Not just dramatic improvement in existing product development processes, but discovering what was previously impossible to see

The benefits of upfront computational fluid dynamics – faster time to market, reducing physical prototypes, ensuring quality in the early design stages – seem fairly evident.  But, they still need to be proven, as Bruno Fairy knows very well.

Fairy is a simulation and analysis engineer working at Parker Hannifin’s Hydraulic Controls Division in Warwick, UK.  But, perhaps his most influential role is for Parker Hannifin Corporate, for which he supports worldwide use of CAE tools.

With annual sales exceeding $10 billion, Parker Hannifin is the world’s leading diversified manufacturer of motion and control technologies and systems, providing precision-engineered solutions for a wide variety of mobile, industrial and aerospace markets.  The company has increased annual dividends paid to shareholders for 53 consecutive years, one of the top five longest-running dividend-increase records in the United States’ S&P 500 index.

Managers at Parker Hannifin recognize that delivering design jobs on budget and on schedule is no longer enough to ensure survival in an economy where margins are tight – and getting tighter all the time.  To gain any real competitive advantage it is necessary to demonstrate an ability to identify opportunities that add value beyond the traditional cost and time savings.

Fixing problems early

For Bruno Fairy, adding value means integrating virtual development technologies such as CFdesign upfront CFD software into the product development mix.  CFdesign gives companies such as Parker Hannifin a clear picture of how designs will perform before making major commitments of time and resources.  This reduces time-to-market, enables experimentation with more design options, and helps ensure reliability.

“Reliability is at the heart of everything we do for our customers,” says Fairy. “If workers have to stop because a machine is not working properly, huge amounts of money can be lost. If we are able to deliver reliable systems everybody wins: Parker maintains its reputation for excellence and builds a better rapport with its customers, while our customers benefit through less downtime.”

Reliability may seem like an obvious goal for any manufacturer, but the real significance of the approach only becomes clear when one considers the cost of fixing a problem with a system or a component.  The cost escalates the further along the development path the manufacturer travels – from the design stage to the procurement stage, the production stage to shipment and, finally, on site with the customer.

“It costs little to fix a defect at the design stage, but as you go along the development process of your system or component, the cost of fixing a problem spirals very quickly,” says Fairy. “If you want to make sure your component or system is as cost effective as possible you need to identify and fix defects as early as possible in the design cycle.”

CFD simulation shows pressure results of hydraulic fluid flowing inside a Parker Hannifin valve.

Reducing reliance on physical prototypes

Parker’s Hydraulic Controls Division-Warwick designs and manufactures some of the best in class and sophisticated hydraulic equipment for the defense industry.  Maximizing reliability in the field involves knowing how that equipment will react if, for example, it is over-pressurized.  Building and testing a prototype under these circumstances is not only expensive and time-consuming, it can also be dangerous. Until recently, however, companies had no choice but to build and test physical prototypes.

“Consider vehicle manufacturers or aircraft designers,” says Fairy.  “They need to know what will happen when a car crashes or a landing gear fails.  Testing prototypes of cars and planes to see how they behave in real-world conditions was, until recently, the only way engineers could begin to design in reliability. CFdesign software allows us to do virtual product development: simulation and analysis of the CAD model to reduce the number of surprises as early as possible in the design cycle, when it is easier and cheaper to fix any defects.”

The ROI methodology

While Fairy knew the benefits of upfront CFD from experience, he had to prove it with an ROI methodology and test cases before Parker divisions felt safe in investing time and money in CFdesign.  He says designing an ROI methodology to show the value of upfront CFD is relatively straightforward.

“You know how much your contract is worth, and from this you can work out a ratio of how much the project costs per day or over a whole year.  You also know your labor rates for engineering, manufacturing, testing, prototyping and approximately how long will be spent on the project for each category.  This allows you to see clearly the financial benefits of getting the product in early.

“Of course you also need to include the cost of the software you are using. You divide the cost of the software seat by the typical number of usage hours in a year and that gives you a cost per hour.  Factor in the number of hours it was used on a project and that gives you the cost of using the software.  You also have to assume it will take a few design cycles before you get to the final version of a product system.  Of course each cycle will cost less than the first cycle because you are building on what you know rather than starting from scratch each time.”

Dramatic, irrefutable evidence

The results from trial projects using CFdesign were dramatic and irrefutable as measured by Fairy’s ROI methodology.

In an example where engineers within Parker’s Domnick Hunter Division needed to rationalize a product range to increase competitiveness and therefore market share, the engineering team was able to demonstrate an 80-percent development cost-saving and a 50-percent development time-saving using virtual product methods over conventional methods.

In another example, Bruno’s ROI comparison of an older project highlighted 90 percent development cost and 80 percent development time wasted because, at the time, virtual product development methods were not available and conventional methods were employed.  In today’s economic climate these numbers are impossible to ignore and provide concrete proof, if it were still needed, of the real-world benefits of frontloading CFD.

Simulation shows flow pattern results of hydraulic fluid inside a Parker Hannifin valve.

Solving the previously unsolvable

The intangible for which there is no real ROI methodology is the ability to solve a problem that could not be cracked in any other way.

Fairy cites a wheel-loader spool design project as an example.  Before implementing CFdesign software, the engineering team could not entirely solve an instability issue after 18 months of effort.

“The instability was thought to be linked to cavitation inside the manifold or valve, something you cannot see very easily with conventional testing methods,” says Fairy.  “Even with considerable expertise and past experience of similar situations it was very difficult to pinpoint the cause of the problem.

“In this situation, the engineering team would engineer a solution, manufacture a prototype, assemble the prototype on the customer’s system, measure system behavior, analyze the results, and then re-engineer a solution.  As with any complex hydraulic system, it doesn’t take much to cause valve or manifold instability.  They kept on trying but literally could not accurately locate where the issue was.  It was impossible to visualize.”

The design, build, test method was eventually replaced by a virtual approach when Parker Hannifin invested in CFdesign software.

“Within about four weeks of attending training, the division was able to develop a model, analyze it and see something that made sense,” says Fairy.  “We designed the solution on CAD, ran the CFdesign simulation again to check that it was yielding better results, and eventually made a prototype, tested it on-site with the customer and it worked.  The feedback from the operator was that it was the best wheel loader he had ever driven.”

In this case, the simulations also showed that engineers at Parker had been doing things right all along.  They simply needed a deeper level of insight into what was going on inside the product in order to solve the problem.

Two strands of ROI 

Fairy’s experience has shown there are two major strands of ROI from virtual simulation tools such as CFdesign – the one derived from vastly improving existing processes and the one from having the ability to visualize the source of previously unsolvable problems or the possibilities of new efficiencies.

Although it is easier to prove the ROI for improving existing processes, the ability to gain new insight into the inner workings of parts and assemblies might just be the more valuable in the long run as it allows Parker divisions to build a knowledge database that can be used by future generations. 

###