Upon receiving my MBA, I worked for Hewlett-Packard as a process engineer.  In the first six months I was assigned my first project: to design a state-of –the-art printed circuit manufacturing and assembly facility.

First, I established the scope of the project, completed a WBS, and then planned the project using the Critical Path Method.

As I considered the risks it became apparent that one activity would be particularly challenging.  But it was an activity over which I had little control. There was nothing I could do except to keep my fingers crossed and hope for the best.

My boss, Ken King, was an incorrigible inventor.  He enjoyed the challenge of finding new ways to solve old problems. In fact he already held several patents one of which was the “Heel-Free Safety Binding” a novel ski binding that was popular in California and Colorado.  But, as the manager of manufacturing at a Hewlett-Packard plant he seldom had the chance to exercise his need to tinker.  

For some reason, he felt that my project would be his big chance.

The printed circuit manufacturing and assembly process begins with the manufacture of the printed circuit board, then proceeds to component insertion, then soldering, and finally cleaning. It was the cleaning process that caught his imagination.

He argued that it made no sense to have two or three people “wash” these boards.  It was the “old” way.  If a dish washer could wash dishes, why couldn’t we design a machine to “wash” boards? Sounded reasonable to me.

For three weeks he worked day and night on the design. The boards would be fed by conveyor into a drum, similar to a 55 gallon drum, and then washed with trichloroethylene, a colorless non-flammable liquid with a chloroform-like odor.  It is often used as a degreasing solvent for metal parts, and is rather toxic. So the challenge would be to design a reasonably airtight machine with adequate ventilation to prevent the fumes from invading the assembly area.

Useful clauses Gypsum cardboard Decorative furnish
While he was busy designing the washing machine I was busy designing the manufacturing and assembly facility.  Two months later my job was done, but Ken was still building, tinkering and tweaking.  The truth was that I never saw a happier man.

Then two months became three and four.  At every test the washer failed:  boards were jettisoned out the end and crashed to the floor; the conveyor motors failed; the hydraulic system had to be redesigned; the jets continually clogged; and the machine leaked.  But thanks to a truly gifted and confident tinkerer, these problems were addressed one-at-a-time.

I remember that Monday morning when Ken declared that the machine was finally ready.  After lunch the ribbon was cut and production began.  In one room the boards were drilled,  in the next components were inserted, then the boards were plated, and finally they were “launched” into the world’s first completely automated printed circuit board cleaner.  In theory, it was a “well-oiled” manufacturing and assembly process.

We were proud.

Then it happened.  Within fifteen minutes the room filled with toxic fumes of trichloroethylene.  People started coughing, covering their noses, and running for cover.  Within five minutes Human Resources demanded that the entire building be evacuated and the fire department called.
Then, Dave Packard (in those days Dave often hung out in the plant) dropped by to see what had happened. You can only imagine his response!

No, the machine was not as air tight as we had expected, and the trichloroethylene did find its way out of every hole and crevice, but Ken confidently proclaimed that within a few days the problem would be solved.

He tried, but finally realized that no one would tolerate another trial; enough damage had been done. We had failed.   By Monday of the following week the machine was dismantled, loaded on a truck, and shipped to the local dump.  

It was several days before we had an effective (manual) board washing facility in operation.  The project was now about four months late and considerably over budget.

For me, it was certainly not the best of beginnings, yet both Ken and I managed to survive.  Ken eventually left HP and purchased an old New England mill on a river, complete with a non-operating water wheel which had been used to power the machinery in the mill some 100 years before.  His wife renovated the mill and opened a very successful restaurant.  But the water wheel was too much for Ken to ignore.  This time, however, the tinkerer prevailed.  The water wheel was restored and eventually was used to generate electricity which he then sold to the local utility company.  

Tinkering with water was clearly more forgiving than tinkering with trichloroethylene.

Lessons Learned

•    It is difficult to get in the way of someone else’s’ enthusiasm.  Especially if that person is your boss.
•    Overconfidence at the beginning of a project is hard to detect because we often want to believe in miracles.
•    Risk assessment needs to look beyond the confidence expressed in the project
•    It is difficult to abandon a project when you are promised that the problem is under control and will be solved.
•    People involved in a project will find it difficult to abandon it even when the evidence suggests that the likelihood of a successful outcome is remote.