Two titanium process blowers at a major U.S. chemical manufacturer were exhibiting excessive vibration levels. Both were employed to move wet chlorine gas, each increasing the gas pressure by approximately 6.5 pounds per square inch (psi) while circulating a volume of 57,500 cubic feet per minute (cfm).
When one of the blowers was shut down for examination, the inboard bearings showed little damage, but the outboard bearing (closest to the fan impeller) showed evidence of severe damage. Robinson Industries of Zelienople, Pa., and No Bull Engineering of Delmar, N.Y., investigated the problem, in coordination with the chemical manufacturer's own engineers.
Each of the blowers was driven by a 3,000-horsepower (hp), 1,780-revolutions-per-minute (rpm) constant-speed AC motor with a metal disk coupling. The blowers were configured with two plain sleeve bearings and an overhung blower rotor inside an extremely heavy-duty blower housing. The blower rotor was fabricated entirely of titanium Grade 7, and the housing was 6 inches (in.) thick, made of mild steel with titanium wallpaper to provide effective corrosion resistance. The entire blower and motor assembly rested on an integral sub-base grouted to a heavy concrete foundation.
Lubrication for the blower and motor bearings was supplied via a circulating oil system providing filtered and constant temperature lubricant. The blower and bearings were monitored by means of radial and axial vibration proximity probes and platinum resistance temperature detectors (RTDs). Blower volume was controlled by a variable inlet vane damper of titanium Grade 7 construction. A multi-ring carbon shaft seal with a titanium Grade 7 housing construction ensured no chlorine gas escaped into the atmosphere.
Damage to the bearings
Each blower system contained four bearings: a motor float (inboard) bearing; a blower float (inboard) bearing; a fixed motor (outboard) bearing; and a fixed blower (outboard) bearing. After shutdown, inspectors found no damage to the fixed motor (outboard) bearing, inboard (float) motor bearing or the inboard (float) blower bearing, but did find extensive damage on the outboard (fixed) bearing on the blower side. The bearing liner was heavily chipped, and pieces of the babbitt had broken away from the liner casing. In addition, the rotor shaft had rubbed against the titanium carbon ring seal housing, damaging the shaft and the seal. These observations led Robinson Industries and No Bull Engineering to focus on bearing loads.
A comprehensive rotordynamics model was created, consisting of the motor shaft, metal disk coupling and the blower shaft. This model was a departure from all previous analyses of this system, which had treated the blower and motor as separate, independent systems from a lateral rotordynamics standpoint. Robinson Industries and No Bull Engineering decided on the comprehensive rotordynamics model after making a number of observations, most of which centered on the large angular stiffness of the metal disk coupling connecting the motor and blower systems.
The comprehensive rotordynamics model was employed for a bearing static loading analysis, which was performed with a proprietary computer code. The analysis examined the effect of blower float (inboard) bearing stiffness on the load distribution among the four bearings. Bearing stiffness was held constant for all but the blower float (inboard) bearing. Perfect coupling alignment was assumed. Results showed that as bearing stiffness for the blower float (inboard) bearing decreased, bearing load held fairly constant for all bearings except the blower float (inboard) bearing. In the latter case, bearing load decreased dramatically.
This analysis exposed a fundamental problem with all previous analyses, which treated the blower and motor as independent systems. When treated as an independent system, the blower is a statically determinate system, meaning that regardless of the blower float (inboard) bearing's stiffness, its load remains 80 pounds (lb). However, when the blower and motor are treated as a single interdependent system, they become a statically indeterminate system, meaning the flexibilities of the supports and the shafting determine the load distribution.
Because three of the bearings are significantly stiffer than the fourth ," the blower float (inboard) bearing ," those three have a tendency to reduce the load on the fourth. In other words, this analysis confirmed that the load on the blower float (inboard) bearing would be well below the previously calculated 80 lb. The lightly loaded blower float (inboard) bearing would be one potential cause for excessive vibration.
Next, an unbalance response analysis was performed on the same comprehensive rotordynamics model to determine how the four bearings would respond to strategically placed forces of unbalance. In particular, the blower wheel's unbalance was taken to be 10.9 ounces per inch (oz./in.), in accordance with a G2.5 level of precision (Air Movement and Control Association International Inc. [AMCA] Standard 204-96).
In addition, an unbalance response analysis was conducted with a maximum coupling unbalance (180 degrees out of phase with the blower wheel). The blower wheel itself remained at an unbalance of 10.9 oz./in.
The following changes were introduced with satisfactory results in lateral rotordynamics analyses and in practice:
The stiffer metal disk coupling was replaced with a more flexible gear coupling.
Tilting pad bearings were designed to replace the blower's plain cylindrical bearings because the latter are particularly susceptible to lightly loaded conditions. On the other hand, with geometric preload and offset pivots, tilting pad bearings will generate significant stiffness and damping even with no load.
Alignment procedures were simplified. The motor shaft was aligned parallel to the blower shaft and 0.004 in. lower than the blower shaft, allowing for sufficient thermal expansion of the motor frame without overload on the bearings in the cold condition. Previously, misalignment had been 0.012 in. between motor and blower shafts.