On paper the composite bearingless rotor seemed too good to be true: a wind turbine rotor that enabled the blades to change pitch without bearings in the hub. And the wind turbine would passively use aerodynamic forces to orient the rotor downwind of the tower. It was the height of simplicity and would be cheap to build. What could go wrong? The short answer: everything. Eventually the nearly 400 wind turbines using the concept in California during the Great California Wind Rush of the early to mid 1980s were scraped off the face of the earth for scrap. And therein lays a sprawling tale.
Some of the leaders in the American wind industry cut their teeth trying to make these wind turbines work, or replace them with something that would. Brian Smith, Kevin Jackson, Paul Migliore, and Bruce Hammett all went on to build careers in the wind industry. Some, such as the erstwhile wind developer Phil Cruver, became notorious.
Our story begins in the mid 1970s in the heart of the military-industrial complex. With the memory of the 1973 Oil Embargo fresh in peoples’ mind, the US government launched a program of research and development in wind energy. The federal agencies awarded contracts to the companies it mistakenly assumed would be the best placed to build wind turbines: the aerospace industry. And the kind of aircraft that most closely resembles a wind turbine is a helicopter.
UTRC
Enter United Technologies whose subsidiary, Sikorsky, built helicopters. The engineers at the conglomerate’s Research Center (UTRC) were steeped in helicopter lore.[1]
In 1975, Petrus A.M. Spierings, a Dutch engineer was working for Marvin Chapin Cheney Jr. at UTRC. Spierings, who goes by P.A.M., and Cheney, who goes by Kip, together developed a concept for a wind turbine adapted from the tail rotor of Sikorsky’s UTTAS helicopter.[2]
In Sikorsky’s tail rotor, two continuous spars of composite materials (fiberglass) pass through the hub, providing a four-blade rotor. The tail rotor operates at high speed providing stiffness to the composite blades and a control spider twists the torsionally flexible blade roots to change pitch as required without the need for bearings in the hub.
My thanks to the many colleagues who were consulted during the research for this article: P.A.M. Spierings, Ken Starcher, Brian Smith, Bruce Hammett, Paul Migliore, Kevin Jackson, Arne Jaeger, Klaus Rockenbauer, Tim Olsen, and Thomas Spiglanin.
UTRC won a grant from the Energy Research and Development Administration (ERDA) in 1975 to develop the Composite Bearingless Rotor (CBR) wind turbine.[3]
Subsequently, Cheney and Spierings issued a report to ERDA on the design, including a description of a 4.5-foot diameter scale model they tested in UTRC’s low-speed wind tunnel.[4] The design at this stage incorporated a torsionally flexible spar or flex beam that accounted for 15% of the rotor span. The flexible spar effectively replaced the bearings of a conventional wind turbine with variable pitch blades. However, the blades on the UTRC wind turbine would not pitch to feather. The blades would pitch only a few degrees toward stall in response to a “moment at the outboard edge of the flexbeam” twisted it elastically. The moment was produced by a pendulum mass and strap.[5]
The rotor was designed with a flex beam that produced coning to passively yaw itself downwind of the tower, eliminating the need for a mechanism to yaw the wind turbine about the tower as it hunted for the wind.[6]
Further cost savings would accrue from the use of pultruded fiberglass blades, which could be produced industrially rather than through the time-consuming and expensive hand lay up of fiberglass blades then used on Danish wind turbines.[7] Though less costly, pultruded blades have a constant planform and no twist, reducing their aerodynamic performance relative to tapered and twisted airfoils. Pultruded blades are also more flexible than the stiff blades made from hand lay up. This latter feature proved significant.
There was no active overspeed control other than stall on the Cheney-Spierings’ design. In high winds or in a loss-of-load accident, the pitch weights would passively twist the blades a few degrees toward stall, dumping power. The slender, flexible blades would also progressively cone downwind of the tower as wind speed increased like the fronds of palm tree in a hurricane, shedding load.[8]
Sikorsky’s UTTAS tail rotor design resulted in a 50% reduction in maintenance over conventional tail rotors. Cheney and Spierings’ expected similar savings from their CBR wind turbine. It was not to be.
The slender flex beam, the pendulum, its attached strap, and the pultruded fiberglass blades became the hallmarks of UTRC’s design.
Cheney and Spierings filed a patent for the design in mid 1976. The patent was granted in early 1978.[9] Tellingly, the patent required 11 pages to explain how the concept worked. In principle, the design was simple. In execution, it was complex.
It’s noteworthy that both the patent and the scale model used four blades, not two blades for which the design, and its copy, are known.[10]
Cheney and UTRC subsequently won a grant from the Department of Energy (DOE) to develop an 8 kW wind turbine for residential applications using the CBR concept.[11] The now two-bladed rotor with a diameter of 9.45 m (31 ft) operated downwind of the tower in free-yaw. One prototype was installed at the Rocky Flats test center near Golden, Colorado. Another two turbines were installed in Wisconsin for Wisconsin Power & Light.[12] Another unit may have been installed on Moon Island in Boston Harbor.
During December 1980 the two 8 kW prototypes in Wisconsin failed during an ice storm.
In 1979, UTRC won a further grant to design a 15 kW version. This unit would use a 14.6 meter (48-foot) diameter rotor. Unlike the 8-kW turbine, which was installed on a guyed tower, the 15-kW version would incorporate a free-standing cantilevered tower. The 15-kW version also integrated the gearbox into the bedplate or main frame of the nacelle.[13]
The integration of the gearbox into the main frame of the nacelle became a key characteristic distinguishing American-designed wind turbines in the DOE program with wind turbines from their Danish competitors.
It’s often forgotten that the small wind turbines designed for residential applications under the DOE program used relatively large rotors to power relatively small generators. For example, UTRC’s 8 kW wind turbine used a rotor nearly 10 meters in diameter. Contrast that with the Carter 25 from the same period which used a 10-meter diameter rotor to drive a 25 kW generator.
Wind turbines in the DOE program were intended to be used where people live, often areas with low winds. Thus the wind turbines had high specific areas and low specific power. In fact, the specific power and specific area of UTRC’s 8 kW and 15 kW were similar to that from designs of the famous German engineer, Ulrich Hütter, who had spoken with ERDA and then DOE during the early days of the US wind program.[14] It would be three decades before the US wind industry would again field wind turbines with relatively high specific area and low specific power for use in low to moderate wind regimes.
Wind-Assisted Pumping
In the mid 1970s, the Alternative Energy Institute (AEI) at West Texas State University and the US Department of Agriculture’s experiment station in Bushland, Texas worked extensively to develop wind-assisted irrigation pumping on the high plains. The area of the Texas Panhandle where they’re located has also seen extensive oil development, requiring electrically-driven pumping on stripper wells.
In mid 1981, UTRC installed a third-generation version of its CBR 8 kW turbine northeast from Borger in the Texas Panhandle with AEI. The turbine was intended to power one of the 100 Phillips 66 stripper wells at the site in a demonstration of wind-assisted pumping. The wells were connected to the grid and the wind turbine was connected on the Phillips 66 side of the meter.
The turbine operated for 18 days before a coupler between the rotor and the gearbox failed for a lack of lubrication. The rotor ran unloaded and the pitch to stall system worked as intended, saving the turbine from destruction.[15]
The Borger unit was then moved to AEI’s test field near Canyon, Texas south of Amarillo. There the turbine was running under load in light winds when the main shaft failed and the rotor fell off. The main shaft was replaced, a new keyway cut, and the turbine returned to service where it was eventually damaged by lightning.[16]
Prototype Failures
Around this time a UTRC turbine may have failed on Moon Island in Boston Harbor. I couldn’t find any online references to it. I remember the scuttlebutt about something like this happening, but my memory could be faulty. Nevertheless, failures of the prototypes embarrassed UTRC’s parent company and they canceled the program.
Cheney, “who was very entrepreneurial,” says his colleague Spierings, took the design and left UTRC to develop the concept independently under the trade name Windtech.[17]
California’s Wind Rush & the CBR
The scene now shifts from a few prototype turbines installed here and there to Cheney’s attempt to commercialize the CBR technology in the hot house atmosphere of California’s Great Wind Rush from 1981 to 1985.
There are three major passes in California where wind development was concentrated: Altamont in the north, Tehachapi in central California, and San Gorgonio in the south. However, commercial wind development was attempted elsewhere, including the Salinas Valley, Pacheco Pass near Los Banos in central California, and the Carquinez Strait of Solano County in northern California.
Despite the chronology I’ve followed until now, we’ll work our way down the state beginning in the far north with the Carquinez Strait. Documentation is sketchy until the California Energy Commission (CEC), through its Wind Project Performance Reporting System, began issuing annual reports in 1986. From 1981 to 1985 we’re dependent upon news accounts and the memory of participants from 40 years ago.
Carquinez Strait
On a bluff overlooking I-680 and Suisun Bay between Benicia and Fairfield, California, Michael Powell installed five of Cheney’s Windtech turbines on land owned by Solano County near the freeway. The design now featured a 15.8-meter downwind, two-blade rotor driving a 75 kW generator. Powell’s company, Wind Watt, signed a 30-year contract with PG&E, the local utility, and was generating $3,000 to $5,000 per month in revenue during the first year. It was not to last.
During a stormy night, one of the Windtech turbines threw its pendulum strap onto I-680. A motorist drove over the strap where it pierced the floorboard of his car. Cheney’s insurance covered the damage, but the county got cold feet about potential liability and revoked Wind Watt’s permit. Cheney agreed to make some modifications to the turbines, but there was a delay in getting the job done. In the meantime, the county panicked and had all the turbines removed without consulting Powell and Wind Watt.[18] This was a harbinger of events to come.
I don’t know when the turbines were actually installed, or removed. The California Energy Commission (CEC), through its Wind Project Performance Reporting System, notes that the turbines were in operation throughout 1985. So they were likely installed in 1984 at the latest. They operated through the third quarter of 1986, and were gone by 1987.[19]
Pacheco Pass & Windtech Clone
One of the painful lessons of wind’s early days was that if you were going to copy a design, copy one that works. American engineers of the day often complained that all Danish wind turbines looked alike. They did, but there was a good reason why they all looked alike. Their turbines not only shared components, such as blades, gearboxes, and generators, but they also shared a design approach (three blades upwind of the tower) because it worked.
Cheney’s CBR design was unique, but it was also in the public domain. UTRC had contracted with the federal government and developed a wind turbine using public funds, including both an 8 kW and a 15 kW wind turbine. Consequently, the design was in the public domain. By the early 1980s, however, the track record for Cheney’s CBR design wasn’t good. He had already suffered several failures of various prototypes. This subtle detail was lost on one southern California wind promoter, Phil Cruver.
Sometime around 1983 Cruver and partners installed 30 Windtech clones on the eastern slope of the Pacheco Pass near Los Banos, California. The site he chose overlooks I-5, California’s major north-south artery.[20] The site and the wind turbines were clearly visible to both northbound and southbound drivers on I-5.
The wind turbine was a “poorly implemented” copy of Cheney’s UTRC design, says Kevin Jackson of Dynamic Design Engineering. Jackson, who went on to build a career correcting the mistakes of others, says Cruver’s clone “lasted only a matter of weeks” at the Los Banos site. He should know. He was brought in by Paul Migliore to repower the site under the banner of WestWind Industries.
WestWind, located in East Sacramento, was a spin off from the University of California at Davis where Migliore taught aerodynamics. WestWind was a product of the times. It provided consulting engineering services as well as metal fabrication to the states burgeoning wind industry. Their replacement for Cruver’s take on Cheney’s CBR design was a three-blade, upwind rotor with active yaw, much like the Danish wind turbines of the period.
The replacement turbines were installed and operated briefly before the investors bailed out of the project. No production was reported from 1985 to 1988 when the project disappears from the CEC’s reporting system.
The site wasn’t particularly windy and the project was simply too small to make it work financially says Jackson.[21]
Tehachapi Pass
No Windtech clones were installed in the Tehachapi Pass. At one point there were 127 of Cheney’s Windtech turbines in Tehachapi, or more than half of the total eventually installed in California.
When I arrived in Tehachapi in the spring of 1984, there were Windtech turbines prominently installed at three sites: Arbutus, Windridge, and Zephyr. Arbutus had installed 81 Windtech turbines on the south-facing slope of Pajeula Peak overlooking Hwy 58 probably in 1983. There was also a small group (4) of Windtech turbines on the Windridge site just off of Tehachapi-Willow Springs Road and another cluster of 30 machines on the Zephyr site beyond Oak Creek Energy Systems’ site.
I don’t remember the Windtech turbines operating much if any on the Windridge site. I have a photo where they appear to be operating alongside a cluster of Windmatics that were still there in 2024. The Windtech turbines were gone by 1986 according to the CEC’s reports.
The CEC’s 1985 report shows another five Windtech turbines at the Cannon site atop Cameron Ridge. I don’t remember them, but it was a long time ago.
The Zephyr site, in contrast, was quite memorable. You didn’t need to be a meteorologist or a wind farm designer to see that the turbines were the definition of poorly sited. Zephyr’s site was a small hill on the east side of Tehachapi-Willow Springs Road just south of Oak Creek Energy Systems’ wind farm. Zephyr gouged out several benches into the hill to install their wind turbines as though they were an Appalachian contour stripper cutting benches to get at the coal. It was a mess. After four decades, the cut banks are still visible on Google Earth.
Suffice it to say, it was a lousy site for a temperamental wind turbine. Zephyr installed the turbines in 1983 and reported to the CEC in 1985 that they had numerous problems with the machines and were seeking a retrofit. I have photos of some of the turbines in operation in the spring of 1984 with a cluster of Carter 25s in the background. They stopped reporting to the CEC in 1987. There were five Windtech drive trains lying abandoned on the ground in the spring of 1997, a decade after Zephyr stopped reporting.
The Arbutus site was a different story. It was particularly windy in a windy pass and there were enough wind turbines to justify trying to keep them running. By 1985 they had 81 Windtech turbines, according to the CEC’s report. They added another seven turbines in 1987.
Windtech set up a shop in Mojave, California to repair and retrofit the more than 100 of their machines in the Tehachapi Pass. They even had an engineer on site at the time. But by 1989, Arbutus reported to the CEC they were operating only five of the remaining turbines.
The five remaining Windtech turbines on the Arbutus site were likely those retrofitted by Hal Romanowitz, a consultant to Oak Creek Energy Systems who had his finger in a number of wind projects in California. Romanowitz’s retrofit followed that pioneered by Migliore in the Pacheco Pass. Romanowitz also used a three-blade rotor up wind of the tower. These proved troublesome themselves and Arbutus reported no production from these turbines from 1992 to 1994 when Arbutus stopped reporting altogether.
San Gorgonio Pass
The wind rush forever changed the face of the San Gorgonio Pass. It eventually contained one of the largest concentrations of wind turbines in the world in a narrow neck of land sandwiched between the San Bernadino mountains on the north and Mount San Jacinto on the south.
Not that the pass was pristine in the early 1980s. It is a major corridor between the Los Angeles Basin and California’s interior deserts. The massive I-10 freeway courses through the pass as does the railway, and transmission towers, the latter carrying electricity from the coal and nuclear plants in Arizona to the near insatiable demand of Los Angeles.
The pass is also the gateway to Palm Springs, once the playground of the rich and famous and still home to politically influential residents.
The wind resource is concentrated on either side of I-10, and on the broad flat of the Whitewater Wash. During the wind rush, developers began snapping up parcels throughout the pass and throwing up wind turbines as fast as they could once Riverside County’s permitting log jam was broken in 1983.
You couldn’t pick a more combustible mixture of flamboyant, fly-by-night developers, hasty installation of unproven wind turbines, and the politically powerful who woke up one day to see the desert covered with flailing wind turbines. It was chaotic, and it became a major headache for the budding American wind industry.
And some of the most prominent of the flailing wind turbines were those using the composite bearingless rotor.
Cheney’s Windtech turbines were installed in a mass of wind turbines of every sort on a one square mile parcel on the bed of the Whitewater Wash. While they suffered the same problems as elsewhere in California, they were far less noticeable and drew much less ire than their clone.
The highly visible disaster in the Pacheco Pass with a Windtech clone was just a prologue for Phil Cruver’s main act in the San Gorgonio Pass. Cruver’s company, International Dynergy, ironically dubbed his clone the Windshark. Like Cheney’s wind turbine, it featured two pultruded blades downwind of the tower with a torsionally flexible spar.
Dynergy developed two main sites: Maeva on the north side of I-10, and Cabazon on the south side of I-10 at the mouth of the pass. Cabazon was the first wind farm that eastbound travelers would see entering the pass. They also added a couple of dozen of their machines to the mass of wind turbines on the Whitewater Wash.
To get a flavor of the times, here is a passage on Dynergy by Peter Asmus from his book Reaping the Wind.
“Two projects by International Dynergy, Inc. of Palm Springs—at Cabazon, one of the most conspicuous sites, and at Maeva—became lightning rods of controversy. The company sold 300 turbines to limited partnerships at Cabazon but erected only 150, and half never worked. Maeva fared no better, generating only 12% of its projected output. Bank of America pulled out of the Maeva operation with $1.5 million in unsettled debts, and the IRS paid a visit. True or not, the folklore is that before the visit, employees were seen attaching helicopter blades to machines that had not been operating for months. Worse still was Cabazon’s location right alongside Interstate 10, where its heaps of debris were visible to tourists and large numbers of passers by. These two projects helped fuel the public impression that the industry was made up of tax farms, not wind farms.”[22]
Peter Asmus, Reaping the Wind: How Mechanical Wizards, Visionaries, and Profiteers Helped Shape Our Energy Future
The GM of Wind
The disaster that Dynergy became was a far cry from Phil Cruver’s dream of becoming the General Motors of wind. He’d contracted with Westinghouse for production of the drive train and electrical controls. The rotor was designed and produced by a consulting company, possibly in Los Angeles.
Bruce Hammett joined Dynergy in early 1984. He, like others, tried to make the turbines work. According to Hammett, the UTRC design had been copied many times. None mastered it, including its inventor. “So many things could go wrong, and they did,” he remembers. Hammett lasted a year when he left to start his own company supplying parts the growing industry.[23]
Another industry veteran, Brian Smith, was also lured into the maelstrom. After graduating with a Masters from the University of Massachusetts’ wind program, he joined Dynergy as their first staff engineer.[24] Smith, who went on to become NREL’s wind laboratory program manager, was hired to set up manufacturing of the Windshark turbine at the Cabazon site. Cruver’s idea was to become a vertically integrated manufacturer, like General Motors.
By now the Windshark turbine parted with Cheney’s nacelle design and began sporting a nacelle with a shark fin on the top and bottom.
One of the many problems with Cheney’s CBR design was that the turbine became unstable in turbulent wind. The rotor wouldn’t always stay downwind of the tower. Sometimes it would “walk” upwind and with the flexible pultruded blades, a tower strike was inevitable, sending the blades flying and destroying the turbine.
On the surface, it would appear that the fins could help keep the rotor downwind of the tower. However, to Smith’s knowledge, the shark fin was never tested to see if it made the wind turbine more stable. “It was probably more for marketing,” he says.[25]
Smith’s recollections of those heady days were captured in “Wild West of Wind,” a chapter in an NREL house publication. Here are three passages that describe what it was like working for Dynergy while trying to stay alive.[26]
“A shadow flashed over Brian Smith’s shoulder as he walked the 210-acre Maeva Windpark near Palm Springs, California. It was 1984 . . . A minute later, another silhouette flashed behind him. Startled, Smith looked up, then turned to his boss who shrugged and said, somewhat too nonchalantly for Smith, ‘Those are blades flying off the turbines.’ The 20-foot-long fiberglass projectiles were airborne daggers. ‘We have a fix in,’ the boss said, ‘but for now, the wind turbines have to keep turning to sell power—even though the technology is a little shaky.’
“Even so, damage was often unavoidable. According to Smith, “If there was a good ‘blow’ [their term for a high-wind day], you’d come into work and it wasn’t a question of, ‘Is there a turbine running away?’ or ‘Is there a turbine on the ground?’ It was, ‘How many?’” Indeed, an inside joke was that one turbine brand, the Dynergy American Windshark 80, had been dubbed a “sand shark” because the early models tended to leap off the tower onto the desert sands below.
“Once, Smith got a call from a site where several machines were ‘running away.’ He drove to the desert, along with some company executives, who tasked him with replacing the controllers as a means of recapturing the renegade machines. ‘The winds were howling,’ Smith recalled. ‘I’d have to be upwind of the turbines, hiding behind structures, with new controllers. When the winds lowered, I’d climb up to the control panel, take out the old controller, and put in a new one,’ he said, hoping that a rotor didn’t fly apart as he stood exposed.”
Ernie Tucker, “The Wild West of Wind,” Clean Energy Innovators: NREL People Working to Change the World
Smith left Dynergy in April 1986 and headed for greener pastures in the Altamont Pass. By then Dynergy was headed toward inevitable default.
Dynergy Dies
That’s not the way it was supposed to be. In December 1983, Cruver was promoting his vision in Palm Springs’ local newspaper, the Desert Sun. Dynergy, Cruver said, employed 24 and expected to add 80 more in early 1984 as the company expanded manufacturing. Everything looked rosy from his perspective.
His first site, Maeva, was 210 acres containing 58 of the company’s distinctive Windsharks. Cruver told the newspaper Dynergy planned another 100 turbines in early 1984. That never came about. It appears Dynergy only installed 98 turbines on the site, not 158.
Cruver’s pitch wasn’t unlike the other hustlers of the day. He claimed that his wind turbines were “a long-term revenue producing piece of property” with a life expectancy of 30 years. This was at a time when it was difficult to keep turbines like his operating for days at a stretch let alone years.
Investors bought the $125,000 machine with a $20,000 down payment, Cruver explained, getting the remainder from a bank. “The bottom line is you’ll get all your money back in tax credits and tax benefits,” Cruver crowed.[27]
It didn’t take long for things to turn sour. While the California wind industry was notorious for poor safety practices, Dynergy was in a class of its own.
In the fall of 1985, two men were seriously injured on one of the Dynergy sites. Andreas Romero, 22, and Mario Botello, 23, were in a basket suspended from a crane to repair the brake on a Windshark turbine.[28] The rotor was not properly secured—if it was secured—and the rotor began to turn. Winds were gusting to 25 mph at the time. The rotor cut the cable, and the basket dropped 50 feet to the ground with the men in it. Both men fractured their pelvis. Romero suffered serious head injuries. Botello suffered two broken ribs and a punctured lung.
Tellingly, in a report on the accident, the newspaper wrote that the turbine’s brake was inadequate for the design.[29]
Previously, Art Gomez slipped on the bed of a boom truck, landed on his head, and died at a maintenance yard on Dynergy’s Cabazon site.[30]
The injured workers eventually filed suit to recover their medical costs.[31]
The Maeva site was on land administered by Riverside County. However, Dynergy’s Cabazon site was on US Bureau of Land Management (BLM) land. In April 1986, BLM shut the site down for violating the terms of its lease. Dynergy would not be allowed to resume operations until it met BLM’s requirements. If the turbines didn’t operate for a year, BLM would declare them abandoned and order their removal.
By mid to late 1986 Dynergy was effectively defunct and preparing to walk away from its wind projects. In December 1986 Dynergy began negotiating with its investors for them to take over the 300 wind turbines at the Maeva and Cabazon wind farms and another two dozen turbines on the Sandberg lease in the Whitewater Wash.
Dynergy claimed it would only cost $1,000 to $6,000 per turbine to return them to service for at least five years. Dynergy didn’t have the funds to make the repairs itself. Nor could it pay its 100 creditors, including Southern California Edison (SCE). Dynergy owed the utility $100,000 for their interconnection.
In January 1987, Riverside County held a hearing on whether to revoke Dynergy’s building permit for the Maeva site. If they revoked the permit, the county would then order removal of the turbines.[32]
By late spring 1987, Riverside County revoked the operating permit for Maeva. Meanwhile a group of investors had taken over ownership of 60 of the 98 turbines that were on the site. SCE had shut off electricity to the site in the late summer of 1986 for failure to pay for their interconnection. The investors were considering moving the wind turbines from the Maeva site to the Cabazon site leased from the BLM. However, the Cabazon site had been closed by the BLM since mid 1986. If that didn’t work they were also considering moving the turbines to the Sandberg site.[33]
The owners were not going to restart the Maeva project. Riverside County planned to rezone the land, noting that it had been previously rezoned in error. According to the county, the Maeva wind farm was too close to West Palm Springs Village, a group of rural homesteads.
Consultants were evaluating whether the turbines at Maeva, Cabazon, and those on the Whitewater Wash could be repaired. All told, there were 300 Dynergy wind turbines and about 200 of those were in the investor group seeking a remedy. Owners of the remaining 100 turbines had to either repair the turbines themselves or remove them.[34]
By late July it was all over but the legal wrangling. 119 Dynergy investors filed a $1.9 million suit against the company for breach of their promises. Specifically, the investors argued that Dynergy did not repair or replace damaged turbines within a certain time frame.
Dynergy was effectively insolvent and “largely inactive.” Cruver and company were pinning their hopes on a new wind turbine, the Sumitomo 180.[35] It too was not to be. That’s a story for another time.
The CEC’s WPPRS
Tax abuses, broken and abandoned wind turbines, and permitting issues in all three major passes became a political problem for California’s new conservative Governor, George Deukmejian, and the California Energy Commission (CEC) in the early to mid 1980s. The eventual result was the implementation of a state law requiring wind farm operators to report their performance to the state.
The CEC’s Wind Project Performance Reporting System (WPPRS) went into effect in 1985, the last year of the lucrative federal tax credits. The state issued its first report the following year for the calendar year 1985. The reports were then issued annually well into the 1990s. They became the barometer of how the technology was evolving, what was working, and what was not.
The CEC’s system required operators to report quarterly on the number of wind turbines installed, those newly installed, nominal generating capacity, swept area, projected generation, and actual electrical generation.
The concept was revolutionary. Those outside the US may think this odd. The wind turbines had been built and sold on the basis of subsidies from the public purse. However, in the American context, the information the CEC required was confidential, top-secret, and proprietary. The last thing an American business wants to do is tell the government—and the world—how well or how poorly they’re doing. I remember the hue and cry from those in the industry at the time, but they grudgingly complied.[36]
All of this is by way of introduction to the data on the performance of the two wind turbines with CBR technology: Windtech, and Dynergy’s Windshark. The data only begins in 1985. By then, some of the many different kinds of wind turbines in California had failed so we don’t have a complete picture. But it’s the picture we have.
For example, the 30 Winsharks installed in Pacheco Pass had already been replaced by WestWind Industries turbines. In the case of Zephyr in the Tehachapi Pass, they only reported that they were looking for a fix of their Windtech turbines, which, apparently, they never found.
So the statewide tally for 1985 was 212 Windtech turbines had been installed. Dynergy reported only 176 of the supposedly 300 wind turbines they had built. With the 30 from Pacheco Pass, we have a record of 206 Dynergy clones. Altogether, there were some 400 wind turbines in California that were using the CBR technology in 1985. The bulk of them were gone by 1988, if not sooner.
CBR Performance
Unlike a helicopter that must be maintained constantly for safe flight, a wind turbine must operate for days, if not months, unattended. It’s a power plant that must work whenever the wind is present.
Modern wind turbines do, and even some of those installed in California during the early 1980s were on the right track. These were not Windtech and Windshark turbines. The CBR concept reached a dead end in California. The wind turbines that used it were unreliable and unsafe. Their performance reflects this.
The CEC’s annual WPPRS summaries reported both capacity factor and annual yield. Windtechs performed better than Windsharks, but not by much in either metric. Their performance was dismal relative to another American manufacturer, US Windpower, and to a well known Danish manufacturer, Bonus. Both of these wind turbines averaged a capacity factor of 17% to 26% and delivered an annual yield from 550 kWh/m² to 900 kWh/m². Neither Windtech or Windshark produced capacity factors greater than single digits and annual yields of little more than 100 kWh/m².
Dynergy’s investors stopped reporting generation in 1987. Windtech hung on longer. They reported generation through 1991.
After nearly two decades, the technology had reached a dead end and was gone.
Well, not quite all of the technology, just the CBR technology.
Cheney’s 5-Blade Windshark
The composite bearingless rotor had reached a dead end, but the pultruded fiberglass blades that were used with the technology had not. The promise of cheap pultruded blades hadn’t died quite yet. Maybe if they could be used with a different overspeed control strategy, such as air brakes or flaps, they still had promise.
Flaps, air brakes, and spoilers had been used before on wind turbines in California.[37] Windmatic had used spoilers on both its 14s and 15s. Northern Power developed flaps, or what they called ailerons, on their prototype teetered two-blade 250 kW turbine developed for DOE. But no one had used them on the constant planform of a pultruded blade.
This is where the National Renewable Energy Laboratory (NREL) steps in. The industry was evolving rapidly and NREL was looking for an innovative medium-size turbine design that would be competitive with designs from the Danes and the Germans. Consequently, NREL launched a procurement for Next Generation Innovative Components (NGIS) that would improve the performance and reduce the cost of components.
NREL envisioned a 400 kW wind turbine driven by 33-meter diameter rotor. Developing a medium-size turbine from scratch was beyond NREL’s purview, so they decided to test a scaled version of the concept.
Blades are a critical and costly component of wind turbines. NREL wanted to explore reducing the cost of a wind turbine by using pultruded blades. The blades were indeed cheap. Cheney had used pultruded blades from Morrison Molded Fiberglass on his turbines for $4/lb. This compared, he argued, to the $10/lb of blades made by hand lay up.[38]
Though the blades would be cheap to make, they are not ideal for wind turbines. Blades made from straight pultrusions suffer a 12% performance penalty compared to blades with twist and taper.[39]
NREL contracted Cheney’s PS Enterprises in the fall of 1994 to help NREL “field test a dynamically scaled version” of a 400 kW rotor using pultruded blades. In a bit of understatement, one report acknowledged that “The history of turbines employed with this technology during the 1980s was not good. Problems included tower strikes, upwind running, and stall flutter.”[40]
However, Cheney would abandon the CBR concept for overspeed control and instead use what the reports call a “spoiler-flap” on the pultruded blades. Further, the test bed would not use two, three, or even four blades. No, in a striking departure from previous wind turbines, Cheney proposed using five of the flexible pultruded blades on a 15.5-meter diameter rotor. The assemblage would be mounted on an existing Dynergy Windshark chassis at the Cabazon site in the San Gorgonio Pass.[41]
Unfortunately, there was no longer electricity at Dynergy’s Cabazon site. So NREL had to bring in a portable 155 kW diesel generator. Testing began in November 1996.
It did not go well—or last long.
On 22 December 1996 in “relatively high winds” of about 15 m/s the generator shaft failed and the turbine immediately went into overspeed, “reaching speeds 200% of normal.” Because there was still power to the turbine, the spoiler-flaps didn’t deploy automatically. The flaps had to be manually activated, bringing the rotor to what was thought a safe speed. However, only three of the five flaps were operating as designed. The flap on the #5 blade didn’t deploy. And the flap on the #1 blade was oscillating between a partially closed position and its deployed position. When the rotor sped up, the #1 flap would deploy fully, but as the rotor slowed down it would return toward its stowed position.[42] The rotor continued to cycle in this manner for 24 hours when one blade failed at the root, the resulting imbalance then threw the rest of the blades destroying the turbine.[43]
The turbine was destroyed so soon after testing began that NREL collected only 70 minutes of performance data. Normally weeks of data are necessary to produce a power curve. The data they did record showed that the turbine produced 70 kW at 19 m/s.[44]
Due to the destruction of the test turbine, NREL chose not to extend testing further.[45]
The study concluded that the five-bladed rotor “demonstrated stable operation” despite its ultimate failure. They also found that proper rotor design “ensures yaw stability and avoids tower strikes,” the bane of both Windtech and Windshark turbines. Importantly for NREL, there were no “fatal flaws” in the concept and the use of pultruded blades in this manner could cut rotor cost by 55% to 74%.[46]
Cheney, Migliore, and colleagues published a series of technical papers in 1999 and 2000, summarizing their work on the shortened field test and on pultruded blades.
After a quarter century of development, this brought to an end any remaining link to the original Composite Bearingless Rotor concept hatched by Cheney and Spierings at UTRC during the heyday of ERDA and DOE’s wind program.
[1] Another subsidiary of United Technologies, Hamilton Standard, designed and built the WTS-4, a 4.2 MW wind turbine installed at Medicine Bow, Wyoming in 1982. It did not fare well either.
[2] Sikorsky’s UTTAS design evolved into the Army’s Black Hawk helicopter. “Sikorsky S-70 UTTAS Prototype – Igor I Sikorsky Historical Archives,” accessed August 7, 2024, https://sikorskyarchives.com/home/sikorsky-product-history/helicopter-innovation-era/sikorsky-s-70-uttas-prototype/.
[3] ERDA was the forerunner of the Department of Energy.
[4] M. C. Cheney and P. A. M. Spierings, “Self-Regulating Composite Bearingless Wind Turbine,” vol. 7, 1976, 328–48, https://ui.adsabs.harvard.edu/abs/1976ssst….7..328C.
[5] M. C. Cheney and P. A. M. Spierings, “Self-Regulating Composite Bearingless Wind Turbine,” Executive Summary, Energy Research and Development Administration Division of Solar Energy (East Hartford, Connecticut: United Technologies Research Center, September 1976), https://www.osti.gov/servlets/purl/7257566.
[6] Downwind turbines were all the rage among American engineers at the time for their presumed cost savings.
[7] Bergey Windpower and Storm Master also used pultruded blades at the time. Bergey used the technology successfully for decades. Storm Master, in contrast, didn’t make the technology work.
[8] It’s important to note that engineers of the day thought stall was sufficient to protect wind turbine rotors in a loss-of-load accident. Early Danish turbines in this period also lacked aerodynamically activated overspeed controls. It was only after the loss of several wind turbines during a noteworthy storm that Danish owners began insisting on fail-safe overspeed controls. Most Danish wind turbines opted for pitchable blade tips. Windmatic chose air brakes. In the US, Enertech and ESI added tip brakes to their wind turbines for overspeed protection. UTRC’s pultruded blades may have precluded the use of tip brakes though NREL’s revision of the design in the 1990s incorporated air brakes.
[9] Marvin Chapin Cheney and Petrus A. M. Spierings, Wind turbine with automatic pitch and yaw control, United States US4083651A, filed August 17, 1976, and issued April 11, 1978, https://patents.google.com/patent/US4083651/en.
[10] Four blades are clearly more expensive than two. However, the use of four blades may have made the commercial version of the turbine more dynamically stable. We will never know if the use of four blades would have made a difference in the design’s success. We do know that when NREL revisited the design in the 1990s, they used five blades.
[11] M. C. Cheney, “Development of an 8 kW Wind Turbine Generator for Residential Type Applications. Phase I: Design and Analysis. Volume II. Technical Report” (United Technologies Research Center, East Hartford, CT (United States), June 25, 1979), https://doi.org/10.2172/5384412, and R. B. Taylor and M. C. Cheney, “UTRC 8-kW Wind System: Phase II – Fabrication and Test. Executive Summary” (United Technologies Research Center, East Hartford, CT (United States), February 4, 1981), https://doi.org/10.2172/6149882.
[12] Earl H. Gilmore, “Use of Wind Power to Assist in Stripper (Oil) Well Pumping” (Canyon, Texas: Alternative Energy Institute, West Texas State University, August 31, 1981), https://www.osti.gov/servlets/purl/6607736.
[13] R. B. Taylor and M. C. Cheney, “UTRC 15-kW Wind-System Development. Phase I. Design and Analysis. Volume II. Technical Report” (Rockwell International Corp., Golden, CO (United States). Rocky Flats Plant; United Technologies Research Center, East Hartford, CT (United States), December 1, 1981), https://doi.org/10.2172/6811444, and M C Cheney, “UTRC 15 kW Wind System Development: Phase II – Fabrication and Testing,” February 1982, https://www.osti.gov/servlets/purl/6069601.
[14] U. Hütter, “Past Developments of Large Wind Generators in Europe,” in Wind Energy Conversion Systems Workshop Proceedings, vol. NSF/RA/W-73-006 (Washington, DC: National Science Foundation, 1973), 19–22.
[15] Gilmore, 1981, Use of Wind Power to Assist in Stripper (Oil) Well Pumping.
[16] Ken Starcher, “UTRC/Dynergy/Windtech?,” email, June 9, 2024.
[17] Petrus A. M. Spierings, Composite Bearingless Rotor History, Telephone, August 20, 2023.
[18] Mike Powell, “Windtech Kip Cheney Solano County 1980s,” email, July 21, 2014.
[19] “Results from the Wind Project Performance Reporting System: 1985 Annual Report” (Sacramento, California: California Energy Commission, August 1986), https://web.archive.org/web/20150906082646/http://www.energy.ca.gov/wind/documents/1985-1993_reports/WPRS_1985_P500-85-013.pdf, and “Results from the Wind Project Performance Reporting System: 1986 Annual Report” (Sacramento, California: California Energy Commission, January 1988), https://web.archive.org/web/20150906080857/http://www.energy.ca.gov/wind/documents/1985-1993_reports/WPRS_1986_P500-87-019.pdf, and “Results from the Wind Project Performance Reporting System: 1987 Annual Report” (Sacramento, California: California Energy Commission, August 1988), https://web.archive.org/web/20150906081401/http://www.energy.ca.gov/wind/documents/1985-1993_reports/WPRS_1987_P500-88-005.pdf.
[20] There are two other wind energy sites in the Pacheco Pass of historical interest. The Romero Overlook Visitor Center was the site of a very early DAF-Indal Darrieus wind turbine that stood overlooking the San Luis Reservoir in the late 1970s or early 1980s. On the southwest side of the reservoir there was a wind farm at Dinosaur Point. This site has since been repowered and is now called the Gonzaga Ridge wind farm.
[21] Kevin Jackson, Kevin Jackson WestWind 20240607, Telephone, July 6, 2024.
[22] Peter Asmus, Reaping the Wind: How Mechanical Wizards, Visionaries, and Profiteers Helped Shape Our Energy Future (Island Press, 2001), p. 116-117. Asmus book remains the most authoritative as well as the most entertaining exposé of the Great California Wind Rush.
[23] Bruce Hammett, Dynergy Windsharks Cabazon and Bruce Hammett, Telephone, February 2, 2024.
[24] Amir Mikhail joined Dynergy shortly after Smith. Mikhail went on to work on the Sumitomo design and eventually he designed wind turbines for Zond.
[25] Brian Smith, Brian Smith Dynergy Days 20240621, Telephone, June 21, 2024.
[26] Ernie Tucker, “The Wild West of Wind,” Clean Energy Innovators: NREL People Working to Change the World (Golden, Colorado: Alliance for Sustainable Energy, June 2022), https://www.nrel.gov/docs/gen/fy22/83177.pdf. p. 32-36.
[27] Amanda Covarrubias, “Firm Works to Harness the Wind,” Desert Sun, December 31, 1983.
[28] This was a common practice of the day, but is no longer permitted because of this and many other similar accidents, including in Europe.
[29] Sanford Nax, “Windmills: More Safety Sought OSHA Officials Seeks Industry Standards,” Desert Sun, November 22, 1985.
[30] My notes only say that in 1984 Art Gomez died on a Dynergy project in the San Gorgonio Pass. However, Brian Smith has provided more detail. Brian Smith, Art Gomez’s Death, E-mail, September 4, 2024.
[31] Peter Hass, “Second Worker Files Turbine Suit,” Desert Sun, January 15, 1986.
[32] Sanford Nax, “Ailing Company Asks Investors to Take Over Wind Farm Operations,” Desert Sun, December 20, 1986.
[33] Sanford Nax, “New Owner Wants to Keep Turbines: County, Machines Must Be Dismantled,” Desert Sun, May 19, 1987.
[34] Sanford Nax, “Rezoning Proposal to Force Removal of Troubled Windmills,” Desert Sun, July 3, 1987.
[35] Peter Hass, “Windpark Developer Hit with ‘Friendly’ $1.9 Million Suit,” Desert Sun, July 21, 1987.
[36] Reporting requirements have since changed and the data is no longer available to the public. The wind industry eventually got its wish.
[37] Flaps, air brakes, and spoilers are embedded lengthwise along the blade. They differ from tip brakes and pitchable blade tips that are added at the end of the blade.
[38] M C Cheney et al., “Analysis and Tests of Pultruded Blades for Wind Turbine Rotors” (SERI/NREL, July 19, 1999), https://doi.org/10.2172/12142, p. 13.
[39] M. Cheney and P. Migliore, “Feasibility Study of Pultruded Blades for Wind Turbine Rotors,” in 2000 ASME Wind Energy Symposium (2000 ASME Wind Energy Symposium, Reno,NV,U.S.A.: American Institute of Aeronautics and Astronautics, 2000), https://doi.org/10.2514/6.2000-61, p. 9.
[40] Cheney and Migliore, 2000, p. 1.
[41] Cheney and Migliore, 2000, p. 8.
[42] Other wind turbines of the period also suffered from the tendency of aerodynamically activated tip brakes or pitchable blade tips to deploy during an overspeed event, but as the rotor slowed down they would start to return to the operating position and the rotor would again speed up. A related problem occurred when the overspeed devices would deploy on two blades sufficient to slow the rotor, hindering the third device from deploying.
[43] Cheney et al., 1999, p. 50.
[44] Cheney et al., 1999, p. 51. Note that wind turbines of the day were often rated at a wind speed of 15 m/s (34 mph) and not at 19 m/s (43 mph).
[45] Cheney et al., 1999, p. 14.
[46] Cheney et al., 1999, p. 82.
