“I made a mistake.”

by Dave Schinke

It was around the period of time that  a design for the first Model 80 IVR system had been settled on. I was at my desk  and Jim Walls came to me with a stack of 11” x 17” drawings. They were circuit designs for circuit boards, backplane wiring for the cabinet, physical slot arrangement for each circuit board, power supplies, etc.  Jim Walls, a good friend, was a circuit designer with deep knowledge of  Bell System equipment, and long experience in the company.

“I need your signature so we can release these to manufacture,” Jim said.

“Why me? Shouldn’t the hardware group supervisors sign off?” I answered.

“Oh sure, they have signed off, but we need someone from the software side,” Jim replied.

Well, the Good Book says something along the lines of “Trust the hardware guys as you would trust yourself.”   After briefly glancing at the drawings, I signed off on the package to be released to manufacture.

Time passes as those drawings become reality, and the first Model 80 is ready to be delivered to Bell Labs from Western Electric.  Circuit boards were made at a facility in North Carolina, and the cabinet, power supplies, backplane wiring, and final assembly were done at the Columbus factory.

Since I was going to be a person doing the customer installation, my supervisor asked me to “Check it out and see if it’s ready to go.”  Two strong men with some heavy-duty moving dollies brought it to the lab, and pushed it on a couple of planks up the step(s) of the raised floor. I set to work “Checking it out.”  I have forgotten all that may have been done but it probably included: loading software on the processor. The circuit cards probably had self-diagnostics at start up.  The speech database needed to be loaded and a test call  IVR transaction needed to be executed.

Finally, I got to the point of making a phone call into the unit.  The call was answered, but no audible sound output, NADA.  Don’t remember all that I did from the software side to see what might be wrong. Checked connectors and looked at the rear backplane to see if anything obvious was loose. Just couldn’t find anything that was wrong.  Well, it was Friday afternoon, and I had to meet my carpool so I went home very frustrated and puzzled about the situation.

On Sunday I told my spouse that I needed to go to the lab in the afternoon because we had a problem with something not working, and I needed to find out what it was.

I pulled out the stack of equipment drawings thinking I could trace the voice path through the system and maybe find something.  How I did it physically is lost to my memory, but I could have done that with a tone generator or a touch tone held down on the phone while I used an oscilloscope to trace the audible signal through the system wiring. I worked from pin to pin on the backplane following the signal, and found that the signal died going from one board to another.  Looking at the drawings for the two circuit packs involved, I found a mistake in the backplane wiring connection.   The signal output side of one board was wired to the ground side of the next board instead of the signal input side. This same type of miss connection occurred at 4 places, two on the upper equipment shelf, and two on the lower shelf.  I had found the problem of “no sound”.

First thing Monday morning, I took the stack of drawings and walked across the corridor from our lab to the Western Electric manufacturing engineers’ bull pen.

As I walked-in I said: “I made a mistake.”

Several responded “Well, that’s unusual.  Normally Bell Labs tells us what we did wrong!”

I said “I signed off on these drawings, and I will take responsibility for a mistake.”

After a bit of joshing, I sat down with the engineer that was handling this product.  We looked at the drawings and agreed that the 4 places that I had isolated appeared to be wrong, and should be corrected.   I asked, very nicely, if someone could come to the lab and fix the machine I was testing to see if it solved the problem.  They agreed, and after lunch, the engineer and a technician came with tools to work on the wire-wrapped connection backplane.  I had disconnected the power to the machine, so they went zip-zip, zip-zip and fixed the wiring in a short time.  I thanked them and said I would test the fix and let them know if it worked.

When they left, I connected the power, fired up the machine, made a phone call and it  worked.   We had sound!  Made a phone call on every line to make sure, and IT WAS FIXED.

A take-away from this story is what can be accomplished in a large organization if a person is patient, persistent, avoids placing blame, and works cooperatively with others toward a solution.