Once upon a time in Iraq

This month's installment of Tales From the Front takes us to Iraq, circa 1962, where a control engineer had to solve an instrumentation cabling problem that was somewhat less easy to do in practice.

Share Print Related RSS

B

ack in 1962 I worked on a project in Iraq – we were installing I&C on gas turbines, pumps and oil/gas separation facilities. The instrumentation cabling between sensors and instruments, etc., was all Mineral Insulated Copper Covered (MICC) cable that consisted of a copper sheath with a number of conductors (as few as one and as many as seven) held and separated inside by a chalk-like magnesium-oxide insulation medium.

The problem with this cabling was that if an open end was left exposed the insulation, being hygroscopic, would absorb moisture from the air (even in Iraq) reduce its resistance and render it useless as an insulator.

Terminating this cable was quite tricky because you had to make certain that a loose end was never left exposed for too long. When installing multiple runs of this cable, the electricians had developed a bad habit (as I eventually discovered) of just leaving ends unprotected at the point where they were supposed to be terminated (hopefully with sufficient spare length) and continuing on with the multiple runs.

To deal with the contaminated insulation problem they left you had two choices: one was to keep cutting the end of the cable until you reached a portion where an insulation test showed it was measuring several MOhms or more, or, if you had insufficient spare cable available, get a blow torch and heat the last couple of feet until you got a good reading.

During the time I was testing/commissioning the whole installation I eventually came across a problem where a pump bearing lubrication leakage alarm (A drain pot with a small orifice and a level switch in it) was giving us intermittent alarm signals (because the mechanical people were running the turbines without all of the protection, this condition did not shut down the drive turbines as it should have done).

         


“I eventually discovered that the cable to the level switch had never been terminated properly but had been tucked away behind the switch in such a way that its end was hanging out the other side of the pump.”

The point was eventually reached (because none of my assistants had any success in identifying the problem) where I really needed to solve the problem.

To get to the site each morning we would leave before dawn, travel the 50 miles by Jeep and arrive in daylight, about 7:00 am. I would begin troubleshooting the intermittent alarm problem first thing, but it would always go away after a couple of hours.

After several days of this, I eventually discovered that the cable to the level switch had never been terminated properly but had been tucked away behind the switch in such a way that its end was hanging out the other side of the pump.

It obviously had not been connected to the level switches and, because it all was the responsibility of the electrical contractor, I had noticed the cable end but immediately ignored it. When I got down to brass tacks and really got serious about fixing the problem I removed the level switch cover and found, to my amazement, that the cable had never been terminated, hence, no wire ends were present at the switch terminals! As I had previously noted, the drain pot did not contain any moisture so the switch was not being normally activated.

What I eventually discovered was that because of the of way the pumphouse was oriented onsite, after sunset, this particular cable end was exposed to the cool night air. The resultant near short circuit operated as a level switch contact “Make” (I had nothing to do with the design of this project so Fail Safe operation was not a consideration) and that would set off the leak alarm.

Each morning the sun would come up and eventually shine around the back of the pumphouse — directly on the cable’s end. By about 9–9:30 the cable would get sufficiently warm to dry out the insulation, relieve the short and deactivate the alarm. The associated alarm lamp in the control room remained energized until someone pressed the Accept button, but there was never any immediate indication that the alarm had become inactive. I must say that while it was quite easy to describe the problem and its solution in these few short paragraphs, it was somewhat less easy to do it in practice!

W. Brown, MIEE, Control Engineer, Katy, Texas.


Tales From the Front are selected from reader submissions. Do you have a story about plants or process control to share with the readers of CONTROL and ControlGlobal.com? We'll make it worth your while if your column is published. For details, e-mail wboyes@putman.net.

Share Print Reprints Permissions

What are your comments?

You cannot post comments until you have logged in. Login Here.

Comments

No one has commented on this page yet.

RSS feed for comments on this page | RSS feed for all comments