On Criticality
Jun. 16th, 2009 12:12 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
aliarasSo, I do this reactor thing, and it turns up a good number of instances of Did Not Do The Research (WARNING: TVTROPES) in movies/fiction. For the most part, I ignore them and watch the thingy. Sometimes, they make it 'better' (OH GOD ATOMIC RAY GUN WHY). However, there's one that bugs me enough that I want to post about it: the word 'critical' does not mean what some scifi authors think it means.
Before we get into critical, let's start at the bottom. A subcritical nuclear reaction is one that isn't going anywhere. The number of neutrons born each 'generation' is less than the one before it, so it's steadily going down in power. If there's a neutron source in there somewhere, the power approached won't be zero, but it'll still be low compared to your reactor's max power. An analogy: pulling the control rods out on a subcritical reactor causes power to increase in a way that kinda feels like dragging a heavy stone block over a rough surface - you pull a bit, the block moves a bit obligingly, you stop pulling, the block stays right where it is. You don't really go very far.
A critical nuclear reaction ALSO doesn't go anywhere! In a critical reaction, the number of neutrons born each 'generation' is equal to the one before it. Power stays stable at wherever it's at. This could be 5W, 230kW, or 1GW - the power is stable, the operator is sitting back with rod control in auto trying not to be bored. The powers a reactor can be critical at are limited by the design of the reactor and regulated by the government. This is why "the reactor is critical, sir!" *panic* provides so much headdesk material for me. Reactors are supposed to go critical! It's their job! Tell me why the reactor being critical now is a problem, then I'll panic. Maybe some Bad Guy got onto the spaceship and started it up when it's supposed to be powered down and oh god, he can now do things he isn't supposed to be able to. Maybe it's critical at a power level it -can- be at, but really -shouldn't- be. But the fact that it's critical at all? Yawn.
Then there's supercritical. When a nuclear reaction is supercritical, power is increasing. The number of neutrons born each 'generation' is higher than the one before. This can also not be a problem. The rate at which power is increasing is described as the period, which is defined as the time it takes for power to increase by a factor of e. If the period is pretty long, the reactor isn't ohmygodrunningaway with you. To get a reactor to a certain power, you have to go supercritical at some point. Now, it doesn't take very much from critical to go supercritical (any extra reactivity will do it), but we're still very much in the realm of under control. Returning to the analogy of the block, pulling control rods on a critical reactor is like putting the block on really slick ice and giving it a push, and pulling control rods on a supercritical reactor is like skating up and pushing more. It'll keep going until something stops it.
A particular species of supercritical, and where we start getting into nervous territory, is prompt critical. Remember period? When the reactor is prompt critical, it's really really fast. Prompt critical is so named because at that point, the reactor is critical on prompt neutrons alone, which is hard to explain without getting into a bunch more nuclear physics stuff. But the moral of the story is, period is really fast, and this is where we get into not really under control reactions. So if the officer came onto the bridge going "Sir! The reactor is prompt critical!" I would agree with the panic and start wondering how.
Prompt supercritical is pretty much the above, but more. Now, some reactors are designed to do this - they're pulse reactors. They shoot a control rod out and their reactor gets to a really high power really quickly, then the control rod drops back in and everything's groovy. But for a reactor that's not designed to do that or at the wrong time, going prompt critical/supercritical can be a Really Bad Thing (see: SL-1 accident)
And that's my ramblings for the day. All comments and criticism welcome, particularly if I messed something up in the writeup.
Before we get into critical, let's start at the bottom. A subcritical nuclear reaction is one that isn't going anywhere. The number of neutrons born each 'generation' is less than the one before it, so it's steadily going down in power. If there's a neutron source in there somewhere, the power approached won't be zero, but it'll still be low compared to your reactor's max power. An analogy: pulling the control rods out on a subcritical reactor causes power to increase in a way that kinda feels like dragging a heavy stone block over a rough surface - you pull a bit, the block moves a bit obligingly, you stop pulling, the block stays right where it is. You don't really go very far.
A critical nuclear reaction ALSO doesn't go anywhere! In a critical reaction, the number of neutrons born each 'generation' is equal to the one before it. Power stays stable at wherever it's at. This could be 5W, 230kW, or 1GW - the power is stable, the operator is sitting back with rod control in auto trying not to be bored. The powers a reactor can be critical at are limited by the design of the reactor and regulated by the government. This is why "the reactor is critical, sir!" *panic* provides so much headdesk material for me. Reactors are supposed to go critical! It's their job! Tell me why the reactor being critical now is a problem, then I'll panic. Maybe some Bad Guy got onto the spaceship and started it up when it's supposed to be powered down and oh god, he can now do things he isn't supposed to be able to. Maybe it's critical at a power level it -can- be at, but really -shouldn't- be. But the fact that it's critical at all? Yawn.
Then there's supercritical. When a nuclear reaction is supercritical, power is increasing. The number of neutrons born each 'generation' is higher than the one before. This can also not be a problem. The rate at which power is increasing is described as the period, which is defined as the time it takes for power to increase by a factor of e. If the period is pretty long, the reactor isn't ohmygodrunningaway with you. To get a reactor to a certain power, you have to go supercritical at some point. Now, it doesn't take very much from critical to go supercritical (any extra reactivity will do it), but we're still very much in the realm of under control. Returning to the analogy of the block, pulling control rods on a critical reactor is like putting the block on really slick ice and giving it a push, and pulling control rods on a supercritical reactor is like skating up and pushing more. It'll keep going until something stops it.
A particular species of supercritical, and where we start getting into nervous territory, is prompt critical. Remember period? When the reactor is prompt critical, it's really really fast. Prompt critical is so named because at that point, the reactor is critical on prompt neutrons alone, which is hard to explain without getting into a bunch more nuclear physics stuff. But the moral of the story is, period is really fast, and this is where we get into not really under control reactions. So if the officer came onto the bridge going "Sir! The reactor is prompt critical!" I would agree with the panic and start wondering how.
Prompt supercritical is pretty much the above, but more. Now, some reactors are designed to do this - they're pulse reactors. They shoot a control rod out and their reactor gets to a really high power really quickly, then the control rod drops back in and everything's groovy. But for a reactor that's not designed to do that or at the wrong time, going prompt critical/supercritical can be a Really Bad Thing (see: SL-1 accident)
And that's my ramblings for the day. All comments and criticism welcome, particularly if I messed something up in the writeup.
