Category Archives: Screaming Circuits Blog

EMS provider Screaming Circuits offers tips and thoughts on design and manufacturing

Major Major and Standard Standard

Posted on by

We ask for your bill of materials, Gerber and centroid files to assemble your PCBs. All those pieces of information are necessary to properly program our machines to place your parts. That’s pretty standard stuff, but did you know that when the Gerber format reference book was first published, Jimmy Carter was President of the United States, Russia was the “Soviet Union” and Voyager 1 was well inside the Solar System?

Use of the format has been going on even longer. Yeah. It’s been around a while. For some reason, it has been very difficult to get everyone to agree to and use a standard file format. Gerbers really don’t have enough information in them to do the job properly, but it is the standard. Hopefully not for too much longer. How many of you reading this were even born when Gerber was new?

There are a number of formats around that are better than gerber and Screaming Circuits will accept many of them. First, your CAD software probably will export an “ASCII CAD file”. This is a good format. Some export ODB++, which is one of the newer formats, again a good choice. One of the newest standards is the IPC-2581. It’s been around a few years and is now getting a lot of attention. If you happen to use Eagle CAD, you can also send us the Eagle “.brd” file.

IPC-2581 includes the best of ODB++ and GenCAM. It has all of the fab data, assembly data, netlist and BOM. Everything needed in one convenient file. My understanding of the format is that you can exclude portions of the data set that you consider proprietary. You can learn more about the format here. There’s more background information on the subject at PCD&F magazine.

Duane Benson
Where’s Henry?
I need an inductor.

http://blog.screamingcircuits.com/

QFN Custom Stencil Layer in Eagle

Posted on by

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.

http://blog.screamingcircuits.com/

Through-Hole to SMT

Posted on by

Thanks to a comment from Michael yesterday, I think everything is now cool with my Geiger counter. I had left the AT2313 default fuse setting at clock/8. That dropped the RS232 speed from 9600 to 1200 and it made the clicking sound into more of a tone, which just didn’t sound right for a Geiger counter. I still need a good radiation source though. I think I’ve picked up just a few clicks of background radiation, but that could just be wishful thinking.

Wishful thinking or not, that’s not the point. The point is that this was an example of migrating from through-hole parts to SMT. I managed to get virtually everything into SMT. The connectors, the power switch, the buzzer, batter holder and fuse clips for the tube stayed through-hole. Although I’m sure I could have all but the battery holder and fuse clips into SMT had I wanted to. I tend to keep switches and connectors that will get a lot of use as through-hole just for the extra staying power. If they aren’t used frequently, then SMT is just fine.

There are a number of things to consider when switching from thru-hole to SMT:

  1. Everything is smaller, so you can fit more in the same space or the same in less space. I took advantage of the extra board area to add in a RS232 line driver so I could connect directly to a serial port. I also added in a power-on LED.
  2. Everything is smaller so your layout is more critical. Most PCB houses will build 8mil trace and space as standard process these days. That gives you a lot of flexibility in squeezing your routing into tight areas, but it doesn’t give complete freedom. You have to be core careful because you frequently do have to route a bunch of traces into a pretty small area. When you get into the really fine pitch parts, like 0.5 or 0.4mm center to center, you have to be extra careful.
  3. Some parts are dimensioned in metric and some in SAE units. If all are one way or the other, it’s easy. But when you’ve got both, you may have to tweak with your grid spacing off and on to make sure your traces are centered in the SMT pads they connect to. It usually isn’t a horrible problem, but it can make even spacing more difficult and can make you more likely to violate a design rule.
  4. You don’t have automatic “vias” on each component leg so routing can be more difficult. You’ll likely have to spend more time tweaking the part locations and the trace routing to get a decent layout. A lot of times everything’s too close so it’s not practical to just plant a lot of vias all over.
  5. Hand soldering is less or not practical. Some people do hand solder some pretty tiny parts, but it’s not practical in more than isolated cases. If you’re a hobbyist or on a tight budget, this might limit you to through-hole or some of the largest SMT parts. For commercial work though, SMT is the way to go.

Some things to think about. But what do you get in return? Typically lower cost – especially if you want your design to go into volume manufacturing. You also get access to the newest parts that only come in SMT packages. And, many designs are space constrained, so you can cram more in while still keeping your board size down.

Duane Benson
I shot a neutrino into the air
And where it landed I already knew

 

Geiger Counter Without the Muller?

Posted on by

My Geiger counter project has been on holiday for a while. When I originally ordered all of the parts, I ordered fuse clips (to hold the tube) with solder lugs too big to fit in the holes and a trim pot (VR1) too small for the SMT pads. I moved on to other things for a while and just now got around to ordering the correct parts and soldering them in. I’ve verified that everything works except the tube. Apparently, S-13BG GM tubethe specific tube I bought (SI-3BG) is not very sensitive.

Other than not knowing if it will actually detect radiation, everything seems to work just five with Mighty Ohm’s original Atmel code. The only difference from his instructions is that the RS232 is 1200 baud instead of 9600. I’m not entirely sure why that is. The source code specifies 9600 and I have an 8MHz resonator just like his kit. I’ll worry about that later. At least it works.

It will beep if I touch my fingers to each lug of the geiger tube, but I haven’t been able to detect any naturally occurring particles. There are a couple of possibilities.

  • I had to choose a different transistor for Q1 and different diode for D1 because I couldn’t find those specific parts in SMT. Maybe the gain or some other performance spec is too far off.
  • The type of tube I bought is not sensitive enough so I just need to find a stronger radiation source.
  • I don’t have VR1 set right to give a high enough voltage to trigger the tube.

I’m going to try a 100:1 voltage divider to a unity gain current amp to measure the voltage and go on the hunt for a hotter (but still safe – I hope) radiation source. I might go back to eBay and buy a different tube too. Lastly, I’m going to get out some data sheets and look at my subs again. Maybe try to find something even closer to the original. Once I’ve verified that it all works, I’ll make the design files available as open source.

Duane Benson
If the Alpha is the tough guy, why can it be stopped by a single sheet of paper?

http://blog.screamingcircuits.com/

BGA Pads with Vias

Posted on by

Via eyeballs

No. This isn’t a closeup of an owl face.

There is still some debate on how best to create a land pattern for a 0.4mm pitch BGA. We recommend solder mask defined pads at that pitch. But that’s not really what this post is about. Although this land pattern uses non-solder mask defined pads which can encourage bridges. If you need to cross a river, encouraging bridges is good. If you’re trying to make a board work, they are not.

In the case of the two BGA pads shown, I really doubt you would have to worry about bridging. That’s because the solder ball would most likely be sucked off the BGA due to the capillary action of the via in the middle of the pad. You most likely wouldn’t get bridging. You most likely wouldn’t get any contact of any kind at all. This will not work.

Duane Benson
Hoot. Hoot.

http://blog.screamingcircuits.com/

Speaking of Small Packages …

Posted on by

T’was a a dark and stormy night when the news came through. Joe Layout had been both dreading and preparing for years. But it had always been little more than rhumors from a far off land. It was a looming threat, always dancing in the distance, but never quite real.

Until now. 1.27mm, 1.0mm, 0.8mm, 0.5mm, 0.4mm … and now … drum roll please 0.3mm pitch.

I just got an email announcing an Amkor 8 x 8mm 368 ball BGA at 0.3mm pitch. Yikes.

There’s still some controversy over the best way to make a 0.4mm pitch BGA land pattern. Some say says you need to use solder mask defined pads. Some say you still need to use the non-solder mask defined pads. Now we throw something 25% smaller into the mix. The image isn’t to exact actual scale — because I don’t know how big your monitor is — but the parts are in relative scale from 1.27 pitch to 0.3 pitch.

Duane Benson

If you can’t see it, you shouldn’t eat it

http://blog.screamingcircuits.com

Open Source: What Is It Good For?

Posted on by

I’ve written about open source hardware (OSHW) a few times before. Like this and this. I’ve understood open source software for quite some time and over the past few years have been starting to get what open source hardware is all about. It is different than open source software.

With software, your tangible product is essentially intangible. Your acquisition and distribution of an open source project can be virtually free. Not so with hardware. Someone has to physically build something, which costs time and money in parts and labor. Really though, all that means is the proliferation of an open source hardware product just takes a little longer. If you look at it as the design being open source more than the actual product, then it gets to be more and more similar to software.

While open source software has moved into real business, hardware is still more closely associated with the hobbyist community. That is changing though. TI’s Beagleboard is serious stuff from a serious company. Some of the hobbyist catering OSHW companies are growing to or have grown to the point of being serious businesses (Adafruit, Sparkfun).

This all begs the question: “What is open source hardware good for?” Let’s divide and conquer. Or, at least, divide and explain.

  • What does it do for innovation?

History is rife with stories of great inventions that were not commercially successful because the inventor was a good inventor but was a lousy business person, didn’t have access to funding or just didn’t have the drive to build, promote and sell the product.

With OSHW, companies that do have the drive, funding and know-how can pick up an open source project from a developer that doesn’t.  There are none of the IP concerns that sometimes keep big companies from taking on product from independent inventors. Great products that otherwise would stay hidden can make it out in the world.

Some OSHW companies, like Adafruit, compensate the designers whose product they sell. No marketing or selling expense for the designer and yet money comes in to them. Much reduced design expense for the seller, yet they can build a business.

  • What does it do for small companies?

It’s another way to jump-start design or production of products that will fund the small business. It can reduce the barriers to entry. People who are good at designing but not so good at selling can still earn money. People who are not so good at designing but good at selling can earn a living. People who are good at both designing and selling — they have the best of both worlds and can earn a living. Products that would otherwise stay in obscurity can more easily make it to the world.

  • What can it do for big companies?

The answer to this question has been the longest in coming, but there are more and more answers showing up. Take the Beagleboard from Texas Instruments. It got a new processor (the OMAP) out into the hands of their customers quickly. It was a great promotional tool. The software side of an organization could get started with the processor without having to wait for the hardware folks to design, layout and build the hardware. The hardware folks could see how the part and its accessories work in real life.

OMAP users could get a jump-start on complex tasks like escape routing. The manufacturing folks could get some insight and practice into assembling the package on package processor / memory combination. design cycles are short enough as it is. Companies that want to use the TI processor get professionally designed short-cuts. TI gets to sell more processors quicker. Everyone wins.

Duane Benson
It doesn’t mean destruction

http://blog.screamingcircuits.com/

Reference Designators

Posted on by

A while back, I wrote about reference designators relative to family panels. Family panels can cause problems because often times, each individual circuit layout will have reference designators that start at the same place.

For example, circuit A, down in the lower right corner of the panel, will have resistors R1, R2, R3… Looking at the other three circuits on this hypothetical family panel, all of them also start their reference designators with R1, R2, R3… That’s bad. It can lead to confusion and wrong parts on the board in the wrong spots. If we see this here at Screaming Circuits, we may spend some extra time and sort through it manually or we may ask you to fix it first. Fixing it here is a labor intensive and risky process. It’s bad news.

Anyway, to the point of this post: In the original post, I listed one wrong way and three right ways. There are two other wrong ways not in the original post, which I’ll list here.

Wrong way number one: R1-1, R1-2, R1-3. Bad. Most assembly software will interpret a dash as meaning a range. It will see “R1-3” as equalling “R1, R2, R3”. That can be bad.

Wrong way number two: Leading zeros. Don’t do “R1, R01, R001”. The leading zeros are stripped and that can cause all of those the be seen as “R1”. Just don’t put leading zeros in your reference designators.

Duane Benson
Corrigan says Long Beach is actually in Ireland

ESC Robot Attendees

Posted on by

We’ve been hanging out in Boston at the ESC show. Yesterday was a busy day with lot’s to look at and lot’s of folks at our booth. Some human, some not. The Freescale people, just a few booths down, brought along a tele-presence robot.

The little guy wandered here and there a bit. It’s a cool concept, but I think it didn’t quite live up to it’s potential. They really should have registered it as an attendee. It did have a little card hanging around it’s neck, but I don’t see why it couldn’t have had a genuine show badge. If it were me, I would have had it actually stopping by booths, talking to people and collecting goodies.

Still, it was a fun demo and, presumably, an example of Freescale chips in action.

Duane Benson
I salute our new robot tradeshow overlords

http://blog.screamingcircuits.com/

Small Open Vias

Posted on by

Pad parts change and so do vias. Our standard policy here is that open vias in pads are bad. We from time to time recommend ways to plug them. Generally, you have several options. Like this post shows. However, with vias in the pads of really small parts, those solder mask options will probably not work. Solder mask generally isn’t put down with enough precision to cover holes on tiny pads, and further, the solder mask would probably mess with the clearance. On the left is an example of a small QFP with open vias in the pads. Those are some small vias.

So, if solder mask isn’t going to work, what QFN center void open vias will? Filling and plating over them. That’s what will work. You really only have two choices: fill and plate, or live with a bunch of voids under the part and solder slopped on the bottom side of the PCB.

Here on the right are two illustrations representing the issue. In the top half of the image on the right, I’m representing the vias with copper plugs and plated over at the board fab house. As with all parts of this sort, there may still be tiny voids. IPC and the manufacturer will have guidelines on the maximum allowable voiding. On the bottom, you see what happens with the vias left open. You get two problems: big voids and solder on the underside of the PCB.

Certainly there are some applications where this doesn’t matter. That’s why there is a second choice: “Live with a bunch of voids and slopped solder.” If you can’t live with voids and solder slop, you have to bite the bullet and pay the extra for a PCB with filled vias. Board houses that do this have a variety of materials to use including copper, electrically conductive epoxy and thermal conductive epoxy.

Duane Benson
Please sir, may I have some more voids?
No! No voids for you!

http://blog.screamingcircuits.com/