Navigation
Home
News
Newsletter
Knowledge Bases
Web Workshops
Tutorials
Directory
Books
Maintenance-Tips
Conferences
Forums
Photo Albums
Reliability Radio
Benchmark
Calendar
Gadgets
Jobs
Twitter
Network Links
Archives
en Español
XML/RSS Feeds Advertise

Promoting predictive maintenance for equipment reliability and cost effective “smart” maintenance by Brian Thorp

Your predictive maintenance program is up and running. You have had some good finds and saves, but are you putting all of the information you collect to use? There is a lot more useful information in that data than just lubricant quality, trends, and failing components. Aside from the reliability aspect that predictive maintenance can provide, there are significant cost savings potentials available though effective "smart" maintenance; Doing the right thing, at the right time, and for the right reason. 

So what is effective or smart maintenance? Of course you would consider CMMS, preventive maintenance, planning/scheduling, and maintenance procedures to name a few. Webster's collegiate dictionary describes it as: Effective: producing a decided, decisive, or desired effect, and Maintenance: the act of maintaining, the upkeep of property or equipment. This paper will primarily focus on the additional information that predictive maintenance can provide to help you work towards achieving effective "smart" maintenance.

Oil Analysis and Condition Based Maintenance (CBM)

Most electric motors have small volume lubricant reservoirs and thus, oil analysis is generally done for equipment condition rather than lubricant quality. Why not use it for both? Our oiled motors where originally set up with a one year frequency PM for oil changes. Even with good lubricant handling practices there where still significant swings in the ISO code cleanliness. The high spike was generally right after an oil change with the cleanest being right before an oil change.

We have all heard about the dirt in circulation of a hydraulic system and the damage it causes. A motor may not have the same flow as a hydraulic system, but the slinger rings do a great job of circulating wear debris within the bearing reservoir. This of course causes wear to the shaft journal and the slinger ring showing up as iron, chromium or brass in the oil analysis. Unfortunately most motors do not have circulation systems or filtration to remove wear debris once they are there, they must slowly settle out to be stirred up again during an oil change. FIGURE 1 graphs show the before and after of CBM on oiled motors for particle counts.

As can be seen by the above graphs, once CBM was established for the oiled motors the particle counts dropped noticeably and remain low.

The cost savings associated with the CBM for the oiled motors was substantial as well.  There is approximately 550 gallons of oil for all of our oiled motors. The actual cost of the oil times 4 (industry average real cost) equals, $15,400 per year. While this may not be that much money per year, when you take into consideration we can run three to five years without changing oil, you are now looking at $46,200 to $77,000 total savings over this time period. The bottom line is: Smart maintenance, less human intervention, cleaner running motors, less chances for oops and ingression of debris = cost savings.

There are many other pieces of equipment that can benefit from CBM for "smart maintenance". We have three Cooper-Joy, three stage centrifugal compressors which ran for five years (40,000 hours) without an oil change. When this equipment was being installed the O.E.M. recommended a 5,000 hour oil change frequency. After contacting the manufacturer and qualifying our oil analysis program the 5,000 oil change was waived based on oil analysis for CBM. These three compressors now have the oil changed during the five year rebuild. This is a savings of over $110,000 (4 X's cost, industry average real cost) for the five year period. FIGURE 2

Another piece of large reservoir equipment is our 25,000 CFM blowers which provide air to convert our scrubber slurry to synthetic gypsum. These had O.E.M. guidelines of 5,000 oil changes that were waived based on oil analysis for CBM. The oil was changed at the 40,000 hour interval based on a RVPOT which indicated the oil was at 25% of its original value. With three of these blowers, this is a cost savings of over $184,000 for the five year period (4 X's cost, industry average real cost). FIGURE 3

Our twelve coal ball mill gear reducers are the last equipment to be referenced for CBM oil changes. Originally set up for six month oil changes on a PM, these accounted for almost 750 gallons of oil per year. We now average 2.5 to 3 years before an oil change is needed based on oil analysis. These gear reducers are filtered and cooled which has a significant impact on the running time before an oil change. While this equipment still requires a six month PM to attend to other items such as couplings and breathers, the cost savings over a 2.5 year period is over $50,000. 

As can be seen by the references on these 18 pieces of equipment, we get an annual savings of almost $80,000 per year, and that’s just by not doing un-needed oil changes. 

With companies trying to work with fewer people, CBM can free up those people to do their work more effectively, "smart maintenance". Do not misunderstand; the benefits of CBM do not replace preventive maintenance. PM's are still a vital function for equipment reliability; in fact they compliment and make possible some of the benefits that can be accomplished through CBM. 

Root Cause Analysis (RCA)

Root cause analysis is an important part of a functional predictive maintenance program.

Not only should you be able to predict when something is failing, but more importantly figure out WHY it failed to prevent it from happening again. Sometimes a minor modification to the equipment or how the equipment is operated makes all the difference.

Henry Ford once said," Failure is the opportunity to begin again more intelligently". So why not learn from your failed equipment to see if you can prevent it from failing again.

We have had several pieces of equipment that originally listed automotive lubricants, 5W 40, SAE 10, 20 or 30 for the OEM preferred lubricant. I have also found several OEM listings for new equipment with out dated lubricant specifications or name types. So why is this mentioned in an RCA discussion? After several failures from varnish and sludge problems on reciprocating compressors that were using SAE 30 engine oil, these were changed to turbine oil ISO VG 100 and the problems have gone away. Another new piece of equipment, a lobe type blower, specified 5W40 synthetic motor oil. The vendor was contacted to get approval to change the lubricant to an industrial type since the equipment was under warranty. They would not approve any other lubricants than what were listed.

We ran the 5W40, but had to increase the change frequency from six months to three months based on sheer down and viscosity decrease. Fortunately one of the specialty lubricant companies got their product approved for use in the blowers. Once this was accomplished, their product along with a comparable product from our lubricant supplier where run in two different blowers as a test; There were no noticeable differences between them. I contacted the OEM with the test results and they accepted another industrial oil for use with the blowers. Since the change of products and sufficient time to flush out the old 5W40, we have increased the oil change frequency back to six months and continue to test which may prove to allow longer service yet.

Another case for RCA revealing an oil related problem occurred in some of our motors with submerged coolers and hydraulic unit heat exchangers that were made of copper.

Due to consolidation efforts we were using AW hydraulic oil in some of our motors with submerged copper coolers and copper was trending up significantly on ICB wear metals.

We were experiencing a similar problem with some of our hydraulic systems with copper heat exchangers and increasing copper trends. Of course with the motors we could have switched to turbine oil, but we had to stay with AW oil for the hydraulic systems. To stay with consolidation efforts we switched to an ashless, inherently biodegradable AW hydraulic oil. With the removal of zinc or ZDDP from the lubricant and system the copper trends stopped increasing. It took a couple of flushes to get everything stabilized but the corrosion of copper was ended. We have now been able to run extended oil changes of up to five years with no negative affects. Figures 4 is a submerged cooler in one of our vertical motors. Note: The copper tubing in figure 4 looks as though it has been acid washed. Knowing what your components are made of and the possible effects of certain additives on certain metals, definitely helps with failure analysis.

Another great example of a few minor changes to up grade equipment came when our effluent processing facility was modified to produce synthetic gypsum from our scrubber slurry. This change in operating process placed some of the equipment in different operating parameters which of course caused problems. The plant modifications were well worth it since it reduced our land fill use by up to 75%. One gear reducer was failing every 3 to 5 weeks from the ingress of fly ash which looked like lapping compound and was wearing the gears and bearings to failure. With some basic improvements such as an increase in lubricant viscosity, bearing isolators, kidney loop filtration, and better breathers we have improved the time between failures to better than 3 years. The average cost per failure was $5,000. The total cost for the gear reducer upgrades were less than $5,000 and a complete replacement system would have cost about $100,000. See Figure 5 (Before and after shots) The picture on the left is after 5 weeks run time of about 8 hours per day. The picture on the right was taken after 14 months run time when an output shaft broke due to bending fatigue. As can be seen there is quite a difference.

Another opportunity for RCA came with our 25,000 CFM blowers mentioned earlier under CBM for oil changes. These blowers generally require a 25,000 hour inspection/rebuild. We experienced an inlet control and guide vane binding which took one of the blowers out of service at 15,000 hours run time. These blowers operate near our wet scrubbers and thus have the opportunity to ingress sulfuric acid, hydrochloric acid, and sodium chloride mists. FIGURE 6 shows the inlet hub and FIGURE 7 the control vane condition.

The original two stage air inlet filtration which existed of a cellulose two inch pre and six inch post filter were replaced with a three stage filtration set up. The new set up utilizes a synthetic media, three ply pre-filter, a twelve inch v-pak type which has almost twice the original surface area and the final filter, a twelve inch activated carbon element. The activated carbon element was used to neutralize the acid mists and tested at 2,000, 4,000 and 6,000 hours to determine the depletion of activated carbon and assure it was having an affect on the acidic air entering the blower. The testing indicated the activated carbon was nearing depletion between 5,000 and 6,000 hours. The average run time per year is 4,500 to 5,500 hours (these blowers are rotated) so a yearly PM was established to change the activated carbon elements. 

With the original air inlet filtration set up, both filters needed to be changed on a monthly frequency for a cost of almost $49,000 per year for three blowers. Even with the increased cost of the better filters, the new three stage set up has the pre-filter changed monthly, the second filter changed every two months, and the carbon filter changed once per year based on activated carbon depletion. The three stage filtration cost is only $39,000 per year for three blowers. Not only is this an operating cost savings of $10,000 per year but we are getting better air filtration plus acid mist removal. This should also increase our inspection/rebuild frequency, which is another significant cost savings as well. 

Well most of this article has been related to lubrication related PdM, (hey I'm the oil guy) what about the other technologies and the information they can provide. Vibration data can provide a wealth of information about rolling element bearings, gear sets, coupling problems, misalignment, resonance, and soft feet to name a few. 

The inner bearing race in Figure 8 is an example of specific component identification through vibration analysis. This 1,200 HP motor is located directly underneath hydroclone which separates and mixes water with limestone before it goes into the ball mill. A significant leak almost buried this motor in lime slurry. The next month's vibration route indicated a problem with the motor inboard bearing inner race. As part of our PdM program we are present during motor disassembly for routine and known problem motors. Once the evidence has been disturbed or accidentally thrown out, root cause analysis becomes hard if not impossible to perform. In this case the root problem, an occasional overflow or leak cannot be eliminated, so a roof was built over the motor and gear reducer to help minimize another bearing failure due to lime slurry. 

We had a crack develop in one of our coal ball mill ring gears. It was first identified through vibration data and monitored closely for almost a year until we were able to schedule the gear replacement. This allowed normal expediting of parts and workforce rather the added expense of doing it on an emergency basis. We are also able to identify ball mill gear reducer gear and bearing problems early enough to schedule and plan the replacement for a time of least economic impact. At times the failure might progress faster than expected but at least we have everything ready, which removes the un-known factor from the equation. 

Infrared is used primarily on scans of our electrical switch gear, and as a back up technology for oil and vibration data. Our scanning of electrical switch gear has been slowed slightly due to the new regulations regarding high voltage equipment. We are currently testing so view ports and other ideas to get back on track with it.

Ultrasound is also used in areas where we don't have vibration routes established for quick diagnosis, air leaks, valve leak through and many more.

While the vibration, infrared, and ultra sound technologies provide invaluable information to the benefit of effective "smart" maintenance, it is much harder to place a dollar value on their benefits.

In conclusion; hopefully this paper has provided some insight into the many additional benefits provided by a functional predictive maintenance department and the data they collect on a routine basis. With a little creative thinking, or thinking out of the box many everyday problems can be corrected or eliminated, resulting in significant savings. Even when it's hard to place a dollar figure on something, the result of effective or "smart" maintenance places your company in a winning situation. Figure 9, is what most of us would rather be doing when we are not at work, rather than worrying about what might break next. It's not that all failures can be predicted, but at least you are ahead of the game when you have a functional predictive maintenance department.

Bio on Brian Thorp

Brian Thorp has been involved with mechanics and maintenance for over 32 years. His range of knowledge includes automobiles, trucks, heavy equipment, and power plants. In his earlier years, he has received extensive training in engines, transmissions, gear trains, hydraulics and pneumatics. His current position is as a Predictive Maintenance Technician, with Seminole Electric Cooperative Inc, where he is responsible for the lubrication and analysis for a combined total 1300 MW coal fired power generation plant.

His current certifications include, ICML, Machine Lubricant Analyst level II , Level I Infrared Thermographer, Level I Vibration Analyst, Level I Certified Airborne Ultrasound Inspector and he previously attained STLE, OMA level I. He has also attended numerous seminars on root cause failure analysis, failure analysis and laser alignment.