The bowling balls

10 November 1972: CERN’s Bent Stumpe places an order for 12 bowling balls for a total cost of 95 US dollars. Although not evident at first sight, he is buying the heart of some of the first tracking devices to be used in the SPS control room. Today, Bent Stumpe’s device would be called a desktop mouse…


The first order for 4 bowling balls later changed to 12 balls. The bowling balls became the heart of Bent Stumpe's mouse.

Almost 40 years ago, the web, Wikipedia and Google did not exist and it was much more difficult to know whether other people in other parts of the world or even in the same laboratory were facing the same problems or developing the same tools. At that time, Bent Stumpe was an electronics engineer, newly recruited to work on developments for the SPS Central Control room. One of the things his supervisor asked him to build as soon as possible was a device to control a pointer on a screen, also called a tracker ball.

The heart of the device was the ball that the user would move his hand over, while the cursor followed the corresponding movements on the screen. “We needed very round, well balanced and smoothly moving balls and we thought that bowling balls best met these requirements”, recollects Bent Stumpe. "The SPS control room needed three such devices, plus a prototype, and so an order was placed for four bowling balls. As the firm only wanted to sell a minimum of 12 balls, the order was changed accordingly. This led to discussions with the purchasing office, which required an explanation for such an unusual material request."

This tracker ball was among CERN's first tracking devices. The bowling ball is clearly visible. (Picture 1)

Pointing devices, the ancestors of the tracker ball, already existed at CERN in the 1960s. They were used in the bubble chamber film-scanning devices in the DD Division, where Bent Stumpe worked before joining the SPS Division. “Having conducted careful searches on the web, I realize today that other people in the world had come to similar solutions before me”, he says. Recently, while he was looking for his own tracking device at CERN, Bent found another similar tool that was apparently used by another unit at CERN more or less during the same period. “It has very similar dimensions and seems to use the same optical principle but the designers apparently found another solution to allow smooth movements of the ball”, he explains. While in Bent’s device the big central ball slips over some very small balls to make the movement almost friction-free, in the other device the central ball moves over an air cushion. “The problem in this case is that one needs a compressor to make it work”, says Bent.

The first mice that industry developed and commercialized in the 1980s, years after these first developments at CERN, were based on the same optical principle as the one designed by Bent. However, they were much smaller because everything was miniaturized compared to the original CERN design. “Our device was very big and the box that contained all the mechanics and electronics was installed under the console so that the user could only see the ball”, he explains.

Another CERN tracker ball possibly older than Bent Stumpe's. If you happen to know who made it, please contact us! (Picture 2)

CERN’s mice worked for more than 20 years. Since the conversion of the old SPS control room into the CERN Control Centre, Bent has lost track of the tracker balls built at CERN. If you happen to know where to find the SPS Tracker Ball (picture 1) or who designed the other tracker ball shown in picture 2, please get in touch with us: you'll be featured in the Bulletin and you’ll make Bent very happy!

The inner workings of Bent Stumpe’s tracking device

The x-y movements of the ball are transmitted to two bearings (one for each direction), which then transmit them to the incremental encoders. The encoders and the rest of the electronics then send a stream of digital signals to the computer telling it the direction and the speed of the movement.

by CERN Bulletin