MyDATA smt machines

Last week, several of us from Screaming Circuits visited the APEX show in Los Angelas. We visited our friends from Sunstone (PCBexpress). They had a pretty nice booth setup there and were demoing an upcoming version of their PCB123 CAD software. Pretty impressive. You should check it out.

We also stopped by the MyData booth. We use their pick and place machines in our shop. Here at Mydata_01005_business_cardScreaming Circuits, we regularly work with components down to 0201, but the machines are terroretically capable of going down to 01005 passive parts.

I gave them my business card and they wrote "MYDATA" in 01005 resistors with a My12 - one of our machines - and then laminated it. Check out the photo. The top of the "T" is about 2 milimeters across and is made up of five parts.

Duane Benson
It might be a small world after all

Embedded Systems Conference

Mark your calendar for April 3, 4 and 5.

We're going to the Embedded Systems Conference in San Jose and we'll be in Booth 3037 on April 3-5.Esc_sm_logo  Come see us. It'll be fun. You can get a get a free Screaming Circuits pen or a strongly motivationalist poster. Maybe even something better.

Follow this link to get a free admission to the exhibits. When you register, be sure and enter our priority code "X164."

Duane Benson

Tombstoning

February has been very lean for blog posts. In the first few months, I was averaging about one every other day. I've put down just four in February, though - five, if you count this one.

While down at the Apex show in Los Angeles last week, I heard mention of tombstoning. Tombstoning small SMT parts isn't something we see a lot of, but it does happen. If you're not familiar with the problem, essentially, it means that the part has stood up on one end, like a tombstone. One end of the part will be soldered and the other is up in the air, not soldered.

I understand that tombstoning also refers to the practice of jumping or diving off of high, shear cliffs into not necessarily well known waters. That's not what I'm talking about and I certainly wouldn't recommend it. It's probably a lot worse than the tombstoning I'm discussing here.

It is much more likely to occur on smaller parts, such as 0402s and 0201s than it is with larger parts. There are a couple of causes, some process related and some design related. The process related causes are mostly up to us, so I'll just cover a couple of the design related causes.

Essentially, what happens is that the solder on one of the two pads melts and captures the part end before the other side melts. Surface tension from the melted side can then pull on the part and stand it up.

The keys to reducing the chances of tombstoning then are to make sure the board surface is level and make sure that both pads on each side of the part heat evenly. HASL (Hot Air Solder Leveling) boards often have uneven surfaces. If the pad on one side is higher because of HASL solder deposits, tombstoning is more likely to occur. This is one of many reasons that immersion gold and immersion silver surfaces are becoming more popular.

An even bigger factor is the amount of copper. If one pad is larger than the other or if one pad is connected to larger traces than the other, the extra copper will act like a heat sink, slow the melting on that side and may cause tombstoning.

Tombstoning_2_1 In the simulated case of tombstoning image on the left, the adjacent via sinked heat away so the solder on the other pad melted first, popping the part up on that side.

If you are using 0402 or 0201 size parts or if your designs have had tombstoning problems, first we would recommend that you avoid HASL finish boards. Second, look carefully at the layout of your small parts. Make sure that the traces connected to each side are equal in width and connect the pad at a similar spot; ideally centered on the X-axis.

Sometimes the problem can be inside the board as well. If an inner layer has a large area of copper the extends under one pad of the part and not the other, that area may heat a little slower than the other side without inner layer copper.Tombstoning_1_1

In the second simulated case on the right, here, one of the  solder pads had too much solder paste. The extra thermal mass, again, caused that side to melt a little slower.

Following these guidelines won't guarantee to eliminate tombstoning, but it will reduce the chances of it happening. We'll do our part with the process related causes and hopefully, you'll never see a tombstoned part.

Duane Benson
No, don't jump!

Other things do happen

Looking back, February has been a slow month for posts. It's been a busy month for everything else and I'm sure the two are related. I put out three posts and all of them were related to via in pad. That is a common challenge, but we do still see other difficulties too.

One of the most common little issues we run across is undocumented parts substitiutions. We'll receive a a parts kit and a Bill of Materials (BOM) and find mismatches. In most cases, there are a few common causes:

  • A standard part is called for on the BOM, but a lead-free part is sent.
  • Capacitors are sent with different voltage ratings than specified in the BOM or schematic
  • Resistors are sent with a different value than on the BOM or schematic

In most cases, none of those will cause assembly problems. However, any of the three issues may cause delays because we need to verify that the substitution is okay. The best way to avoid needless delays is to make sure that your BOM is current to the kit you send.

Duane Benson

Filling VIA in pad - Assembled

[Editor's note: This post is the third and final of a three-part post about a rather difficult and undersireable via in pad PCB. This board violates industry-wide recommended practices. The right thing to do in a case like this is to redesign the board without vias in the pads or have them plugged at the board house. We did get this one to work, but it wasn't a sure thing. Some designs require vias in the pads. If you have one of those, get the vias plugged and plated over at the board house. If you don't, you can ask us to try what we did here. It may work. It may not. There are no guarantees in a situation like this and in may incur additional labor charges. Call first for situations like this.]

It's long back to the customer, so it's probably about time I finish this series. If you didn't read the prior posts, we received the board with large vias in some of the BGA land pads, we filled the vias, and now, we've assembled the board.

9x13_via_in_pad_bga_land_assembledThe top view shows the Bluetooth radio module BGA happily secure on the board and the inset shows the underside of the board with all of the solder still in the via holes where it should be. The vias were plugged with high-temperature solder prior to assembly. Once plugged, the the board was assembled just as it would have been if did not have vias in the pads in the first place.

Again, please don't ever do this. We were lucky in this case. If you absolutely must put vias in pads, read all of our other via in pad posts for recommendations on how to properly handle them.

Duane Benson

Via in pad methods

Here is another method for dealing with QFN heat slugs with thermal vias. While the best option is to plug the via holes with metal early in the board fab process and then plate over with copper, there are a number of alternatives. We've discussed several of them in previous posts.

Qfn_w_soldermasked_vias Here we have another option. This is for a 9mm X 9mm QFN on a board using an ENIG (Electroless Nickle Immersion Gold) finish. The larger size of this part gives a little more flexibility than some of the real small parts. In this case, while the vias are left open, the area around the vias is masked to keep solder away.

This method can work well on larger QFN and QFP parts, but it is absolutely critical that your solder paste stencil be segmented and not put any solder over the masked parts. (see this post). If the stencil is left fully open or deposits solder paste over the vias, that solder may go down and mess up the bottom side of the board. With a well designed stencil, this may be an easy and reliable method for dealing with your thermal vias.

Duane Benson

Filling VIA in pad - After

[Editor's note: This post is the second of a three-part post about a rather difficult and undersireable via in pad PCB. This board violates industry-wide recommended practices. The right thing to do in a case like this is to redesign the board without vias in the pads or have them plugged at the board house. We did get this one to work, but it wasn't a sure thing. Some designs require vias in the pads. If you have one of those, get the vias plugged and plated over at the board house. If you can't do that, you can ask us to try what we did here. It may work. It may not. There are no guarantees in a situation like this and in may incur additional labor charges. Call first for situations like this.]

Here we are again with our board that has giant via holes in the middle of several BGA land pads. (If you haven't yet, read the original post here)

Take a look at the call-out in the picture now. You can see what the via holes look like once filled 9x13_via_in_pad_bga_land_filled with high-temperature solder plugs. Now, when we apply the solder paste, those plugs will act as the substrate. The solder balls will bond properly and no solder will wick down and drip out the other side of the board.

For a second run, we would probably recommend re-spinning the board to have the vias plugged, but for a prototype, dealing with it like this will deliver a perfectly reliable board.

There are a variety of processes available at board fab houses to properly deal with vias in pad. You can have the vias capped with solder mask on either the top or bottom. We don't like this option. If the vias are capped on top, the BGA may not get enough of a mechanical connection. If the vias are capped with soldermask on the bottom, expansion of trapped gasses can pop the cap, causing the solder paste to run out as though there wasn't a cap at all.

The board house can plug vias with high-temperature solder just like we did. Some will fill vias with a conductive epoxy or insert a small copper peg. These are all decent options. The best option is to do one of these three and then plate over with copper. If that is done, an assembly house can't even tell that the vias are there, so the process and reliability will be unaffected by the vias.

Duane Benson
Happy BGAs R good

« January 2007 | Main | March 2007 »