The Case of the Stuttering Tape Drive

In 1987, I worked as a field support engineer for a company that provided fully integrated automatic test equipment (ATE) systems for use in a variety of military aviation programs in the United States and its allies. I was sent to a posting in Cold Lake, Alberta, to provide initial cadre training and maintenance services for the ATE systems used on the newly transitioned CF-18 Hornet.

The building was quite new. The lower third of it was somewhat below grade with earthen berms surrounding other parts. The skin of the building was metal. I suppose that was a response to electromagnetic pulse protection, as well as esthetics and economics. The prairie is cold and windy, which wreaks havoc on buildings. There were very few windows, and most of those were long strips that followed the second- and third-floor office area.

One of the tasks I was typically assigned was troubleshooting the program development system (PDS). The PDS consisted of a militarized Harris H-100 minicomputer that hosted a nine-track tape drive, a Hewlett-Packard 10Mb removable disc pack, and a rebranded CDC5550 300Mb "washing-machine" style disc drive where you lift the lid and screw the disc pack on to a spindle and press the start button.

The PDS was used to develop test program sets (TPS) that would later be tested on the ATE. Typical problems included things like compilation failures, printing problems, and tapes that were somewhat less than interchangeable. The system ran very nicely and was not an unusually burdensome support item.

To make the code, compile, and test process more convenient, we moved the PDS system to a location on the third floor. Almost immediately after the move, we began to receive an inordinately high number of visits from users and support requests for the PDS. The specific problem was that the nine-track tape deck was having increased difficulties reading and writing tapes.

Sometimes it would take numerous tries to get a file off the tape or write one to it. This continued despite numerous attempts at cleaning the head and capstan wheel, replacing filters, and using new tape stock. We ran full diagnostic suites for the deck. The deck continued to display the same problematic behavior.

After several days of digging into the problem, I found myself back in the office space that housed the PDS, checking again for the umpteenth time for anything that I might have missed. The airman who normally operated the system had left a portable radio, tuned to a local station, turned on and placed on the sill of a window that overlooked the southeast-facing end of the airfield. It was a spectacular view in winter -- lots of snow and the faintly rolling countryside of the prairie. In the distance, you could see the base industrial complex and communications areas.

While running yet another series of diagnostics, I heard "BRRRRaaapp!" come across the radio, and at that instant, the tape deck transport stopped moving!

I looked up at the radio on the sill, and I noticed that outside there a very large white geodesic shaped dome. In an spark of insight, I opened the base phone book, and sure enough, it was the location for a surveillance squadron. I called the operator on duty and explained the problem I was having. I asked about the transmission status. As you can imagine (and rightly so), I received very oblique responses to my questions. I opted for another approach. I asked if the operator had a live PPI display and if he could tell me either yes or no to a single question: "Are you painting me?"

He acknowledged that he was. I started the transport moving again. After a couple minutes of waiting, I spoke into the phone "Now," and the operator responded "Yes." The transport had stopped moving. I thanked him for his time and presented my findings to the facility commander.

The PDS was moved into an office space nearer the core of the building, and the cabinet was oriented so that the transport and the tape heads in it were facing away from the site, rather than directly at it. No further problems with the tape transport or the PDS!

This entry was submitted by Chris Spacone and edited by Rob Spiegel.

Chris Spacone is employed as a broadcast television maintenance engineer for GDMX, a Warner Brothers company.

Tell us your experience in solving a knotty engineering problem. Send to Rob Spiegel for Sherlock Ohms.