this accident has the benefit of hindsight, but I still have trouble believing this fundamental temperature-pressure relationship wasn’t emphasized to the civilian operators before the TMI-2 meltdown. I don’t know. Perhaps I too would miss it in all the excitement of a unit trip and reactor scram.
Or perhaps not. Perhaps the naval reactor training was far superior. Think about this: Starting with the USS Nautilus, the United States Navy has operated approximately 269 enriched-beyond-bomb-grade pressurized water reactors. Not a single one of them has melted down despite the extreme conditions in which they are operated.
Has a civilian nuclear plant ever taken a 30 degree down angle while raising load from barely above self-sustaining all the way to torpedo evasion power in a matter of seconds? Or have a captain-like figure routinely wander into the engineering spaces and purposely cause a reactor scram to see how the operators respond? Nope. Not a one. Those plants are designed to make nice steady power, so to purposely trip one offline for training while supplying the grid will cause hundreds of thousands of dollars of monetary damages to the company and maybe even risk causing a blackout.
The US Navy indeed has an impressive track record of reactor safety and reliability considering the rigors their nuke plants are put through. Much of that safety record I blame not just on training, but on the selection of the pressurized water reactor design for naval use. Sure TMI-2 was a PWR, but they performed risky maintenance while online, inhibited the backup coolant systems, sustained a mechanical failure of a safety relief valve, and the operators did not recognize the signs of the loss of coolant casualty. That’s four links of a chain leading to disaster, any one of which if missing would have Three Mile Island continuing to operate in obscurity. We’d never have heard of the place.
Yes, that poor PWR was running at top load for an extended period of time, scrammed, did not have backup cooling systems in place and then suffered a loss of coolant casualty. Yet it still didn’t explode or release significant radioactive material into the environment. Think about it. The pressurized water reactor is a sound design.
Now since we have just discussed how the loss of coolant and formation of a steam bubble in the core can melt down a reactor, let’s move onto the next reactor core ingredient to see just exactly what we’re melting: the cladding. The main purpose of the cladding is to keep the nasty bits contained inside the core and not released into the coolant or environment. Uranium is a very hard, heavy metal (I love heavy metal), and it is only marginally radioactive. You can handle uranium fuel pellets or plates with your hands without the need for shielding if they have not yet gone critical.
However, once uranium fissions, the products are various highly radioactive elements, and not necessarily metal. They are nasty, nasty things. You are left with neutron rich isotopes of krypton, strontium, molybdenum, technetium, iodine, xenon, cesium, promethium, samarium and other radionuclides with mass numbers clustered around either 95 or 135. Because these isotopes have many more neutrons than their stable siblings, they emit a lot of radiation trying to slim their way down to stability. The kicker is that not only are they highly radioactive, but also a few of them like to be hoarded by your body. You say you want some examples? I will give you three.
Surely everyone knows that potassium-iodine pills are handed out after radioactive fallout has been detected. The idea is to saturate your thyroid with stable Iodine-127 so that your body doesn’t accumulate the highly unstable fission product Iodine-131. That isotope is a beta and gamma radiation emitter, and concentrated in your body, it may lead to thyroid cancer. With a half-life of eight days, I-131 is a threat for about 56 days after the fallout from the atomic bomb just you detonated or the reactor you accidentally and explosively melted down.
A lesser known nasty nuclide is Strontium-90. It is a beta radiation emitter, but it has a longer half-life measured in years. The body thinks it’s calcium, as strontium is also in the group two alkaline earth metals. It’s actually right below calcium on the periodic chart. Strontium-90 is deposited in your bones, like calcium, because your body is dumb. This may lead to leukemia. With a half-life of 29 years, Sr-90 will be a threat for about 200 years after the fallout you created.
Perhaps in the middle of meddling radionuclide recognition is Cesium-137. This is a beta and gamma emitter. Since it is in the group one alkaline metals, your still easily tricked dumb ass body thinks it’s potassium and distributes it as such to your soft tissues like muscle, liver and red blood cells. This may lead to various types of cancers. With a half-life of 30 years, Cs-137 is a threat for about 210 years after your fallout.
So, your first line of defense from these nasty creations and the cancer they are trying to cause for you is the cladding around either civilian reactor fuel rod pellets or military reactor fuel plates. The qualities in a cladding that we are looking for are low cross section for neutron absorption, high strength, high temperature resistance, high thermal conductivity and high corrosion resistance.
There are a few different materials that have been used, like stainless steel, magnesium and aluminum alloys, but most American reactors, whether civilian or military, use an alloy of zirconium for cladding. It’s an exotic and expensive material, but it meets all of those above requirements. As long as you don’t melt down your reactor, the zirconium alloy cladding will keep all of your nasty ass fission products within the plates or pellets, and none will enter the coolant. At least not appreciably. I mean, there is something called diffusion. Just nothing significant gets through.
There is, however, one tiny little noteworthy detail about zirconium you should probably know if you think that you could possibly melt down your reactor. Yeah around 1650*F (900*C), zirconium has an exothermic reaction with water which produces zirconium dioxide and hydrogen gas. This happened at Three Mile Island (either no hydrogen explosion or one little one perhaps) and Fukushima Daiichi (several rather large hydrogen explosions). So, don’t melt down your reactor! This segues us nicely into the last reactor core ingredient.
Neutron thieves! If your reactor starts misbehaving, you want to be able to shut it down instantly by interrupting the nuclear chain reaction. This is done by rapidly inserting a material with a high microscopic cross section of absorption of neutrons. Let’s compare thermal neutron capture microscopic cross sections of various nuclides. If you remember, the units are called barns. The symbol is a lowercase sigma (σ), and it is a measurement of area equal to 10-24 cm2.
Hydrogen-1/Protium (0.3σ)
Hydrogen-2/Deuterium (0.0005σ)
Hydrogen-3/Tritium (0σ)
Boron-10 (3,835σ)
Boron-11 (0.006σ)
Carbon-12 (0.004σ)
Oxygen-16 (0.0001σ)
Magnesium-24 (0.05σ)
Aluminum-27 (0.2σ)
Chromium-52 (0.8σ)
Iron-56 (2.6σ)
Cobalt-59 (37σ)
Nickel-58 (4.6σ)
Zirconium-90 (0.01σ)
Xenon-135 (2,720,000σ)
Xenon-136 (0.3σ)
Samarium-149 (42,080σ)
Samarium-150 (104σ)
Gadolinium-157 (254,000σ)
Gadolinium-158 (2.2σ)
Hafnium-176 (23σ)
Hafnium-177 (373σ)
Hafnium-178 (84σ)
Hafnium-179 (41σ)
Hafnium-180 (13σ)
Lead-208 (0.03σ)
Uranium-235 (681σ)
Uranium-238 (2.7σ)
Plutonium-239 (1017σ).
I realize that I put up a lot of numbers, but there are a lot of different materials in the reactor for a lot of different reasons. I’ll break this down into reactor fuel, reactor construction materials, reactor coolant (and moderator), and neutron absorbing materials (poisons! thieves!).
The fuel is easy. As you know, we use U-235 for the fuel, and it has a nice fat neutron cross section. Most of the absorbed neutrons lead to fission and not neutron capture. Plutonium-239 is similar, yet a bit hungrier. Notice non-fissile U-238 has little appetite for thermal neutrons.
Reactor plant materials contain a lot of iron, nickel, chromium, carbon, zirconium, lead, and cobalt. These materials have low cross sections with the notable exception of Cobalt-59. The issue with structural materials is not neutron capture, but a phenomenon called neutron embrittlement. The constant bombardment of neutrons changes the crystalline structure, hardening these materials, which then limits their service life due to the possibility of sudden catastrophic failure. That includes the reactor vessel, which in a submarine is limited to the life of two reactor cores, or roughly 35 years at a regular operating tempo.
Now regarding the significance of Cobalt-59, it’s not that it’s a significant neutron thief or that embrittlement becomes an issue. No, as a component of stainless steel alloys necessary for hardening, the major concern is that it becomes activated after capturing a neutron. Cobalt-60 is not a fission product, yet it is a nasty, nasty bit because it is a material used outside the cladding. This necessary evil is the main source of gamma radiation when the reactor is shutdown, and it is the main source of reactor coolant contamination. With a half-life of 5.3 years, it is dangerous for at least 37 years after neutron activation.
Concerning the light water coolant and moderator, or hydrogen and oxygen broken down to its constituent components, they generally do not have an appetite for neutrons. But these elements pass right through the reactor core and are subjected to intense neutron bombardment. (Yeah, that’s sort of in the job description of a moderator.) So, both elements do capture neutrons a bit despite the undesirability of that in reactor operations.
When a protium hydrogen atom in the coolant absorbs a neutron, it becomes deuterium, making heavy water. It is not radioactive. Heavy water is harmless to humans unless you drink enough of it to replace close to half of your body water. (That would be impressive. I would actually like to see your refrigerator stocked with bottled reactor coolant. I prefer sparkling, by the way.)
Deuterium has slightly different chemical properties than protium hydrogen, and this resultant heavy water once inside you tends to inhibit eukaryotic cell division. Whatever that means. (I’m not a doctor.) Probably means you’ll be shooting blanks. But for the love of god, just wear a condom and don’t drink the still or sparkling reactor coolant.
As for the oxygen, well, it’s a different and significant, yet short story. It’s a very, very short story that takes less than a minute. Oxygen-16 subjected to an intense neutron flux will absorb a fast neutron and eject a proton, becoming Nitrogen-16 through a scattering reaction. The N-16 is highly excited and radioactive, rapidly undergoing beta decay back to O-16. This releases intense beta and gamma radiation in the process. It is the primary source of our gamma radiation while the reactor is online. The good news is that N-16 has a half-life of slightly more than seven seconds, so in just under 50 seconds after activation, it’s almost completely gone. (That’s why Co-60 then becomes the main source of gammas when shutdown.)
I should probably talk about half-life misconceptions here. I see in media the notion that something with a half-life of thousands or millions of years is worse than something with a half-life of seconds or minutes. Well, the longer the half-life, the more stable it is. Radioactive material with a half-life of seconds is absolutely lethal. If you are around it while it is still decaying, you will die of acute radiation poisoning. The good news is that you only have to wait seven half-lives for it to decay away to basically nil.
In contrast, something with a half-life of millions of years isn’t reacting much and will therefore not be too radioactive. (Like bananas and rocks. They are radioactive.) Now a half-live of days, months or years? That’s the shit you have to worry about. Chronic radiation sickness and cancer causing crap right there! But it doesn’t sound too cinematic to say, “It has a half-life of eight days! Eight days!!!”
Now, let’s talk about the neutron thieves. Afterall, this last reactor ingredient topic was started due to the need to shut down our misbehaving reactor. There are a lot of materials used in the commercial world, but I’m going to cut to the chase and state that naval reactors use hafnium for nuclear chain reaction stopping control rods. That is not classified. (It’s like saying a naval reactor uses uranium for fuel. I can write that.) In my cross-section list, various isotopes of the same element are listed, particularly hafnium. No, I didn’t fall asleep on the keyboard. These were listed for a reason.
Despite having only a modest neutron cross section, the Navy uses hafnium for control rods because it has desirable mechanical properties such as high strength and corrosion resistance, but more importantly, each hafnium nucleus can absorb five neutrons before becoming saturated. (Hence the inclusion of various isotopes of it in my list of cross sections. Told you I didn’t fall asleep on the keyboard.)
Also, hafnium is kind of free. It’s a byproduct of zirconium mining. Coincidentally, those two elements are found in the same ore. Interesting that these two materials critical to naval reactor construction are found together like that, and without separation, they would be absolutely useless at their respective jobs. You can’t have a cladding that absorbs neutrons, and you can’t have control rods that are transparent to them.
Going back to the cross-section list, you can see that there are some absolute monster absorbers. Boron-10, Samarium-149, Gadolinium-157, and that pitbull-off-the-chain, Christ-in-a-chicken-basket, crazy ass gas Xenon-135. You’ll find all three in naval reactors, but only one of them is purposely added to the core. Boron-10 is a stable nuclide that is added to the fuel plates as a burnable poison. It’s a one and done neutron absorber. (Notice that Boron-11 is full.)
Strategic placement of burnable poison helps even out reactivity. Reactivity? Well, my notes from Naval Nuclear Power School were classified and thus confiscated, shredded, and burned upon graduation. Of course, if I go by memory and botch the “key words and tricky phrases” from Power School, I’m an asshole. So, let’s just use the US Nuclear Regulatory Commission definition here:
Reactivity is a term expressing the departure of a reactor system from criticality. A positive reactivity addition indicates a move toward supercriticality (power increase). A negative reactivity addition indicates a move toward subcriticality (power decrease).
Boron-10 strategically placed in the fuel plates temporarily lowers localized reactivity and allows a higher U-235 concentration to be added to that area of the fuel plates. U-235 obviously adds positive reactivity. So that fuel and poison reactivity balance ultimately extends the life of the reactor core. This Boron-10 is a desirable poison purposely added to the reactor by the manufacturer. Samarium-149 and Xenon-135, in contrast, are undesirable poisons born of fission products for a life of mischief.
Like Boron-10, Samarium-149 is stable. Therefore, the only way to get rid of this fission product is to let it do what we don’t want it to do: steal neutrons. Oh well. But hey, at least Sm-149 is a relatively rare fission product, making up about 1.1% of those products. It does affect reactor operation by adding negative reactivity, but not significantly so due to its scarcity.
The same cannot be said about Xenon-135, however. It most definitely affects reactor operation in a significant way. How much so? Let me put it this way: xenon’s behavior and the resultant human actions taken trying to compensate for it are the main contributing factors in the destruction of Chernobyl-4. Is that significant enough for you?
Yeah okay, but how does a poison—something that absorbs neutrons, something that adds negative reactivity, something that moves the reactor towards subcriticality, something that reduces reactor power—how does that something cause a supercriticality accident!?! Yeah… we’re going to have to discuss xenon production and elimination, and something called the xenon pit.
Xenon-135 itself is a rare direct fission product, even rarer than Sm-149. Xe-135 is only about 0.25% of that direct fission product yield actually. Instead, the overwhelming majority of its production is the result of beta decay of Iodine-135. If you recall, beta decay is when a neutron transforms into a proton while releasing beta and sometimes gamma radiation.
Comprising about 6.3% of the yield, Iodine-135 is the second most common fission product. But it doesn’t last too long. You see, with a half-life of 6.6 hours, I-135 is quite unstable. It’s all decayed away in about 46 hours. But that beta decay produces Xe-135, the greatest neutron absorber the world has ever known. It is the undisputed heavyweight champion of poisons.
While Xenon-135 production is based almost entirely on decay of the fission product Iodine-135, there are two ways to eliminate the xenon. Like I-135, Xe-135 is also very unstable. It has a half-life of 9.2 hours, so within 65 hours, all of it will decay away into relatively stable and mostly non-neutron absorbing Cesium-135. (Cs-135 has a half-life of 2.3 million years and a neutron capture cross section of 8.3σ.) However, as Xe-135 is the most powerful neutron absorber in existence, the only time it will decay away into Cs-135 is when the reactor is shutdown.
When the reactor is online, Xe-135 is removed by neutron capture. It then becomes Xe-136, which not only has little appetite for neutrons, but with a half-life greater than the age of the universe, it is effectively stable. (Uh… half-life greater than the age of the universe? How the hell did they measure that!?! Will there be a scientist in two sextillion years verifying its half-life mumbling, “morons” to his or herself? Off by three days! Three days!!!)
So, let’s focus on online reactor operation for the moment. This means that Xenon-135 poison production is based upon decay of its short-term fission product parent I-135, and its elimination is based upon the strength of the neutron flux burning it away. We can simplify this by stating that its production is based upon average reactor power (over a period of time factored on the half-life of I-135), and its elimination is based upon instantaneous reactor power. Basically, it’s the past verses present reactor power.
If reactor power is stable for some period of time, the neutron flux and Iodine-135 production are stable. Therefore, the xenon poison production and elimination rates reach an equilibrium. It’s when reactor power changes that things become a bit more complicated.
If reactor power is raised, the neutron flux and Iodine-135 production are also raised. Since the neutron flux burning off the Xenon-135 is raised instantaneously, yet the decay of the increased I-135 production takes time to produce more Xe-135, the poison is now burned away faster than it is being produced. This results in temporarily lower overall xenon poison concentration until the new equilibrium is reached.
If reactor power is lowered, the neutron flux and Iodine-135 production are also lowered. Since the neutron flux burning off the Xenon-135 is lowered instantaneously, yet I-135 produced at the higher power levels has yet to all decay into Xe-135, the poison is now being produced faster than it is being burned away. This results in temporarily higher overall xenon poison concentration until the new equilibrium is reached. (Note that a large neutron thievin’ Xe-135 poison concentration is called a “xenon pit.”)
Outside of steady state operation, it is up to the Reactor Operator to compensate for the negative reactivity added or removed from the reactor due to xenon poison concentration. They do this by adjusting the control rods. Like the poisons, control rods also add negative reactivity.
So, if the concentration of poison increases (adding negative reactivity), to balance the changes in reactivity, the control rods would need to be withdrawn (removing some negative reactivity). If the concentration of poison decreases (removing negative reactivity), to balance the changes in reactivity, the control rods would need to be inserted (adding more negative reactivity).
Due to their low enrichment, most civilian reactors do not have enough positive reactivity in their fuel to overcome the negative reactivity added by Xenon-135 if they operate at high power for a significant amount of time and then reduce power to a very low level. These reactors will “stall” due to the xenon pit. The reactor will shut down and the operators will need to wait as much as an entire day for enough of the Xenon-135 to decay into Cesium-135 in order to commence a startup. This dead period is called a “xenon precluded startup.”
So, let’s talk about Chernobyl-4, shall we? They were running at very high power for days, weeks, or perhaps months and needed to do a particular safety test at a very low power. Instead of being cleared to do the test when it was scheduled, the dispatchers needed at least half-power from the unit to maintain the electrical grid demands. The low power test was rescheduled for the night shift when electrical demands were low.
While at half-power, the neutron flux was also reduced to half, and the Xenon-135 burn off rate was similarly reduced to half. Yet the Xenon-135 poison production kept humming along as if they were still at full power due to the time delay of the Iodine-135 decay rate. Chernobyl-4 was entering a xenon pit just in time for their safety test.
On the night shift, they initiated the test by reducing power even further, which of course reduced the xenon burn off rate even further as well. But think about the xenon production rate! It had only been a few hours since they were screaming along at top load, making crazy amounts of Iodine-135. Therefore, the Xe-135 production rate due to the abundance of I-135 all decaying into that poison was still very, very high. The reactor stalled. It shut itself down during the test.
But the safety test! The safety test needs to be completed! We need to complete this safety test whether it’s safe to do so or not, damn it! Remove some negative reactivity! Raise the rods! Raise them all!
RBMK-1000 reactors at Chernobyl have 211 control rods. The operators began withdrawing groups of rods, but reactor power would barely budge. They kept pulling rods. Reactor power remained stubbornly low due to the xenon pit. Eventually 205 of the rods were withdrawn. Fully. Yes, 205 of the 211 control rods were fully withdrawn. That’s an incredible amount of negative reactivity to remove! After a while, reactor power began to slowly increase due to the neutron flux burning away the xenon and due to the decay of xenon into cesium. Reactor power began to increase less slowly.
The neutron flux strength was now starting to make way burning off the xenon. And then reactor power increased not slowly at all. In fact, it began to increase exponentially. With all but six control rods fully extracted, reactor power was now out of control. The spooked operators scrammed the reactor, but it was too late. Reactor power was last seen at ten times its maximum rated output, and that’s only because the instrumentation was the pegged to the top of the range. It was likely even higher.
The exponential increase in reactor power vaporized its coolant, creating a massive boiler explosion. It spewed reactor fuel and all those nasty fission products like Iodine-131, Strontium-90, Cesium-137 and others into the air to be deposited as fallout all around Europe and Canada. The world’s worst nuclear disaster had just been written into history. Hopefully you now understand how Xenon-135 poison can significantly affect reactor operation.
But what about in my S6G reactor plant? Well, if one pictures commercial reactors as big lumbering airliners, think of military reactors as nimble and highly powered fighter jets. In the S6G reactor, there is no such thing as a stall or xenon precluded start up. I mean, seriously? On a fucking warship? On the god damn tip of the spear? Fuck. That. Shit. No fucking way! Yeah, due to naval reactors’ insanely high beyond-bomb-grade U-235 enrichment, we do not stall. And we start up whenever we god damn please. So, bring on the xenon. We will fucking burn you.
Well, we will as in us group of nukes back aft will. Me individually would never control reactor power like a Throttleman or pull out the control rods from the source range to the power range like a Reactor Operator. I would never calculate the Estimated Critical Position outside of Naval Nuclear Power School. But maybe now like me, you understand some of the factors used in calculating that ECP. In an earlier chapter, I wrote this:
The ECP calculation takes into account where the reactor is in its lifespan, the power history before the last shutdown, the time since the last shutdown, the temperature of the coolant, and probably a bunch of other crap I have long since forgotten.
The lifespan takes into account how much Uranium-235 fuel and Boron-10 poison have been spent. Reactor power history takes into account the Samarium-149 poison produced and more importantly, the Iodine-135 produced. Time since last shutdown takes into account how much Iodine-135 has decayed into Xenon-135 poison, and also how much of that Xenon-135 has decayed into Cesium-135. Reactor coolant temperature takes into account the density of the moderator, as the coolant is the moderator. All of those factors affect reactivity.
We of course are able to remove negative reactivity by withdrawing the control rods. And I’m pretty sure that no matter how thoroughly you fuck up your calculations, it will never tell you to pull a Chernobyl and pull 97% of your rods, fully. Cross-referencing the ECP tables at that point probably leads you to an entry stating to go fuck yourself, fully. That ECP calculation is to determine how much negative reactivity to remove in order to balance the positive reactivity of the fuel and reach criticality safely.
I was taught all of this. I was tested on all of this. I was reminded frequently of all of this. And there I was in front of Queen La Chiefa spouting off much of that knowledge I had learned in order to qualify for Engine Room Lower Level. The watch that has the least to do with the reactor.
Seriously. Was I supposed to be thinking about all of this reactor theory while wiping up oil dripping from the turbines in Upper Level down to me in Lower Level? Should one always think about the beauty of the reactor when being directed to clean a particularly grimy area of the Engine Room? Like, do hospital janitors need to learn a little brain surgery, just in case?
All the nuke chiefs and officers kept repeating to us all that “reactor safety is our primary concern.” Only a few days before this qualification interview, Queen La Chiefa had reminded me of this primary concern for the umpteenth time during a particular four hour deep cleaning session called a “field day.”
As a child, a field day was a good thing. Fun school games with potato sacks and three legged races. We didn’t have potato sack races in the Navy. We cleaned. That’s a field day. Four fucking hours a week, usually on a Sunday, and much more intense than the bullshit post-watch clean up. Even the way they woke us up to have a field day was insulting. It gave Chief of the Boat a fucking boner, I swear. He’d come into berthing turning on all the lights and enthusiastically shouting something stupid like this:
“Drop your cocks and grab your socks, it’s field day! Wooooo!”
You could hear the giddiness in his voice as he yelled that stupid rhyming field day announcement. What an asshole. I hated the Chief of the Boat. He looked like an angry Muppet with a cheesy porno mustache glued on. I’d get dressed and filter back to the Engine Room, barely awake and hating life. I was too new and didn’t yet realize the early bird back there gets snagged by Queen La Chiefa to do the bilge cleaning. But I had always remembered Article III of the Code of the US Fighting Force:
“If I am captured, I will continue to resist by all means necessary.”
Oftentimes the best resistance was simply a filibuster. Talking means you’re not cleaning. So, I’d ask Queen La Chiefa questions before going down into the bilge. The goal was to lead him somewhere and then irk him so that he “corrects” me with his words. The more we talked, the less I cleaned. I knew if I asked enough questions about field day, it would somehow in a roundabout way lead back to our primary concern: reactor safety. Then I could hit him with a question that would get under his skin.
“So Chief, if reactor safety is really our primary concern, then why is there a battle short switch? I’m not so sure reactor safety is actually our primary concern then due to the existence of that switch.”
The “battle short” switch removes all reactor protection so that we don’t scram while we need some torpedo-evading propulsion in a hurry. Death is on the line! You know, during battle with those godless commies. At least they were godless commies when the 688 class were designed. I suppose, however, they were merely agnostic kleptocrats by the time I joined the Navy. I digress. Only the Captain can authorize use of the battle short switch. Obviously, Queen La Chiefa knew all that. I was just getting you up to speed. Mildly irritated, La Chiefa responded to me without this expository paragraph of delay.
“You know damn well why we have the battle short switch! It would be futile to save the reactor if it results in losing the ship!”
“So, one could argue that the ship’s safety is our primary concern, and therefore reactor safety is kind of just a distant secondary concern.”
“Reactor safety is YOUR primary concern, shipwreck. God help us all if the ship’s safety was up to you.”
“Aye, Chief. Reactor safety is our primary concern.”
“Exactly.”
“Until it isn’t.”
He glared at me. I further qualified my statement.
“Except for me, because reactor safety is always MY primary concern.”
“Look. For every little tiny thing you intend to do here in the Engine Room, before you take that action, you better be asking yourself, ‘Am I affecting reactor safety?’”
My cheekiness went too far, for he was now completely irked. Mocking superiors is one thing. Mocking the system engrained into the superiors’ psyche is an entirely different thing. And now he was onto my skylarking. This was a field day, and I was a non-useful body just standing around. This irked him further.
“Get in that damn bilge! It’s not going to clean itself!”
A surefire way to gauge if you won an argument with a superior in the Navy is how swiftly he or she makes you clean something. My victory was indeed swift. I squeezed down below the deck plates in Condensate Bay to wipe oily water out of the bilges. For each tiny little swipe of the Kimwipe brand oil rags I took, I asked myself out loud,
“Am I affecting reactor safety?”
<swipe>
“Am I affecting reactor safety?”
<swipe>
“Am I affecting reactor safety?”
<swipe>
I don’t think Queen La Chiefa heard me, and that’s probably a good thing. Or if he did hear me mocking his system, perhaps it didn’t bother him all that much. Perhaps with me being a well-known field day scofflaw, La Chiefa was thinking,
“He may be trying to provoke me, but in order to accomplish that, at least he is actually cleaning for once.”
But you know what? At the time, I kind of hoped he did hear me. In my little incensed mind, I was duped. It took nearly two years of advanced training to get to my first submarine, the bulk of which were courses in nuclear physics, chemistry, and thermodynamics. I studied hard for it. My test scores belied proof of this. I was motivated. I wanted to get to the fleet. I mean, do you realize how ridiculous someone wearing a sailor suit inside a classroom looks? But not just for that reason. Those damn recruiters made a job in the nuclear field seem glamorous, as though I would be among the elite and envied.
Yet once actually in the fleet, reality hit. I took out trash, cleaned bilges, wiped up oil, humped cables, chipped away old paint and lost brain cells applying fresh coats of paint inside the fucking sewer tube with absolutely no ventilation just like any other enlisted scumbag. What the fuck does this have to do with nuclear physics!?! Why the fuck did they send me to all that god damn training? That shit must have cost a fortune! Hundreds of thousands of dollars perhaps. What a waste! I mean, what do they say to a guy who graduates in the top ten percent of Naval Nuclear Power School?
“Get in the fucking bilge, shipwreck!”
At least that’s what it always seemed like. Yet there was a method to the Navy’s madness. Not that I would admit this to Queen La Chiefa, but it is a privilege to go on watch after a field day to see a nice clean station and not some shithole. And there must be some sort of psychology behind it. If you don’t care about maintaining a clean watch station, what else do you not care about? What else would you let slide?
And of the insane amount of nuclear physics, chemistry and thermodynamics courses they made me endure in order to be a trained bilge monkey? It’s easy to lose focus of the big picture when your superiors force you to do unpleasant tasks, such as waking up from a deep, post-jerkoff sleep to clean for four hours straight. But this seemingly mad method of the Navy has resulted in not a single one of their 269 or so reactors melting down.
Despite not having any radiation warning signs in my rather non-nuclear Engine Room Lower Level post, I too had a role to play in reactor operation and safety. The main thrust of this chapter is to explain who controls reactor power and why that is. So, we’re going to get back on that course. In doing so, I hope that the reader agrees that the reactor is not just an amazing feat of engineering, but that you get a sense that it is almost like a living breathing creature.
Eh, maybe not. That sounded pretty stupid. What am I, a fucking diggit? Let’s try this again. Blah, blah, blah, back on course, yeah… Okay. In doing so, it will also show the part I played. I mean, fuck the Throttleman. This is my memoir after all. Yeah, we’ll see how a trained bilge monkey such as myself has a hand in reactor plant operation and safety. That knowledge will be a byproduct of explaining how the energy released from splitting atoms ends up spinning the big pinwheel-looking thing hanging out the ass of the boat so we can punch holes in the ocean and go drinking in strange and foreign lands.
Next up in the nuclear propulsion operation explanation is the reactor plant proper. Before moving on to the reactor plant, a quick review of the reactor core is in order: The fuel plates are 97% enriched Uranium-235 with strategically placed Boron-10 burnable poison all clad in zirconium alloy. Pressurized light water is used for a moderator and coolant. The most significant concern to reactor safety is drawing a steam bubble in the core due to excessive temperature or a loss of coolant and/or pressure. Control rods used to shut down the reactor are made out of hafnium. The main operating challenge under normal conditions is the buildup of Xenon-135 poison inside the fuel plates. All of this in a McGuffin about the size of two refrigerators.
Okay. Now the plant proper. So, this little reactor core is placed inside a thick ass cylindrical stainless-steel pressure vessel located at the bottom of the submarine’s reactor compartment. The reactor vessel is piped up to redundant steam generators (aka the boilers). These steam generators are located high in the reactor compartment and connected to reactor’s pressure vessel with seamless all-welded piping to ensure there are no leaks. Even the reactor coolant pumps are hermetically sealed units and are welded to the piping. The pressure vessel is topped with a massive lid that is bolted on and then seal-welded.
After the fuel plate cladding, this fully sealed reactor plant is the second layer of containment. Should the reactor melt down, the reactor vessel, piping and components will contain all the nasty bits. Therefore, all of the welds are visually inspected, subjected to dye penetrant testing for cracks, and then checked more thoroughly with either x-rays or ultrasonic testing. After plant completion, the systems are hydrostatically pressure tested to verify their integrity. The reactor plant is fucking tight!
At the top of the reactor pressure vessel are the control rod drive mechanisms, which with all the cable penetrations make you realize that something is definitely going on inside that vessel. That’s no simple water tank. The control rod drive mechanisms inside are incredibly elegant designs in and of themselves. I can’t find too much information about US Naval reactor control rod drive mechanisms online (other than the rod material being hafnium), so you’re going to have to take my word that the engineering that went into the drives is beautifully simple and simply beautiful.
I can, hopefully, admit in the interests of concerned taxpayers that these drive mechanisms are fail-safe. The control rods are spring-loaded and require electricity to be withdrawn, the absence of which will almost instantaneously insert all the poles in the holes. (Poles in the holes being our informal term for scram.) Apparently, there are reactor designs where the rods have to be pushed up into the core from below. More amazingly, there are yet other designs which require electrical power to insert the rods, the absence of which will cause them to fail in place. Yikes.
Along those lines is the answer to something else you may be wondering. Perhaps you were questioning why I mentioned that the reactor is mounted down low, and the steam generators are mounted up high. It’s a passive safety feature which takes advantage of the difference in density of water due to temperature. Surely, you’ve heard the phrase “heat rises.” This applies to reactor coolant as much as anything else.
The reactor is the heat source, and the steam generators are the heat sink. If all the reactor coolant pumps fail, the plant was designed to remove the reactor’s decay heat through natural circulation. Hot coolant from the reactor loses density and tends to rise up to the steam generator where it is cooled, gains density and tends to sink back down to the reactor. The momentum builds and soon there is quite the flow rate. This natural circulation action actually provides ample shutdown cooling to prevent a meltdown.
If you think this passive safety design feature is a pants-load of pure pro-nuclear propaganda on my part, look up the S8G reactor of the massive Ohio (726) class ballistic missile submarines. At the time, those were the quietest submarines ever built. So quiet that to find one, I am told you had to look for “holes” in the natural background noise of the ocean. Why am I talking about how quiet the Ohios are when I’m supposed to be talking about reactor safety?
Because 1970s vintage reactor coolant pumps at the time of the Ohio’s conception were relatively noisy. This is due to the fact that they are directly mounted to the submarine’s steel structure unlike the rubber-mounted equipment in the Engine Room. Neutron bombardment would destroy those soft sound isolation mounts, so without those mounts, the reactor coolant pumps contribute significantly to a nuclear submarine’s sound signature.
With exactly that in mind, what they did in the Ohio class was take advantage of the natural circulation phenomenon so that they could shut down their coolant pumps while the reactor is critical. Yes, they can safely run a reactor online with absolutely no coolant pumps. Natural circulation works perfectly well for the typical levels of reactor power needed while on patrol. This makes them lethally silent. Only when those ballistic missile behemoths really needed to boogie did they have to kick on the comparatively noisy reactor coolant pumps.
My significantly smaller attack boat and its S6G plant used coolant pumps during all reactor critical operation, but natural circulation would work just fine during emergencies with only decay heat to remove. I mean, think about it. If you realize that it works perfectly well for a much larger S8G reactor while online, then I don’t think it’s a stretch to believe that it works just fine for a smaller S6G reactor in an emergency while shutdown.
When does natural circulation not work well in a reactor plant? Are you thinking about the bubbles? Because if you are thinking about the bubbles, you’re spot on. A steam bubble in the core will stop natural circulation dead. But what if I told you that a steam bubble is something so advantageous to pressurized water reactor operation that we want to and actually do draw and maintain a bubble? Yes, it’s true. We absolutely do draw and strive to maintain a bubble… just not in the core.
Steam bubbles were mentioned when discussing the meltdown at Three Mile Island. So perhaps you remember a piece of equipment called a pressurizer and know where I’m going with this. That is where we want to draw and maintain that steam bubble. But why do we want any bubble at all? Because a nice fat steam bubble in the pressurizer acts like a shock absorber.
You see, liquids like reactor coolant—aka water—are virtually incompressible. So, if the plant was filled completely “solid” with liquid coolant, any thermal expansion from the reactor’s heat would multiply into ungodly pressure, possibly rupturing the piping, pumps, steam generators or even (gasp!) the reactor vessel. Similarly, if the coolant dropped a little bit of temperature, it’s possible that the pressure would drop to saturation, and then the dreaded bubble in the core would form.
This pressurizer is a vertical cylindrical vessel connected to one of the reactor coolant “hot legs.” At the bottom are electric heaters and at the top, a spray nozzle fed from one of the reactor coolant “cold legs.” The hot leg is the piping from the reactor to the steam generator, and the cold leg is the piping from the steam generator to the reactor. The heaters heat the coolant inside the pressurizer to saturation temperature to draw and maintain the bubble.
Once the bubble is formed, the entire reactor coolant system will maintain the saturation pressure for the temperature maintained in the pressurizer. Looking at the not classified civilian Westinghouse PWR figures of 2235 psig (146 Bar) coolant pressure and 547*F (304*C) coolant temperature in the core at hot standby conditions, it tells me that the heaters maintain about 653*F (363*C) in the pressurizer.
If pressure gets too high, the spray nozzle valve is opened and supplies cooler reactor coolant from the cold leg. Cold is a relative term here as the coolant is still several hundred degrees. This spray will cool and collapse the bubble a bit, reducing the pressure due to the saturation temperature-pressure relationship. If pressure still rises too high, there are numerous pressure relief valves in the pressurizer system to dump the excess into a retention tank. I’ve never seen that happen.
The pressurizer also acts as a surge volume when reactor coolant temperature changes. As I said, the bubble is like a shock absorber. Coolant from the pressurizer will flow in and out to keep the reactor coolant system and core completely filled. If the pressurizer level gets too low, it in turn can be filled using one of the reactor coolant charging pumps. If the pressurizer level gets too high, coolant can be released through discharge system to a retention tank or even overboard.
DISCHARGE REACTOR COOLANT OVERBOARD!?! INTO THE OCEAN!?! WHERE THE DOLPHINS SWIM!?!
Yes. Where the dolphins swim. But relax. Reactor coolant is ultra-pure water and as you know, less than a minute after passing through the core, it is no longer radioactive. Remember, that radioactivity is from those neutron-activated seven-second half-life Nitrogen-16 isotopes beta-decaying back into regular Oxygen-16.
Okay fine, but what about the Cobalt-60 eroded away from system components? That’s in the coolant and takes nearly forty years to decay away! True, but it doesn’t dissolve into the coolant and is therefore removed before discharge through filtration and ion exchanger beds. Your beloved cetaceans are safe and will not give birth to three-eyed dolphin puppies.
As for the humans, we are protected from radiation by multiple layers of shielding. The primary shield is around the reactor and consists of several inches of lead to stop gammas, a water tank surrounding the reactor to moderate escaped neutrons, and a boron lining to capture those neutrons. What about the betas? No need to worry about the betas. Beta radiation can’t even penetrate your clothing, let alone the reactor coolant piping.
The secondary shield is built at the front and rear ends of the Reactor Compartment, but not the sides other than around the tunnel linking the cone to the Engine Room and the roof to protect those walking topside. It is also lead lined for gamma protection, but instead of using water for neutron shielding, several feet of boron impregnated polyethylene plastic is used instead. Polyethylene’s chemical formula is C2H4, so there are plenty of hydrogen atoms in it for moderating the escaped neutrons so that the boron can capture them.
Note that the Reactor Compartment is the third layer of containment. Should the reactor melt down, the fission products will obviously escape the first layer of containment, which is the fuel plate cladding. I’ve said that those nasty creations will then be contained by the second layer of containment, which is reactor coolant system. But I’ve also mentioned that TMI-2 melted down due to a loss of coolant casualty.
So, let’s say a reactor coolant pipe ruptures, and we can’t isolate the reactor from that fissure. Pressure in the reactor drops down to saturation, and a bubble forms in the core. We were all on crack that day and can’t use alternative means to cool the reactor. The fuel plates overheat and melt down. Now there is a direct path for the fission products to escape both the first and second layers of containment. Yes, but not the third!
The sealed-up Reactor Compartment will then contain all those little cancer-causing isotope assholes. Take note that if the Reactor Compartment hatch is opened when shutdown, both the Engine Room hatches to topside (in the aft escape trunk) and the Engine room hatch to the cone (at the end of the reactor tunnel) have to be shut so that we still provide this third layer of containment. Thems the rules. You best be minding them.
And that there is the reactor plant. The reactor core is inside the pressure vessel, which is connected to the steam generators with a seamless fully welded piping system. Coolant is circulated through the core and steam generators by multiple redundant reactor coolant pumps. System pressure is maintained by the pressurizer using electric heaters and spray nozzles to regulate the saturated steam bubble inside the pressurizer so that one does not form in the core. This reactor coolant system is also known as the primary loop and therefore all of the aforementioned components are referred to as primary system components.
Now what of the reactor power? All this emphasis on nuclear safety is great for all the tree-huggers and all, but there’s got to be some red meat for all the military enthusiasts too. This reactor is powering a warship after all. And this warship was designed to hunt and kill enemy ballistic missile submarines before they can turn the United States into an apocalyptic wasteland. When these Los Angeles class boats were constructed, they were the most advanced and effective hunter killers ever built.
With that in mind, we should revisit the primary loop’s purpose. I had emphasized cooling the reactor and preventing a meltdown. It’s basic reactor safety, shipmate! But wait. The real purpose of the coolant system is to transfer reactor power to the secondary loop where it can be used to propel the ship and create electricity. Atomic energy is not glamorous. It is actually the lowest form of energy. It’s heat. But when this heat is used to make steam, it is quite useful. Enter the secondary loop.
The secondary loop is a heat engine. Specifically, a Rankine cycle external heat engine. If I go into the level of detail for the secondary loop as I did for the primary loop, there would be no room left for my memoir. So, let’s keep this brief. It’s all a balance. Since the secondary loop is a closed system, heat and work added to the system must equal heat and work extracted from the system. A Rankine cycle has four major components which maintain that balance and adhere to the laws of conservation of energy: boilers, turbines, condensers and pumps.
Boilers add large quantities of heat to the system, turbines extract large quantities of work from the system, condensers extract small quantities of heat from the system, and pumps add small quantities of work to the system. If you include work extracted by friction and heat extracted by ambient losses, this is a perfect balance.
In our case, the boilers are our steam generators. The heat added into the secondary loop to generate this steam is exactly the amount of heat removed from the reactor coolant. You see, there is also a conservation of energy balance inside the steam generator. Note that the reactor coolant does not mix with the boiler water. They are separated by the boiler tubes, with the reactor coolant being inside the tubes and the boiler water outside the tubes, which is called the shell side.
High-pressure saturated steam is sent from the steam generators to the ship’s service turbine generators and the propulsion turbines. This is where heat energy is converted into rotational mechanical energy to spin generators and the ship’s screw. We are extracting work from the system to create electricity and propel the ship. The turbine’s low pressure and temperature exhaust steam is thoroughly exhausted now, excuse the pun, and then sent down into the condensers.
Unlike in a steam locomotive, the exhaust steam in a power plant or ship is recovered and reused. We then need to get it back into the boilers or steam generators. It is damn near impossible to pump steam in any practical sense. Therefore, we remove a little bit of heat from the secondary loop by using the Main Sea Water system. This drops the turbine exhaust steam temperature to just below saturation. And that will obviously condense it back into water.
Note that the seawater does not mix with the steam. The seawater is pumped through the insides of the condenser tubes while the steam and condensate are outside on the shell side. The steam condensing into water on these tubes drops down and is collected in tanks at the bottom of the condensers called hotwells.
Unlike steam, water is easy to pump! So, we add a little bit of work into the system with feedwater pumps. Once those pumps feed the steam generators, the Rankine cycle is complete. Hopefully you’ll consider that secondary system explanation mercifully brief yet thorough enough to understand the basics. Here’s a quick review of the four secondary system components and their heat balance:
Boilers: large heat input (Qi)
Turbines: large work output (Wo)
Condensers: small heat output (Qo)
Pumps: small work input (Wi)
Qi + Wi = Wo + Qo
Qi – Qo = Wo – Wi
Now as a Machinist’s Mate, these secondary system components were my primary responsibility. When off watch, I performed routine mechanical maintenance on these systems. This routine maintenance was boring yet vital to ensure their proper operation. There’s a smidgen satisfaction knowing this, but even more satisfaction going to the bar afterwards and bitching about it. While on watch in the Engine Room, I operated those systems in which I had been maintaining.
This is all well and good, but I have yet to tie it all together into an integrated plant ah-ha moment of realization. I mean, how the fuck does the Throttleman control reactor power, and just how the hell does a bilge monkey Machinist’s Mate affect reactor safety? Let’s put this all together and keep an eye on the moderator. The moderator is the key to the beauty of the reactor.
Taking it from the top, the reactor core generates heat from fission, which is transferred from the fuel plates to the reactor coolant. This coolant is circulated by reactor coolant pumps into the primary side of the steam generator where it transfers its heat to the boiler feed water on the secondary side of the steam generators in order to produce steam. The cooler reactor coolant then enters the reactor core ready to remove more heat from it. This is the primary loop.
Steam produced in the steam generators from the heat added by the reactor coolant is fed to the ship’s service turbine generators and the propulsion turbines. The amount of steam fed to the two types of turbines is regulated by the governor valves and main engine throttle valves respectively. Work is extracted from the steam and exhausted to the condensers where heat is removed to condense it back into feed water. The feed water is returned to the steam generators with pumps, to be boiled again in order to replenish the steam that had just been consumed by the turbines. This is the secondary loop.
Now it’s time for some action. Let’s say we’re lurking around somewhere silently answering an All Ahead One Third bell as requested by the engine order telegraph system. All Ahead One Third is the terminology for minimum speed. Then some fucker drops a torpedo in the water right on our ass. We need to seriously scoot! Oh yeah, we will be answering an All Ahead Flank bell immediately. That’s what we call our maximum velocity.
Hash Brown the Throttleman is about to save our asses!
The speed of the ship is determined by the rpm of the screw. The screw is what is converting the rotational mechanical energy of the propulsion turbines into kinetic energy. The submarine stops accelerating and maintains a constant speed when there is a balance between the work the turbines extract from the heat engine and the work imparted by the screw into the sea. So, we really need the turbines to extract, like, all of the work so that we can accelerate to flank speed and outrun that torpedo!!!
The work extracted from the heat engine by the turbines is determined by how far the throttle valves are opened. We want maximum power from the turbines to outrun a torpedo, so the Throttleman fully, yet carefully, opens those valves. More steam is admitted to the turbine and more work is extracted to turn the screw at a higher rpm and make the ship go faster. Everything is a balance, so in order to remove more work from the heat engine, we need to add more heat into it.
This means the production of more steam for the turbines extracts more heat from the reactor coolant, resulting in a drop in reactor coolant temperature. We now need more reactor power to heat that reactor coolant back up to temperature in order to keep up with the increased production of steam that the Throttleman is demanding. More reactor power means we need more positive reactivity. So, the Reactor Operator has to pull up the control rods to unblock the positive reactivity of the fuel and increase reactor power, right?
Wrong.
He doesn’t have to do a god damn thing. The reactor increases its power all by itself. No Reactor Operator input is necessary. He could be fucking passed out on his panel and the reactor still responds, “Yeah fuck that torpedo, I got this.” And when the Throttleman cuts the power to the main engines by closing in on the throttles after successfully outrunning that torpedo? The reactor responds by cutting back on its own power output, again without any Reactor Operator input. Seriously. No bullshit. The fucking reactor acts like a living, breathing creature. It really is quite amazing. “We outran that bitch torpedo, huh?” the reactor asks, continuing, “I knew we would, but maybe someone should wake up my operator.”
Now that we know that the reactor is basically operating itself, why do we even need a Reactor Operator in the first place? Should we rename him Reactor Hall Monitor? Or since his most critical job is to maintain proper reactor coolant temperature and pressure, maybe he should simply be called the Bubble Maker? As long as he makes the bubble in the pressurizer, that name works for me.
Hmm. Are you detecting a little bit of jealousy towards the Bubble Makers while I’m all the way down in the belly of the boat as the Engine Room Lower Level watch-stander? Yeah maybe. Those damn bubble making twidgets being the only ones allowed to sit at the Reactor Plant Control Panel while the reactor is critical are perhaps the purest form of nuke. Then maybe the squats are the next purest because they maintain the reactor chemistry, do the radiation surveys and are in charge of decontamination. And even the electricians get to control reactor power when they qualify as Throttleman.
I suppose that finding out you will never be allowed to actually control the reactor itself stings a little after those two years of the nuclear training Peepayleenay. I mean, the recruiter does not supply you with that information when you are selecting your nuke specialty. Since I’m a motorhead, the mechanical specialty rating of nuclear Machinist’s Mate appealed to me, particularly because it was possible to get selected for the welding school. I didn’t know how to weld before joining the Navy. Now I know how to weld exotic superalloys used in the reactor plant such as Inconel-600. And I knew it well enough to pass ultrasonic testing of my nuclear grade welds.
To tell you the truth, the sting of not selecting the rate of nuke Electronics Technician when signing my enlistment papers and not being allowed to become a Reactor Operator in the fleet was actually blunted by relief once actually in the fleet. It’s that little fucking box. What a relief it was knowing I wouldn’t be stuck in it! I would never be stuck in Maneuvering! I was free to roam the expanse of my part of the Engine Room. With all that space, how could I be jealous? That beautiful reactor can go fuck itself. Besides, it practically operates itself.
I mean seriously. I hated people. I was a loner. I liked being underwater away from everyone and everything. So as a misanthrope, I belonged way down in the secluded Engine Room Lower Level. This was a perfect match for my personality. Once in the fleet, I was certainly no longer jealous of the Reactor Operators.
Besides, I had a piece of the nuclear propulsion plant action too. Sure, it was a piece of the secondary loop, but that piece was critical to reactor operation too. I operated the systems that extracted heat from and added work to the Rankine cycle heat engine. I also operated the auxiliary systems that supported the components which extracted the work from the heat engine.
We can’t go from All Ahead One Third to All Ahead Flank without the Lower Level adjusting the speed of the Main Sea Water pumps, kicking on more Main Condensate Pumps, and adding more cooling to the propulsion turbine’s lube oil system. Everyone had their part to do. It makes me smile thinking of all the times I was in Shaft Alley when hearing the “ALL AHEAD FLANK” announcement by the EOOW.
Drop the logs, go to the ladder-well and slide down into MSW bay with your feet outside the ladder rungs, adjust the speed of the MSW pumps, take a step forward, wing open the lube oil temperature regulating valve, walk into Main Condensate bay and kick on extra main condensate pumps, and then go back and monitor your oil temperatures while you hear the main engines spool up and feel the boat list over a bit from the enormous torque of the prop shaft and screw. It was all a dance, and the nuclear training pipeline was quite the choreographer.
Since I went through the Nahvee Nookay Peepayleenay, I knew how failing to do my job properly affected reactor safety. Take the Main Sea Water system for example. If I operated that system improperly and failed to provide adequate cooling to the main condensers, turbine exhaust steam will back up from lack of condensing action, over-pressurize the condensers, and trip the main steam cutout valves. Now the steam generators are bottled up and the reactor core has no cooling. Similarly, I could starve the steam generators of feed water by failing to operate the condensate system properly. I could destroy the main engines and ship’s service turbine generators by failing to properly operate the propulsion lube oil and SSTG lube oil systems respectively.
Maybe it’s a bit clearer now why the Navy put me through all that crazy long and expensive nuclear training. The Navy brass didn’t want just any old bilge monkey to climb out from cleaning below the deck plates and then operate their precious little systems in Engine Room Lower Level. Queen La Chiefa was right to remind me of our primary concern. Maybe I didn’t need to ask myself “Am I affecting reactor safety” while wiping up oil in the bilge, but I sure as hell did when I had my grimy little fingers on the Main Sea Water pump speed controls.
Am I turning this switch the right direction? Am I going to increase cooling or decrease cooling by going clockwise? Did the Engineering Watch Supervisor order me to secure the port or the starboard Main Sea Water pump? Is my hand on the port or starboard pump controls? It’s an extra second of thought before acting to make sure you are taking the correct action.
This is important because that action might be the first link in a chain of failures that could lead to a nuclear meltdown. So, you’re god damn right I’m going to ask myself if I’m affecting reactor safety before acting. I don’t want to be in some documentary as the asshole that caused the Navy’s first reactor meltdown. Now, after reading about my role in the nuclear propulsion plant and all that advanced training I passed with flying colors, you must have a question or two for me. I bet I know what they are:
Fuck you and your stupid Engine Room Lower Level! How the FUCK does the reactor know when to increase or decrease its power by itself without the Reactor Operator?!? And why the HELL do you keep saying that the Throttleman controls reactor power when you just said the reactor controls its own power!?! All that asshole ever did was just change the turbine speed!
Was I close? I think I was close. Okay. First of all, the Throttleman controls the steam demand. And I told you, reactor power follows steam demand. So how in the world does it do that all on its own? Well, I also told you to keep an eye on the moderator, didn’t I? Yes, I believe I did. And way, way back at the beginning of this really long chapter, I said there was something beautiful about the balance. I think I even called it elegant. I may have gotten a little carried away there. Let’s just stick with beauty, and the most beautiful balance is in the moderator.
You remember that the moderator’s job is to thermalize neutrons so that the Uranium-235 can absorb them and fission. You also remember that this action is similar to a mosh pit with everyone bouncing off each other. And you should remember that the reactor coolant is the moderator, which means its water. Well, the moderator’s effectiveness is extremely sensitive to changes in temperature, and that’s because the density of water is extremely sensitive to changes in temperature.
Hot water is like a mosh pit with a hundred metalheads on the general admission floor while the opening act is playing. Cold water is like when the metalheads pour over the walls from the seating area after Slayer comes on and there are now a thousand metalheads thrashing around in the pit. It’s all about the density. If you were running full bore into the pit, you might make it some distance with the widely spaced unknown opening act band fans. But the densely packed Slayer fans will moderate you right into the fucking floor almost immediately. Yes, it’s all about the density.
So, when that Throttleman admits more steam to the propulsion turbines, this increases the steam flow out of the steam generators. Extra steam flow out of the steam generators means more feed water flow into the steam generators. The extra energy extracted from the reactor coolant needed to boil the increased rate of relatively cool feedwater flow means extra heat removed from the coolant. This reduces the coolant temperature, thereby making the coolant denser. If the coolant entering the reactor core is denser, that means the moderator is denser. A denser moderator is more effective at neutron moderation. This means adding colder coolant to the reactor adds positive reactivity. The reactor becomes supercritical and power increases. The power increase then adds more heat to the coolant, which then becomes less dense, which reduces the effectiveness of the moderator, removing positive reactivity, and finally the now critical, no longer supercritical reactor levels off at a higher power.
When the Throttleman cuts back the steam admitted to the propulsion turbines, this decreases the steam flow out of the steam generators. Reduced steam flow out of the steam generators means reduced feed water flow into the steam generators. The reduction of energy extracted from the reactor coolant needed to boil the reduced rate of relatively cool feedwater flow means less heat removed from the coolant. This results in an increase in the coolant temperature, thereby making the coolant less dense. If the coolant entering the reactor core is less dense, that means the moderator is less dense. A less dense moderator is less effective at neutron moderation. This means adding hotter coolant to the reactor removes positive reactivity. The reactor goes subcritical and the power decreases. The power decrease then adds less heat to the coolant, which then becomes denser, which increases the effectiveness of the moderator, adding positive reactivity, and finally the now critical, no longer subcritical reactor levels off at a lower power.
I may have just removed a bit too much mystique from reactor operation with that explanation. Maybe you’re less in awe now that you’re standing behind the curtain with me. Living, breathing creature my ass! Some engineer figured that shit out, and it’s not that beautiful! Fine, but if you’re going to stand behind the curtain with me, at least use the proper terminology. This particular moderator feedback action inside our naval reactor is called the negative temperature coefficient of reactivity.
This baked-in-the-cake design feature is crucial to the safe operation of the Navy’s pressurized water reactors. An overheating reactor will reduce its own rate of fission. This prevents runaway reactor power. Not all reactor designs are like this. And even some reactors with a negative coefficient of reactivity do not follow steam demand all by themselves.
Well what kind of asshole reactor doesn’t fucking follow steam demand!?! Okay so before answering that, let me first admit to my bias towards pressurized water reactors. This is due to my experience operating four of them. It’s like rooting for the home team, which in this case is the Pittsburgh Pressurizers. The naval PWRs were first developed by the Bettis Atomic Power Laboratory in West Mifflin, PA, a suburb of Pittsburgh.
So, who then is the challenging team? The Chicago Boilers. These are the boiling water reactors (BWRs) first developed by the Argonne National Laboratory in Lemont, IL, a suburb of Chicago. Boiling water reactors? Are you shitting me!?! I know, right? Those crazy ass fuckers actually make steam in their reactor vessels on purpose! Okay, it’s not really that crazy. Allow me to explain.
Steam is a poor thermal conductor, but the continuous boiling of water is very effective at removing heat. While this is completely different than drawing a static steam bubble in the core of a PWR, it is still so counter to my training and feels so unnatural that it makes me shudder to think of it. It works fine though. I mean, I have a few examples of BWRs to share:
Chernobyl-4 (Soviet model RBMK-1000 reactor)
Fukushima Daiichi-1 (General Electric model BWR-3 reactor)
Fukushima Daiichi-2 & 3 (General Electric model BWR-4 reactors)
Shots fired! Cheap shots fired actually. My bias is showing. No boiling water reactor has ever melted down in the United States, and GE built roughly 35 of them for the domestic market. (Second in popularity only to Westinghouse’s PWR.) Those were indeed cheap BWR shots considering one was a flawed Soviet design, and the three others were all taken out by a tsunami. A tsunami which would have destroyed a PWR built along the coast of an earthquake prone nation just the same. Whatever. Call me a bigot, but I’m still no fan of a reactor that purposely boils off its own damn coolant.
To the unbiased, there are plenty of advantages and disadvantages of both PWRs and BWRs. That’s something which should be further researched if you find the differences interesting. In particular, the Soviet RBMK boiling water reactors, which have graphite moderators. I know very little about how that works. Graphite is carbon, an atom of which has roughly twelve times the mass of a neutron. So that’s like playing pool with a standard cue ball yet using bocce balls instead of the regular solid and striped pool balls. I don’t really get it. Crazy Ivans.
But I suppose the real reason I don’t like BWRs is that they don’t seem like the living, breathing creatures to me, as the beautiful PWRs appear to be. I mean, you’re behind the curtain now, but don’t you find the self-control of PWRs eerily like a primitive life form? No? You don’t? Whatever. You’re probably a control freak BWR lover then. Makes sense because those BWRs can’t control themselves. Someone or something has to adjust water flow into the core to change reactor power. I can’t make a joke about who controls reactor power in a BWR because it’s the Reactor Operator.
That dude or dispatch computer is adjusting the water flow to the core to control reactor power, and then there are these logic controllers responding by adjusting steam flow to the turbine to maintain a certain pressure, which then controls the turbine generator output. That’s not funny.
PWRs are funny. We can joke about PWRs because they adjust their power output all by themselves. That allows us to bust the balls of the Reactor Operator sitting there on his ass not controlling reactor power. Damn Bubble Makers! Okay, I told you I’m no longer jealous of them. So why act like it? I hereby formally withdraw my request to rename the Reactor Operators the Bubble Makers. No, that was just a passing phase. Besides, the pressurizer pressure control through electric heaters and spray valve operation are done automatically. Like reactor power, the Reactor Operator sort of just watches it.
I know. Maybe we should call them something cool like the Xenon Fighters. Because that’s something they manually do. There’s no actual xenon or samarium meter on the Reactor Plant Control Panel, but that doesn’t matter. The reactor maintains its power, compensating for the additional negative reactivity of the poisons by a reduction of coolant temperature. Colder coolant as we now know of course adds positive reactivity. Therefore, poisons manifest themselves as that drop in reactor coolant temperature.
The Reactor Operator’s keen eye is trained to spot this. He will manually raise the control rods to reduce negative reactivity in the core, thereby keeping the average reactor coolant temperature in the “green band.” When poisons are burned away a bit, he may then have to shim in the rods to keep everything in the green band. Okay. Settled then. We’ll call them the Xenon Fighters. But to maintain balance, let’s just break their balls one more time:
So, I ask you, in a nuclear submarine, who controls reactor power?
You know this now. You know it’s not the Reactor Operator. You know it’s really the Throttleman at the Steam Plant Control Panel. You know this because you know without any input from operators at all, reactor power follows steam demand. It’s a beautiful thing. And obviously that steam demand follows the Throttleman’s inputs to the propulsion turbine throttle valves. So yeah, the Throttleman is the guy that actually controls reactor power. That’s what I told Queen La Chiefa as we wrapped up my qualification interview. He was pleased with my answer and explanation.
But hold on a second. The Throttleman is just answering bells by following that old-timey engine order telegraph system. The other end of this engine order telegraph system—where the order originates—is up in the Control Room. Hmm. Think about that for a minute. So, who really controls reactor power?
Shit. I guess I should have told Chief Queen it’s actually the coners…