Control panel of the historical cyclotron:

Peter van Leuffen - A history of the first cyclotron

Peter van Leuffen in front of the historical cyclotron |
Peter van Leuffen (born 1946) was one of the first staff members who worked on BV Cyclotron VU’s most important machine: their first cyclotron. It became operational in 1964, and two years later van Leuffen started his career as a young electrical engineer with this exciting, new piece of technology.

Gigantic electronic apparatus
"I didn't actually see the cyclotron coming in. And I also missed out on the first particles that were accelerated. But since then I have pretty much been involved in every single electrical component and every quirk of this marvelous, gigantic electronic apparatus.
This machine was operating at the forefront of what at that time was viewed as the 'Ultimate Future': the atomic age. The cyclotron itself was built by PhD students and other researchers from Eindhoven University of Technology (TU/e). Early in the 1960's Philips started to get involved. With Philips and the people from the TU/e, a total of nine of these cyclotrons were built.
The cyclotron was a huge machine. Its pole diameter was 1.4 m, and it was one of the first cyclotrons beyond the classic, much smaller range. Its magnets were made from 25 t of low carbon steel with two copper plated poles. With all the electronic equipment included, it probably weighed about 55 t. It was located inside an inner vault with concrete walls and doors about 2 m thick to shield the surroundings from the radiation present when the machine was running."

Acceleration of ion
"The exciting part for me was that electronics rules almost everything within a cyclotron: it all starts with transformers that apply a high-frequency AC voltage to two electrodes. Then, in the space between the electrodes, there is a vacuum chamber with an ion source. This source produces either positive or negative ions depending on the configuration. A positive ion is accelerated toward the electrode with a negative potential. Because of the perpendicular magnetic field present in the electrodes, the ion is deflected into a semicircular orbit until it appears again at the gap between the two electrodes. The next time the ion appears at the gap between the electrodes, the alternating voltage applied to the electrodes has been reversed. In its current path, the ion will be accelerated again, this time with a negative potential. Each time an ion crosses the electrodes, it is boosted to a higher energy level. As a result, the radius of its orbit increases. This produces the typical spiral path of the particles.
The acceleration continues until the particle escapes from the electrode. At that moment it is extracted into the target chamber. All the subatomic particles together will then form a beam that can be used to bombard a variety of target materials to produce radioactive isotopes."

Operation by hand
The control panel of the historical cyclotron machine in Amsterdam |
"Back in 1970, there were no microprocessors or microcontrollers to guide this process. Everything was operated by hand. The numerical calculations were all done manually as well. The operators controlled the particle beam by calculating the required power voltage and frequency for the electrodes and for the magnetic field so that it would be possible to center the beam within acceptable limits. Our cyclotron was used as an instrument in experimental physics, therefore, the operational power could be regulated to a value between 3 and 40 MeV. Without computers or automatic control systems, the operators needed to have expert knowledge of the machinery because behind every operational dial, switch, or button, there was a piece of electronic or mechanical hardware."

Handmade electronics
"The electronic hardware was a wild mix of new, refurbished, or custom-made electronic equipment. Many of the components had been recycled from old Philips X-ray machines. So, although the first transistors were available in the early 1960's, there weren’t any in our cyclotron. Everything was built with vacuum tubes (the glowing bulbs similar to the ones used in old television sets or radio amplifiers). The high voltage switches were large Bakelite door handles. The cabling was made of — mostly recycled — old copper wires with black or brown burlap insulation. It was — even at that time — not what you would call 'state-of-the-art' electronics. Everything was handmade. And we had to improvise a lot. But it worked, and we were excited to be working on one of the few large particle accelerators in the world."

From tubes to transistors
"I specialized in transistors. I just came out of school and was, of course, excited about the possibilities of the new technology. The engineers here were debating the pros and cons of tubes versus transistors in high-frequency electronic equipment. In terms of frequency response, distortion, and noise measurement, transistors were at least equally as good. And they were far superior in terms of life span, power consumption, and reliability.   
So, we wanted to transform the electronics of the cyclotron from tubes to transistors. However, that was easier said than done. In 1970 Philips decided that the field of particle physics would never turn a profit, so they stopped building cyclotrons (read why by the then director of Philips Natlab, Casimir). The only option we had was to take every individual electronic unit, reverse engineer its functionality, and then rebuild it from scratch using transistors."

a Philips' design manual of a cyclotron |
Do-it-yourself
With seven years of operational experience and access to Philips' design manuals we decided to give it a go. Our team consisted of seven electrical engineers, five mechanical engineers, and a physicist. The task was a formidable one: would it be possible to build a safe, reliable, manageable transistor-based equivalent to all this complex high-frequency, high-voltage equipment?
In 1972 we started replacing the high-voltage control units of the (radio) tubes with transistors and analog amplifiers. The high voltage of 50 kV is necessary for accelerating the particles. Also, the transformer for the main magnet coil was replaced by one we bought in Denmark. It was an electrical motor that would function as a DC-voltage generator, turning 380 volts AC to 250 volts DC with 300 ampere. These motors were huge, loud, and prone to failure. But these improvements turned out to work very well, until today."

Complicated issues
controllers of the historical cyclotron machine |
"During this kind of replacement work, we had to deal with extremely complicated issues. For instance, the electrodes were cooled by a special oil-based system. At one time, the container for the oil leaked and that almost gave me a heart attack. You can’t just wipe the oil off: you have to deal with high doses of radioactivity, and the machine is operational at high vacuum. Anyway, the oil had been spread all over the cyclotron. And there was no easy way to get it out. In the end, we attached coils to the cyclotron so that we could heat it. The idea was to slowly evaporate all the oil out of the cyclotron. It took us weeks, but we succeeded. After that we were determined to rebuild the whole cooling system based on a system using xylene. The advantage of that was that if the xylene would ever leak in the machine again, you would be able to evaporate it easily.
Later on, we built some of the first programmable logic controllers (PLCs) to make some of the control systems regulate frequencies and voltages automatically. These controllers were extremely sensitive, and they would often fall to pieces. But we were able to make it work, and with that we were able to control the beam with extreme precision."

Experimental fever
"What I liked most about all this was that you get infected with a kind of experimental fever—you want to understand things, to make solutions that work. If it takes day and night to fix things, then you will do it because your heart and soul are in that machinery. Fortunately for us, there was plenty of time for such exploration and development as long as we were part of the VU University Amsterdam.
However, all that ended in 1986. The Dutch government decided that with the new large accelerators at CERN, there was no place for a particle accelerators in the Netherlands anymore. So, our department was closed. All the staff was dismissed."

Commercial venture
"But together with seven other people, I refused to accept the dismantling of the cyclotron! That would have been an unacceptable destruction of intellectual and financial capital. So we got together and brainstormed about what we could do with the cyclotron. We did a feasibility study on operating a cyclotron as a commercial venture.
The prospects didn’t look very good — but there was a small chance of earning our keep with bulk production of radiochemicals such as thallium, gallium, and rubidium. None of the raw materials or basic equipment for that were available, so we had to develop everything from scratch. And then we started producing:  24 hours a day, 7 days a week.
For me, that was one of the most stressful periods of my life. We only had one cyclotron. And it was our lifeline. We needed to keep it going. Even with too little time and resources available to do the maintenance. I think we have been lucky that it worked out so well."

Worth the risk
"By operating the cyclotron so intensively, we gained an enormous amount of experimental knowledge. And that knowledge gives us a clear advantage even if we use modern cyclotrons that are fully operated by computers today.
When I look back at this transition from a purely academic setting to the commercial operation of our cyclotron, I am just very proud that we managed to do it. It was an adventure and a risk. But we made it."