As tech firms keep reducing product development cycles to stay competitive, simulation before going for a prototype is important to save time and improve product reliability – which ultimately results in cost savings. This article looks at the latest in simulation
Today, apart from designing innovative and reliable systems, engineers also have to ensure that they do so within a tight budget and a short time-frame. In this cutthroat competition, the primary driver to a successful product is the speed of getting the product to market before competitors. Simulation tools give engineers a very valuable aid—the ability to quickly select the most promising designs before committing to a prototype.
Testing while designing
While it is possible to come up with prototypes and do real-world tests on them, the cost implications are going to be sky-high. Simulation tools give you the ability to simulate your product on the computer and see how it would behave in a real-life situation by allowing you to model different physical characteristics and coupling them simultaneously.
An important factor that has been stressed so many times is to increase the test coverage so that it does not get limited just to test during production but also extends to the design phase.
According to a study conducted by researchers at NASA Johnson Space Center, findinga product defect during production is 21 to 78 times more expensive than during design. Increasing the test coverage during design can bring about a dramatic reduction in the relative cost to repair defects.
Addressing development cycle and cost issues
Traditionally, the product development process in any industry has a cycle of design, detailed design, prototype development, prototype testing and design rectification. This cycle goes on until the desired functionality and performance are achieved.
Being able to reuse simulation models for implementing algorithms by employing automatic code generation technologies enables design teams to leverage the effort invested in creating simulation models, reduces the overall product development time, and saves expensive and lengthy design iterations by mitigating coding and interpretation errors”
— Prashant Rao, technical manager, application engineering & training, MathWorks India
“This whole development cycle is very time-consuming and costly, as multiple iterations of physical prototypes are required to arrive at a refine prototype. If you take the case of a pump, our tool helps in simulating the various phenomena that happen inside the pump, such as the centrifugal force, the pressure produced, the flowof the fluidcreated by the pump, the heating up of the motor when in use and the electromagnetic phenomena based on which the motor performs. The combination of electromagnetism, rotation, vibration, thermal variances and other related phenomena can be simulated and this will help you to estimate what the best design should be,” explains Gautam Dutta, country manager, ANSYS.
With the demand for shorter product release cycles, while keeping costs as low as possible, development teams are challenged to bridge the gap between system designers who develop new algorithms and engineers who implement algorithms in software or hardware.
“Embedded software implementations are written in ‘C’ language and executed on digital signal processors or microcontrollers. Hardware implementations are written in hardware description languages (HDLs) such as VHDL or Verilog and programmed onto FPGA or realised on ASICs. Being able to reuse simulation models for implementing algorithms in ‘C’ or HDL languages by employing automatic code generation technologies like Embedded Coder and HDL Coder enables design teams to leverage the effort invested in creating simulation models, reduces the overall product development time, and saves expensive and lengthy design iterations by mitigating coding and interpretation errors,” explains Prashant Rao, technical manager, application engineering & training, MathWorks India.
There are two big advantages to doing a simulation before actually building the design and testing it.
“The biggest of these advantages is money. Designing, building, testing, redesigning, rebuilding and retesting for anything can be an expensive project in itself. Simulations take the building/rebuilding phase out of the loop by using the model already created in the design phase. Most of the time simulation testing is cheaper and faster than performing multiple tests of the design each time,” explains T. Anand, managing director, Knewron.
“The second biggest advantage is the level of detail that you can get from simulation. Simulation can give you results that may not be experimentally measurable with our current level of technology or available instruments in hand. Additionally, you can set the simulation to run for as many time steps as you desire and at any level of detail. The only restrictions are your imagination, programming skills, etc.”
Apart from cost savings, the second biggest advantage of simulation is the level of detail that you can get from simulation. Simulation can give you results that may not be experimentally measurable with our current level of technology or available instruments in hand”
— T. Anand, MD, Knewron
Besides, design firmsbenefitfrom time-to-market reduction and savings on expensive resources. Since many tests may call for destructive mechanism, costly components cannot be used frequently. Simulation helps to save a lot on that part. It also saves a lot of time in re-iterations being done physically and thus improves time-to-market.
Addressing system complexity
With system complexity increasing rapidly, and in some cases even exponentially, simulation models that capture and mirror the behaviour of systems have also become extremely large.
“Simulation models are increasingly being used to communicate and share design specifcations between project partners or between original equipment manufacturers (OEMs) and suppliers. Large corporations are using model-based design across product development processes spanning development and application centres across the globe. To collaborate, they have really large models—some with millions of blocks that are created or assembled by hundreds of engineers. A simulation environment has to be able to handle huge models that are developed by hundreds of engineers from diverse domains across geographically dispersed markets,” explains Rao.
Virtually every organisation is facing challenges to make equipment that are more energy-efficient than their predecessor. I remember talking to a key executive from a product manufacturing company, who said that today he has to build a product that has half the size and twice the rating of those made a mere five years ago”
— Gautam Dutta, country manager, ANSYS
The new MATLAB Desktop features a Toolstrip that puts key features and commonly-used functionalities up-front, a gallery of pre-built MATLAB apps, and tools that enable customers to create and share apps of their own. Simulink platform for model-based designs now features the capability to step through as well as rewind a simulation. Stateflow also incorporate the capability to create state-transition tables and use MATLAB as the Action Language.
As far as energy is concerned, virtually every organisation is facing challenges to make equipment that are more energy-efficientthan their predecessor.
“It should take up less energy and do more work. I remember talking to a key executive from a product manufacturing company, who said that today he has to build a product that has half the size and twice the rating of those made a mere fiveyears ago. That’s the kind of challenges energy and rotor machinery manufacturers and product developers are facing,” explains Dutta.
The industry has grown such that an engineer might have to work with a multi-disciplinary approach to successfully deliver cutting-edge products. Simulation tools now feature add-on modules that enable engineers to speed things up.
When working on a product, you do not want to spend a lot of time learning a new software and ensuring what you have built in your software is accurate. So we have add-on modules where we do all the customisation for our customers. Then they can use these models and very quickly come up with solutions that they are looking for”
— Vineet Dravid, managing director, COMSOL Multiphysics
“When you are working on a product, you do not want to spend a lot of time learning a new software and ensuring what you have built in your software is accurate. So we have a lot of add-on modules where we do all the customisation for our customers. Customers can use these models and very quickly come up with solutions that they are looking for,” explains Vineet Dravid, managing director, COMSOL Multiphysics.