This group brings together the best thinkers on energy and climate. Join us for smart, insightful posts and conversations about where the energy industry is and where it is going.


Tour of NuScale Control Room and Test Facility

Disclosure: I have a small contract to provide NuScale with advice and energy market information. That work represents less than 5% of my income for 2014.

On October 20, 2014, I had the opportunity to visit NuScale’s facilities in Corvallis, OR. Though the company now has offices in three cities, Corvallis, the home of Oregon State University, is the place where the NuScale Power Module has been conceived and refined.

Unlike the other participants in the race to develop small modular reactors that can be licensed and sold in the United States, NuScale is a focused start-up company with many similarities to other high tech start-ups. It grew out of a university research project, has been through several rounds of capital raising, maintains a working relationship with the university that initially nurtured its development, and has located its offices in an available vacant building with a motivated landlord.

Front door of building housing NuScale's Corvallis design office

Front door of building housing NuScale’s Corvallis design office

In NuScale’s case, the available building was in excellent, move-in condition. It is part of a multi-building campus owned by HP that was once a bustling center for designing and manufacturing personal printers and their associated ink cartridges. In 2011, NuScale was one of the first tenants to occupy space in its building, which was already set up as an engineering design office. NuScale’s occupancy was part of a local reuse program. Desks, cubicle dividers, and even chairs were ready and waiting for the new creative occupants.

My first stop was in Jose Reyes’s office for a quick update. Jose is NuScale’s Chief Technical Officer and one of the company’s cofounders. I’ve known him for several years, our first official interaction was when he and Paul Lorenzini were my guests on Atomic Show #100 in August 2008.

Reyes began by telling me that NuScale’s head count is up to 380, including contractors serving as staff augments. About 200 of those work in Corvallis. As a fully owned subsidiary of Fluor, NuScale receives ongoing investments through the budgeting process. There are numerous positions still open; the company has recently been adding several people every week.

So far, the total project spend has been about $230 million. NuScale and the US Department of Energy recently finalized their cost-sharing agreement through which the DOE will provide grants of up to $217 million over a five year period to assist in defraying the additional costs that are imposed on the leader of any new nuclear technology development, subject to annual appropriations.

In the US, the first-of-a-kind leader must pay the Nuclear Regulatory Commission costs associated with reviewing and producing positions on new concepts. All followers get to use those established positions without incurring the $279 per professional staff hour fees or having to pay its own people to devise the concepts and provide the technical justification that results in NRC approval. In the case of NuScale, some of the ground breaking concepts include natural circulation and control room designs that enable different staffing concepts than used in existing reactors.

NuScale’s testing and licensing program are moving along. During the next year, Reyes expects that his team will be submitting a number of topical reports to the Nuclear Regulatory Commission. In NRC lingo, there is a significant difference between a topical report and a technical report. Companies submit technical reports as a way to keep the NRC informed about various aspects of a design, but the NRC is under no obligation to comment or respond.

In contrast, the NRC reviews and comments on topical reports. If appropriate, the NRC may formally accept a topical report as suitable to be used in a licensing and provide limiting conditions under which the report may be used. The topical report process will provide the company with better understanding of the NRC’s position on certain key issues.

According to Reyes and the schedule that has been submitted to the NRC, NuScale should be ready to submit a high quality application before the end of 2016. Since the company first notified the NRC that it intended to submit an application for a design certification in 2008, it should be apparent that working through the licensing process in the United States is no job for people with an impatient, “git ‘er done” mentality. It should also be apparent that the current situation must be improved.

NuScale has begun developing its supplier base. As a Fluor subsidiary, it has access to a large, world-wide supplier network, but some of the components of the NuScale Power Module will require expansion of that supplier base. Reyes indicated that the company believes there is sufficient capacity for key items for building one or two power stations in a reasonable period of time, but additional investments in facilities will be required to enable a higher throughput.

The company has hosted a couple of supplier days already; most of those have taken place in the northwest US, which is where NuScale intends to concentrate its deployment efforts.

NuScale control roomThe next stop on the tour was a visit to the control room prototype. It is in a large room with the same footprint as will be available in the actual power stations. The ceiling is a bit lower; the room is limited by a ceiling height available in a commercial office building.

The main control panels are arranged in a horseshoe configuration with thirteen individual desks and screen groups. There is one desk/screen group for each unit and a larger one to operate and monitor shared systems like circulating water and the large pool in which all of the modular containments will reside.

One of the primary goals of the facility is to develop the concept of operations and human factors program to support the rule exemption request that the company will need to submit. The current rule that specifies plant operations has no provisions for more than 3 units on a site or more than 2 units controlled from a single control room. NuScale is not yet certain how many operators it will need, but its initial position is to attempt to reduce the work load to the point where it is manageable by an operating crew that is the same size as the one used at existing nuclear facilities.

Because I have not signed any non-disclosure agreements with the company, the screens NuScale showed me during my visit were generic, but they provided a good understanding for the direction that the company is taking to streamline operations, automate functions where desirable and provide clear, understandable indications.

NuScale DisplaysMy tour guide and I had a good discussion about NuScale’s plan to provide intelligent alarms that do not overwhelm operators with unnecessary noise and distraction. As an example, he mentioned that some control rooms provide dozens of alarms during a typical turbine trip even though all of them are reporting conditions that are expected to happen whenever the turbine goes off-line.

Several experienced senior reactor operators have joined the company’s team and are working with the system designers to achieve a complete systems approach that takes into account indications, alarms, layout, controls and human skills.

The final tour stop required a short drive to the OSU campus, where NuScale’s test loop is hosted. Reyes began that part of the tour with some background information about the testing and scaling techniques he and his team learned as contractors for the Westinghouse AP600 and AP1000 passive cooling testing program. He described the program schedule and the way that they used temporary trailers, university students and contractors to achieve a six-day, 24-hour per day testing regime to produce valid results in a cost effective manner.

NuScale poolThen I had the good timing to be one of the first people to tour an in-progress modification of the NuScale testing facility, which was undergoing a major revision as the result of recent system redesigns. Having a good familiarity with mPower’s Integral System Test (IST) facility, I was surprised at the substantially more compact test loop that Reyes and his team had designed and built. The facility was inside an existing laboratory building that is about 30 feet high and has a tall garage door similar to what you might see at a fire station.

Reyes described the effort involved in producing reliable scaling computations and explained that there are two different paths that can be taken to build system test facilities. One is to use relatively simple, well-known calculations and produce a full height, reduced diameter facility. The other path involves more work up front and some specialized computational techniques, but results in the ability to reduce scale in other dimensions while still providing valid results predicting full scale system performance.

The full height path is much more expensive in terms of component construction and installation; it often results in a special purpose building. On the plus side, the one-of-a-kind facility will create some temporary engineering, architecture, construction and manufacturing jobs.

As a start-up led by a professor at a major university, NuScale could obtain skilled engineering services for the up-front calculations relatively cheaply. It did not have the capital to invest in manufacturing and housing a full height test loop. As a focused start up, the company had no established divisions vying to contribute their particular core competencies to a shiny new project that had top-level support from a large, long-established corporation.

They also did not have local civic boosters interested in creating new jobs thinking of ways to add work or offering to spend economic development money for facilities targeted at attracting additional businesses.

Jose Rod Test FacilityNuScale’s new test loop will be ready for operations in early 2015. It still needs a few finishing touches along with the installation of lagging (insulation) before it can begin operations. However, the trailer that will house the control room for the facility has arrived, most of the piping connections have been made, and instrumentation installation is well under way. The revised test loop will provide NQA-1 quality data from a data acquisition and control system taking inputs from more than 500 instruments.

After an informative visit, I headed for the next stop on my whirlwind visit to the Pacific Northwest. That leg involved a journey on one of the most scenic interstates in America, especially for someone who is obsessed with ultra-low emission energy production systems. More to follow.


Bob Meinetz's picture
Bob Meinetz on Nov 1, 2014 6:15 pm GMT

Rod, though NuScale describes their safety systems as “passive”, they don’t appear to be more so than a traditional PWR. From NuScale’s website:

Following a LOCA [Loss of Coolant Accident] or other condition resulting in an actuation of the ECCS [Emergency Core Cooling System], heat removal through the containment vessel rapidly reduces the containment pressure and temperature and maintains them at acceptably low levels for extended periods of time. Steam is condensed on the inside surface of the containment vessel, which is passively cooled by conduction and convection of heat to the reactor pool water. Since the containment vessel is evacuated to a low absolute pressure during normal operation, only a small amount of non-condensable gas will be present inside the containment vessel.

They don’t elaborate on how “heat removal through the containment vessel” will be accomplished but it is apparently non-passive, and supposed passive cooling by conduction and convection with pool water would be non-existent in a true LOCA when the pool water leaks or boils away.

What am I missing here?

Nathan Wilson's picture
Nathan Wilson on Nov 2, 2014 2:29 pm GMT

One of the underlying philosophies of nuclear safety for new reactor designs is that the likelihood that the coolant will be lost from a pool is much less than the chances that coolant will by lost from a system of pipe, valves, and other equipment (the IFR fast reactors uses a coolant pool, as do some salt cooled reactor concepts).  

Integral PWRs (in which the core, pressurizer, and steam generators are in the same vessel, rather than being in separate vessels connected by pipes and welds) are another option for reducing the incidence of LOCAs (loss of coolant accidents).  

The NuScale reactor design has both: it uses integral steam generators and has a below-grade coolant pool for even more safety.

My understanding of  “heat removal through the containment vessel” is that it truly is passive heat conduction thru the steel walls to the water in the pool.  For the Emergency Core Cooling System (ECCS) to work however, there are some valves that have to open (maybe passive check valves?).  

DHRS: When forced coolant flow from the steam turbine is lost, assuming the (active?) valves switch, they have a set of condensers that Removes Decay Heat (via the steam generators) into the pool via natural circulation.  If that fails:

ECCS: if the steam generators stop removing heat for any reason, the passive over-pressure vent valves open, letting reactor steam into the containment vessel.  It is this steam that condenses on the inner surface of the containment because the outer surface is in contact with the water in the pool.  When the water level in the reactor falls to a certain level (due to boil-off going to the containment), the (passive?) recirc valves let water flow back into the reactor (gravity?), thus setting up a natural circulation, which keeps the core covered in water and dumps heat into the pool.

The pool has enough water for 30 days of boil-off, after which the heat production rate is low enough that air cooling is adequate. (I don’t think big reactors can have enough water inventory for this, but maybe the AP1000 can air cool sooner because of the large surface area containment?).

Bob Meinetz's picture
Bob Meinetz on Nov 3, 2014 3:29 am GMT

Thanks Nathan, obviously a lot of consideration has gone into this design.

In my view, its complexity is worrisome. These steps seem all designed to limit the effects of a meltdown, which like any other solid-fuel reactor is possible, and its safety ultimately depends on the integrity of the water pool. In the case of a leak in the pool from sabotage or a seismic event, what’s to stop a NuScale reactor from spiraling out of control in an unpredictable and catastrophic manner?

Michael Keller's picture
Michael Keller on Nov 3, 2014 11:46 pm GMT

As I understand the design, there is a hot reactor vessel sitting in another vessel (containment) that itself is sitting in a pool of water.

What happens if the containment suffers a failure (failure of attached pipe) and the operating reactor vessel is inundated by water from the pool? Seems to me you may end up with a series of high pressure steam excursions playing out over time. Unclear where the reactor (and containment) relief valves would be venting, although I suspect it is into the main building, which is not a containment structure. Logic suggests that this may lead to substantial off-site releases and potentially cascading failures at adjoined units.

Attempts to get at the NUSCALE safety information with the NRC generally leads to documents with vast amounts of “redacted” blank spaces. Hardly inspires confidence in the design.

Am also a little puzzled by the hoopla concerning the control room. Nothing but a bunch of LCD screens; really not that big of a deal. I will say, however, having managed process plants with lots of units, one control room is marginally OK, so long as things do not go “south” at the same time. The ability of a few folks to handle a large mess at a number of units is limited.

Big combined-cycle plants (3 combustion turbine & boilers) have pretty small control room staffs, but the machines generally do not create that much mischief when events go south. It’s not like they will completely contaminate the surrounding countryside.

My hunch is need about three folks per pair of reactors in the control room, excluding the lads out in the plant. One of the three would be expected to be out and about in the plant, from time to time.


Get Published - Build a Following

The Energy Central Power Industry Network is based on one core idea - power industry professionals helping each other and advancing the industry by sharing and learning from each other.

If you have an experience or insight to share or have learned something from a conference or seminar, your peers and colleagues on Energy Central want to hear about it. It's also easy to share a link to an article you've liked or an industry resource that you think would be helpful.

                 Learn more about posting on Energy Central »