The Common Parts Library

The two most common causes of delay in small volume manufacturing here at Screaming Circuits (and presumably, others like us) are component availability, and footprint mismatches.

We don’t substitute parts without your approval for a number of reasons. I’ve written about those reasons a few times before. (Here, here, and here.)

Incorrect footprints can lead to a host of headaches as well. (Read more herehere, and here.)

Until recently, I haven’t seen a lot of progress toward solving these problems for the hordes of engineers that don’t have big support departments at their disposal. In fact, with the proliferation of newer, and small, component packages, and evolution of the supply chain, it’s really gotten worse.6a00d8341c008a53ef01bb089b708f970d-120wi

However, there are a couple of Knights in Shining Armor riding in to try and solve both problems. The Common Parts Library (CPL), created by Octopart, aims to create a list of components with the highest probability of being available and staying available (there are no guarantees where component supply is concerned).

The other exciting entrant is SnapEDA. SnapEDA has a massive, and growing, library of component footprints. I’ve used their footprints with good success for high pin-count devices, and other parts with complex packages. It can save a lot of time and give better confidence that all of the pins go to the right functions.

Duane Benson
Map makers put fake roads in as copyright traps
These folks don’t do that. Nice.

Design Tradeoffs

As many EMS firms are trying to grab a bigger piece of the design services market, one thing we’ve noticed when we tour their digs is how much more relaxed those designers appear. They are different in terms of setup – some sit in open cubicles, others have individual offices, and still others share a common but separate office, akin to a bullpen – but no matter the configuration, the occupants come across in control and unrushed.
Contrast that to the OEM designers we speak with, who almost uniformly come across as harried.
We’re not sure why this is. Perhaps those at EMS sites are more confident in their job security, knowing that more designs are being shipped their way each year, while their OEM counterparts feel under the gun, worried that their bosses, having already outsourced fabrication and (in many cases) assembly, might at any time let design go, too.
Even so, those designers who responded to our annual salary survey overwhelming were employed by OEMs. Does that suggest EMS designers are significantly fewer in number, harder to reach, or just less interested in filling out a survey? We don’t know.
What we do know, however, is that designers are as not easily compartmentalized as they once were. More have advanced degrees and increasing responsibility. They have become integral, even if more than one-third of respondents still worry about their jobs.
About three-fourths of those who responded were based in the US (probably because the survey was conducted only in English). Most of them have more than 20 years’ experience, suggesting that cost-cutting measures elsewhere aren’t decimating the field.
The average annual US household income was $63,000 in 2011. Given that, designers are doing well. Some 73% of respondents indicated their salary exceeds $60,000, with 17% revealing salaries topped $100,000. For comparison, the median income for a bachelor’s in engineering is $82,712. And most continue to get raises in line with or exceeding the average US raise of 2.8% last year. After the roller coaster of 2008-10, stability is welcome.
Keeping up with the Joneses is one thing. Keeping up with technology is something else. More than one-third of our respondents again said maintaining their technology fluency is their biggest challenge. That’s understandable – as consulting editor Jan Vardaman notes (pg. 20), advancements in everything from wiring materials to substrate systems are ahead. Moreover, an impending shift to copper pillar offers exciting possibilities for tighter silicon and package routing, but with those come the headaches of greater crosstalk and signal integrity issues. Technology, like life, is about tradeoffs.

One more note on the salary survey. Of those designers who recommend or approve products or services, only 78% get to weigh in on CAD tools. While we understand why some designers are out of the loop on this – many EMS companies buy tools as directed by an end-customer, user be damned – it’s still jarring in this day and age that those tasked with such a critical job don’t get a bigger say in how they perform it.

(I would be remiss if I failed to add that senior editor Chelsey Drysdale conducted the survey, compiled the data and wrote the report.)

January Issue Now Available

Hi, and Happy New Year!

Our January issue is now available. Highlights this month include a profile of APCT, the Silicon Valley board shop; a recap of Productronica (including highlights of the new fab equipment); a fabricator’s take on rebuilding America’s manufacturing base; plus some great technical columns on centroid files and designing flex boards.

Here’s the link to the online version, or if you’d prefer the digital version, click here.

Happy reading!

ValueProto from Sunstone

I think I’m done with the Geiger counter layout. Now I just need to get the thing built up to see if it works. I’m pretty sure, but you never know. I have an idea … I’ll build a prototype. And … I’ll build it in as self-serving a way as I can. How might I build a prototype in a self-serving manner?

First, I’ll use my company (and our partner Sunstone Circuits) to build it. Second, I’ll write about it here. Technically, you’re not really supposed to review your own stuff, but I really don’t get to order things very often. I know all about Sunstones’s PCB fab services, but I haven’t used their ValueProto service so I’m using this as an opportunity to do so. This PCB looks like it should work for their “ValueProto” service as well as with Screaming Circuits’ “SimpleProto” service. Small quantity, no leadless parts. Perfect for the simple and value services.

DPAK in gieger I actually made a small change since I last wrote about this design. The particular high-speed, high-voltage transistor in the original design isn’t available in an SMT package. I could have still left that one part in through-hole, but I didn’t want to so I didn’t. This is one of the reasons I understand the difficulties of parts substitution. I found two similar parts. One in a SOT-23 and the other in a larger DPAK. I really wanted the smaller package, but the specs of the DPAK part were closer. The DPAK is quite a bit bigger than the SOT-23, but it fits.

When I pulled up the layout to take this screen capture, I notices that the “Q1” label was slightly on the big pad for the DPAK. That’s not good. When I find a last minute error like that, I usually take that as a sign to go back and give everything another once-over. I’m going to do that tomorrow, so stay tuned.

Duane Benson
Same bat-channel. Same bat-time. (Different real-time though)

Random Via-In-Pad Myth #5

Myth #5: When you need thermal vias, more is better, bigger is better

Hmmm. Logically, this would seem to be the case. There are limits though; especially if you want a reliably assembled product. Older parts with heat slugs easily accessible for bolting on heat sinks didn’t have this issue. Just bolt on a piece of metal and maybe blow a fan across it. It’s different with a lot of the new, Padinvia smaller surface mount packages. Many have a heat slug on the bottom, which requires carefully placed thermal vias to a copper pad on the underside of the board.

An extreme case of flooding the land with vias can be seen in this illustration here on Padinvia_alt the left. In terms of assembly, you can hack this together for a prototype, but it’ll never fly in a production environment.It would be much better to use fewer smaller vias and have the center land covered with solder mask except where the metal on the chip is exposed, as in the illustration on the right.

Duane Benson
Place one carrot seed in each via and cover it with planting soil