APR Atomic Journal - Elk River part 2

In the first installment of this series, we discussed the origins of the Elk River Power Reactor (as it was called at the time) using as source material many original documents and sources. We conclude the story of the Elk River plant as shown in these documents with this installment of the APR Atomic Journal.

Construction underway

Our previous installment on Elk River ended with the announcement (in September, 1958) that Chicago Bridge and Iron had been selected to construct the containment for the Elk River plant. Other long term contracts (known as "long lead time" contracts) began to be awarded at the same time; Sargent and Lundy, Inc. were brought in as architect-engineers and the Maxon Co. of Dayton, Ohio was hired as the construction contractor. In the first quarter of 1958, the estimated total cost to the AEC for the Elk River project was almost $11.5 million, with the Rural Cooperative Power Association providing facilities costing about $1.75 million. The AEC cost included a fixed ceiling contract with ACF Industries to develop and construct the reactor plant, train operators and perform testing. At that time the plant was expected to be completed by October of 1960 (AIF Forum Memo magazine, May 1958.)

Groundbreaking for the new plant (actually, for the reactor plant portion at the site) took place on August 4th, 1958. Construction of the containment began in October, 1958. As noted before, the nuclear portion of the plant was entirely owned by the AEC; interestingly the AEC advance budgeting as announced in mid-1959 did include an earmark for $100,000 for operating funds for the year 1960, as opposed to construction funds, as it was considered at that time that the plant would be operational sometime in that year.

Vendor upheaval

In the middle of 1959, a major change occurred to the project; ACF Industries decided to exit the reactor business completely. In this transition for ACF it was decided to sell the nuclear products operations, but not the ERCO operations (the nuclear products and ERCO divisions had been merged together by ACF Industries on April 1, 1957.) ACF sold its entire nuclear business to Allis-Chalmers Manufacturing Company, including not only the Elk River project but the various test and training reactors it had under construction. This was the second of two rapid and somewhat major transactions in the nuclear energy field; also in 1959, General Nuclear Engineering Corporation was bought by and merged into Combustion Engineering, Inc.

Allis-Chalmers at this time had one power plant contract, which was for a boiling water reactor for Northern States which would become known as the Pathfinder Atomic Plant. Allis-Chalmers immediately integrated the ACF operations into its own, and in fact the 1960 Allis-Chalmers annual report supplement makes absolutely no mention of work underway as "ex-ACF" or any such thing; it is all A-C so far as this document is concerned.

Below we see our first photo for this installment. This is a previously unduplicated UPI Telephoto dated February, 1960.

This photo shows the arrival at the Elk River site on February 11, 1960 of the reactor pressure vessel on what appears to be a standard 40-foot railroad flat car. This pressure vessel was fabricated by Pacific Coast Engineering Company, of Alameda, California. The control rods on this reactor plant are bottom entry; thus, the lower end of the pressure vessel is at left. According to the caption on the UPI Photo (in the APRA collection) the vessel weighed 72 tons, and was to be placed in a minimum of 8 feet 3 inches of concrete shielding. The caption also states an expected date of completion of November, 1960.

Elk River plant details

Let's now briefly take a look at the design of the reactor plant in some closer detail. Below is an illustration from "Boiling Water Reactors," Andrew W. Kramer / Addison-Wesley 1958 as published as part of the 1958 Geneva Conference set in the APRA collection.

This illustration shows the reactor vessel in good detail. The reactor plant at Elk River was conceived originally as being convertible; although it was completed and operated only as an indirect-cycle BWR plant, it was convertible to forced circulation with the use of installed, but capped and welded, nozzles as indicated in the drawing above. None of the recirculation equipment was ever ordered.

The Elk River plant was rated at 22MWe; the thermal output of the reactor itself was rated 58.2 MWt, while the coal-fired superheater was credited with 14.8 MWt for a total thermal power of 78 MW.

In the photo showing the arrival of the reactor vessel, the nearly complete containment is visible behind it; below, also from "Boiling Water Reactors," is another illustration showing equipment layout inside the containment with the reactor plant complete.

In this view we can see the equipment labeled "evaporator," which in any other plant would be called a steam generator. This plant's use of the indirect cycle was intended to take advantage of the fact that the steam would carry no radioactivity into the turbine building (the turbine generator that Allis-Chalmers was designing for Northern States was very heavily shielded; that for Elk River was not.)

The control system for the plant used, as its primary input, the pressure of steam exiting the superheater. Variations in this pressure caused immediate but fine motion of the nine control rods, made of boron-steel.

This reactor was one of two early reactors (the other being Indian Point 1) which used uranium and thorium fuel to explore the thorium fuel cycle. The initial fuel loading for the plant was 195 kg of uranium and 4200 kg of thorium dioxide, all in pellets inside stainless steel tubes.

Construction, testing, delays

As has been the case with most nuclear plants, unexpected delays began to push the actual startup date back, then pushed it far off. Although the plant itself was practically completed by fall of 1961, almost a year off the initial target, problems with the reactor itself - fuel element defects and problems with the stainless steel pressure vessel internal cladding - pushed anticipated startup as perceived in late 1961 to probably late 1962. (Atomic Energy Deskbook, Hogerton, 1963.)


Above is a post card view (APRA collection) showing the Elk River plant in a state of partial completion in 1959. The reactor plant containment is visible on the right, the turbine building on the left. The interposed superheater section is not yet constructed.


Above is a photo taken from a very similar angle, which was included in the 1960 Allis-Chalmers annual report. This report states that construction at this time was virtually complete, and the final appearance of the plant is now established - except for the fact that, for whatever reason, the A-C art department has airbrushed the signage off of the front of the building which reads "Rural Cooperative Power Association - Owned by Those it Serves." (An almost identical photo, but with the sign intact, appears in the Atomic Energy Deskbook.)

Below, a photo I used to introduce this series but which is nevertheless fairly stunning. This is the cover photo of the 4th quarter, 1962 edition of "Allis-Chalmers Electrical Review."

In all of the old material thus far available, this is the only photo this author has seen that shows Allis-Chalmers technicians (note the A-C emblems on their coveralls) at work inside one of their plants; this is the initial fueling of the Elk River reactor. The caption inside reads as follows: "ELK RIVER REACTOR achieved sustained nuclear reaction in the 4th quarter, 1962, with 41 of its 148 fuel elements.... The station, with its coal fired superheater, is gradually being brought up to its full 22,000 kwe capacity by mid-1963."

The tractor enthusiasts' magazine "Old Allis News" issue of Summer 2007 contains a very interesting first-hand article which includes information from three operators at Elk River. (This great article was written by Mark Rusk and originally appeared in the Upper Midwest A-C Collectors' Club Newsletter.) One very interesting fact appears in this article - the statement by operators that, from the very start, it was known that there was a small amount of coolant leaking into the containment. Clearly, then, the problems with the PACECO-manufactured pressure vessel had either not been fully rectified, or perhaps not even fully discovered. Since the leakage was very small, operation was continued and the plant given its operating license at rated power.

Commercial operation - but only for a short time

The 1964 Annual Report to Congress of the Atomic Energy Commission (published January, 1965) includes the following notation about Elk River:

"During its preliminary test program, the AEC-Rural Cooperative Power Association's boiling water reactor at Elk River, Minnesota, attained its full design 58.2 thermal megawatt, 23,000 electrical kilowatt (ekw) {APR note: This must be gross kwe} power level and a full power, 28 day warranty run was completed March 21. {1964.} This was followed by a 60-day transition run to prepare the reactor operators for plant operation after transfer from Allis-Chalmers to RCPA. In November, the reactor was shut down for visual inspection of the control rods. They were found to be serviceable, and the plant was returned to power operation."

Thus, as we've just read, the Elk River plant was finally fully operational and on the grid roughly seven years after the first project, which saw one vendor dropped and the second one bought out and merged mid-project. The official date for initial power operation was July 13, 1964; the official date for commercial power operation was June, 1965.

In another section of the same "Old Allis News" is a one-page interview with former Elk River technician Donald R. McRae who intimates some of the developmental problems that slowed the Elk River plant's becoming fully operational. Most interesting of these is the fact that the plant's protection and alarm system was originally designed and constructed single-channel, causing many spurious trips and scrams. Technicians, according to McRae, convinced A-C to purchase and install "two out of three" coincidence equipment to make operation reliable.

According to both Old Allis articles and many other sources, the leakage of coolant into the containment remained a constant problem. Although the plant did generate power reliably for several years, it was shut down January 31, 1968 for a detailed investigation into the nature of the leakage. The investigation revealed small or hairline cracks in the vessel and piping, which would have necessitated heavy rework or replacement. This date became the official date for the end of commercial power operation of the Elk River Plant, because the RCPA did not elect to continue operation of the plant after the official AEC demonstration period which would in all likelihood have required the purchase of the plant from the AEC. (A similar set of events, along with problems, dictated the end of life of another well known rural power reactor, that being the one in Piqua, Ohio in that the end of the demonstration period did not lead to purchase by the municipality.)

At this point the Elk River plant was essentially doomed; what remained was the decision process to dismantle it. The first step was to defuel the plant; this was accomplished by September 16, 1969 which is the official AEC date indicating all fuel was shipped offsite. On March 15, 1971, the decision to dismantle the plant immediately was announced by the AEC and RCPA; on March 23, the AEC amended the plant operating contract to include dismantling. The plan for dismantling the plant was developed immediately and issued October 18, 1971; the order to begin dismantling was issued June 5, 1972. The work took roughly two years, with the final site survey being issued July 23, 1974 and the entire Elk River project termination order being issued September 30, 1974. "Rural America's First Atomic Power Plant," as the sign on the pressure vessel had indicated and as RCPA and the AEC had advertised, was gone - in the end, doomed essentially by material problems, economical viability with continued operation by, and eventual ownership by, RCPA not considered.

EPILOGUE

The site still sees use today, by a descendant of RCPA; Great River Energy now owns and operates plant, today known as the Elk River Energy Recovery Station and which now burns refuse, instead of coal and oil, to generate energy. Even with the addition of new equipment in the place of the long-gone reactor containment, the station is immediately recognizable even today as seen below in a photo taken by Great River Energy.


1:05 PM Eastern Saturday September 24, 2011
ATOMIC POWER REVIEW

Read More

APR Atomic Journal - ELK RIVER 1

-APR introduces a new recurring feature to its pages - the APR Atomic Journal. This non-uniformly cyclical feature will cover many historical and technical aspects of nuclear energy for both laymen and experts. I sincerely hope you enjoy it.-

Above: Initial fueling of the Elk River Power Reactor by technicians of the Allis-Chalmers Manufacturing Company, late 1962.

THE ELK RIVER POWER REACTOR, part 1.

The nuclear power plant commonly known as "Elk River" was hailed in the early 1960's as "Rural America's first atomic power plant" by the Rural Cooperative Power Association, or RCPA, who was to operate the plant and supply power from it to its customers in and around Elk River, Minnesota. Atomic energy was still very new and very much anticipated, and this plant was a test to see, among other things, if a small reactor plant could be built and successfully operated as a part of a comparatively small electric utility's system.

I have been quite fortunate to acquire a large amount of historical material covering the early years of the atomic age in the United States, and while the majority of the focus is on more well known plants (Shippingport, Fermi, Dresden) the Elk River plant appears fairly regularly in this early material. It is for this reason that I present the following narrative details, in order to supplement historical information already available in print and on the internet. Most of what you are about to read has been generally unavailable until now.

THE BEGINNING - AMF ATOMICS

The first negotiations to build what became Elk River started in 1957 between AMF Atomics Inc. which was a division of AMF (formerly American Machine and Foundry) and the Atomic Energy Commission (AEC.) The plan was to construct a small boiling water reactor operating on the indirect two-loop cycle (also known as the "closed cycle boiling water principle"), which would be owned by the Atomic Energy Commission. The utility involved, RCPA, would purchase steam from the plant (thus from the AEC) and would also provide the steam portion of the plant, as well as the operating staff of the entire plant once turned over from the reactor vendor, AMF. (In reality, a conventional steam plant already existed, having been built in 1950; RCPA would purchase additional equipment to install in the plant to accommodate the nuclear plant built nearby.)

We do not have any solid data on what the plant would have looked like. We do however have a design from an AMF Atomics sales brochure from that time which roughly matches the designed output and may well have been the basis for the opening of AMF's negotiations with the AEC.

Above, an illustration showing AMF Atomics' Closed Cycle BWR plant. As with all photos on this site, you can click to enlarge the photo.

This plant was to operate with a closed boiling water cycle on the primary side, with flow by natural circulation. According to the AMF information, the entire 20 MWe plant is contained in a concrete block 16 feet wide, 60 feet long and 50 feet deep. No external containment to this other than weather covering was to be provided. The concept of the plant was simplicity; stated advantages over a pressurized water design were the following: 1. No pumps or valves in the primary, saving cost and maintenance. 2. Lower primary pressure, allowing lighter pressure vessel. 3. High heat transfer rates and more effective use of heat transfer surface. 4. No radioactive steam supplied to turbine building. 5. Heat capacity of primary heat exchanger adds lag ("flywheel effect") to system, relaxing requirements on control system.

This brochure (bound in a large volume of various reactor brochures and reprints which we obtained from a former Bettis Atomic Power Laboratory employee) states that while the cost of such a plant per kwhr wasn't competitive with large central station plants, it was competitive with rural plants.

Now, looking at the drawing we can see one very interesting feature; this plant has a very primitive pressure suppression containment design - and an early one too, considering that the AMF data is from September 1955. Below the reactor cavity can be seen what is called the "Shielding water drain tank and containment quench tank." This is connected to the steel containment shell inside of which are the reactor, steam generator and all primary piping by a three foot diameter vent pipe. The design of the containment was such as to limit internal pressure under any circumstance whatsoever to 5 psig.

This plant would have been built, theoretically, adjacent to or near the existing Elk River Station and supplied a newly installed turbine in that building with steam. However, before any material was purchased, AMF was out of the running.

ACF WINS THE BID

The March 1958 edition of the Atomic Industrial Forum's FORUM MEMO magazine details the fact that in October 1957, AMF Atomics and the AEC had reached an impasse on the grounds that AMF could not agree on a price ceiling (knowing how the PDRP was working at this time, probably a "turnkey" price) which led to the AEC making an open offer for bids on the reactor portion of this project. In a later issue of the same magazine, RCPA President O. N. Gravgaard told the AIF that AMF had withdrawn its proposal for the project and had decided to get out of the power reactor business. However, RCPA had already ordered most of the conventional (non-nuclear) equipment for the project, causing concern to the company; as the AEC had opened up for bids, the company elected to construct another coal fired boiler on site and supply the steam equipment with this new boiler until an atomic plant was built on the site.

Taking a tangent for the moment, it is interesting to note Mr. Gravgaard's statement that AMF Atomics was getting out of the power reactor business, because it wasn't really in it, having received no other orders for power reactors even though it was fairly heavily advertising them. AMF did later on through 1958 however begin work on a project to construct exactly the type of reactor plant shown earlier in Cuba; this was later announced as having been dropped due to the unstable political situation in that island nation. With that cancellation ended any prospect of AMF Atomics building a power reactor.

Getting back to Elk River... Two bids were received by the AEC, one from General Electric and one from ACF Industries' Nuclear Products-ERCO Division. According to the AIF Forum Memo, other firms expressing interest to bid were American Radiator & Standard Sanitary Corp., Nuclear Development Corporation of America, and two other smaller companies with no nuclear experience. The deadline for proposals was February 3, 1958 and in March the AEC announced that ACF's bid had been approved. At the time of acceptance of ACF's proposal to begin a discussion that would lead to a fixed price ceiling, the basic schedule was for construction of the plant to begin in the Spring of 1958, critical testing at the end of 1959 and full operation of the plant on RCPA's grid early in 1960. RCPA agreed to operate the plant for five years.

ACF BEGINS WORK

ACF Industries formed its own nuclear engineering section in 1954 in order to allow it to enter the field first with tank type research reactors, and then later with power reactors. At the same time, ACF purchased a company called the Engineering and Research Corporation ("ERCO") and operated it as a division. April 1, 1957 saw ACF merge its nuclear engineering section of ACF Industries and its ERCO Division into one unit, which it then called the "Nuclear Products Divison-ERCO." This firm had several tank type research and training reactors under construction and more planned when it won the bid to construct the Elk River Plant's new atomic addition.

At the time that ACF won the bid, the original concept as outlined by Rural Cooperative Power Association (RCPA) stated that the plant would be completed in late 1959 for a total cost of $6.2 million. The growth of the cost, which killed the AMF bid because it kept going up, reached $8 million with no fixed ceiling and a completion not until 1961. ACF promised completion in 1960 at a cost of $11.4 million.

ACF immediately engaged Sargent & Lundy (Chicago) as the architect-engineer firm for the project, and the Maxon Construction Company of Dayton Ohio as its general contractor.

PLANT DESIGN

ACF's design for the plant was functionally identical to AMF's, in that a boiling water reactor was to be employed operating in a closed loop, with natural circulation; steam was to be generated in multiple heat exchangers which would then supply steam to the RCPA plant. Since the plant was designed for superheated steam it was necessary to incorporate in the new construction a separate coal-fired superheater to arrive at the proper steam conditions. We will now look at some early ACF advertising artwork showing the plant design.

Above (and again, click to enlarge) we see an ACF illustration which appeared as part of an advertisement in the May, 1959 issue of NUCLEONICS. From this ad: "The boiling water reactor concept is ideally suited for application in electrical generating plants such as the one ACF is now building for the Rural Cooperative Power Association at Elk River, Minnesota." The ad states that ACF was prepared to supply such plants at ratings up to 50 KWe (although the Elk River installation was to be a much more modest 22 KWe.)

This illustration clearly shows two horizontal tube and shell type steam generators connected to the reactor without forced circulation pumps; the steam outlets to, and feed return from, the generators is clearly visible as are the feed preheaters. A ring near the top of the vessel internal to the shroud is for emergency cooling water injection and/or borated water injection.


Above, detail of the cover illustration from an ACF advertising brochure. As with all illustrations, click to enlarge. This illustration shows essentially the Elk River plant design, and very largely the containment design actually built, but it is attached to a generic steam plant; for example, the control room in this illustration is between the reactor containment and the turbine building, which space at Elk River was actually occupied by the superheater plant. It is interesting to note that the turbine generator set illustrated here appears to be an Allis-Chalmers "centerline at floor level" machine. Notable in this illustration are six horizontal steam generators.

This brochure is undated, but indicates that ACF had already built two research type heavy water tank reactors (MIT, and Italian National Committee) and was building four light water tank type reactors in the USA, in Holland, and in Sweden. The brochure indicates that ACF had designed power reactors as well with outputs up to 150 MWt. This brochure is thus probably immediately pre-Elk River.

The ACF Horizons magazine issue of August 1958 - roughly five months after ACF won the Elk River contract - shows the following illustration.

This is the first illustration that shows the plant at Elk River essentially as it was built. The existing steam plant can be seen furthest from the point of view, with two stacks atop it. Between this plant and the reactor plant containment is the coal fired superheater. The containment is of wholly conventional, for that time, form and shape.

The October 1958 FORUM MEMO magazine mentions that in September 1958 ACF had awarded a contract to Chicago Bridge & Iron to build the containment for the Elk River plant for $500,000 which was to begin immediately. The June, 1959 issue of POWER REACTOR TECHNOLOGY gives the following details on the containment: The plate material was A201 steel (A300 heat treatment) of either 0.50 or 0.70 thickness depending on location; "A" numbers are ASTM standard. Gross internal volume of the containment was 405,200 square feet. Design pressure was 21 psig and test pressure was 26.25 psig with a specified leakage of 0.1, expressed as percent of volume (measured at STP) per psig.

NEXT APR ATOMIC JOURNAL: Construction, testing, operation and shutdown of the Elk River reactor. You don't want to miss the next installment!

7:20 PM Eastern Thursday September 8, 2011
ATOMIC POWER REVIEW

Read More