Accident at kakrapar atomic power station 2016

In summary, there were reports of a coolant leak at the Kakrapar atomic power station on March 11th, 2016. The plant's emergency core cooling system was activated to maintain adequate cooling, and containment venting may have occurred. This incident has potential impacts on all PHWR/CANDU type plants, as the cause of the coolant channel failure is still unknown. The pressure tubes in CANDU reactors have had problems with delayed hydride cracking, and the Kakrapar units have been in operation since 1993 and 1995.
  • #1
Hiddencamper
403
182
Hi everyone.

On March 11th 2016, there were reports of an alleged coolant leak at the kakrapar atomic power station. The reactor is an Indian PHWR.

Articles at the time suggested the first trip on the reactor was high containment pressure, which signaled to me some sort of loss of coolant accident (LOCA). Articles in the last few days confirm that there was a definite steam condition in the reactor compartment, with cameras in the area having nothing but a wall of steam.

While the unit is now in "cold shutdown", defined as average reactor coolant system temperature less than boiling point, it appears the plant's emergency core cooling system (ECCS) was required to ensure adequate core cooling was maintained. Additionally there are reports that containment venting occurred, although radiation levels in a 5 km radius were not seen to increase to an extent indicating fuel failure.

This could have potential impacts to all PHWR/CANDU type plants, as the cause for the rapid failure of a coolant channel is unknown at this time. In the mid 80s one of the CANDU units at Pickering had a similar failure, and the coolant channel tube materials and manufacturing/installation methods were changed to prevent this from recurrence.

The unit at kakrapar atomic power station had its coolant channels refurbished in 2011, so this failure was after 5 years, and was allegedly a rapid failure.

The purpose of this post is to spur some discussion on the event, link articles as we find them, and hopefully have an interesting thread like we had with the Fukushima accident.
 
  • Like
Likes DrDu
Engineering news on Phys.org
  • #2
There are a number of news articles discussing the accident. Here is one from Nuclear News.

Since not everyone on the board would be familiar with CANDU style reactors I'll give a bit of information below:
- Pressure tubes ~10 cm diameter contain ~10 MPa pressure D2O coolant and fuel bundles together referred to as a channel.
- Adjacent channels have the coolant and fuel moving in opposite directions to 'average' the behavior.
- The coolant for each channel is supplied separately by a 'feeder' at each end of the channel (note there are therefore 712 feeder pipes for this particular reactor)
- The feeder pipes collect at inlet and outlet headers which are circulated to and from steam generators

There are a couple of things that make loss of coolant accidents interesting in CANDU type reactors:
- Positive void reactivity coefficient. If you loose coolant in a single (or few) channels, the reactor power wants to climb initially, which can potentially make the situation worse. This is countered by a negative fuel temperature coefficient which eventually dominates over the void effects limiting the size of the power pulse.
- Because each channel has independent feeders on both ends, the consequences of a single channel break are somewhat limited. This means the whole reactor doesn't lose coolant, just a small subsection of it. In this case 1/306 of the fuel in the reactor. If the fuel gets really hot and melts (which is unlikely) it will fall into the moderator which can supply a massive amount of heat capacity. Depending on how and where the leak occurs there may be damage to other near by fuel channels or it could be limited to a single area.
- In the past a CANDU fuel channel failed during operation due to hydrogen uptake. A pressure tube ruptured along the length leaking coolant into the space between the pressure tube and calandria tube (provides a gap between pressure tube and moderator to prevent heat-loss to the moderator). The reactor was repaired and returned to service.
 
  • #3
I take it as a failure of a pressure tube, which according to the nuclear new article is composed of a Zr-Nb alloy, probably Zr-2.5 Nb, as opposed to Zr-2 (Zircaloy-2).

The Nuclear News article does not mention the location of failure, e.g., internal to the core or at the end fittings.

Pressure tubes have had problems with delayed hydride cracking (DHC) in which temperature cycling allows for cyclical precipitation and dissolution of Zr hydrides resulting in the development of cracks in the pressure tube.

Zr-Nb alloys generally tend to have lower corrosion rates and lower hydrogen pickup fractions than Zr-2, which is a principal reason that Zr-2.5Nb replaced Zr-2 as the material of choice for pressure tubes. Zr-Nb also has a lower rate of irradiation growth than Zr-2.
 
Last edited:
  • #4
One of the Indian press reports said the failure was in a feed pipe. Unfortunately there was no further detail as to how this relates to the overall installation.
 
  • #5
  • #6
Thank you for this excellent summary document.
One thing that leaps to the eye is that the design requires a very large number of small pipe joins. With the pipes made of special alloy and the feed lines converging on the very compact calandria core (see P16 of the report), it poses a very demanding manufacturing quality control challenge. Even verifying the joins is difficult. If the cause was indeed a feed pipe failure, it will not be an easy fix.
 
  • #7
Each cooling channel (of 306 total) has two joints, for a total of 612 joints.

Each component, i.e., vessel and piping, that serves as a 'coolant pressure boundary' is made to nuclear grade specifications. Erosion and corrosion are two material degradation issues that must be monitored over the service time of each component. Corrosion is inherent, but erosion generally requires some threshold coolant velocity depending on turbulence. Pipe bends and junctions can be more susceptible. Thermal fatigue is another concern.

Kakrapar units 1 & 2 went into commercial operation in 1993 and 1995, respectively. They are rated at 235 MWe gross (220 MWe net), with a thermal rating of 801 MWt with 3672 assemblies/bundles. World Nuclear Association reports each unit at 220 MWe gross (202 MWe net).
http://www.world-nuclear.org/information-library/country-profiles/countries-g-n/india.aspx

The plant is in Gujarat state on the western side of India.

Dr A Gopalakrishnan, who headed the Atomic Energy Regulatory Board from 1993 to 1996, said the situation in Kakrapar might be more serious than what we are being told.
http://news.yahoo.com/nuclear-leak-kakrapar-may-more-060000162.html

The leak is somewhere in the Primary Heat Transport (PHT) system, which would include the pressure tubes in the core, the headers, and piping between the core and steam generators, and the tubing in the steam generators.

It is most likely that one or more pressure tubes (PT) in the reactor (which contain the fuel bundles) have cracked open , leaking hot primary system heavy-water coolant into the containment housing .
http://www.dianuke.org/kakrapar-nuc...oing-loss-coolant-accident-dr-gopalakrishnan/

I would think it be rather straightforward to determine if the leak was through a pressure tube into a calandria tube. If outside the core, it could be a leak in a stainless steel component, which might imply stress corrosion cracking (perhaps from attack in sulphide inclusions, or sensitized material) or thermal fatigue, or some combination.note by Dr. Gopalakrishnan (http://www.countercurrents.org/sundaram130316.htm):

The Kakrapar Unit-I nuclear reactor in Gujarat is undergoing a moderately large leakage of heavy water from its Primary Heat Transport (PHT) system since 9.00 AM on March 11,2016. From the very limited information released by the Nuclear Power Corporation of India Limited (NPCIL) and the Atomic Energy Regulatory Board (AERB) of the government , as well as from the conversations I had with press people who have been in touch with nuclear officials, few inferences can be drawn.

Till 7.00 PM on March 12,2016 , the DAE officials have no clue as to where exactly the PHT leak is located and how big is the rate of irradiated heavy water that is leaking into the reactor containment . However, some reports indicate that the containment has been vented to the atmosphere at least once , if not more times , which I suspect indicates a tendency for pressure build up in that closed space due to release of hot heavy water and steam into the containment housing . If this is true, the leak is not small , but moderately large , and still continuing.
 
  • #8
The Times of India March 20th report on the incident includes this comment by Sekhar Basu, chairman of Atomic Energy Commission, who said a "feeder piper carrying heavy water leaked and all radioactivity was confined within the reactor building".
I would certainly expect that the feeder pipes, which have several bends to reach their assigned pressure tube, would be more vulnerable than the actual pressure tubes. The welds joining the feeders to the pressure tubes would be another potential weak point. With hundreds of such tubes and pipes involved, ensuring adequate quality is not easy.
 
  • #9
Seems the accident is now in a somewhat messy cleanup stage, but was non trivial.
It involved spillage of 70 tons of heavy water, according to this report: http://www.dianuke.org/nuclear-accident-gujarat-enters-third-week-lethal-gamble-continues/
The documents author is a prominent anti nuclear figure in India, so his views surely color it, but his skepticism about the official statements seems warranted.
I had not known that the CANDU design had positive void reactivity and was so relatively graceless in failure.
 
  • #10
etudiant said:
Yes - the article is problematic.

For example, the statement " These reactions are taking place in timescales of nano to micro seconds and human reaction times are too slow for effective action," while accurate with respect to moderation, it fails to acknowledge the delayed neutrons, which are delayed by fractions of a second up to about 80 seconds. Delayed neutrons are important with respect to power maneuvering since it allows one to change power over several minutes with slight increases in reactivity.

Secondly, "If both these shutdown systems had not automatically worked for any reason, it would have been goodbye to South Gujarat." No, this is not correct. Failure of both system would have endangered the plant, but not South Gujarat. The statement is disingenuous and irresponsible.

The following statement indicates the author is not careful with proofreading.
The coolant heavy water is kept under pressure to prevent boiling. This heavy water is circulated through the reactor in 306 pressure tubes. The temperature inside these pressure tubes is around 2500 to 2900 centigrade.
I believe he meant 250 to 290°C. The melting point of Zr-alloys is around 1850°C, so the temperature is certainly not going to be 2500°C to 2900°C. Uranium dioxide melts at around 2840°C, so they certainly are not going to operate a core with molten UO2.

Earlier in the article, it is correctly stated, "The temperature of the coolant inside the core varies from 249oC at inlet to 293oC at the outlet. As no boiling is allowed inside the core, pressure is maintained at 87 kg/cm2."
 
Last edited:
  • Like
Likes mheslep and mfb
  • #11
Thank you, Astronuc, for your considered comment, which injects more realism into the discussion. The event still leaves an uneasy feeling though.

Presumably the venting of heavy water steam as happened is radiologically benign provided the fuel remained intact, as it was claimed to be the case here.
Still, given the multitude of high pressure pipe joins inherent in the CANDU design, it seems like an accident that could easily repeat. If in another such instance all does not go well and the ECCS does not work properly, it seems potentially very messy.
 
  • #12
Based on the statements by KAPS site director L K Jain, this is a PT break not a feeder break. The bundles in the channel were removed and were all in tact; so no fuel bundle damage. We need further metallurgical data on the cause of the PT break. This will have an impact for all operating CANDU reactors.
 
  • #13
The only statement by LK Jain that I could find was his Mar 11 disclosure of a 'small leak in the Primary Heat Transport system'.
As the site manager, LK Jain should be the best informed, but his characterization of a 70 cubic meter leak as 'small' suggests some bias.
Is there any subsequent report that adds more specific detail?
Otherwise I'd consider the statement implicating a feeder line made by Sekhar Basu to the Times of India on March 20 as most authoritative public comment.
 
  • #16
Thank you, that is a more recent disclosure and it does specifically point to one of the coolant channels, rather than a feeder line.
Does rather pose a concern about the risks at similarly designed installations elsewhere.
 
  • #18
Does CO2 have some particular use in a PHWR? Or are they referring to atmospheric CO2?
 
  • #19
mheslep said:
Does CO2 have some particular use in a PHWR? Or are they referring to atmospheric CO2?
While responding to the thread on nuclear reactors with channels, I found an article that mentioned that CO2 is used between the pressure tube, in which the fuel is placed and through which coolant flows, and the calandria tube. The calandria tube keeps the heavy water moderator contained in the calandria.

From the NDTV article cited by etudiant
The AERB then ordered that all the tubes made out of a special alloy of zirconium-niobium be checked on the outside, to their surprise, they discovered that the contagion of the "nodular corrosion" or what in layman's language can also be described as "small pox-like" was very widespread in many of the 306 tubes.
Nodular corrosion is a type of localized corrosion, usually in water cooled systems with susceptible materials and certain aggressive water chemistries. For it to happen on the 'dry' side of the pressure tube is interesting. The pressure tube is the hotter surface with respect to the calandria tube.

I'm wondering if it was localized hydriding, which could make the Zr-Nb alloy susceptible to local oxidation. Later on in the article:
The source of the carbon dioxide was further back traced and it seems only the Kakrapar plant was sourcing its gas from a "Naptha cracking unit" and possibly it has some contamination of hydrocarbons.
Hydrocarbons would be a source of hydrogen.

Some CANDU references
http://www.iaea.org/inis/collection/NCLCollectionStore/_Public/28/021/28021870.pdf
https://canteach.candu.org/Content Library/20044203.pdf
 
  • #20
Astronuc said:
found an article that mentioned that CO2 is used between the pressure tube...
Okay, though it seems odd choice instead of say, nitrogen, argon.
 

1. What happened at the Kakrapar Atomic Power Station in 2016?

In March 2016, there was an accident at the Kakrapar Atomic Power Station in Gujarat, India. A leak was discovered in the primary coolant channel of one of the reactors, leading to a partial meltdown of the reactor core.

2. Was anyone injured or killed in the accident?

Fortunately, no one was injured or killed in the accident. The Nuclear Power Corporation of India Limited (NPCIL) reported that all safety protocols were followed and the radiation levels at the site were within safety limits.

3. What caused the accident at Kakrapar Atomic Power Station?

The exact cause of the accident is still under investigation, but it is believed to be due to a failure in the reactor's coolant system. The primary coolant channel was damaged, leading to a loss of coolant and subsequent overheating of the reactor core.

4. Was there any environmental impact from the accident?

According to NPCIL, there was no significant environmental impact from the accident. The radiation levels at the site were within safety limits and there was no release of radioactive materials into the environment. However, a thorough environmental assessment is still ongoing.

5. What measures have been taken to prevent future accidents at Kakrapar Atomic Power Station?

Following the accident, the reactor was shut down and a full inspection of the plant was conducted. The damaged reactor is being repaired and upgraded with additional safety features. NPCIL has also implemented new safety protocols and increased training for their employees to prevent future accidents.

Similar threads

  • Nuclear Engineering
Replies
15
Views
3K
  • Nuclear Engineering
Replies
5
Views
1K
Replies
2
Views
2K
  • Nuclear Engineering
Replies
2
Views
2K
Replies
14
Views
8K
  • Nuclear Engineering
Replies
27
Views
7K
Replies
6
Views
3K
  • General Discussion
Replies
29
Views
7K
  • Art, Music, History, and Linguistics
Replies
1
Views
999
Replies
10
Views
2K
Back
Top