Screaming Circuits: May 2010

Don't Do It

Friends don't let friends wire-wrap.

Quit with the wire-wrap already.
Do people still do this? Do you still do this? When's the last time you did some wire-wrap?

Digi-Key still sells an assortment of wire-wrap wire and wrapping tools, including the simple little hand tool that I own priced at - holy mackerel! - for $34.35! I should sell mine on ebay. Somebody must still be doing this if they've got such a nice selection. Amazing.

Duane Benson
Who wants spaghetti?


A lot of small quantity PCBs come individually routed these days, but when they get too small or come in larger quantities, panels can be very nice to have. When you have your PCBs panelized, what's your preferred method? And why?

We get questions about this reasonably often: "What's the best way to panelize my boards?" For our shop, we have some guidelines on how to go about it (make sure to follow the specific guidelines from whatever manufacturer is assembling your boards), but the guidelines don't specify whether you should use V-score or tab routed. That's a decision left to you.

V-score section What if you don't know? Well, it depends then, but you can easily eliminate a few options. For example, if you have curves in your board outline, you can't V-score. V-scoring only works in straight lines. With curves or odd shapes, you have to use tab-routed. If your outline is a pure rectangle, V-scoring tends to require less board-edgBoth tab and 
v-scoree so you can get a bit more out of your PCB real estate. But it's more difficult to deal with on very thin boards and V-scoring leaves a rougher edge after snapping the boards apart.

Two of the key disadvantages of tab-routing are the greater waste area and the nubs that stick out after separating the boards. You can leave the nubs, sand them down or use a clean-up router.

Here's my take on it: A) If it truly doesn't matter, use whichever method is less expensive or that you think looks prettier. B) If you have curves or other odd shapes, you'll probably need to go with tab-routed. C) If your boards are rectangles and you can deal with the less-smooth board edge, go for the V-score.

Duane Benson
Tab. Not Diet Coke.

1.0, 0.8, 0.5, 0.4, 0.3...

As pitch gets smaller, things get more difficult. Did that blinding flash of the obvious hurt your eyes? Mine too. It's not always just more difficult though. Sometimes it's different as well.

Like BGAs, for example. With 0.5mm and larger pitch, we and pretty much everyone else recommend NSMD (non solder mask define) pads. However, once the pitch drops down to 0.4mm, some manufacturers are LGA recommending solder mask define pads to prevent bridging between solder balls. Make sure your fab house can nail the mask registration.

With LGAs and QFNs, IPC recommends NSMD pads for 0.5mm pitch and larger. Once the LGA or QFN (DFNs too) pitch drops below 0.5, does everyone suggest solder mask defined pads like with BGAs? No. Actually, IPC tells you to go ahead and remove the solder mask web between your land pads with 0.4mm pitch parts. With spacing that close, the solder mask can't reliably stick to the PCB. That would be messy. As to why a solder mask defined pad would not be a good thing, I can only speculate at this point. My assumption is that without the solder ball like the BGA has, the whole land area is needed to ensure a good solder joint.

Duane Benson
Fight Uni!

Weekend Wondering - is anything really new?

Last year, Screaming Circuits started closely following the Ti OMAP processor and it's package on package (POP) form factor. The OMAP 35XX processors are very nice on their own, being a very hi-performance ARM jobby, but the package on package made it something for us to take real notice of. POP had been done before, but this seemed to be the first broad-audience application of the technique.

But is it really the first common package on package application? I've seen some unintentional package on package, like the capacitor-under-connector pictured on the right in this post. That's not so fun, but it can POP old style keep your board size down - as long as you don't want it to actually work. And then, many, many network cards used to have a chip placed right under their ROM socket. Would you call that package on package?

How about the old Ti SN754410 motor driver? It's pretty common for robot POP 754410 builders to stack a pair of them to get two amps of drive from a pair of one amp chips. That's probably more a case of actual package on package than is the network card example. Maybe the network card should be called "package under package". I know, the OMAP is a BGA, but that might actually make it easier to manufacture. With the 754410, all of those leads have to be hand soldered. The OMAP, we just put it on our machines and they do all the work for us.

The other "new, but really old" subject I'm thinking about is cloud computing. Yes, it's the newest rage in the software and application world right now, but is it really that new? Or has just the name been changed so some pundit can claim to have invented a new concept? I learned software development in a cloud computing environment - back in the early 1980's.

Duane Benson
The story you are about to hear is old; only the names have been changed to protect the egos.

PCB Edge Clearance

In my continuing saga of answering the question "what are the real limits?", I'll spend a little time giving my thoughts on edge clearance. That makes the question for the day: "how close can I put parts to the edge of a PCB?" It's a good question but, regardless of what IPC says, surprisingly difficult to nail down.

The problem is that there are a number of valid answers all governed by the phrase "it depends." I'm not going to leave it at that though. I could. But I won't. Not today anyway. It really does depend on a number of factors, but I think it can be nailed down a little tighter than that. You've got to start by considering a few things:

  1. First, our old friend IPC-7351A does specify a keep-out area (AKA Courtyard Manufacturing Zone) on the part footprint. Keep stuff out of that area. This includes the board edge. Don't have any part of the component's keep-out area hanging over the edge of the PCB.
  2. Second, look at your PCB mounting arrangement. Make sure the keep-out area does not interfere with any mounting screws, and that includes any washers you may be needing. If your PCB mounts with slots or rails, make sure the keep-out area doesn't interfere with any of the rails or slots or case edges.
  3. Check with your manufacturer about their specific line limits. Many manufacturers have specific edge clearance limits based on their assembly line.

Now, take your application and consider those three items. Whichever gives you the biggest number is the Eagle_corner edge clearance you need to follow. Based on your mounting scheme, you may have different clearance requirements on different parts of the board. Still, make sure that for each section of the board, you use the biggest number from above.

Now, in the world of prototypes, things are a little different. At Screaming Circuits, we sometimes build up boards with parts right up to, and in some cases, over the pc board edge. You just have to ask yourself two questions for your prototype:

  1. Is there enough copper to give a good electrical and mechanical connection to the PCB?
  2. Am I likely to knock the part off the board with my handling?

If the answer the #1 is yes and #2 is no, then go for it. Of course, in the prototype world, you can always accept the risk yourself and go for it whatever the answers are. That's up to you.

Duane Benson
No worries. The green patches will burn off in the reflow oven.

IPC Says What?

I recently wrote about spacing, IPC standards and such. James M. commented on the post with a question:

"I run into this issue all the time. Assembly houses love to say that they can make anything that meets "IPC Standards for Placement" but when pressed no one can provide the exact number that gives this information. I know of IPC7351A, in fact I own a copy. And the keepout is clear there. But that doesn't influence things like component placement grids, how close things can get to the edge of a board, etc.... Are there other specific IPC standards that do?"

After reading James' comment and re-reading my post, I kind of realized that there isn't a lot of actual content in that blog post. Certainly not much actionable data. And, upon further thought, I think perhaps, all of us assembly folks kind of use that phrase "can make anything that meets IPC Standards for Placement" as a bit of a cop out. Not totally - we do have to set some limits on what we can build, but yikes! I've been trying to navigate the morass of different standard numbers (Maybe I should say "plethora" instead of "morass") and I don't know how anyone that doesn't specifically live the standards for a living could easily find those kind of answers. But us manufacturers really do owe it to our customers to come as close as we can to giving definitive limits that don't take a week of reading to interpret. I think I have to do this in stages.

First, we have three things: IPC-7351A for land patterns and IPC-A-600, covering the PC board workmanship. Then, we have IPC-A-610, covering our workmanship when we build and inspect the board assemblies. Those are the key standards that we live by.

There are a few other questions, such as edge clearances and things like that. I'll dig some more into that one later, but one thing to note is that for the Screaming Circuits prototype service, we don't require any edge clearance, nor do we require panels, rails or fiducials on our full-proto service. I hope this helps.

Duane Benson

Why Those Castle Walls?

When I pick a subject, I seem to get moderately obsessed with that one topic and I'll end up covering it over and over again. I don't mean to. It just tends to happen that way. Castellated mounting systems seems to be the obsession of the day right now as evidenced by this, this and this very post.

Over on the Circuits Assembly blog, where my posts also show up, Rick posted a question about the form factor:

"I do not completely understand the purpose… would you expound on the functionality?"

It is an odd duck of a form factor and at first glance, it doesn't make a lot of sense. But upon digging into it a bit, I can see where it comes from. Generally, this form factor is used for modules, such as GPS units, RF devices and POL (point of load) power supplies.

  1. First, each of those things are not really chip-able as complete units. They are self-contained functional blocks and really need to be on a PCB.
  2. In the past, most modules of the sort had thru-hole pins for mounting, but SMT allows for less expensive manufacturing when used on a design that is all SMT
  3. The castellations, or half-vias, allow for a fillet on the outside of the module to improve the mechanical reliability of the connection.

It's really not that difficult of a form factor to use. But since it's fairly new, you're unlikely to fined the CAD library footprints to be pre-made. Make sure you read the datasheet carefully when creating the footprint. Make sure you leave pad both under and outside the part.

Duane Benson

The "real" Parts Spacing

Last week, at the Embedded Systems Conference, a gentleman came up to me and asked me about component spacing. I started rattling off things about IPC standards when we interrupted me and said: "Not IPC. I want to know the real rules - how close together can I really put my parts."

Part clearance 1 That's a difficult discussion to have. By our policy here at Screaming Circuits, we guarantee our service if the spacing and all of the other things meet IPC standards. Then we'll build it to either IPC-A-610 Class II or IPC-A-610-Class III standards. But he pointed out that whenever he asks us to do something non-standard, we pretty much always say we'll do it.

So, really... How close is close? Well, we want it to have IPC spacing. You will get a better product if you stick with IPC spacing, but yes. We often do weird things. There are limits though. In image "A", there's a little passive part under a connector that needs Part clearance 2 to be flush to the PCB. That's not gonna work. Sure, maybe for a one-off when you just need to see if the design works and if the leads on the connector are long enough, then maybe it will work and you can avoid an extra board spin. Maybe. But I wouldn't count on it for something like this.

The board in image "B" was a little luckier. The SOIC sits up high enough for the little passive to fit snugly underneath. The bigger passive is too close as well. In the face-paced world of prototyping, sometimes the "we'll make it fit" approach can get the job done, but, really. Don't go there intentionally. You want a design that will hold up to your testing and handling. You want a board that's mechanically ready to go so you can focus your testing and revision efforts on the actual circuit.

These problems are reasonably easy to avoid. Just make sure your CAD footprints have the IPC keep-out distances correct. If in doubt, get the actual part and a micrometer and measure it out. You could save a board spin and lots of dollars by doing that.

Duane Benson
Measure it with a micrometer
Mark it with chalk
And cut it with a Type III Phaser rifle

Using Package on Package (POP)

Back in August of 2009, we at Screaming Circuits assembled our first package on package chip set. We did a number of test components first to tune the process and then built up some Beagleboards. We've done a few more since then. It's not yet a high-demand item, but it is getting more popular.

Amkor POP sm At the ESC (Embedded Systems Conference) last week, we had a number of folks stop by our booth and ask about how to use the part. The OMAP processor from Ti that comes in a POP form factor is a great high-performance part, but I think a lot of designers are still intimidated by it. Really, though, there's nothing special about designing with POP.

It's a 0.4mm pitch BGA and that gives some challenges with escape routing and PCB masking, but those are standard BGA-type issues. For escape routing, go to, download the beagleboard reference manual. They have their version of the escape routing in the book. And, check out this post for some advice on the BGA footprint.

That's pretty much all there is to it. The memory chip just plops on top of it, so as a designer, you don't have to worry about that. Just do a good 0.4mm pitch BGA layout and your POP will come out just fine.

Duane Benson
Does Kellogg make Raspberry flavored Package on Package Tarts?

Circuit Design ECOsystem

Years and years ago, I was a product manager at In Focus, the projector manufacturer. It was a great time to be in the display industry. New technology was being invented left and right (and center and back, and some over in that far corner too). Competition was still reasonably light and we were ahead of most of it.

It was always interesting to take one of the early overhead projector-style displays through airport security. Laptops were rare at the time, let alone a big clear display that looked like a see-through touch-pad computer, but without the computer. But that's not the point.

Back in our engineering department, we had the electronics engineers, a few folks to work on firmware, a layout specialist, documentation specialists to deal with all the documentation (duh), purchasing people to buy the parts and PCBs, technicians build up the prototypes, manufacturing people to get the pre-production and production going. And here,s the contrast today. Quite a few engineers I talk to these days have to do all of those jobs except final production. That wouldn't be too much of a problem except that while all of those jobs were being assigned to the engineer, everything got more difficult. Parts got smaller, timelines shrank, competition got more fierce, clock speed increased and a lot of formerly company functions, got out-sourced. It's a lot of work and a lot of ground for that engineer to navigate.

A couple of companies; Digi-Key, NXP, National Instruments, Sunstone Circuits and Screaming Circuits (my company), have gotten together to form the Circuit Design ECOsystem; a cross-company organization designed to help that design engineer get a design from inside the brain to the market.

NXP makes components and is creating library components for the CAD software made by National Instruments and Sunstone. Sunstone allows quoting and ordering of Screaming Circuits assembly service on their website and Screaming Circuits does the same with Sunstone PCB fab. Digi-Key is working to improve the data-flow to Sunstone's PCB123 CAD and streamline the parts procurement process to Screaming Circuits.

It's still early in the process, but the idea is to take the, now fragmented, design to manufacture process and make it easier for the electrical engineer to get through - to remove roadblocks, add in new services and improve communications to make it easier to produce a quality product.

« April 2010 | Main | June 2010 »