John Miller, our resident supercap expert, recalls the early days of ultracapacitor work in the U.S. – including the days when he shook the streets of Las Vegas, burnt out an army truck, and more.
On 10 October 1990 the U.S. Department of Energy published a solicitation for the development of ultracapacitor technology intended for use in electric vehicles. The solicitation, published in Commerce Business Daily, was for four years of funding to develop prototype devices that were rated at 20V and would store about 1.8kJ of energy. The goal was an energy density of 4 Wh/kg. This solicitation marked the beginning of serious U.S. Government involvement in the development of large electrochemical capacitors.
Before this annou-ncement some open work had already been done on developing large capacitor devices. While at SOHIO (Standard Oil of Ohio) the author of this article had been involved in some of it, funded by ‘Star Wars’ defence system money. Our programme eventually developed an electrochemical capacitor device much larger than anything available at the time. As its primary purpose was simply to demonstrate capacitor technology to the military, however, it did not go on to become a commercial product.
At the time of that original solicitation the largest electrochemical capacitors commercially available in the West were Japanese products built by Nippon Electric Corporation (NEC), the licensee of the SOHIO technology mentioned earlier. These devices, state-of-the-art for the period, stored in the order of 15J and were rated at 5.5V. Although not particularly powerful, they demonstrated exceptional cycle life.
Maxwell Technologies was awarded the contract resulting from the DOE solicitation and began a programme. First located at Auburn University, Alabama, this programme created a device based on an aqueous electrolyte and a composite electrode made of nickel and carbon fibres. Early results were reported in December 1991 at the International Seminar on Double Layer Capacitors and Similar Energy Storage Devices in Deerfield Beach, Florida.
At that same meeting NEC described a 1,000F, 5.5V device which was basically a scale-up of its smaller devices. The continuing motivation for this development was the perceived need for stored energy in electric vehicles. Unfortunately the NEC capacitor had very thick electrodes, in the order of 6mm in thickness, making it very slow in response. Its time responses were in fact more like those of a battery than a capacitor. Such devices were never commercialised.
Also at the Deerfield Beach meeting Panasonic described a 500F single-cell electrochemical capacitor rated at 2.3V, and Asahi Glass Co. described a 240F, 2.5V cell. These both used a spiral-wound construction that, again, had been developed most probably in response to the publicised needs of the American automobile industry for electric vehicle technology.
Devices like these represented the known state of the art until the December 1993 capacitor meeting, again in Deerfield Beach. There Dr Alexander Ivanov presented a paper describing capacitor technology so advanced that most of those in the audience summarily dismissed it. He described capacitors – not just in development but already in production – that were used for starting 3000HP locomotive engines. These PSCaps were up to 60kJ in size, rated at 12V or higher, used bipolar construction, and were capable of delivering many thousands of amps of current.
Dr Ivanov in fact gave two papers – one on the technology and the other on applications, both with co-authors. They presented advances well beyond anything available elsewhere at the time. The devices he described were not only very large but very powerful – much more powerful even than batteries. In Russia they apparently were already being used to replace batteries for cranking large engines like those of locomotives.
Powerfully impressed by the Ivanov presentation, I inquired into obtaining a sample PSCap capacitor to test whether the devices actually functioned as described. This took some time to accomplish. The capacitor finally sent to me was a 64V rated device of some 40kJ. The format was a right cylinder approximately 9 inches in diameter and 17 inches long, weighing about 70 lbs, with an electric terminal at each end. It was in general both much larger than anything previously encountered and of a very different and unfamiliar design.
Testing proved, however, that the device functioned exactly as reported in terms of both energy and power performance. I presented these results in a paper titled ‘Technical Status of Large Electrochemical Capacitors’ at the Twelfth International Seminar on Primary and Secondary Battery Technology and Applications at Deerfield Beach in March 1995, showing that the device indeed had exceptional capabilities compared with those being designed and constructed in the West.
Interest continued in the product and I was eventually invited to visit Dr Ivanov and his organisation in Moscow, which I did in December 1995. I met him and his associates, learned more about their technology, and additionally enjoyed an opportunity to learn more about Russia and its people. My visit to the Ivanov organisation, the beginning of my own interaction with Russia’s technology community, led eventually to contacts with several other Russian companies.
Appreciating the availability of the Russian technology and its remarkable degree of advancement, I actively sought out opportunities to put it to use. One of the first was in a city transit bus; in fact it was the first gas-electric hybrid power system in the U.S. with capacitor energy storage. This was a demonstration vehicle created by a group including NASA, Lincoln Electric (Motor Division), Bowling Green State University, the Flexible Bus Company and the Regional Transit Authority of the City of Cleveland, Ohio.
The city provided the 40ft city transit bus that served as the platform for conversion. The storage system was composed of PSCap capacitors ordered from Dr Ivanov’s group. The energy storage system involved twenty modules, each rated at 200V, and having total stored energy in excess of 1MJ. The modules were configured to create 400V operation. At the time this was the largest assemblage of capacitors in the U.S.
The project was completed in October 1997 and performance testing was done at the test track of the Transportation Research Center in East Liberty, Ohio. Publications on the hybrid vehicle reported substantial energy savings and emission reductions. This bus was the forerunner of a great number of hybrid buses now on the road in the U.S. that use capacitors for energy storage. This project, built on PSCap technology, was the first of its type in the country.
The hybrid demonstration bus was a serious project conducted by a group of well-respected organisations. But the author of this article also saw prospects for this capacitor technology in a very different and arguably less serious venue – the car audio market. At the time of the bus project, 1F electrolytic capacitors were often used in car audio systems to ‘stiffen’ the voltage rail and thus deliver increased power when needed to produce particularly loud base notes. Without this additional power from the stored energy, headlights could dim in time to the music, and the engine might even stall.
A 1F capacitor was generally deemed sufficient for a 1kW power supply, a rule of thumb generated more than anything else by the fact that there were no larger capacitors readily available. A demonstration system for this purpose might in fact put two or more capacitors in parallel. This application needed more stored energy than was rapidly available.
With an idea to use the PSCap for the same application, I became involved in a project funded by Boss Audio, a major supplier of aftermarket car audio equipment, including amplifiers, speakers and other associated items. The project planned to implement capacitor energy storage in a large demonstration vehicle featuring Boss Audio components. It was to be built on the chassis of a Brinks armoured truck of the type used for transporting money to and from banks. The particular vehicle to be used had 4 inch thick glass windows and heavy doors, and was heavily-built overall. The demonstration vehicle would be developed to feature in a show booth at the yearly Consumer Electronics Show in January 1997.
A Brinks truck proved somewhat difficult to obtain, owing to general security concerns over their potential uses; eventually, however, Boss Audio succeeded in purchasing one. Our first step was to ‘gut’ the truck, preparing it to hold the audio system that would feature the Boss components. Work continued through the late summer and on into early December 1996. The showpiece audio system included 108 speakers in total, including sixty 15 inch bass woofers. Its amplifiers had a dynamic power of 120kW and continuous power in excess of 20kW. Along with all this it also had the appropriate midrange speakers and tweeters. The sixty big speakers and the bank of amplifiers together made an impressive sight.
To power all this we had a bank of batteries and ten associated modules of the PSCap. Five of the PSCaps were of a special design for extra high power and five were the standard design, totalling 1,266F of capacitors inside the truck. With so much current flowing, special bus bars of solid copper weighing nearly 200 lbs were needed to connect the storage system to the amplifiers. The schedule for completing the whole system was very tight, given that it was a one-off and involved extensive implementation. The involvement of my company, JME, after designing the capacitor system was to demonstrate the advantages of using the PSCaps to increase the sound pressure level that could be achieved in the truck. This involved measuring the total power delivered by the energy storage system when power-peaks were reached in typical ‘boom-boom’ music with and without the PSCaps.
The time limitations were made even more difficult by the fact that the truck had to be driven to Las Vegas, Nevada, in time for the Consumer Electronics Show. This meant that to meet the schedule it must leave for the show no later than the last week of December. JME was on 24-hour call to make the final testing at any moment, complete with a storage oscilloscope, shunt resistors and other test equipment to measure performance and to give the system its final tuning. Test results were to be used in PSCap promotion in the car audio market.
One evening at about 11.30pm the call came that testing should begin at once. Despite the lateness of the hour we packed up our instrumentation equipment. We were particularly aware that it would be the first occasion that this massive system had been turned on. After installing our shunt resistor in the power bank, along with an oscilloscope to measure voltage and currents, we turned the system on and began playing typically obnoxious car stereo music, replete with heavy booms and reverberating rhythms.
For obvious reasons we set the volume low at first. But in the presence of a system that size, seeing just how much it can do is a temptation beyond human power to resist. Once the volume was turned up, we began to conduct measurements, finding that an amazing 10,000A was being delivered by the capacitor system during the heavy ‘booms’. An immediate and unexpected consequence, however, was that burglar alarms in businesses up and down the street began to go off. The hour being well after midnight, we promptly turned down the volume in anticipation that the police would soon arrive to investigate the disturbance.
This ended our first test of the Boss Audio demonstration system. The vehicle did indeed get to Las Vegas on time and was a tremendous hit at the Consumer Electronics Show. It featured the ten large PSCap capacitor modules, painted specially for the installation and designated ‘flux’ capacitors. This was several years after the film Back to the Future came out, so the flux capacitor was indeed something special. Figure 5 shows promotional art work developed for the flux capacitor. Unfortunately for Dr Ivanov’s group, but probably fortunately for most U.S. automobile drivers, this application for the PSCap was never commercialised. Without doubt this was a largest ‘fun’ system which JME was involved in that made use of the PSCap.
On a more serious note, the PSCap was originally designed for engine cranking. One other American market for it that seemed at the time particularly ripe for development, however, was use in connection with the stationary generators maintained as part of the U.S. telephone grid system.
Upon inquiry JME was invited to demonstrate such a capacitor system to AT&T (formerly the American Telephone & Telegraph Company). This demonstration took place at an AT&T operating station in New Jersey, one with three large diesel generators set up especially to power the relay station should utility grid power fail. Each generator system included a 16-cyclinder, approximately 20ft long, 1,500HP Cummins diesel engine directly connected to a generator without any clutch mechanism. Normal starting for these engines required a cabinet full of aircraft Ni-Cad batteries that provided 28V for cranking. Each engine used two 12V starters connected in series.
At the invitation of the engineering manager of the facility, I arrived with a storage oscilloscope, shunt resistor, cables and two PSCaps, both special high-power devices together storing about 40kJ of energy. The AT&T staff displayed open amusement that we expected to start one of their engines with nothing but two capacitors each 9 inches in diameter and about a foot tall. As we set up, the three engineers at the facility maintained a visibly sceptical attitude, plainly not believing that a power source as small as this could do the job. We continued with the setup, however, connecting the PSCaps in series to provide the 28V needed for cranking.
The control room was the typical glass-walled enclosure off to one side, where the operator could view the operation and fire up the engines or perform other tasks connected with the power electronics of the station. The demonstration began quite properly by cranking one of the engines with the Ni-Cad battery system to ensure that it would, in fact, start. We recorded the current and voltage during the start and found it very impressive – several thousand amps. I must admit that at that point I felt some doubt about our chances of success but did my best not to let it show.
We connected the PSCaps for charging and, using the battery system, charged them to 28V. We then set up the capacitors in the starting system as the sole source of cranking power and signalled the control room operator to start. The engine, much to my relief, started right up. Figure 6 shows current and voltage profiles measured during this engine start. Notice that the peak current was almost 7,000A. The system voltage immediately dropped to about 10V and then rebounded to just above 12V at about 200ms. The engine was up and running within one second.
At that point, the noise level in the room, owing to the engine operation, was very high. This was particularly unpleasant, given that none of us was wearing ear-protection equipment. When, however, to remedy this the operator went to turn the engine off, he discovered that he could not do so. The engine would idle, but it simply would not shut down. What had happened, apparently, was that the system as a whole had begun to operate at the discharged, depleted capacitor voltage, about 12V. This had thus become the only voltage available for all of the control circuits, including the one designed to shut the engine off. In the end, we were only able to shut the engine down once we had recharged the capacitors from the batteries. The ‘last laugh’ of the day was ours. Not only had we managed to start the massive engine with the ‘tiny’ capacitors, but we had even for a while taken control of it away from AT&T.
I had been working with the U.S. Army for some time on engine cranking systems for five- and seven-ton trucks. The Army wanted, of course, better cranking performance at low temperatures. JME had been quite active in using PSCaps for this kind of application, and a number of the devices had been ordered and were being tested by the Tank & Armament Command in Warren, Michigan. Because the U.S. Army felt uncomfortable about buying equipment from a Russian company, I was asked to do so and eventually bought quite a number of PSCaps for this purpose.
Test results were generally very positive, with excellent cranking at low temperatures. I did, however, get a concerned call one afternoon saying that one of the trucks being tested had been destroyed by fire, with the capacitor believed to have caused this. Upon asking for further details, I learned that the capacitor had been installed under the driver’s seat in the truck cab. The protocol for the test was to start the truck in the morning, let it idle all day long, turn it off at the end of the day, and turn it on once again the next morning – essentially a mild endurance test of the starting system.
This had been going on for several months when, on the day of the incident, the truck had started normally but sometime during the afternoon it had caught fire.
It was completely consumed by the flames; a total loss. A preliminary examination showed that the fire had originated in the cab and strongly suggested that the capacitor was the cause. The PSCap was extracted from under the seat deep in the burnt-out remains and shipped to me for a post-mortem. Was the PSCap the cause of the fire and, if so, why?
Briefly, although the capacitor was indeed the source of the high temperature that caused the fire, a closer look revealed that it had in fact already been seriously damaged through improper installation. The securing clamps encircling the capacitor had been overtightened and crushed the case, causing shorting between the grounded case and the internal current collectors.
Thankfully for JME and the future of the project, the specified installation procedures had in fact warned of such a possibility. The effect of a 150A, 24V alternator dumping current continuously into that short was certainly more than enough to generate precisely the high temperatures that had caused the fire. The destruction of the truck led to the development of new mounting hardware for further tests that were subsequently performed. The fire did not end the programme, but it certainly held it up for a time while the problem was identified and solutions to it worked out.
The PSCap was the first large electrochemical capacitor product available in the American market. As indicated, it saw some important early applications, including the demonstration of a hybrid city transit bus, a massive car audio system, and engine cranking for military equipment. That this technology has had such a historically significant impact from early on is owed largely to its robustness and its extraordinary power performance.