Supply chain
Good Tech Instruments' dry pump monitoring solution takes risk out of semiconductor manufacturing
Press release

The fabrication of semiconductor devices comprises a sequence of processing steps including etching, lithography, diffusion and thin film. How to ensure all process equipment is in tip-top shape is critical to semiconductor manufacturing.

For example, if the dry pump of the low pressure chemical vapor deposition (LPCVD) system or ion implanter used in the diffusion process provides not enough vacuum or even fails, causing back pressure pushing air and excessive particles into the chamber, the entire lot of wafer could go to waste. As to the thin film and etching processes, inadequate vacuum could also result in electrical anomaly or wafer defect, incurring tremendous loss to manufacturers. However, it may be challenging to prevent problems of inadequate vacuum or pump failure.

LPCVD systems and ion implanters operate under high vacuum conditions created by dry pumps, consisting of rotor motors and Roots vacuum pumps. In general, monitoring systems provided by dry pump suppliers only keep track of conditions including pressure, flow, temperature and electrical current but not vibration, so users do not have access to complete information and thus have difficulty predicting failure. Specializing in vibration measurement, Good Tech Instruments has introduced a dry pump monitoring solution in 2018 to help semiconductor manufacturers eliminate problems that have long been troubling them.

Wen-Tze Hsu, technology manager, Good Tech Instruments, notes that users expect two things from their dry pumps. One is the ability to predict failure and prevent downtime. The other is reliable data proving that good pumps can maintain long lifetime, which translates to reduction in maintenance costs. Even in the case that the pump is returned to the vendor for service, there must be guarantee that the serviced equipment is compliant with maintenance requirements.

Users demand precise estimation of equipment lifetime to bring down operation and maintenance costs

Hsu explains in more detail that users are accustomed to assessing machine health by monitoring electrical current, which is a lagging indicator, meaning when the current gets too high, the machine could shut down any time. The approach is too risky and does not comply with the principles of predictive maintenance. Hsu also speaks about the estimation of equipment lifetime, which is generally done based on statistical data and people's experiences. When a dry pump has been operating for a predetermined number of hours, it will be sent back to the vendor for service regardless of its condition. The user may incur unnecessary costs by sending a perfectly good pump for maintenance. Furthermore, the serviced pump returned from the vendor generally goes through an acceptance check based only on electrical current. Such a practice can hardly guarantee third-party maintenance quality, likely to increase the risk of shortened lifetime.

To address these problems, Good Tech Instruments has developed software programs and algorithms, implementing assessments based on ISO 10816 specifications on vibrations of rotodynamic machines, broadband value management and spectrum overall value (SOV) management and transforming complex signals into intuitive pump health indicators for easy viewing. They can effectively complement traditional monitoring systems. Even if the user does not have Fast Fourier Transform (FFT) spectrum for energy analysis, Good Tech Instruments' solution can still provide some insights.

Good Tech Instruments' monitoring solution supports assessments in accordance with the vibration limits defined in ISO 10816 for Category I pumps. It can show physical vibrations under certain bandwidths according to different machine types so that users can instantly know if their dry pump is in a good, normal, warning or dangerous condition. It also supports assessments based on the less-noticed type II variations, meeting the trend of quick and nimble management.

MCM-100 computes frequency spectrum with fine resolution, helping capture minor signs of defect

SOA is a summary of frequency spectrum analyses that combines measurements of misalignment, dynamic unbalance, axial bending, mechanical loosening and rotary friction – the five major indicators of abnormal vibration – and presents them in easy-to-understand numbers for users to stay on top of machine conditions without hassle. Moreover, Good Tech Instruments' solution additionally monitors pitting corrosion (broadband high-frequency vibration) and powder turbulence (low-frequency vibration) that conventional vibration monitoring solutions cannot capture, thereby minimizing the false alarm rate.

Hsu sheds light on how Good Tech Instruments is able to deliver high quality monitoring with such refined precision. The secret is ADLINK's MCM-100, which is the foundation of Good Tech Instruments' monitoring solution. Featuring a compact form factor enabling quick and easy installation, built-in data acquisition (DAQ) card, high-performance processor offering computation power, multiple ready-to-use driver programs, IEPE/ICP sensors as well as interface ports including USB, COM, Ethernet and 4G, MCM-100's all-in-one design significantly simplifies system integration efforts for Good Tech Instruments.

With four input channels, MCM-100 can monitor four axes at the same time. It is particularly noteworthy that MCM-100 features high 24-bit resolution and captures high-frequency signals at a very high 128kS/s, so it is capable of supporting a wider range of sensor inputs and performing more refined frequency spectrum analysis, compared to conventional 12-bit to 16-bit solutions and 20kS/s sampling rate. Coupled with Good Tech Instruments' mathematical modeling working behind the scene, the collaborated solution delivers outstanding vibration analysis with the results displayed in intuitive and user-friendly visual presentation. Users can thereby assess the health condition of their dry pumps and optimize equipment lifetime based on data management. Furthermore, they will be able to conduct predictive maintenance at reasonable costs, effectively evaluate the quality of third-party maintenance and most importantly, significantly lower the risk of unexpected downtime.

DIGITIMES' editorial team was not involved in the creation or production of this content. Companies looking to contribute commercial news or press releases are welcome to contact us.

© 2020 DIGITIMES Inc. All rights reserved.
Please do not republish, publicly broadcast or publicly transmit content from this website without written permission from DIGITIMES Inc. Please contact us if you have any questions.