Screaming Circuits: Layout


Freescale KL03 and PCB123 at 0.4mm pitch

Small component packages seem to be a recurring theme with me. It's understandable, I guess. Super tiny packages are becoming more and more common and we build a lot of product with them.

The smallest we've built is 0.3mm pitch. Those aren't common enough to be considered standard - they're still an experimental assembly - but not many chips use them yet. 0.4mm, on the other hand, is something we see on a pretty regular basis.

What's so tough about that?

The biggest challenge with these form-factors seems to be footprint design and escape routing. I can see why. There really isn't room to follow any of the standard BGA practices. You can't fit escape vias between the pads and you can't put vias in the pads, unless they are filled and plated over at the board house. Filled and plated vias are the easiest way to go, but it can make for an expensive board fab.

KL03 WLCSP20 on a US Lincoln Penny

1-DSC_0008One of my side-projects involves trying to make the smallest possible motor driver. For this project, I've chosen the Allegro A3903 driver. It's a 3mm X 3mm DFN (dual flatpack no leads) with 0.5mm pitch pads and a thermal pad in the middle. The microcontroller will be the new Freescale KL03 32-bit ARM in a 1.6mm X 2.0mm WLCSP (wafer level chip scale) package. It also comes in a 3mm X 3mm 0.5mm pitch 16 pin QFN. Without an expensive PCB, that may be my only option.

Pick your CAD package

I'm using the newest version (5.1) of Sunstone Circuit's CAD package, PCB123, but the principles here will apply to any CAD software. If you don't already have a copy, download PCB123 V5.1 here.

If you've got fast Internet, you're done now, so go ahead and install it. You'll need the manual too, which you can get here.

I need to eat now, so stay tuned for Part 2.

Duane Benson
Nerfvana - It's like Nerdvana, but with more foam darts.

VTP - Very Tiny Parts

FreescaleKL03A while back, I wrote about a new ARM Cortex M0+ chip from Freescale. It's not the first M0+, but I do believe that it's the smallest. I've been checking stock off and on and finally found the smallest package to be in stock and available to ship.

I actually bought a couple of different types. First, there's the WLCSP 20. It's got 32K FLASH, 2K SRAM and an 8K bootloader. The real kicker is that the package is only 1.6mm X 2.0 mm. I also got a few in the QFM 16 package, which is a bit more workable at 3mm X 3mm.

Finally, I bought a Freedom development board with th 4mm X 4mm QFN 24 package. The dev board is hardware compatible with Arduino shields, so that will make for some interesting possibilities.

Anyway, here at Screaming Circuits, I'm most interested in that 1.6mm X 2.0mm package to see how easy (or difficult) it is to use - see if there are any particular layout challenges. The other stuff is just for after hours play time.

Duane Benson
I'm not a number. I'm a free development board!
(Free, as in named "Free...", not free as in "don't cost nothin")

Cost Reduction in Design - More Advice for Makers

If you're looking for the absolute, cheapest possible assembly service, you'll need to look outside of North America. If you really need a decent price with good quality and good service, you can keep your gaze West of the Atlantic and East of the Pacific.

Like everything else in the modern world, design decisions can have a pretty big impact on your cost. So, lets take a look at some design decisions that can make your manufacturing more affordable.

  • Accept longer lead times

Lead times are one of the biggest factors in electronics manufacturing. Screaming Circuits can turn a kitted assembly job overnight, but it costs a lot of money to do that. Screaming Circuits also has a 20 day turn-around that is much, much more affordable. Accepting longer lead times on PCB fab will drop your cost as well.

  • Avoid leadless packages like QFNs and BGAs

We build tons of QFN and BGA boards - even down to 0.3 mm pitch micro BGAs. That's great if you need those packages. However, since all of the leads are underneath, we have to x-ray every part. That adds a bit of cost to the process. If you can, stick with TSSOPs and other parts with visible leads.

  • Use reels, or 12" or longer continuous strips

Tab routed multi panel 1024We will gladly assemble parts on strips of almost any size. But, to save costs, use full or partial reels or continuous strips of at least 12" long. It costs us less time to work with reels and continuous strips, and we pass those saving on.

  • Stick with surface mount

These days, thru-hole components tend to be hand soldered. That costs more than machine assembly, so use surface mount wherever possible. Surface mount components tend to be less expensive than thru-hole too. If you do need a few thru-hole parts, this is an opportunity to put in a little sweat equity by soldering the thru-hole yourself and save a bit of money.

  • Panelize small boards

We can work with really tiny boards individually, but sticking with a larger size makes the job easier, and, again, we'll pass those saving on. If your PC board is smaller than 16 square inches, panelize it. We put in less labor and you get a price break.

By sticking with Screaming Circuits, you get the same care and quality that we give to boards going up into space, down into the ocean, and everywhere in between. By sticking with Screaming Circuits, you get a known turn-time; not an "about ..."

By following these guidelines, you get a decent price and really good quality and service.

Duane Benson
That would be telling

Choose Your Package Wisely

As I mentioned in my prior blog, there are reasons to consider different packages than just physical size.

Sometimes it is just space available on the PC board, but there may be other considerations as well. One of the first to consider with really small size packages, is the capability of your manufacturer. Not all assembly service providers can TI ESD CSP 007 croppeddeal with super-duper small parts.

That's a paperclip next to the little ESD protection chip in the photo on the left. At Screaming Circuits, we can go down to 0201 passive parts and 0.4 mm pitch BGAs. We've even done a few 0.3mm pitch BGAs, but those are pretty rare still.

Some manufacturers stop at 0402 (or even 0603) parts. If that's the case with your manufacturer, then you'll need to eliminate sizes smaller than their limit (or find someone else to build the board).

Cost might also come into play. It probably won't be enough of a factor to worry about during prototyping, but it may be worth looking at for volume production. Sometimes the smaller form-factors add cost. Sometimes the part value you need may not even be available in the smallest packages.

TI TPS62601 front and backIf both cost and size are significant drivers, weigh the cost savings from reducing the PC board area against any additional cost with smaller packages.

Noise can factor into your package choice too - especially regarding bypass capacitors on high speed chips. You want your bypass capacitors as close to the power and ground pins as possible. The higher the speeds, the more important this is. Dropping your package size down to 0402 or 0201 can make it easier to put the caps closer to power and ground pins.

Duane Benson
You don't need to ask Alice because your parts aren't ten feet tall

QFN? QFP? QFwhat?

The QFN (quad flat pack, no leads) has become my favorite integrated circuit package. It's very compact, yet is easier to use than a micro BGA.

Micro BGAs of 0.5mm and smaller pitch become a bit more difficult and costly with more than two rows of pins. At those geometries, escape routing can involve plugged and plated vias which adds complexity and cost to the board fab. QFNs can be almost as small, but have all of the pins exposed around the edges - so, no need for escape routing.

One thing that's important to note, is that despite sharing the first two letters (Q and F), the QFP and QFN footprints are not interchangeable. We do, from time to time, see boards laid out for one along with the other form packaged part. Arduino w QFN and QFP

Take a look at this PCB layout clip from the Arduino Leonardo. It has both footprints on the board. You can see how much bigger the QFP package is.

They put down both footprints because the Atmega32U4 chip used in the Leonardo sometimes has supply issues in one package or the other. This gives them the flexibility to use either without making any changes on the board.

You might consider this as an option if you have the space for a QFP and are concerned about the available of one package variant or the other.

If you do, there are some very important things to check out:

  • Make sure the pin-outs match. Some parts will vary the pin-out a bit between packages or have extra pins on one or the other.
  • Make sure the extra space won't cause noise problems. Generally, you want bypass caps as close as possible to the supply pins. This amount of extra space probably won't be a problem when using a QFN, but in some designs, it might.
  • Make sure the board won't be in an environment where unsoldered pads will be a problem. Some harsh environments could attack the unsoldered pads. If that's the case, consider conformal coat.

Duane Benson
We're always being pushed and shoved by people trying to beat the clock
But we like it - it's what we do

Warped PC boards

So... You just got a nice big PC board back from the fab shop. You set one on your desk to admire only to discover that it's warped. What do you do?

There are two primary types of causes of board warping: process related at the fab or assembly shop, and layout related issues. If it's warped before assembly, it's between fab and layout. If it's flat before assembly and warped, after, it's most likely between layout and assembly - although, sometimes a fab problem won't show up until a pass through the reflow oven at your assembly partner.

Determining the root cause is generally a bit of an iterative process. It's tempting to start right off with your fab or assembly partner, but you need some information before giving them a call. You'll need such things as the amount of warpage per inch, board size, and thickness. With that, you need to take a good look at your design and consider copper pours, component size, and component placement.

With that information in hand, you can make your phone call. If the board is warped before assembly, call your fab shop. If it's flat pre-assembly and warped post assembly, call your assembly house.

The shop you call will want to talk over your design to help you pinpoint the cause. If you can rule out a design issue,then you need to talk with your partner to determine whether it's a fab or assembly issue and next steps to take care of you.

 Here are a few design issues that could contribute to warping:

  • Uneven copper pour. Copper and FR4 are a good match relative to thermal expansion, but they aren't exact. A large pour on one side or corner of your board can lead to warping due to dissimilar expansion characteristics. This could cause warpage either at the fab shop or the assembly house.
  • Components with large thermal mass grouped together on the board. This would be more likely to cause problems during assembly than during fab. The thermal mass will act as a heat sink for that area on the board, which can lead to uneven expansion and uneven soldering.
  • A board that's too thin for the size or number of components could lead to warping at any stage.
  • Odd shapes or large cut-outs could also lead to warping at any point.

There may be other, more obscure causes, but those are the main design related causes. If it's none of those, talk with your partner.

Occasionally, design requirements lead to a board that is essentially non-manufacturable. Hopefully, you never have this situation, but if you do, make sure that thickness, component location, pours, or cut outs really, really, really, need to be the way they are.

If you absolutely, positively can't change anything, go back and try again. Then you can to look for heroic means to get the board fabbed and built.

Slight warpage might go away when the board is mounted. Just be careful with that. Some components may not stay securely soldered when you flatten it.

The board may need a special fixture during assembly to prevent warping. This will likely cost extra, but if you can't change your design, and still need it built, it may be your best option.

Finally, if nothing works, you may need to look harder at the design, or look for a new fab or assembly house. We all like to think we can do just about anything, but every shop has its limits, and on rare occasion those limits can be difficult to spot.

Duane Benson
What if Godot was late because he was waiting for John Galt?

0.4mm Pitch BGA is Awesome

I recently had a conversation with a friend about 0.4mm pitch BGAs. The specific part is the Freescale FreescaleKL03KL03 ARM Coretex-M0+ microcontroller in a 1.6mm x 2mm, 04.mm pitch package. That's a 20 ball wafer scale BGA form factor.

I don't have an actual part to photograph next to a grain of sand, but trust me (or don't), it's really small.

Ti 0.44 pitch dimensionsThe challenge, and the reason I suggested a QFN form factor instead, is the costs
involved. If you have the extra budget money for more expensive PC boards, then go ahead and use this form factor. You probably won't be able to use this package in cost constrained situations.

The simple reason is that you can't escape route the inner six pins without using super small vias between pads, or in pads and filled and plated over. The page on the left is from a Ti doc, but any variations in geometry will be minor.

You can see that you can't put a trace between the pads. Maybe a 2 mil trace, but maybe not. There just isn't much room. The recommended method is to put micro vias in the pads and have them filled and plated over at the board fab house. Never put a via in a micro BGA pad unless it's filled, plated over, and flat.

Duane Benson
There are more things in heaven and earth, Horatio, 
Than are dreamt of in your philosophy.
But open vias in pads aren't one of them

VLV - Very Large Vias

I recently received a question over on Twitter. Tomaž Šolc, AKA avian2 asked:

"@pcbassembly What is your opinion on the "one big plated drill in QFN ground pad" pattern? pic.twitter.com/M9ZLftpuo0"

From Avian2 Ban_N62IEAAm3acI answered back: "Bad for machine assembly, okay for hand assembly." That's definitely true, but it's worthy of a bit more explanation. Here's the photo avian2 included along with the question. We're looking at the side opposite of where the QFNs are mounted. The two big openings in the square gold pads are the big vias (plated drill).

This is often done when hand soldering QFNs. Somehow you get the little pins on the outside edge of the QFN soldered down. Then you turn the board over and poke your soldering iron into the big opening to solder the pad down.

Generally, there wouldn't be any reason to do this with machine assembly, as we do here in our plant. You put a number of small vias, cap them, and segment the solder paste layer (refer to this post and this post). Thus, we would never recommend using big vias like this for machine assembly.

However, I can envision some situations that might call for this. First, there's the hand solder method I mentioned above. Next, there may be some very specific need to expose a lot of the pad to open air for cooling. In general, this is not the best way to get cooling, but maybe in some special case. Third, perhaps you need access to the pad as a test point and don't have enough room to get access any other way.

You wouldn't do any of those three things in a production environment, but in a prototype world, sometimes things happen differently.

Duane Benson
Hurray! Only one day until Mitten Tree Day!

Pads on Ground Plane

Pour-no thermalGenerally, small pads for passive parts are connected  with a single PCB trace of equal size to each pad. That's the right way to do it.

However, sometimes, circumstances dictate a little different approach. The illustration on the upper right here shows something of a worst-case. This is for a snubber (resistor, capacitor pair) between two power planes.

A couple of things will likely happen. The power plane will act as a heat sink, preventing the solder paste on one side from melting, resulting in a poor connection. Or, the unequal melting could lead to surface tension pulling the part up, causing tombstoning.

Pour-with thermalMost designers are aware of that, but sometimes, thermals will be deliberately turned off to allow for better current capacity to and from the large power Mosfets (not shown). If that's the case, make sure that you can turn the thermals (see image on loer right) on or off by the part, rather than just by the plane.

Duane Benson
The rain falls mostly on the ground plane due to static attraction

Super Small Via In Pad

Via in pad is an old issue that still pops up now and then. Our standard answer hasn't changed: No open vias in pads. But one of the questions we get related to the subject is: "What if we make the vias really small?"

Beagleboard U6 viasLogically, that makes sense. In fact, in some cases, the via is so small that it's essentially closed. If it's so small that it really is closed, then it's not an open via. But look close - if it's closed with solder, that solder may melt during reflow leading to an open via.

The images here show some pretty small vias. I believe they're 0.3 mm in diameter.

Beagleboard vias back sideIn the first picture, on the left, it appears that the vias are open. They aren't though. This board (an unstuffed Beagleboard) uses soldermask on the back side of the PCB to close off the vias, as shown in the image on the right.

Our recommended method (se more detail here and here) is to plug the via with copper or epoxy and have it plated over at the board fab house. Next, we'd recommend via caps on the component side. FInally, capping the back side with soldermask, like this example can work, but it comes with the risk of voids. The via caps and also pop open, leading to an open via.

Duane Benson
No more open vias-in-pad, I mean it!
Anybody want a peanit?