Preventing a Trip and Severe Damage: Early Warning of Increased Vibration on Gas Compressor Turbine Driver
Author Sticky
Jacqueline Vinyard
GE Vernova’s Software Business
Palash Verma
GE Vernova’s Software Business
Jun 17, 2025
4 Minute Read
Share
Predict: What did GE Vernova predictive analytics software find?
Industry: Oil & Gas
Region: N. America
Asset Type: Steam Turbine Driver - Single Case
A vibration deviation on a turbine driver for a gas compressor at an oil and gas site in North America was detected by GE Vernova's SmartSignal Predictive Analytics. Specifically, the digital twin of the turbine driver showed a spiking signature for the drive end vibration sensors on the turbine, with frequent vibration spikes noted up to approximately 1-1.2 mils on the turbine drive sensors.
As time progressed, the vibration levels increased to approximately 1.8-2.5 mils. The GE Vernova engineering team, Industrial Managed Services (IMS) sent out a notification to the customer and discussed the detected anomaly during the regularly scheduled call with them. The IMS team suggested confirming whether the sensor was reading correctly and inspecting for oil leakage into the steam seal.
Region: N. America
Asset Type: Steam Turbine Driver - Single Case
A vibration deviation on a turbine driver for a gas compressor at an oil and gas site in North America was detected by GE Vernova's SmartSignal Predictive Analytics. Specifically, the digital twin of the turbine driver showed a spiking signature for the drive end vibration sensors on the turbine, with frequent vibration spikes noted up to approximately 1-1.2 mils on the turbine drive sensors.
As time progressed, the vibration levels increased to approximately 1.8-2.5 mils. The GE Vernova engineering team, Industrial Managed Services (IMS) sent out a notification to the customer and discussed the detected anomaly during the regularly scheduled call with them. The IMS team suggested confirming whether the sensor was reading correctly and inspecting for oil leakage into the steam seal.
Prefer to listen?
Stream our audio version
00:00/00:00
Diagnose & Prioritize: What was the underlying cause and risk urgency?
Following the alert from the IMS team, the customer conducted a thorough inspection and discovered that there was an oil leak in the bearing housing. Initially, this issue was not perceived as significant. However, during a detailed weekly discussion with the assigned GE Vernova engineering team member, it was suggested that the vibration spikes were likely caused by oil being drawn into the steam seal. This diagnosis was based on similar previous incidents where oil intrusion into the steam seal led to significant issues. When oil enters the steam seal, it can undergo thermal degradation due to the high operating temperatures, converting into coke.
This coke can then adhere to the shaft, creating areas of uneven friction and causing intermittent spikes in vibration levels. If unchecked, the continued escalation of vibrations could lead to a trip of the equipment, posing a risk of severe damage.
Upon further inspection, the customer discovered that a damaged oil baffle was responsible for the leak, allowing oil to transform into coke on the shaft surface. This finding led to immediate action, and the site personnel replaced the faulty oil baffles.
Once the turbine was reassembled and brought back into operation without the presence of leaks, the vibration levels stabilized and returned to the expected predictive analytics trends.
This coke can then adhere to the shaft, creating areas of uneven friction and causing intermittent spikes in vibration levels. If unchecked, the continued escalation of vibrations could lead to a trip of the equipment, posing a risk of severe damage.
Upon further inspection, the customer discovered that a damaged oil baffle was responsible for the leak, allowing oil to transform into coke on the shaft surface. This finding led to immediate action, and the site personnel replaced the faulty oil baffles.
Once the turbine was reassembled and brought back into operation without the presence of leaks, the vibration levels stabilized and returned to the expected predictive analytics trends.
Customer Impact
The early notification in November provided significant value to the customer by enabling them to take a proactive approach to maintenance. The customer was able to strategically plan a shutdown of the equipment in December, thereby preventing an unplanned trip that could have resulted in more extensive damage to the turbine rotor due to excessive rubbing.
This early intervention allowed the customer to identify and replace the damaged oil baffles, as well as to clean the affected areas, all before the situation necessitated a major overhaul. By adopting this proactive strategy, the customer ensured that necessary resources and parts were readily available for scheduled maintenance activities. If the issue had gone unnoticed, it could have escalated to a point where the shaft might have bent, leading to significantly higher repair costs and prolonged downtime.
The early detection by the GE Vernova engineering team averted a forced outage and potential catastrophic failure. The turbine was offline for approximately 87 hours, but this planned downtime potentially saved up to six days of operational disruption and an estimated $337,500 in costs. These savings were primarily achieved by addressing the steam seal and bearing housing issues early, thus avoiding the much more costly replacement of the rotor. This proactive maintenance strategy underscores the value of predictive analytics in preventing costly and time-consuming repairs.
Ask about M&D+ Powered by SmartSignal.
This early intervention allowed the customer to identify and replace the damaged oil baffles, as well as to clean the affected areas, all before the situation necessitated a major overhaul. By adopting this proactive strategy, the customer ensured that necessary resources and parts were readily available for scheduled maintenance activities. If the issue had gone unnoticed, it could have escalated to a point where the shaft might have bent, leading to significantly higher repair costs and prolonged downtime.
The early detection by the GE Vernova engineering team averted a forced outage and potential catastrophic failure. The turbine was offline for approximately 87 hours, but this planned downtime potentially saved up to six days of operational disruption and an estimated $337,500 in costs. These savings were primarily achieved by addressing the steam seal and bearing housing issues early, thus avoiding the much more costly replacement of the rotor. This proactive maintenance strategy underscores the value of predictive analytics in preventing costly and time-consuming repairs.
Ask about M&D+ Powered by SmartSignal.
*Avoided costs are based on North American average production loss.
Author Section
Authors