1、99FTM12 I Gearbox Field Performance from a Rebuilders Perspective by: C.D. Schultz, Pittsburgh Gear Company I I TECHNICAL PAPER COPYRIGHT American Gear Manufacturers Association, Inc.Licensed by Information Handling ServicesGearbox Field Performance from a Rebuilders Perspective Charles D. Schultz,
2、Pittsburgh Gear Company The statements and opinions contained herein are those of the author and should not be construed as an official action or opinion of the American Gear Manufacturers Association. Abstract AGMA Standatds have traditionally focused on rated capacity and allowable stress levels.
3、Design life is addressed in various calculations and occasional comments are made concerning component design. The focus of this paper is the actual field performance of over 2,000 different gearboxes that have been repaired over the last 15 years. The information presented consists of a statistical
4、 review of the types of gearboxes repaired and the typical failure modes. Among the problems described are tooth breakage, tooth wear, housing bore damage, seal wear, and lube system failures. Photographs are included along with suggested design practice changes and a proposal for extended life serv
5、ice classes. . a Copyright O 1999 American Gear Manufacturers Association 1500 King Street, Suite 201 Alexandria, Virginia, 22314 October, 1999 ISBN: 1-55589-751 -7 COPYRIGHT American Gear Manufacturers Association, Inc.Licensed by Information Handling ServicesGearbox Field Performance From a Rebuil
6、ders Perspective Charles D. Schultz, PE Vice-president of Engineering Pittsburgh Gear Division of The Brad Foote Gear Works Pittsburgh, PA 15225 Introduction The major focus of the American Gear Manufacturers Association standards activity has been the accurate determination a gearboxs ability to tr
7、ansmit a specified amount of power for a given amount of time. The need for a “level playing field” in this critical arena was one of the reasons the association was formed in the first place. Over the past 85 years AGMA committees have spent countless hours “discussing“ the best ways to calculate t
8、he rating of a gearset, often arguing vigorously over factors that varied the resulting answers by fractions of a percentage point. While all this “science” was being debated in test labs and conference rooms all over the country, our customers were conducting their own experiments through the daily
9、 operation of gear driven equipment of all types. Unfortunately the results of these “test programs” are usually unavailable to the design engineer unless failure occurs during the relatively short warranty period offered on new equipment. My employer, the Pittsburgh Gear Company, has been engaged i
10、n the repair, rebuilding, and field servicing of gearboxes for many years. The record of this activity provides some interesting insight into what happens to gearboxes long after the warranty runs out. See Appendix A While the majority of our customers are in the steel business, equipment from the c
11、hemical, mining, and rock quarrying industries has also been repaired. At one time or another products from most domestic and foreign suppliers have been serviced. The gearboxes cover much of AGMAs history and include most of the designs popular today. One of the drives operated by a local customer
12、was put on-line in 1921 and hasnt missed a day of work yet. This type of performance is exceptional, of course, but it certainly inspires respect for the designers who labored in our trade long before the advent of the computer. It is in deference to those creative engineers that I encourage todays
13、gear designers to avail themselves of any opportunity to study long term performance. Having spent most of the last 25 years engaged in gearbox design and development, exposure to the far less glamorous side of the business has been extremely beneficial to my design work. In this paper 111 try to sh
14、are some of the things Ive learned and how theyve affected my design philosophy. Fa I u re Definit ion Our customers have their own definition of “gear failure” and it has nothing to do with bending stress or durability rating. The average mill superintendent cares about only one thing: Can the eu u
15、ment wo rk todav ? If a little pitting or a small crack appears the user couldnt care less if production can continue. While some of the more sophisticated plants are rapidly moving towards a “predictive maintenance” environment, the vast majority of mills react only to catastrophic breakdowns. Weve
16、 seen some incredible performances by gearboxes run completely without oil for months or missing sizable tooth fragments due to bearing related misalignment. We have seen very few “failures” caused by over rating or misapplication, although overloads due to process line “crashes” and field modificat
17、ions remain a significant problem. AGMAs standards writers and the application engineers can be justifiably proud of their work. plant maintenance crews, however. The most common causes of failure recorded in our data base are lack of lubrication, poor lubricant quality, and debris damage. Tooth bre
18、akage is rarely seen unless bearing damage, extreme tooth wear, or debris are involved. Pitted teeth are usually left un-treated until the drive becomes noisy enough to attract attention or someone becomes alarmed at the metal seen The same pride cannot be shared by the 1 COPYRIGHT American Gear Man
19、ufacturers Association, Inc.Licensed by Information Handling Servicesduring an oil change. While there are certainly design deficiencies that contribute to these problems, most customers would enjoy lower overall operating costs if they did a better job of monitoring the equipments oil quality. For
20、the designer, knowing that, even the most state-of- the-art gearbox will operate in an environment full of soot, sand, water, blunt instruments, and mechanics struggling to keep everything moving can help reduce the anxiety of determining the contact stress to the third decimal place. Desian Practic
21、e Chanaes Observing the maintenance problems first hand has resulted in some changes to my design philosophy. For example, unless a customer insists we no longer use the traditional bolt-on inspection covers. See Photo #1 J Weve found the thin plates almost impossible to seal against moisture and no
22、ticed that many times the debris that inflicted the final damage to the gears was a small capscrew used to attach the plate. Our technicians have found an incredible amount of “stuW inside gearboxes with large covers, including wrenches, screwdrivers, files, and flashlights. Weve often joked that th
23、e bigger the inspection cover the bigger the debris. While handy for gear inspection the covers ought to have a warning label affixed reading “No user serviceable parts in here.” Weve switched to large threaded pipe plugs for inspection ports. See Photo #2 They seal tightly against water, permit qui
24、ck inspection of tooth condition, and are impossible to drop into the oil sump. The pipe wrenches required to remove them are too big to drop though the hole, although the tiny flashlights favored by some mechanics do present a continuing hazard. Photo 2: Threaded Pipe Plug Inspection Port level” se
25、nsors. See Photo #3 Originally wary of putting delicate electrical devices in the rough and tumble mill environment, observing a plant start-up where the only gearboxes run “dry” were the only ones without the sensors was convincing evidence of their value. Weve also been increasing the “robustness”
26、 of the external lubrication plumbing used on both new designs and rebuilds. Lube lines apparently make excellent ladders for climbing, serve as emergency crane hooks, and are easily snagged by passing loads. One of our customers now has us fitting his more susceptible gearboxes with guards made of
27、A plate to protect them from damage. See Photo #For most pressure lines weve switched to high quality hydraulic hose after discovering that even the bravest mechanic wont use them for a step or a lift point. See Photo #5 So far the customers have been very accepting of this change, especially when t
28、hey realize it is much cheaper to make a new hose than it is to repair a damaged pipe. Weve also become believers in “low oil Photo 3: Oil Level Sensor Photo 1: Traditional inspection Cover 2 COPYRIGHT American Gear Manufacturers Association, Inc.Licensed by Information Handling ServicesPhoto 4: Gua
29、rded Lubrication System Photo 5: External Lubrication System with O Hydraulic Hose Fast turn-around on repairs is increasingly important to the mills, many of which have greatly reduced both their maintenance staffs and their spares inventories. Among the most distressing problems we have to deal wi
30、th on a regular basis is the housing bore damaged by a “spun bearing.” See Figure #1 Repairing a damaged housing is very time consuming and costly, especially on special units with multiple split lines or a divided power path. The AISE heavy duty crane spec See AISE Technical Report #6, June 19961 r
31、equires that bearings be mounted in replaceable cartridges see figure 21 due to the difficulty in removing the housings from their lofty perches for re-machining. Superintendents faced with expensive in-place machining of gearbox housings grouted into the floor quickly make sure future purchase spec
32、ifications follow the AISES example. Except for bearings that must float axially it might be worth studying the effects of tighter outside diameter fits on most housings. Where possible anti-rotation pins see Figures #3 and #U seem to be quite effective. CUMHIIN ANT I-Rill AI ili PIN OESIGNS IN CLF
33、Figure 3 and 4 Some of the lubricant loss experienced in the mill can be traced to worn seal diameters. Many output shafts are un-heat treated material whose soft seal diameters wear out rather quickly. After market wear sleeves have proven to be quite effective in fixing this problem, although the
34、units have to be removed from their mountings and stripped of their couplings to install them. Coupling removal occasionally causes enough additional damage to the shafts to necessitate their replacement anyway. We recommend that seal diameters be at least 285 Bhn on all shafts unless wear sleeves a
35、re fitted. See Figures #5 and #6 Chrome plating or induction hardening is preferred by some customers, although these processes add to the turn-around time. For severe duty “wash down” environments special seals such as packing glands are well worth the extra expense. 3 COPYRIGHT American Gear Manuf
36、acturers Association, Inc.Licensed by Information Handling ServicesI l Figure 5 and 6 , We are still trying to develop better ways to keep water out of the sump. Continuous caster gearboxes are particularly prone to water contamination, even when fitted with expansion chambers rather than breathers.
37、 As a result of this problem the mills have reluctantly gone to strict preventive maintenance programs on these critical drives. Through careful attention to details such as chrome plating seal diameters and regularly re-painting the interior surfaces of the housing we have been able to improve the
38、time between rebuilds by 50%. The most amazing thing to me is how well the carburized bevel gears have held up after being lubricated with watered down oil. In similar situations through hardened helical gears destroy themselves in a matter of weeks. failure” mode than tooth breakage on the drives t
39、hat we repair. As noted earlier, customers seldom react to the appearance of pitting, even on carburized gears. Severe wear frequently produces a noticeable change in how the gearbox sounds and draws the interest of the mechanics. Carburized gears typically outlast several sets of bearings unless th
40、e bearings are allowed to deteriorate so far that misalignment results in severe pitting or tooth breakage. In similar applications through hardened gears do not seem to hold up as well and are more frequently replaced at the same time as the bearings. The longevity of through hardened gear sets is
41、adversely effected by the high face to diameter ratios used in some designs which makes them prone to deflection related load distribution problems. In redesigning these gearsets we often use the additional capacity afforded by changing to carburized gearing to Wear is a much more common “gear reduc
42、e the face width thereby reducing the face/diameter ratio to more conservative levels. Basic design problems are relatively rare but when they occur the customer has to live with the results for years. One installation we service has a 491 ratio double reduction gearbox in a very tight location. The
43、 7:l gear sets have face to diameter ratios of over 2.0 and require replacement every 18 to 24 months. We have other “frequent visitors” with multiple input- multiple output configurations that suffer from recirculating power problems. Improved part quality or increased hardness have lengthened the
44、time between rebuilds but the lack of re- design options prevents effective long term resolution. When designing our own “specials” we try to give ourselves room for future upgrades by taking a Conservative approach to ratio selection, face to diameter proportions, and internal housing clearance. De
45、sian for Extended Life While recognizing the commercial limitations imposed upon us by the global marketplace, I believe that we can “design for extended life” by considering the problems likely to be encountered during the 15 to 25 year operating life of the typical process line and addressing them
46、 at the original equipment level. Not every gearbox needs “cartridge bearings“ but all deserve a water-free sump. Following the AISES example of graduated service classes to account for duty cycle, reliability expectations, and life requirement would go a long way towards improving customer satisfac
47、tion with our industry. Within the context of these service classes it would be possible to address detail design issues in a way that provides a consistent, level playing field for all competitors without forcing customers to develop their own in-house specifications. I realize the difficulty of ch
48、anging from a catalog selection system based upon a list of applications and service factors to one that includes non-gear related factors such as housing design. It was just a few years ago, however, that we considered 10,000,000 cycles to be “infinite life” despite knowing that this was less than
49、100 hours of use for the typical high speed pinion. A proposal for service classes is shown in Appendix B. 4 COPYRIGHT American Gear Manufacturers Association, Inc.Licensed by Information Handling ServicesAppendix A Cause of Repair Summary I. 2. 3. 4. 5. 6. Cause of ReDair Description # Orders o/Q Total Inadequate Lubricant Contaminated oil 353 16.1% Low Oil Level Insufficient oil 449 20.5% Bearings Worn or failed bearings 21 1 9.61% Overload Gear or Bearing Damage 38 1 17.4% Preventive Maintenance Precautionary Repairs 596 27.2% Miscellaneo
