The Complex World of Soldering
Industry veteran Happy Holden and Saline Lectronics Senior Process Engineer Cathy Cox discuss the various challenges and issues that users face in soldering, including the lack of a "one-size-fits-all" approach in the process, and some key factors that the PCB assembly industry should consider when it comes to different applications and markets.
Stephen Las Marias: What do you think are the greatest challenges when it comes to soldering in PCB assembly?
Cathy Cox: I would have to say our challenges are temperature. Temperature is a big problem with the flex boards. With the circuit board being so thin, actually getting the right temperature to be able to run them through a reflow is definitely a challenge.
Happy Holden: Picking a final finish that’s compatible with the solder that you picked in the flux. There’s no one answer that fits all. So picking that, and then you might say defending it—when people are selling different combinations—is a pretty big challenge, especially when not all the criteria for evaluation and testing is readily available. I've come to learn that the wetting balance really isn’t reliable to use that much anymore.
Our end-use applications are so diverse that when you talk about any particular subject, it’s tough for people to apply. ‘Does this apply to me or is he talking about the other guy over there?’ For instance, flex assembly and soldering is a particular challenge; but when I think of soldering, I think of rigid boards and I think of standard size thinner ones. When you get to really thick—over 2 mm—type boards, getting lead-free solder up to a milling temperature is a real challenge. There, you have to worry about the materials you built the boards for. It’s a huge industry, and how you carve it up is part of the challenge.
My life for years has been in automotive electronics. The company I work for is one of the highest volume assembler of automotive electronics in the Western hemisphere. Soldering is not really a problem except for new devices, QFNs and things like that that are coming on the market. It’s a process that’s running extremely smooth.
Las Marias: Do you think lead-free soldering continues to be an issue when it comes to reliability?
Holden: The industry has already moved to lead-free soldering. That was done a long time ago. It’s only the military and avionics and a few hold outs that haven’t moved to lead free. I talked to people at the APEX show that were in the military, and they realize that their time is running short because they have fewer and fewer people who support them. If you're going to lead free then, unlike tin lead where there wasn’t much variety, now you’ve got a whole proliferation of solders and different compatibility with final finishes, and different suitability for reliability life. That’s one reason why it’s so difficult because depending on what you’re assembling for, the answers maybe different.
Cox: I totally agree with Happy. At Saline Lectronics, the majority of our customers are doing lead free. Lead free is definitely more of a challenge with flex boards just due to the temperature. Getting good solder joints is definitely a challenge, but it’s a challenge that most companies are up for, so it's good. We need new things in our industry.
Las Marias: Do miniaturization, increased board densities, and shrinking component sizes impact the soldering process from your perspective?
Cox: I don't know if they necessarily affect it from what I can see. We deal with a lot of the 0201s. We're pretty excited to get our hands on 01005s and start playing with them and see if our pick-and-place machines can handle them. We'll need more precise placements to be able to make sure you get a good solder joint. It will be a little bit more challenging for an operator to manually inspect a board with solder joints being so small. But we have our high-tech AOI equipment, so that should not be a problem for us.
Patty Goldman: So, you don’t see very many of those types of components?
Cox: We see very small components all the time, but the smaller they get will be a challenge. I don't think it’s a challenge that’s going to stop us from doing anything. It’s just a different way of looking at it; it’s a different way of manufacturing it. You have to have more precise placement, make sure the thickness on your solder paste machines are proper, and I think it’s going to be awesome.
Goldman: What's your biggest bugaboo, then?
Cox: Our biggest with flex boards would actually be our pallets that we use in surface mount to keep our boards flat during the reflow. We can set pallets up for all different types of boards, but getting one to fit completely flat in there every time, and make sure the components are going to sit flat when they go through reflow, is a challenge. Most of the time, it's the bigger components that we see issues with because there’s a bigger area to have a bubble in the flex board.
Las Marias: Is solder jetting catching up with the paste printing process?
Holden: My last six years have been in high-volume production, and solder paste printing and the integral AOI machines are going to be tough to match in terms of jetting. Jetting may be much more applicable to high-mix, low-volume, but for the high-volume automobile market, mobile phones, consumer products, paste printing is pretty well defined and highly robust. Jetting maybe coming along, but it will have a hard time matching the fancy step stencils and some of the other techniques developed in stencil printing. But it may be totally adequate for high mix. I’ve never seen or we’ve never used any jetting equipment except for inkjet printing or solder mask printing in the PCB area. I don’t have a lot of experience with it, but any kind of jetting like that, be it inkjet, solder mask, or solder, really has not met the needs of volume manufactures to be cost effective, not yet. It’s for the prototype or high mix.
Las Marias: Does solder paste selection or qualification still matter today despite the standard solder types used in certain applications?
Holden: Yes, because they’re constantly improving solder paste and things like that. You're kind of always evaluating tweaks and minor improvements to see what it does, especially when one of our focuses in the past has been voiding and the micro-voiding you had with lead free. We haven't rejected products but we'd liked to eliminate it. Those are constantly minor tweaks in the formulation of the solder paste and the refill profile.
Las Marias: I understand solder voiding is still a big issue in the industry. Cathy, how do you tackle voiding?
Cox: We go with Indium solder paste. We have a really good connection with them. They help us through a lot of our issues with incisions, voids, stuff like that. We have been trying a few things here in our shop with changing the amount of paste that we put on a stencil, removing the mask around the pads so that the air can actually escape between the component and the solder pad. We've seen some success with that. The IPC standard keeps getting more and more vague on voiding under QFNs or even in through-hole components. We took it upon ourselves to kind of set our own standards, making sure that we have anything under 25% under a component or within a barrel fill. With our window pane and with reducing the solder mask around the pads, we’ve been successful in minimizing voiding to about 25% with our QFNs, BGAs.
Our biggest struggle is making sure that if you call a component out, your footprint should actually meet what the manufacturer of the component recommended. We do run into quite a few times here where the customer will call out a part, and it doesn't quite fit, which means that you don't get proper solder joints, you don't get the heel fillet that you want, or the side joints that you want. Then they don't meet IPC standards. It's a struggle to keep going back and forth between the customer and the manufacturer trying to make sure the component still functions. So verifying that you're setting your board up for success with proper pad size would be a good thing for customers to know.
Las Marias: What else do you think should the readers know more about when it comes to soldering?
Holden: Some of the effects that different formulations of lead-free solder paste has when it's reflowed, in terms of interacting with final finish and interacting with land size, is that they don’t spread like tin lead, and we find smaller pad sized footprints because it’s a stronger joint than tin lead. We’ve always watched very closely the Japanese and the JPCA standards, because they’ve been using lead free for 10 years longer than North America. Since a third of all of our output goes to Japanese automakers, we’ve got a lot of cooperation with Nissan, Honda, and Toyota about what they recommend.
Las Marias: Standardization is still an issue?
Holden: It's all a point of view. Like I said, we're such as big industry that I’ve been talking about automotive electronics, which were going for a 15-year useful life—and that’s a lot different than a mobile phone. We virtually test 100% of every lot because the Japanese or the German auto companies insist on that. If we ever have a failure, we have to go back into root cause analysis. A lot of my experience deals with reliability because warranty failures and things like that is a big bug-a-boo in automotive. But depending on which industry you’re in, there's a lot of differences. Most mobile phones don't make it past two years before they've been dropped in the toilet, run over, or just that everybody turns them in for the new models.
You have to support your customers, and not all customers have the same needs. It’s the final performance that our customers required. They weren’t dictating a design or a process, only the final performance, whereas if you're an EMS and you have multiple customers, you have to keep shifting it all the time depending on what the requirements are. We still have wave solder machines. They're lead free, but they’re still wave solder.
Cox: We do too. We do a lot of military and oil and gas that we build predominantly through-hole. We have a lead one and a lead free.
Holden: Lead free is a lot tougher than the tin lead.
Cox: Yes. I was thinking about it and a lot of what I've experienced here in the last couple of years is our customers will sometimes call out to use a lead-free no clean solder on their boards. Then they are upset with us that their boards do not look clean and the solder looks to be disturbed. Really, the industry needs to know having a lead-free solder is going to look more grimy than using leaded solder. They look different. A no-clean solder doesn't necessarily mean that it's going to be clean. You're going to have residue there, but it's not active flux residue. The different types of solder that you run into along the way, whoever you use should talk to you about the way your solder is going to look before you actually put it on the boards. I think that’s a big thing. Our customers come back and say the boards are dirty, but they’re not. They meet IPC standards and they use a no-clean flux, which will leave slight residue on their boards.
Holden: That's one thing I can't ever remember reading or seeing: a breakdown of the soldering industry by what all of the different types of solders look like, in terms of a picture, whether clean or no clean. Then, their interaction with final finishes, and then what industries they are preferred by in terms of a matrix. A lot of people tend to get confused, because if you're talking about the communications industry, that may not apply to the consumer industry; or when we're talking about automotive, that might not apply to medical. You haven't always seen these things. Nobody’s ever come out and given us matrix of what they’re supposed to look like, and maybe what the estimated life is in terms of their interaction with land patterns, with final finishes, and with percent of voiding. That chart or definitive explanation has not been out there.
Cox: I agree. A lot of customers I know say, ‘Oh, if I have the choice between the no clean or the aqueous, I want to go with the no clean because that means my board will be clean.’ But in reality, it means that you don't need to clean the board because the residue that's left isn't active. You will actually have a cleaner board, appearance-wise, if you use an aqueous flux. Then you can wash it and there is nothing left behind. That’s something we struggle with over here a lot with our customers. It's just a miscommunication of what they're actually using, what the product is doing, and how it's going to look.
Las Marias: Cathy and Happy thank you so much again for your time and your insights.
Cox: Thank you.