
5Efficient and reliable UHPLC pump maintenance - Self-diagnostics using Digital Twin
4. Diagnostics 2.0
4. Diagnostics 2.0
UHPLC systems are complex and it is a great challenge for a service
engineer to master all parts of the system, both in terms of operation
and troubleshooting. The UHPLC pump alone requires a great deal of
technical knowledge, which is not always present. This also applies to
the troubleshooting tools that are commonly used. The level of
knowledge and skills of the service engineer is leading for successful
diagnostics and troubleshooting.
Service engineers and maintenance departments are always looking
for better ways to guarantee or restore the performance of UHPLC
systems. This means easier, quicker and more reliable ways to keep the
systems and its modules operational and minimize unnecessary
downtime. In short, they want the best reliability strategy for maximizing
uptime of their UHPLC pumps.
Based on her years of experience with Digital Twin, Spark Holland
concludes that her diagnostic tool for autonomous testing is a valuable
and very reliable addition to existing methods of maintenance and
diagnostics.
A Digital Twin is a highly accurate, virtual working model of an
intended system (product or service) in a given environment.
Within the innovative manufacturing industry, a Digital Twin is
widely used today. A Digital Twin is constructed based on the
same natural laws and principles as a system in the real
world. The way a Digital Twin behaves in its virtual world is
almost identical to its physical counterpart.
Designing with a Digital Twin has great advantages. The
system design can be optimized by engineers at an early
stage, with or without customer feedback, before it is put into
production. Furthermore, it enables several design concepts
to be virtually validated in advance. This saves time and
money, also because fewer physical prototypes are needed.
In short, the engineering process is more efficient while at the
same time it is more thorough. This also applies to further
development and modification of the product/system. An-
other advantage is that the development of hardware and
firmware can run in parallel with the mechanical design,
which again saves time.
A Digital Twin of a product consists of data (the digital base)
and models related to requirements, form, structure, calcula-
tions, specifications and simulations of a product. At Spark
Holland, the mechanical domain of a Digital Twin is devel-
oped at the start of the mechanical design to validate its
ability to meet product requirements. With the system
description that the Digital Twin offers at that stage, engineers
are able to simultaneously start with the design of electronics
and firmware. When the design has reached the required
level of detail, prototypes can be produced and tested. By
collecting data from physical prototypes, for example on flow
and pressure, the last details of the Digital Twin are tuned in
order to represent the physical system as accurately as
possible. With this approach the Digital Twin forms an addi-
tional design layer, which helps engineers analyze a system
more thoroughly and foresee problems before they occur.
In addition to the benefits of developing a Digital Twin during
the design process, the Digital Twin also allows to perform a
multitude of experiments in a short time. Furthermore, a
Digital Twin is capable to consistently and repeatedly simu-
late the system behavior with worn wear parts in a controlled
environment, which is difficult to achieve on a physical system.
These properties allow engineers to develop state-of-the-art
diagnostic tools based on machine learning. These tools help
field service engineers to faster locate worn components
causing an underperforming system, saving time and costs.