Introduction
This report describes a proof of concept test performed on the #2 Wash Press at the Domtar pulp and paper mill in Kamloops, BC Canada.
The Intellinova system was initially installed on the #2 wash press and began collecting data on June 29th, 2020. The system initially was collecting vibration data without a tachometer for speed reference and order tracking due to a connector issue. On July 30th, 2020 the wash press was down for several days and the tachometer was installed and the data started being order tracked. With the data now being order tracked for optimum clarity, outer race bearing fault frequencies were observed on the Southwest Bearing NDE. A once per revolution event was also observed on the Northwest NDE bearing measuring point location.
Mill information
Domtar’s Kamloops Mill uses softwood fiber from third party sawmills in the region to produce high quality Northern Bleached Softwood Kraft pulp and unbleached softwood kraft for customers from North America, China and Southeast Asia.
Kamloops Mill at a glance
- One Fiber Line
- Products: Northern Bleached Softwood Kraft Pulp, Unbleached Softwood Kraft
- Production: 408,000 Air Dry Metric Tonnes Annual Pulp Capacity
- Pulp end uses: Hygiene products, building siding, unbreakable dishware, more…
- Mill Built in 1965
Fig 1. Domtar Kamloops Mill
Conclusion and summary
During the 60-days of monitoring and analysis of the #2 Wash Press, it was determined upon viewing the data that the Southwest bearing on the south roll was indicating stage 2 bearing damage on the outer race. The damage is somewhat early and is mostly seen in SPM HD and HD ENV Filter 4. Progression of the bearing damage will more evident once the BPFO fault frequencies are visible in HD ENV Filter 3. The magnitude of the BPFO fault frequency seems dependent on load. It is recommended to provide a load indicating signal into the SPM Intellinova to compare the bearing damage with the load of the press.
Application description
A pulp wash press is used in the chemical pulp process for dewatering, washing and pressing of the pulp. The pulp wash press is critical in the pulp manufacturing process; a malfunction may cause severe limitations or, in worst case, a complete stop for several days. The design of the pulp wash press is shown in the drawing and photograph below.
Fig 2. Schematic drawing of the pulp wash press
The rollers are driven by hydraulic motors (Hägglunds) and the speed of the rollers is synchronized via a control system. Typically, the RPM ranges from 7 to 9. Synchronizing gears are used on the non-driven end for matching of roller speeds. It is relatively common for other types of pulp wash presses to have two separate RPMs, one for each roller without the synchronization gears. If that is the case it is important to have separate RPM transducers in order to have a properly working order tracking.
Fig 3. Wash Press Syncing Gears
Fig 4. Wash Press Drive End Fig 5. Wash Press Non-Drive End
In most presses, an automatic high pressure water spray system is used for cleaning purposes. This system can be continuously active, or activated on a time based interval. These high pressure water jets affect the shock pulse sensors, creating “ski slope” readings. In most cases, the “ski slopes” are bandwidth limited and possible bearings signals are well above the “ski slopes”. If the “ski slopes” are severe, a triggered approach must be used. A “spray on/off” signal from the control system can be used to trigger the measurements. For the presses described here, a non-triggered approach has been used and the bearing related signals are clearly seen. The forces acting on the rollers are a combination of gravitational forces and nip forces. A good transducer position is indicated in the picture below.
Fig 6. Forces affecting the load zone direction
Background
Domtar is currently testing an Ultrasound monitoring system to catch low RPM failures on equipment. A second comparison test with the SPM Instrument Intellinova system was installed utilizing SPM HD and HD ENV technologies to have low speed asset control and avoid costly downtime.
System setup
The #2 wash press is monitored using an SPM Intellinova INSEN16P system. Four DUO-TECH transducers were installed (1 per bearing). One proximity switch was installed to capture RPM data. The system is taking four measurements per bearing every thirty minutes for twenty-four readings a day for a good amount of baseline data. The number of measurements could be reduced to less once installed permanently.
The following parameters were used in the system setup:
SPM HD :
- Measuring time: Same as FFT
- Short time memory: Time signal and FFT
- Long time memory: Time signal and FFT
- Upper frequency: 100 orders
- Lines in Spectrum: 1600 lines
- Symptom enhancement factor: 5
HD ENV:
(Filter 4 > 5kHz to 40kHz + Filter 3 > 500Hz to 10kHz + Velocity > no filter)
- Measuring time: Same as FFT
- Short time memory: Time signal and FFT
- Long time memory: Time signal and FFT
- Upper frequency: 100 orders
- Lines in Spectrum: 1600 lines
- Symptom enhancement factor: 5
Case Description
During the 60-days of monitoring and analysis of the #2 Wash Press, it was determined upon viewing the data that the Southwest bearing on the south roll was indicating stage 2 bearing damage on the outer race. The damage is somewhat early and is mostly seen in SPM HD and HD ENV Filter 4. Progression of the bearing damage will more evident once the BPFO fault frequencies are visible in HD ENV Filter 3. The magnitude of the BPFO fault frequency seems dependent on load. It is recommended to provide a load indicating signal into the SPM Intellinova to compare the bearing damage with the load of the press.
Fig 7. SW bearing (MP#1), SPM HD trend
Fig 8. SW bearing (MP#1), SPM HD Spectrum
Fig 9. SW bearing (MP#1), HD ENV Spectrum Filter 4
Fig 10. SW bearing (MP#1), HD ENV Spectrum Filter
Fig 11. SE bearing (MP#2), Segmetation
Fig 12. SE bearing (MP#2), Drive Commutation Frequency
Fig 13. SW bearing (MP#1)
System Recommendations
- Continue to monitor the wash press with HD Technologies.
- Confirm bearing geometrics on the Northwest bearing.
- Verify teeth count on Sync Gear (I set up a TSA to assist).
- Could consider filtering some ski slopes, may have risk.
- Acquire a signal from the PLC indicative of load/process to compare to vibration data.
- Confirm 1X harmonics with process/segmentation impact/water pressure.
