Is common sense all that common? If it were, people wouldn’t cause gridlock (if you can’t clear the intersection, wait on your side of the intersection. Easy.). If the red light is out, treat it as a stop sign – everyone gets a chance, traffic moves smoothly and everyone eventually crosses the intersection safely. Hmmm… can you tell traffic was especially nasty on the way to work today? Anyway, we’re here to talk about common sense as it applies to the design process.
Common sense (as well as lots of empirical data) suggests that the earlier a design change is made the less costly it is for the organization. Making changes becomes increasingly more difficult and disruptive as a specific model progresses thru the design process. Making a change can become even destructive after a product has gone to market because it can have devastating effects for the company (both financially and brand-wise). Unless you’ve decided to immerse yourself in World Cup frenzy and stopped listening to the news, then you’ve already heard about what’s been happening at Apple, Sony and Toyota/Lexus. Apple’s had to deal with faulty antenna problems, Sony has had to recall their Vaio laptop while Toyota and Lexus … well you know.
I have seen a lot of information about the cost of making a design change (as I’m sure so have you). If I remember my charts correctly, the cost of making changes increase by a factor of 10 at each step of the design process. In other words, while a change during the initial concept stage might cost $1, by the time it reaches the production phase the same change could cost $10,000+. Ok, so these figures can look rather benign in the grand scheme of things but think about this. How often do you need to make only one change? And do you know how much Engineering Change Orders (ECO) cost at your company? These costs rack up pretty quickly especially if you’ve got a large product portfolio.
One of our customers, Integrated Device Technology (IDT), uses FloTHERM PACK for thermal simulation. Founded in 1980, IDT develops mixed signal semiconductor solutions used in digital media products. IDT launches about 50 new products with new packaging configuration every year. Before using simulation, IDT relied on physical testing to ensure thermal performance. Between 10% and 20% of the initial thermal designs didn’t meet specifications thus requiring additional tests. And in these cases, the product launch was typically delayed by about 6 weeks. Making a major change at this stage also added engineering time and tooling costs. Since moving to simulation, they simulate every new packaging configuration prior to physical testing.
IDT engineers now evaluate thermal performance of several packaging alternatives in order to identify a package configuration that best meets thermal, board real-estate cost and other objectives. In so doing, the organization believes that simulation can help increase their revenues by enabling them to avoid late stage design changes that would otherwise delay product launches. I won’t go into all the technical details here… if you’d like to read a detailed version of this case study, please feel free to follow this link. But I wanted to share something they said because it was rather poignant: “since we have begun simulating every new packaging configuration, we have never had to make a late-stage design change nor has any product been delayed for thermal performance reasons,” said Jitesh Shah, Advanced Packaging Engineer for IDT.
I don’t know about you but I think using simulation is just common sense.
Until next time,
PS. Happy belated Independence Day to my fellow Americans.