Alternative to generators to prevent the Chernobyl disaster

[Phys12]

TL;DR Summary

From my understanding of the Chernobyl disaster; the test they wanted to conduct that led to the disaster was to figure out if the turbine which was producing electricity, after the nuclear reactor had stopped working, would be enough to generate sufficient amount of electricity to keep the water pumps running until the backup generators began pumping the water. My question is, why not just use battery and have a switch to power the water pumps instead of generators (which took 60 sec to start)?

The only material that I used to get this information is Wikipedia: https://en.wikipedia.org/wiki/Chernobyl_disaster#Steam_turbine_tests. I don't know if that's reliable enough...

 

[Rive]

Look up actual regulations and articles related to 'station blackout'.

 

[anorlunda]

y question is, why not just use battery and have a switch to power the water pumps instead of generators (which took 60 sec to start)?

That's so obvious, we must believe that if they could have used batteries, they would. We are talking about very big pumps.

It is very difficult for us to know details about the CCCP rules and regulations applicable at Chernobyl.

But even if they did use batteries, there would still be a need for reactor start/stop tests anyhow.

 

[Astronuc]

My question is, why not just use battery and have a switch to power the water pumps instead of generators (which took 60 sec to start)?

The only material that I used to get this information is Wikipedia: https://en.wikipedia.org/wiki/Chernobyl_disaster#Steam_turbine_tests. I don't know if that's reliable enough...

From further on in the Wikipedia article,

As planned, a gradual reduction in the output of the power unit began at 01:06 on 25 April, and the power level had reached 50% of its nominal 3,200 MW thermal level by the beginning of the day shift.
At this point, another regional power station unexpectedly went offline, and the Kiev electrical grid controller requested that the further reduction of Chernobyl's output be postponed, as power was needed to satisfy the peak evening demand. The Chernobyl plant director agreed, and postponed the test. Despite this delay, preparations for the test not affecting the reactor's power were carried out, including the disabling of the emergency core cooling system or ECCS, a passive/active system of core cooling intended to provide water to the core in a loss-of-coolant accident. Given the other events that unfolded, the system would have been of limited use, but its disabling as a "routine" step of the test is indicative of the lack of attention to safety in the test.

So they deliberately took the reactor outside of its safe domain.

From https://www.nei.org/resources/fact-sheets/chernobyl-accident-and-its-consequences:
What Happened - "The accident, which occurred at reactor 4 of the plant in the early morning of April 26, 1986, resulted when operators took action in violation of the plant’s procedures. Operators ran the plant at very low power, without adequate safety precautions and without properly coordinating or communicating the procedure with safety personnel."

Another point in the NEI article, "The reactor built at Chernobyl is a RBMK reactor, which was never built by any country outside the USSR because it had characteristics that were rejected everywhere outside the Soviet Union. Chief among these was its inherent instability, especially on startup and shutdown."

 

[Phys12]

Look up actual regulations and articles related to 'station blackout'.

I did, this one: .

That's so obvious, we must believe that if they could have used batteries, they would. We are talking about very big pumps.

Yes, but if you look at the video I quotes above, one of the backups used by the plant during a total blackout is a gravity driven water flow which works without any power. Say they didn't have powerful enough batteries, why not use a system gravity assisted system like the one shown in the video?

It is very difficult for us to know details about the CCCP rules and regulations applicable at Chernobyl.

I'm sorry, I don't get the implication behind this statement.

But even if they did use batteries, there would still be a need for reactor start/stop tests anyhow.

It's true that they would still have had to conduct a start/stop test, however, as soon as they had stopped the flow of water into the system simulating a blackout, instead of checking the amount of power the station produced and see if it was sufficient enough to keep the water pumps running, the would just turn on the battery source which would, hopefully, get the pumps running immediately preventing an uncontrolled increase in the reactivity.

 

[Phys12]

From further on in the Wikipedia article,

So they deliberately took the reactor outside of its safe domain.

From https://www.nei.org/resources/fact-sheets/chernobyl-accident-and-its-consequences:
What Happened - "The accident, which occurred at reactor 4 of the plant in the early morning of April 26, 1986, resulted when operators took action in violation of the plant’s procedures. Operators ran the plant at very low power, without adequate safety precautions and without properly coordinating or communicating the procedure with safety personnel."

Another point in the NEI article, "The reactor built at Chernobyl is a RBMK reactor, which was never built by any country outside the USSR because it had characteristics that were rejected everywhere outside the Soviet Union. Chief among these was its inherent instability, especially on startup and shutdown."

Oh, I see. So it wasn't just the fact that the reactor could not get adequate cooling when needed, but also that the operators deliberately ignored safety precautions needed to operate the reactor, in addition to the instability of the reactor inherent to how it was built.

 

[anorlunda]

the would just turn on the battery source which would, hopefully, get the pumps running immediately preventing an uncontrolled increase in the reactivity.

LOL. The word hopefully is seldom seen in nuclear safety procedures.

 

[Rive]

I did, this one:

A sole youtube link does not really feels like adequate attempt to reflect on a provided search keyword...
Anyway. SBO events are part of the design basis, up to a limit (try to find relevant regulatory requirements). There are batteries and other equipment to keep the plant in operational/controlled state in case an SBO, up to a limited length of time.
This does not include the main pumps, especially not the main pumps in RBMK style/size.
No battery is enough for that.

 

[artis]

Just to add, the RBMK is a rather unique design compared to most other commercial reactors in operation, well also more risky than others so requires extra caution and expertise which the operators at that fateful night shift in 1986 simply did not have. but even with safer technology human error and misjudgement can cause tragic consequences as there have been multiple meltdown's in history in smaller test reactors as well as larger and safer designs like the pressurized water reactor in Three Mile Island.All reactors require heat removal after shutdown for some time, chain reaction is stopped but the decaying of created fission elements continues, you can read up about this yourself in google.
Some reactors are more durable and can sit longer without coolant others less, but eventually all reactors need cooling or temps will rise to the level of fuel melting and causing damage.

get the pumps running immediately preventing an uncontrolled increase in the reactivity.

reactivity is mostly controlled by control rods and after a reactor is shut down by insertion of the rods the coolant flow does not affect reactivity, so the reason for the necessity of pumps is not reactivity but heat removal from the core otherwise the core will get damaged by heat.The RBMK is a very large reactor and it's coolant pumps are electric motors and together (each reactor has 8 pumps) 6 are active during normal operation , two are for reserve, I forgot the power rating for each of the pumps but put multiple in operation and that is MW (megawatts) so you would need a huge battery in order to power a station , even when shut down, with say 4 reactors for a minute on battery.
Making a water tank and use gravity as the power source is also next to impossible because the water required is a lot and the tank would have to be very very big and located higher than the reactor itself,
and RBMK reactor is already rather high, so this tank would have to be the height of a highrise/sky scrapper,

using the inertia of the turbine is the most practical and logical solution, and the test itself wasn't dangerous, it's just that the mistakes and blatant disregard the operators had combined with some inherent design risks caused the accident. but this is a topic on it's own and very long.

 

[Dr.D]

I am surprised to hear the 60 sec start time. That seems awfully long. Long ago, with the nuclear power industry was alive and well in the USA, the total time allowed on getting the pumps started on jobs that I worked on was 15 seconds. This was from the first signal to start, crank the engine and get speed stabilized, apply excitation to the generators and again get the frequency stabilized and close the breaker on the pump motors.

 

[anorlunda]

I am surprised to hear the 60 sec start time.

Russian technology differed from western technology in many ways. I learned to change my expectations from "east and west technologies are probably similar" to "probably different."

 

[artis]

It's hard to say why it took so long for the diesels, in all of the info online about RBMK's there is very little or next to none that I have read about the specifics of the diesels and their attached generators.

From what I read they had a problem with respect to covering the dead time between external power switch off and diesels switch on and they couldn't cover the full 60 second time span with just the inertia of the turbine/generator shaft.

@Dr.D can you elaborate what jobs where those were the pumps started in 15 secs? were they in construction or nuclear industry?

 

[Dr.D]

My experience was specifically with standby gensets for the American nuclear industry, although I cannot identify the specific plants.

 

[anorlunda]

This was from the first signal to start, crank the engine and get speed stabilized, apply excitation to the generators and again get the frequency stabilized and close the breaker on the pump motors.

But then the pump motors need some time to ramp up to full speed also.

Edit: And there may be additional time for some valves to open after the pumps reach full speed.

 

[Dr.D]

But then the pump motors need some time to ramp up to full speed also.

Edit: And there may be additional time for some valves to open after the pumps reach full speed.

Yes, there may be additional delays. I am still surprised at the 60 sec figure.

 

[whitehare]

Couldn't anyone reactor draw upon the power being generated by one or more of the other three (soon to be four) nearby reactors to power it for one minute while the diesel generators came up to speed? Perhaps the test made sense for a nuclear plant operating in isolation, but in this instance the motivation for the test seemed to ignore the obvious benefits of being co-located such a short distance from obvious sources of alternate power.

 

[artis]

Surely the test had very little to no chance ever being relevant or necessary, it was simply added to the operating manual by the reactor designer yet historical accounts show that most RBMK reactors came online without even attempting the test.
Chernobyl was no exception as the fourth block came online in 1983 yet they tried the test only months after.

Sure they could have used power from other neighboring reactors, they might as well have simply decreased their own reactor power to the minimum and use their own generated capacity to power the pumps, but the deal here is different, they simulated the very rare case of emergency when they both need to shut down the reactor and when offsite power is also lost simultaneously, imagine a super strong hurricane or earthquake you loose the incoming power and for safety you also need to shut the reactor.
In Ukraine there have never been earthquakes noticeable nor strong hurricanes or anything like that but as I said this was more of a "tick" box and all station directors and engineers wanted to have as good of a result to show higher authorities as they could get.

The irony here is that while trying to prevent a worst case scenario accident by improving safety they managed to create the worst case scenario accident by themselves on a perfectly calm and sunny spring morning,

PS. In fact the emergency cooling was designed for the reactor to be cooled under shut down conditions to remove decay heat, even if this failed the core would meltdown and cause some radioactivity release, what they did with the test is blow up the reactor under 10 to 100 times it's full max power which undoubtedly created an accident that was far worse than any core meltdown could have ever been under a shut down reactor condition.

 

[ElliotSmith]

I not 100% sure about this, but the same test that went horribly wrong at Chernobyl was performed on the other 3 reactors, but the results were inconclusive.

 

[artis]

yes they tried the test also in other NPP's elsewhere like Kursk and Smolensk. In theory the turbine/generator shaft under full RPM would have enough kinetic inertia to be able to power the core pumps in emergency reduced power mode for the timespan between shutdown and auxiliary diesels coming online although I have no real data or calculations for this and it is hard to find something on the web

 

[ElliotSmith]

yes they tried the test also in other NPP's elsewhere like Kursk and Smolensk. In theory the turbine/generator shaft under full RPM would have enough kinetic inertia to be able to power the core pumps in emergency reduced power mode for the timespan between shutdown and auxiliary diesels coming online although I have no real data or calculations for this and it is hard to find something on the web

I assume that the very last time that this test was ever performed was on Chernobyl unit 4?

 

[artis]

Well I have no information on that but I would also think that this is the case.

Well normally this test by itself is/was not dangerous, all they have to do is to stop steam supply to the turbine and at the same time shut down the reactor (because a reactor under load produces lots of steam which now have nowhere to go)
Normally this happens automatically if for some reason both turbines fail for example the reactor is scrammed/stopped , the problem arose because leading up to the preparations for the test a couple of important mistakes were made(although unknown to the operators) and then when finally shutting down the reactor it's design flaws completed the circle of mistakes and the results we all know.

 

[Nik_2213]

"The word hopefully is seldom seen in nuclear safety procedures. "

IIRC, it appears just once, when UK designers added those 'funny' filter assemblies to the top of Windscale's stacks...

And, yes, they did mitigate the core fire's mayhem...

 

[franklyamazed]

LOL. The word hopefully is seldom seen in nuclear safety procedures.

Yes, Exactly, the accident happened because they 'hoped' it would work!

 

[franklyamazed]

That's so obvious, we must believe that if they could have used batteries, they would. We are talking about very big pumps.

It is very difficult for us to know details about the CCCP rules and regulations applicable at Chernobyl.

But even if they did use batteries, there would still be a need for reactor start/stop tests anyhow.

There were 8 main cooling pumps at Chernobyl Unit 4, designed in two cooling circuits, 4 MCP's for each. 3 MCP's were on duty for each circuit with 1 used in reserve. So six working, 2 on standby.

These pumps were massive, powered by 6 kV 5.5 Megawatt electric motors moving 7,000 to 12,000 cubic meters per hour. So, 33 Megawatts of power required just to run the 6 duty pumps. That would be a lot of battery power. Batteries were used to backup the control systems and essential controls, such as the valves for the emergency cooling system.

 

[franklyamazed]

I did, this one: .
Yes, but if you look at the video I quotes above, one of the backups used by the plant during a total blackout is a gravity driven water flow which works without any power. Say they didn't have powerful enough batteries, why not use a system gravity assisted system like the one shown in the video?
I'm sorry, I don't get the implication behind this statement.It's true that they would still have had to conduct a start/stop test, however, as soon as they had stopped the flow of water into the system simulating a blackout, instead of checking the amount of power the station produced and see if it was sufficient enough to keep the water pumps running, the would just turn on the battery source which would, hopefully, get the pumps running immediately preventing an uncontrolled increase in the reactivity.

That video is an advert for Canada's Candu Reactors, and I may be wrong, but I think gravity/convection emergency cooling was designed later.

 

[artis]

I don't know for sure but as a sidepoint I think convection circulation (circulation due to heat differences in different parts of the primary coolant circuit) is mainly possible for cores where the core itself is vertical , that would include a PWR a BWR and also RBMK because each channel is like a small core with its own fuel assembly/heat source and water would tend to rise to the top as it gets hotter even without forced circulation,
Does this work for CANDU cores?

because they are horizontal and from what I imagine the water would get heated evenly across the length of each pressure channel since each channel is horizontal,
i couldn't find a real schematic of a CNADU primary loop but from what I imagine both the inlet and the outlet tubes connecting to each of the horizontal pressure channels are at the same height for the same pressure channel at each end of the core.

Any expert knowledge on this issue? I find it interesting.

 

[Astronuc]

real schematic of a CANDU primary loop

See slide 9 in http://sites.ieee.org/npec-sc2/files/2017/06/SC-2Mgt03-2_Att6-Faya-CANDU.pdf
Outside the core, the CANDU primary system looks very much like a PWR system, which it is more or less, but with horizontal flow channels in the core.