Like a Disaster Movie

This article is part of an ongoing series. The introduction appeared in Machinery & Equipment MRO’s February 2005 issue and the series has run in every issue since. Previous instalments are archived online at www.mromagazine.com. In this issue, we pick up where we left off in our November 2007 issue, as maintenance manager Bob Edwards and his team at the Plentya Paper Company deal with a shutdown of the mill after a bearing seizes, and raises the question, ‘Why didn’t PM prevent the problem?’

The scene that greeted me when I arrived at the mill reminded me of one of the disaster scenes I had seen on the television. There were parts of roll covers and paper-carrying felts all over the place. Both the operators and millwrights were climbing over the machine, trying to remove the broken parts.

“Hi Ted, what exactly happened?” I asked the night shift millwright.

“Well, from what we’ve seen, the drive-side bearing on the suction roll has seized completely. This ripped the paper felt — and that bundled up on the next set of rolls and tore the covering off them — then it seemed to build from there.”

“Any idea what caused it?”

“Keith, the operator, said he heard a kind of rumbling sound, and then there was a loud crash. That’s all I know.”

“Okay, let’s figure out what we need to replace and make sure we have the spares ready. We’re going to need more guys than we’ve got. I’ll start to call some of those who are supposed to be off work tomorrow.”

“I’ve already started that, Bob,” said Ted. “It was obvious that we couldn’t handle this ourselves so I gave the guys a call — there are three coming in tonight and three others will be in at seven o’clock in the morning. I hope I didn’t overstep the mark.”

“I don’t think helping me out can be considered overstepping the mark. Thanks, Ted. I really appreciate your interest and initiative, so let’s get started.”

By the time we had looked at all of the rolls, we found that we had to change out the suction roll and two felt carrying rolls. The good news was that we had spares for each. The bad news was that this was going to take at least 24 hours to complete. Plant manager Joe Davis wasn’t going to be happy about this.

I was right in that assumption. “What happened here, Bob? I thought we were supposed to stop things like this from happening!” Joe was clearly not in a mood for excuses when he arrived at the mill the next morning, so I decided to play for more time.

“I’m not too sure yet, Joe. It will take a little bit more investigation as I don’t want to jump to the wrong conclusion.”

“Well you’d better come to some conclusion by this afternoon, as I’m going to have to tell corporate something.”

By mid-day, we had the damaged roll on the mill floor and the guys were clustered around the bearing.

“So it’s pretty obvious what the problem is,” said Jim, the maintenance supervisor. “This bearing failed, seized up and caused all of the damage.”

“Yes, you’re right. That much is obvious, but we need to figure out what caused it to fail,” I replied. “And we’d better figure that out soon as Joe wants answers by this afternoon.”

“It’s just worn out. You know these things just happen over time,” said Brian, the paper machine millwright.

“If that’s true, then all of the stuff we’ve been hearing about vibration monitoring and predictive maintenance is a load of garbage,” said Pete, another millwright.

“Why do you say that?”

“Well this is one of the rolls that John took readings on and he said it was fine — and that was only a month ago.”

“You’re right, Pete. I think I still have some of those readings in my office; I’m going to take a look.”

When I got back to my office, I found John’s report. I looked at the suction roll and it read “0.01 in./sec, very good.” Pete was right. If it was just age-related wear, then it had happened very quickly and we would need John to monitor the roll every day or so and that would become very expensive. I decided to call John to see if he had seen such rapid deterioration before.

“My considered opinion is that it wasn’t age-related, Bob. Bearings of that size and turning at that speed should have a much longer deterioration curve. I would look for some other reason — some foreign material getting into the bearing or something.”

“Thanks John, I’ll let you know what we find. I don’t know where we go from here as Joe is not very impressed with us right now.”

“Okay, Bob. Just try to get to the root cause and then we’ll talk.”

When I got back to the mill floor, I shared John’s opinion with the group. “Pete, can you go down and check that the filter on the lube system is clean and not damaged. Brian, did you see anything unusual when you took the cover off the bearing?”

“Come on, Bob, there were rollers all over the place — there was nothing usual about it. Though there was one strange thing — there wasn’t a trace of oil but the bearing did heat up, so ….”

“That’s a place to start. Let’s go down to the lube station and see what Pete’s found.”

“Sorry, Bob, but the filter is clean, the oil is clean and nothing seems out of place.”

I repeated what Brian had said. “So let’s take a look at the whole of the lube system — maybe there’s a blockage or a hose busted somewhere.”

“We’d know about that,” said Pete. “All of the lines have flow alarms on them so if they even drop pressure, an alarm goes off in the control room.”

“Let’s take a look anyway, as something must have happened and we’ve got to get to the bottom of it.”

We traced the lubrication line from the central lubrication unit and when we got to the flow meter, we were surprised to see a tag attached to it that read “Not Calibrated.”

“Oh no,” said Pete. “It looks like someone has changed the flow meter and they didn’t calibrate it. That means that we could have starved the bearing of oil without giving an alarm and that could have caused the bearing to seize.”

We met Brian as we continued to trace the line. “Bob, I think I found the problem. One of the connecting hoses to the other side of the paper machine is folded over and wouldn’t allow oil past it — but I don’t understand how it didn’t alarm in the control room.”

We explained what we had found and headed back up to my office to figure out how it had happened.

“Well, I think we know what happened, but we still have to figure out how. Someone must have changed the flow meter and we need to find out why they didn’t calibrate it. The machine must have been down for them to do it, so let’s take a look at recent downtimes on the machine.”

I called Rick, the production analyst, to get a copy of the recent downtime report and we soon found that the only possible time the flow meter could have been changed was two days previous, when the machine was down for almost one hour.

“Okay, Ted was in that night and he’s still around, so let’s talk with him to see what he knows.”

When we found Ted, he knew about the flow meter right away. “Yes, I know all about it — I changed it! It was at the end of the shift and I didn’t have time to calibrate it, so I left a note in the log book. Why, what happened?”

His face turned pale when we told him. “I definitely left a note in the log book for Stan.”

“That could be the problem, as Stan has been away sick for the past two days,” said Pete.

“Okay, as of now, everyone stays on the job until their relief arrives. We can’t afford to have disasters like this because we can’t communicate effectively,” I said.

I wasn’t looking forward to telling Joe about what we had found, and as I was trying to think of a way of preventing this from happening again, the phone rang. It was John.

“So Bob, did you find the problem?

“You’d never believe it,” I replied as I recounted everything we’d found. “And the worse thing is that there is nothing we could have done to prevent the disaster.”

“That’s not quite true,”
John replied. “A lot of companies have online monitoring and have alarms set so that when vibration starts to increase, it shuts down the machine.

“In this instance it was human error that caused the problem, but what would have happened if the hose had burst? You still would have had flow through the meter, but the bearings would have failed.

“What if there is a problem with the alarm or if there’s no one in the control room — there are lots of possible scenarios. The only failsafe is to shut down the machine when there is increased vibration. Remember the P-F curve we talked about? When the time from detection to failure is hours or minutes, then you really need online monitoring in your application. We could set an early alarm that would at least alert you to look at the issue.”

John’s explanation wasn’t going to make the conversation with Joe any easier, but at least there did seem to be a way of preventing this problem from happening again.

Cliff Williams is the maintenance manager at Wrigley Canada in Toronto, Ont., and a consultant with TMS — Total Maintenance Solutions Inc., Markham, Ont. He can be reached at williamscliff@rogers.com.

Key Points

* Communicate

* Communicate

* Communicate