Issue link: https://iconnect007.uberflip.com/i/1067105
48 SMT007 MAGAZINE I JANUARY 2019 independent labs that they work with for qual- ity checking. The distributors are now requir- ing that they are authorized to do destructive testing because of the high risk of counterfeit- ing. We haven't seen any counterfeit material yet, but we are very cautious and watchful for it. Las Marias: What can you advise our readers when it comes to these issues in the supply chain? Martin: They need to forecast. Even if it's not 100%, you need rolling forecasts for 52 weeks. We understand that customers do have that much visibility, but they need to give their part- ners as much visibility as possible. A 52-week lead time is pretty standard. Every BOM has components now that are over 40 weeks long—every one of them. That doesn't mean that we can't still find parts, but when the grey market, available inventory, and distribution channel dries up, there will be no parts. You'll be forced to wait for it. The customers who are in the best position are doing rolling forecasts. They update it monthly out 52 weeks. When their partner gives them an alter- nate component, they need to approve it very, very quickly. Usually, if it's in the grey market these days, it's gone within 24–48 hours. If the customer takes a week to approve it, the inven- tory is gone. The next time we find it, the price may be up. I would suggest that they buffer critical stock components and authorize addi- tional buffer stock on the allocated compo- nents. Then, look at a redesign for the smaller case sizes. If the lifecycle of their product has three to five years remaining, they may want to look at putting an additional buffer for the end of life. If their product lifecycle is eight to 10 years or more, then they really should look at redesigning the small case sizes so that they'll have sustainability. Finally, they need to look at an alternate type of materials. On the MLCC, you can typically replace several ceramic chips with a tantalum chip. Although there are some lead time issues in tantalum, it isn't nearly as bad as ceramic, and there's the newer technology—the poly- mer. If they see their product is 35 volts or less, they need to look at potentially a differ- ent product other than ceramic. They can get a four to one placement on the polymers, but it only works if it's 35 volts or less. On the Vishay CRCW series resistor, if they can take the non- automotive temperature range, you can add a C to the end of the part number and those lead times drop to approximately 12 weeks from 80 weeks. There are a lot of things that they can do to help themselves. Matties: Are there other strategies designers can use? You mentioned the lower case size, which makes sense, but are there other strate- gies the designers can employ to combat this? Martin: Yes, if they generally just look for long- term production, there is a huge technology shift that's happening right now. We're at the beginning stages of it now. I think it's going to accelerate where the industry is going smaller. So, a designer needs to look at using the small- est size parts that they possibly can. Designers tend to use the parts that they're comfortable with. It is not unusual when we see a brand- new design that it has obsolete components. That's one of the reasons we suggest to our customers to let us review their compo- nent list before they finalize the design. We can offer suggestions to them on alternative components given the same form, fit, function, and placement. It's designing for the supply chain. There are a lot of things that they need Panasonic.com Polymer technology, an MLCC alternative. Polymer technology, an MLCC alternative.