Was the Nuclear Control Room Panel Fire at Quad Cities Caused By a Defective GE-Hitachi Switch?

Tags

BWR, cancer, Chernobyl, control panel, control room fire, dangers of nuclear, environment, fire, Fukushima, GE, GE-Hitachi, Japan, NRC, nuclear, nuclear accident, nuclear disaster, nuclear energy, nuclear industry, nuclear power, nuclear power plant, nuclear reactors, nuclear safety, risk management, torus, US NRC, USA

US NRC Photo: Control Room Panels (unknown nuclear power station), Nov-2007; black and white photo of Quad Cities Control Rm at Blog Post Bottom

At Quad Cities Nuclear Power Station: “On February 1, 2017, at 1929 hours [CST], a fire was discovered on the Unit 2 Main Control Room panel 902-3 in the 3E ERV/ADS [Electromatic Relief Valve-Automatic Depressurization System ] valve switch… No automatic isolations/actuations occurred. The fire was extinguished at 1932 and the reactor remained at 100% power. “An Alert was declared at 1938 [CST].”The initiation of the event was attempting to change a light bulb. The cause of the event is under investigation. ” https://www.nrc.gov/reading-rm/doc-collections/event-status/event/2017/20170202en.html The control room is just that – it is to control the nuclear reactors. This is not good. Hitachi centers around the Mizuho group, in trouble a few years ago for lending to the Japanese mob (yakuza): http://www.japantimes.co.jp/news/2013/11/09/business/hitachis-consumer-credit-unit-caught-up-in-mizuho-yakuza-loan-scandal

The fire was in the Electromatic Relief Valve (ERV)-Automatic Depressurization System (ADS) Switch. The relief valve discharges steam to the suppression pool (torus) of the containment, which is important since the steam in a BWR (boiling water reactor) is radioactive, and the radioactive water can be filtered. This keeps the safety valves from opening, which would vent radiation to the atmosphere. See p. 4: http://www.ewp.rpi.edu/hartford/~ernesto/F2011/EP/MaterialsforStudents/Petty/Dresser.PDF

Below, on the right, is the Chernobyl exclusion area centered on the Quad Cities Nuclear Power Station. This is exclusion area only. Chernobyl has seriously impacted areas over 1000 miles away in the UK, Finland and Scandinavia. The US do not eat food fallout zone officially extends out 50 miles and includes an estimated 674,951 people for Quad Cites. See population estimates and zones here: http://www.psr.org/resources/evacuation-zone-nuclear-reactors.html

“To give an idea of the scale of the abandoned area, this map shows how it compares to other places on the planet. See how the Chernobyl exclusion area matches to a city or region you know.“. http://www.greenpeace.org/international/en/campaigns/nuclear/nomorechernobyls/exclusion-zone-comparison-map/

The AP is in trouble for reporting that a milk sample in Belarus, near the Chernobyl exclusion zone, contained 37.5 Bq/kg (l) of Strontium 90, compared to the 3.7 Bq/kg allowed by the government. https://nuclear-news.net/2017/01/03/belarus-nuclear-power-police-state-critique-by-ian-goddard/. Absent from the discussion is the fact that the US allows 160 Bq/kg of Strontium 90 in its food all of the time, since 1998! So, when they tell you that food is ok after an accident, you will know what they mean! 43 times higher than Belarus allows and over 4 times more than near Chernobyl! http://www.fda.gov/downloads/MedicalDevices/…/UCM094513.pdf

Instead of waiting to see the US NRC report, if they ever make one, we speculate on what could have caused the problem – a known defective GE-Hitachi part. They tell us that it was a fire in the main control room panel in the ERV/ADS valve switch. It so happens that a related switch is potentially defective. This could have theoretically caused a short circuit in the panel. This is about Dresden, but Quad Cities has the same reactor types, and is named as having the same potentially defective parts. Regardless, the reading further below gives an idea of the importance of the ERV-ADS switch which caught fire.

“PART 21 REPORT – ELECTROMATIC RELIEF VALVE (EMRV) CUTOUT SWITCH

“Following the return of the actuator that failed bench testing to GEH, on 6/12/15 at 1804 [CDT], General Electric Hitachi (GEH) notified Dresden Station of a potential parts quality Potential Failure of the EMRV Cutout Switch. It has been determined the notification is applicable to DNPS [Dresden Nuclear Power Station], Units 2 and 3.

“The GEH investigation concluded that the EMRV actuator assemblies failed to change state because of the failure of the cutout switch to fully close and provide the appropriate current path. Multiple contributing factors were discovered which could have led to the presence of the gaps in the cutout switch. The most significant of these factors is a change in lever arm positioning causing increased forces in the tension spring which prevent proper closure of the cutout switch. Design changes to reduce wear caused by vibration on the actuators changed lever arm position and also allowed for additional dimensional tolerance which tended to increase force in the tension spring.

“Identification of Facility and Component: DNPS / EMRV Actuator, GEH Part Number 352B2632G001
“Safety Significance (e.g., substantial hazard that is or could be created): Identified condition is a

Potential Substantial Safety Hazard since it could cause affected EMRVs to fail to operate as designed, which could result in a loss of safety function. Potential to affect the Minimum Critical Power Ratio (MCPR), Reactor Coolant System (RCS), Automatic Depressurization System (ADS), and Low Set Relief “Plants with similar GEH cutout switches: Quad Cities Nuclear Power Station.” Part 21 Reference: EN #51386” https://www.nrc.gov/docs/ML1528/ML15280A387.pdf

The Quad Cities fire was in the control panel for the Electromatic Relief Valve (ERV)-Automatic Depressurization System (ADS) Switch. The electromatic relief valve (ERV) discharges steam to the suppression pool (torus) of the containment, which is important because the steam in a BWR (boiling water reactor) is radioactive, and the radioactive water can be filtered. This keeps the safety valves from opening, which would vent radiation to the atmosphere. See p. 4: http://www.ewp.rpi.edu/hartford/~ernesto/F2011/EP/MaterialsforStudents/Petty/Dresser.PDF

“During the theoretical leakage design basis accident the reactor coolant flashes to steam in the drywell, pressurizing it rapidly. Vent pipes or tubes from the drywell direct the steam below the water level maintained in the wetwell (also known as a torus or suppression pool), condensing the steam, limiting the pressure ultimately reached.” https://en.wikipedia.org/wiki/Containment_building

More details are given here for Dresden, which like Quad cities has GE Boil Water Reactors (BWR)-3:
“RESIDUAL HEAT REMOVAL ALTERNATIVES
Electromatic Relief Valves. (Each Electromatic Relief Valve will remove approximately 140 MWth.)
(1) Verify Suppression Pool water level is > 6 feet.
(2) Place LPCI in Suppression Pool cooling (DOP 1500-02).
(3) Open one or more Electromatic Relief Valve(s) as necessary to reduce reactor pressure/temperature, while maintaining the cooldown rate below 100F/hr.
(4) Alternate opening of Electromatic Relief Valve(s) at five (5) minute intervals, in the following sequence to minimize local torus water heating when possible: A, C, E, D, B.
NOTE: The Suppression Pool water level should be above six (6) feet to ensure exhausted steam from the Electromatic Relief Valves is condensed in the Suppression Pool water” https://www.nrc.gov/docs/ML1032/ML103220475.pdf https://www.nrc.gov/docs/ML1034/ML103480873.pdf

More of the US NRC Notice about the fire:
“ALERT – FIRE IN SAFETY RELATED SYSTEM
“On February 1, 2017, at 1929 hours [CST], a fire was discovered on the Unit 2 Main Control Room panel 902-3 in the 3E ERV/ADS valve switch. A reactor SCRAM was not required. No automatic isolations/actuations occurred. The fire was extinguished at 1932 and the reactor remained at 100% power.

“An Alert was declared at 1938 [CST].

“The initiation of the event was attempting to change a light bulb. The cause of the event is under investigation.

“The Senior Resident Inspector has been notified of the event.”

The licensee entered a 14 day Technical Specification Action statement as a result of the damage to the switch.

Notified DHS SWO, DOE, FEMA, HHS, NICC, USDA, EPA, FDA, NNSA (e-mail), and NRCC SASC (e-mail).

* * * UPDATE AT 0040 EST ON 2/2/17 FROM DAVID KNEPPER TO JEFF HERRERA * * *

Notified that the Alert was terminated at 2336 CST on 2/1/17. The licensee stated that the fire was extinguished and an extent of condition walkdown did not identify any additional equipment damage as a result of the fire. The licensee will be issuing a press release.

The licensee will be notifying the NRC Resident Inspector.

Notified the R3DO (Duncan), IRD MOC (Gott), NRR EO (Miller), DHS SWO, DOE, FEMA, HHS, NICC, USDA, EPA, FDA, NNSA (e-mail), and NRCC SASC (e-mail).” https://www.nrc.gov/reading-rm/doc-collections/event-status/event/2017/20170202en.html

Details of the GE Hitachi Electromatic Relief Valve Automatic Depressurization System actuator defect, as submitted to the US NRC from GE-Hitachi:





September 10, 2015 MFN 15-042 R2 “U.S. Nuclear Regulatory Commission Attn: Document Control Desk Washington, DC 20555-0001 Subject: Closure of Part 21 60-Day Interim Report Notification: Potential Failure of Electromatic Relief Valve Cutout Switch” https://www.nrc.gov/docs/ML1525/ML15253A859.pdf

Excerpted from:
“General Electric Advanced Technology Manual Chapter 6.8
Safety Relief Valve Differences”
“6.8.6 Electromatic Relief Valve
Electromatic relief valves are provided to prevent lifting of the safety valves on early BWR product lines. The electromatic relief valves (Figure 6.8-7) are actuated by the solenoid assembly. Steam enters from chamber (A) and passes upward around disc guide to chamber (B). Steam enters chamber (C) through clearance space between the main valve disc and disc guide. The main valve is held in the close position by the steam pressure in chamber (C). Chamber (D) connects the main valve assembly to the pilot valve assembly. With the pilot valve in the closed position, chamber A, B, C, and D are at system pressure. The pilot valve is held closed by the pilot valve spring and the steam pressure in chamber (E). When the solenoid is actuated the plunger head moves downward striking the pilot operating lever. The operating lever opens the pilot valve disc, thus allowing steam from chamber (D) to vent through port (F). The steam is vented from chamber (D) faster than it is supplied to chamber (C) via the clearance between the valve disc and the disc guide. The resultant unbalance of pressure in chamber (B) and (C) produce a force that moves the main valve disc downwards and allows steam to escape. If the actuating signal resets, the solenoid plunger retracts and the pilot valve reseats. The pressure in chamber (C) increases causing the relief valve to reclose. https://www.nrc.gov/docs/ML1126/ML11263A362.pdf

The PORV and Electromatic Relief Valve have similar functions. PORV is apparently for Pressurized Reactors (PWR) like Three Mile Island, and the Electromatic Relief Valves for BWRs like Quad Cities. Problems with the PORV and a light in the control room panel was a major cause of the Three Mile Island nuclear disaster.
“… 00:00:03
Pressure Relief valve (PORV) automatically opens
When the pressure of steam in the reactor core rises above safe limits, the pressure relief valve is designed to automatically open, releasing the excess steam to a containment tank
[…]
00:00:09
Boron and Silver control rods are lowered into the reactor. PORV light goes out, indicating valve is closed.
…. When the PORV light goes out the operators incorrectly assume that the valve is closed. In reality the valve is not only open but is also releasing steam and water from the core. This is now a LOCA (Loss of Coolant Accident)…” “Timeline of Events Occurring During the Three Mile Island Disaster” Online Ethics Center for Engineering 2/16/2006 National Academy of Engineering Accessed: Thursday, February 11, 2016 <www.onlineethics.org/Resources/tmiindex/time.aspx

David Lochbaum writes about the problem of Fires at Nuclear Power Stations: http://allthingsnuclear.org/dlochbaum/fire-at-the-nuclear-plant

This is excerpted from discussion about fire risk at nuclear power stations, and secondary problems which one might not think of:
“For the purposes of illustration, the following discussion focuses on internal or intra-cable shorting behavior in multi-conductor cables to the exclusion of external or inter-cable shorting behaviors. Inter-cable shorting behaviors would face similar considerations.

The input data provided to the EPRI expert panel focused on the likelihood that either of two specific target conductors experienced a hot short from either of two energized source conductors for the surrogate motor operated valve (MOV) control circuit tested. A Fire PRA is typically interested in a somewhat different question. For example, given a MOV where the desired position is “open”, the risk-relevant question would be “what is the likelihood that the valve spuriously closes?” In the context of the experiments, this would correspond to a hot short energizing the one target conductor representing the “close” side of the surrogate MOV control circuit. While the probability of a valve going closed is likely lower than the spurious actuation probability promulgated by the expert panel, one cannot simply divide the expert panel value by two. That is, one cannot assume that a hot short impacting the “close” target conductor is half the likelihood of impacting either the “open” or “close” target conductors. In many of the tests, if one target conductor experienced a hot short then both target conductors experienced hot shorts. This implies a dependency that would not be properly quantified by a simple “divide by two” solution.

Now consider a case cited during the peer review process. The cited case involved a specific circuit configuration on a normally open MOV. Circuit analysis found that if the “open” conductor experienced a hot short first, then permanent valve damage could occur because the fault would bypass the valve limit switches, over-driving the valve into its “open” seats. In this case, the valve would remain open until repaired. However, if the valve experienced a hot short on the “close” conductor first, the valve would close with no valve damage because the limit switches remain effective for this fault configuration. For this example, the questions of hot short timing, and potentially duration, come into play. These subtleties are not currently captured by the methods recommended in Tasks 9 and 10.

As a final illustrative example, consider a case involving a normally closed high/low-pressure interface MOV. Depending on the overall piping configuration and likely given other concurrent failures, a hot short that even partially opened the MOV might lead to permanent damage and rupture of the low pressure piping. In such a case any prior or subsequent shorting behavior and considerations of hot short duration would be essentially irrelevant to the risk calculation. For such a case, the expert panel results might well be the most appropriate interpretation of the available data.

The above examples illustrate the complexity of the problem. Refinement of any given spurious actuation likelihood value requires a clear understanding of the available data, consideration of the specific circuit configuration, and a clear understanding of the Fire PRA application and scenario implications. ” https://www.nrc.gov/reading-rm/doc-collections/nuregs/contract/cr6850/v1/cr6850v1.pdf “Fire PRA Methodology for Nuclear Power Facilities Volume 1: Summary & Overview EPRI 1011989 NUREG/CR-6850, Final Report, EPRI/NRC-RES
This Fire document isn't as useful as it might be because its apparent function is to prove that fire risks aren't that great.

US DOE, HD.6D.125, Central Control room of Quad-Cities nuclear power plant at Cordova, Illinois, 2013 photo.