RSS Feeds
The Case of the Soggy Furs
True Stories
Kenneth Russell, Professor Emeritus, MIT, Cambridge, MA -- Design News, January 12, 2003
|
Most fur-bearing critters love water: Coats made from their pelts do not. Soaked and dried fur coats are rock-hard and pitiable. Hot weather is less destructive than water to fur coats but still is undesirable. People do not much like to wear furs in hot weather anyhow, and as a result many entrust their valuable coats to a fur storage vault.
A Mr. Doe once owned a fur store and basement storage vault in a Boston suburb. One August morning he opened the door to his vault and to his horror found water about six feet deep and rising. The city water pipe had been attached to a 2.0 x 0.75- inch bushing that had failed where it had been screwed into a large gate valve. The failed bushing was on the city side of the meter, hence was their responsibility. The coats were hung in two layers. All the coats in the lower layer were soaked and ruined. The full-length coats in the upper layer suffered the same fate, whereas the fanny-freezers (jackets) above the water level were spared. Over 1,500 coats were ruined—a loss of millions. The owners of the coats were prepared to sue the socks off the furrier, or more accurately, his insurer.
A senior colleague referred this, my first big consulting case to me. The insurer hired a metallurgist (me) and a master plumber so as to get both theory and practice into the case. I was thrilled to be part of the team and set about my study with a vengeance.
The question was, what could cause the sudden and catastrophic failure of the bushing of relatively new plumbing? Good sense says that plumbing should last much more than 20 years and should give some warning, via leaking, before failing. The failure at hand was clearly premature and very sudden.
We found the reason to be not one, but three really dumb mistakes. First, the bushing was made of common gray cast iron and the valve of brass. Connecting such dissimilar metals in the presence of an electrolyte gives a short-circuited galvanic cell of about 1V. Electrons pass from the iron to the brass, and the iron is oxidized to Fe++ ions. Second, the bushing was screwed only about 0.25 inch into the valve when good practice dictated twice that. Only a modest amount of corrosion was needed to cause the failure. Third, the large cathode (brass) combined with the small anode (cast iron) gave accelerated corrosion. The overall result was a short-circuited cell with a current, which steadily corroded the base cast iron while leaving the more noble brass pristine.
As installed the plumbing was a time bomb. The bushing corroded uniformly and rapidly from the end and finally failed suddenly and catastrophically. The failure could have been prevented by use of copper or brass for the small amount of plumbing between the gate valve and meter.
How long should it have taken to corrode the fitting? I used Ohm's law, along with guestimates of the electrical conductivity of tap water and the amount of iron to be oxidized and got a lifetime of about 30 years. My rough calculation thus agreed with the 22-year actual age about as well as could be hoped.
The case went before a master, a senior lawyer acting much as a judge. The plumber and I reported our findings. The lawyer and expert hired by the city did little to either refute our testimony or even to muddy the issue. (The latter strategy follows from the axiom, "If you can't blind them with brilliance, baffle them with BS.") The master found totally for my client. The city appealed and the case automatically went before a judge and jury, who found totally for my client.
The city had to pay up. The amount of the judgment was large enough to require a noticeable increase in the property tax rate. I understand that as a result of this court decision, insurance companies no longer cover fur coats in storage as part of a homeowner's policy.
