PAS2000A Hamburg

After my PAS2000A installation adventures at IBM, the next challenge was installing another PAS2000A system at Valvo - a Philips subsidiary - in Hamburg. This installation went far from smoothly. The system was transported from Eindhoven to Hamburg in a special temperature-controlled truck, as the equipment is sensitive to big temperature variations. Before we left, the propellers of the city hopper with which we flew from Eindhoven to Hamburg were cleared of ice by men with long brooms. When we arrived at the Valvo premises, it was freezing cold. The PAS2000A had arrived a day earlier and was stored in a basement.When we arrived, I noticed the basement door was open, probably all night. A big shock! We measured a temperature of 6°C in the vicinity of the stepper. Together with our customer, we decided to proceed with the installation. The exposure unit was unpacked and placed in a large transport elevator. With the aid of several electrical heaters that magically appeared from somewhere, we slowly brought the temperature to the right value around 20°C. You can imagine that occupying the elevator for almost a day was not much appreciated, as it interrupted the planning of other activities. We tried not to pay much attention to the complaints; what else could we do to save our system?

The next obstacle to overcome was moving the stepper into the cleanroom. The cleanroom entrance was about 1 meter above the floor level. Add to that, the aisle was as wide as the stepper. It was a puzzle to get the heavy stepper moved up 1 meter, turn it, and shift it into the cleanroom. During our pre-installation visit, we discussed what the solution would be. A construction was built in the corridor matching the cleanroom entrance. The stepper was placed upon that structure, picked up from inside the cleanroom, and placed at its final location. Described in a few words, it sounds easy, but when I think about the wobbly stepper on top of the construction, I still get the creeps.

My installation colleague and I were happy that the complicated move-in was successfully completed, not aware of more trouble to come. We connected the electrical cables, pressurized air, and vacuum hoses. The final step was connecting the high-pressure hydraulic hoses through holes in the floor to the oil unit located in the basement underneath the factory. Pushing the "on" button to fire up the system was always a special moment. Would all subsystems come to life? Closely, we watched the monitor for unexpected messages. All was fine. Now we had to switch on the oil system to bring 80 bar of oil pressure to the hydraulic stage so we could check its functioning. I cannot remember who went to the basement and switched the oil system on. What I can remember is that an oil fountain was created at the back of the machine. A pipe was not fixed well, and oil was spreading all around. Oil was dripping through the holes in the factory floor into the space for the routing of cables and hoses. Things could have been much worse if the safety protection had not switched off the oil system. Lucky as we were, the big plastic transportation bag around the system was not yet removed and prevented the oil fountain from reaching other equipment in the factory. As you can imagine, we spent hours on our knees cleaning up the mess we made. Our superb, accurate, and fast hydraulic stage never became popular in the semiconductor world.

After the completion of the site acceptance test, the system was handed over to the customer. Even the lens performance was fine despite the cold. As we did not have any local office and wanted to offer fast responses to solve issues, I was "babysitting" in Hamburg for a period of 3 months, as other colleagues did. Early on Monday morning I flew to Hamburg and back to Eindhoven on Friday evening. I'd like to share two issues I had to solve, as the learnings from them are still relevant today.

For an unknown reason, the wafer stage stopped functioning randomly. I replaced the HP laser, re-calibrated the optics of the interferometer measurement system, cleaned the mirrors, replaced electronics printed circuit boards, and checked the optical path of the laser for any obstruction. Nothing solved the problem expressed in the error message: XY stage control failure (or something similar). Finally, I found the root cause. The PAS2000A continuously monitored the pressurized air. In case the pressure dropped below a threshold, the XY stage was put on hold, otherwise, the airfoot could be damaged. Glitches in the air pressure caused this safety system to interfere. An error message referring to this root cause would have saved me a lot of time and Philips a lot of money. On the other hand, I remember how proud I was to have solved the issue based on the knowledge I had about all aspects of the system.

According to our customer, the light dose control was not functioning well, causing line width variation from wafer to wafer. I had no clue. All my checks and staring at an oscilloscope screen during wafer exposure cycles did not indicate something was wrong with the exposure system. And that was correct. An application engineer found the cause. The lines we exposed were much smaller than they were used to. With these small lines, the resist was much more sensitive to temperature fluctuation of the fluid used for developing the exposed wafers. By this, the cause of the observed variations was determined. An important lesson to diagnose problems "holistically" and have knowledge of the entire litho process!

My career in the field, supporting our customers located in Burlington, Nijmegen, and Valvo, was challenging for me and my home front. Consequently I left field service and became a Service Development engineer. I started writing Service and User documentation, developing and delivering training for our new hires and customers. Another task was influencing and reviewing designs to optimize serviceability. All these activities are now performed by colleagues from large departments all over the world. It makes me happy and humble to realize that I, together with my valued early years colleagues, planted the seeds of these still essential activities within ASML.