Design Costs, But How Much?

Lots of studies find that most of the cost (75%-plus) of an electronics product is determined in the design phase.

Here’s my question: Do those studies take the BoM costs into account as part of the design cost? Or are all the “costs” in this sense value-added (design, fab, assembly, rework, etc.)?

Design Costs, But How Much?

Lots of studies find that most of the cost (75%-plus) of an electronics product is determined in the design phase.

Here’s my question: Do those studies take the BoM costs into account as part of the design cost? Or are all the “costs” in this sense value-added (design, fab, assembly, rework, etc.)?

QFN Custom Stencil Layer in Eagle

It’s been said over and over that you don’t want to leave the solder paste opening wide open for a QFN center pad. A 50 to 75% paste coverage will get the best results. With full coverage, your QFN can end up floating too high and not connecting with all of the pads due to their significantly smaller aperture.
But how do you create a custom paste layer? In Eagle, it’s not terribly obvious, but it is easy. Open the part that you want to customize in the Eagle Library editor. Open up the package for that component. Now, select “i” on the left side and click on the center pad. You might need to turn off the “tcream” layer in order to select the pad.

In the Properties dialog box, uncheck the check box for cream. That will get rid of the standard stencil layer. Now you can use the rectangle tool to add in stencil cut-outs as you want the. Make sure you set the layer for the rectangle to be “tcream” and remember that you are drawing the cut-outs of the stencil, not the blocked part.

Obviously it will be different for every CAD package, but the concept is the same. As is the need to do so.

Duane Benson
The Internet is weird.
There’s actually a website for paste eaters.

Done and Done

LED scroller 005 trimmed And … (drum roll, please) … it works.

I put in a couple of batteries, programmed the MCU and turned it on. It works.

I’m always surprised when something I design works on the first try. This being such a simple design, I probably shouldn’t be surprised, though. I should at least give myself a little more credit.

The unpopulated lands on the board in the photo are supposed to be unpopulated. I left a few things out because they aren’t needed for what I’m doing with this piece now and leaving them off keeps the cost down.

So, what did I learn from the process?

  • If you have a lot of different parts laying around, it’s pretty easy to grab the wrong one.
  • I ran into some variability in the “zero rotation” position in the CAD library land patterns.
  • The whole process is pretty easy, but start to finish, there are quiet a few steps.
  • It’s a nerve-racking wait after sending off a box of parts.
  • Good communications between designer and assembler are very important.
  • Clear documentation from the designer is very important.
  • This was a WHOLE LOT easier than hand soldering all the SMT parts (I’ve done that before).

That’s a good set of educational results. Next time, I think it will be easier.

Note the large diode polarity indicators on either side of the long row of LEDs and by LED D25. D1, the Schottky on the upper right has the same polarity indicator, but it’s in between the pads, under the part. In case you’re interested, I have a 3V supply. The LEDs drop 1.8V and I’ve got a 150 ohm resistor for each. That gives me a theoretical 8 mA per port for a total maximum of 176 mA with all 22 lit up. That’s within spec in the -40C to 85C temperature range but too much when above 85C. I’m not sticking this in an engine compartment or anything, so no worries there.

Duane Benson
0x45 0x53 0x43 0x20 0x62 0x6F 0x6F 0x74
0x68 0x20 0x38 0x32 0x33 0x20 0x20 0x20

Time to Co-Opt Co-Design?

In recent conversations, I’m hearing designers say they are spending enormous amounts of time in meetings. These comments tend to come from folks who work for larger OEMs or ODMs and work on teams spread around the globe.

Certainly there is something appealing to upper management about follow-the-sun design. It maximizes time resources and leverages both the lower labor cost regions of the Pacific Rim and the experienced hands in the West.

But whereas the old model of vertically oriented design and manufacturing had its warts, if designers are getting hung up all day in meetings, as opposed to spending time routing boards, one begins to wonder whether the follow-the-sun model has been taken too far. There’s nothing process-oriented about commiserating with a manufacturing engineer over lunch in the cafeteria, but there is something to be said for being able to talk things over as they occur, rather than being holed up in never-ending CYA sessions.

Is round-the-clock design actually a drag on efficiency and productivity?

Virtual Show, Real Value

We are eagerly looking forward to next week’s Virtual PCB trade show. It’s the fourth year we’ve produced the Web-based event, and we’ve learned a few things along the way.

1. Although the attendees are online, they usually act as if they are in the flesh. There’s plenty of the “how are you doing,” “great to see you,” and gentle ribbing that takes place when we run into each other at PCB West, SMTAI, Apex or one of the other “bricks and mortar” shows. It’s social. (Perhaps that’s why they call it “social” media.)
2. People are polite to the point of near invisibility. Just like a physical show, some attendees do lots of talking, while others never utter a peep. That’s OK. Lurk away. Everyone learns in their own way.
3. Speaking of learning, it’s almost impossible to attend Virtual PCB and not take away something. Nearly 3,000 people registered last year! These are your peers across the entire electronics manufacturing spectrum, from design to assembly to test. The same experts you might see at a physical show — folks like signal integrity expert Dr. Eric Bogatin or reliability guru Werner Engelmaier, will be there, holding court and sharing their wisdom. More than that, it’s a chance to meet folks from all over the world. These are potential future colleagues and employers. Insofar as networking is concerned, it’s tough to beat.

We hope you take a moment to register (it’s free!) at and log-on to Virtual PCB, March 8-9. It’s a fresh way to stay up on our industry — without ever leaving your desk.

Narcissistic Parts

Maybe not completely narcissistic, but at least self-centered.  Or, self-centering. Okay, are you lost now? Am I making any sense at  all? Well, I’m going to say that it doesn’t matter, because the  world-revolves around me.

But what I am talking about is parts that will more or less center themselves during the reflow process. Some parts like BGAs and QFNs tend  to follow the surface tension of the melted solder and tweak themselves  into a more centered position on the land. That’s a good thing.

It’s not always a good thing though. Sometimes that same surface  tension action can work against you. Take this TO-263 (pictured). When it was placed on the land, before reflow, the leads were centered  right in their pads like they should be. The big land for the thermal  pad is set up a little too high though and once melted, the surface tension from the big thermal pad sucked the part up, nearly dragging it  off of the lands for the leads. Bummer days. Here’s another example.

You probably shouldn’t leave the part like this, so here’s a few suggestions.

You could make the thermal pad smaller so that when the metal tab of  the part is centered, the leads will be, too. Cooling needs might dictate  that you don’t reduce the size of the pad, though. If that’s the case,  you could make the bad bigger by extending it down toward the leads, again so the leads will be centered when the body of the part is. You could also mask off the top part of the pad, or put a thin strip of mask  as a solder dam. What you’re doing is making sure that if and when surface tension moves the part, the leads will end up where they are supposed to.

<em>Duane Benson
It is all about me, you know</em>

DesignCon Wrapup

Overall I was impressed with DesignCon. I still see it primarily as a show for chip and IP designers, but I could agree with anyone who says it is a show for electrical engineers involved in the entire design process. To me, the conference was lacking in practical design topics in the PCB space. Granted, there were only about eight sessions in the “PCB Summit” portion of the program, but aside from the Lee Ritchey, Eric Bogatin and Bruce Archambeault sessions it was pretty thin.

I do have to mention a panel I attended called Science Fiction…Is it Really Fiction? The panel was chaired by Gabe Moretti and included Eric Bogatin, Gentry Lee and Charles Pfeil. It may please you to know that, according to Gentry Lee, by the year 2200 we will have unlocked the keys to immortality and be able to choose how long we live. Something tells me I won’t live long enough to see it.

Exhibitors in the PCB space included laminate suppliers, a couple of board shops as well as EDA companies. The thing that impressed me most was the traffic. The show was busy every time I was on the floor. In fact I did not get to see some people I wanted to see because they were always busy with potential customers. Some may think that exhibitors are always glad to see the press, but customers – even potential customers – always come first.

Even though the PCB portion of the show was very small compared to the total conference and exhibition, overall I give DesignCon a thumbs up, with one exception. Somebody needs to talk to whomever put the show directory together. I thought at first it might just be me, but talking to other people the consensus was that this was the most difficult to use directory we’d ever seen. If there was a rhyme or reason to the way the directory was organized I couldn’t find it.

The best thing about DesignCon for me this year was reconnecting with people I haven’t seen in years. Some were people from the publishing world and others were from the EDA industry. It was good to see and talk to them all.

All in all, DesignCon was worth the trip.