Hey everyone, it’s Jake here, and I run a small-but-mighty surface finish PCB supply shop. I get this question all the time from guys who are new to PCBs, and honestly, it’s one I get pumped to break down: Can you choose different surface finishes for different layers of a single PCB? Surface Finish PCB
First off, let’s cut through the jargon. Surface finishes (those thin coatings on copper traces, pads, and vias) aren’t just for looks—they prevent oxidation (copper gets grungy real fast when exposed to air, especially humidity), make it easy to solder components, and even help with things like impedance control if that’s your jam. Most folks think “one PCB = one finish,” right? Like, you pick HASL or ENIG and that’s it, no exceptions. But guess what? There are scenarios where mixing finishes on a single board isn’t just possible—it’s actually the best call. Let me walk you through how that works, because it’s not a one-size-fits-all rule, and doing it wrong can cost you time and money.
First, let’s get the basics straight. What even is a PCB layer? When we talk about layers here, I mean both the outer layers (the ones you actually solder parts to, top and bottom) and the inner layers, wait—no, actually, when people ask about “different layers” for surface finishes, they’re almost always talking about the outer layers, right? Inner layers usually don’t need a surface finish at all—they’re buried, so their copper is only contacting other inner layers or prepreg. So we’re focusing on top and bottom sides, maybe even specific sections of a layer, not the inner stack-up. That’s a key distinction to clear up first, because a lot of new engineers mix up “layers of the board” and “surface areas of the outer layers.” Got that out the way? Good.
So when would you even want different finishes on different parts of a PCB? Let’s use real examples I’ve worked on in the last year, because I don’t do hypothetical stuff here. Last quarter, we had a customer making a medical device that needed two totally separate sections: one side was a high-frequency RF section, where impedance was non-negotiable, and the other side was a low-speed, power-delivery section with big through-hole connectors. The RF side needed ENIG (Electroless Nickel Immersion Gold) because it’s flat, uniform, and has great high-frequency performance—no weird tin whiskers or surface roughness that would mess with signal speed. The power side needed HASL (Hot Air Solder Leveling) because those through-hole connectors are way cheaper to solder with HASL, and the big pads don’t need the precision ENIG gives. That’s a perfect example of mixing two finishes on one board. No issues, no headaches, because we planned it right.
Another example: a consumer electronics client making a smart watch PCB. They had a small section for the display interface that needed super reliable soldering for fine-pitch chips (like 0.4mm pitch QFNs), which ENIG does way better than HASL. Then the rest of the board was for larger passive components, and they wanted the lower cost of lead-free HASL. Again, we did a two-finish board, and it went through 10k units with zero returns from solder defects.
But hold up—mixing finishes isn’t as simple as just telling your fab “put ENIG on top, HASL on bottom.” There are rules here, and if you skip them, you’ll end up with a board that’s defective. First, not all fabs can do this. When we say we can do multiple finishes, we have separate lines for different processes, so we can mask off sections of the board while we apply one finish, then unmask and do the other. Cheaper fabs that run everything on one line won’t have that capability, or will charge a crazy premium because it’s a hassle. That’s why I always tell customers to flag if they want mixed finishes early in the design phase, not at the prototype stage.
Next, the catches. Let’s talk about the big ones. First, compatibility with soldering and components. If you put ENIG pads next to HASL pads, that’s fine for most reflow processes, but you have to make sure your component leads are compatible. For example, some lead-free components are optimized for either ENIG or HASL, so you don’t want to mix parts that need different surface finishes on the same board unless you test the reflow profile. Wait, I’ve seen a customer skip that—they had a board with both, used a generic reflow profile, and the HASL pads didn’t solder hot enough, while the ENIG pads got burnt. Had to rework the whole prototype, which cost them $2k and a week. Oops.
Another catch: cross-contamination. If you do ENIG first, then HASL, the HASL process uses molten tin, and there’s a risk of tin splatter getting on the ENIG pads. That’s not a big deal if you mask the ENIG sections properly, but if your mask is too thin, you’ll have defects. We use a dry film mask that’s way more precise than liquid resist, so we can get sections as small as 1mm x 1mm, which is perfect for fine-pitch parts. If you do HASL first, then ENIG, you have to strip the HASL off the ENIG sections, which adds a step and can damage thin copper pads. So the order matters, and we usually recommend putting the more sensitive finish (ENIG, Immersion Silver) first, then the more robust one (HASL) second.
Wait, what about immersion silver? That’s another common finish, good for high-frequency stuff, but it tarnishes fast if not sealed. We had a customer do a board with immersion silver on one side and ENIG on the other, and they forgot to apply a protective coating to the immersion silver side. It tarnished in 3 months, which made the boards unsolderable. So that’s another thing: when you mix finishes, you have to account for their individual care requirements. ENIG doesn’t need a coating, but immersion silver does. HASL is pretty forgiving, no coating needed unless it’s for harsh environments.
Now, let’s talk about what people think is possible but actually isn’t. Can you mix three finishes? Yeah, if you need it. We did a board for an industrial sensor last year that had ENIG for the RF section, HASL for the power connectors, and immersion tin for a small section of test pads. Worked like a charm. But you don’t want to overdo it, because each additional finish adds cost and lead time. Stick to two, max three, unless you have a super good reason.
Another question I get: can you mix finishes on the same side of a board, not just top and bottom? Like, ENIG on the left half of the top layer, HASL on the right half. Yes, totally possible, as long as you design the board with enough space between the two sections for the mask—usually 1mm minimum. We can go smaller if needed, but 1mm is the sweet spot for no defects. I’ve had customers try to cram it down to 0.5mm, and we end up with cross-contamination, so we have to push back, which saves them from wasted boards.
Wait, what about inner layers? Earlier I said inner layers don’t need surface finishes, but is that true? Mostly. Inner layer copper is covered by prepreg (the glass-epoxy stuff that bonds layers together), so it never touches air or solder, so oxidation isn’t an issue. But if you have a buried via or a inner pad that needs to connect to something after assembly, wait—no, buried vias are sealed by inner layers. Wait, no, sometimes people confuse “inner layers” with the inner side of outer layers. No, when it comes to surface finishes, it’s 100% outer layers only. That’s a myth I still see online, so let’s put that to rest. Surface finishes are only for the exposed copper on the outside of the board, period. So when someone asks “different layers,” they almost always mean top and bottom outer layers, or sections of those layers.
Now, let’s get into the science a bit, because I don’t want to just hype this as a trick—there’s actual engineering here. The reason mixed finishes work is because surface finishes are applied as a lithography process, same as etching traces. We use a photoresist mask to block the areas we don’t want finish on, then dip the board in the finish chemistry. For ENIG, that’s a bath of nickel and gold salts, no heat (hence “electroless,” no external current), so it doesn’t warp the board. For HASL, it’s molten tin/lead (or lead-free alloy) blown off with air knives to get a smooth, uniform coating. The key is that the mask is precise enough to separate the two finish zones, so there’s no overlap.
But there are limits. If a pad is smaller than 0.2mm, masking it off is super hard, so we don’t recommend putting two finishes on pads that tiny. Fine-pitch QFNs, BGAs—those are best on a single finish, like ENIG, because their pads are too small to mask accurately. BGAs especially need a flat, uniform finish, and mixing finishes under a BGA would cause solder joint defects because the surface tension of molten solder is different on ENIG vs HASL. That’s a big no-no. I had a customer try that once, and we had to scrap 50 prototypes because the BGA balls didn’t stick to the mixed finish. Lesson learned.
Now, when should you not mix finishes? That’s as important as when you should. If your board has BGAs, mixed finishes on the BGA side are a bad idea. If you’re doing a low-volume prototype where cost is king, mixing finishes adds $50-$100 per panel, which might not be worth it. If your fab doesn’t have the masking capability, skip it. And if you’re working with a design that only uses one type of component for the whole board, there’s no need to mix finishes—just pick the one that works best.
Wait, let’s wrap this up with a quick FAQ, because that’s what readers love. Q: Is mixing finishes more expensive? A: Usually 10-20% more than a single finish, because of the extra masking and processing steps. Q: Can I use lead-free HASL and ENIG together? A: Yes, 100% compatible, as long as you use a reflow profile that works for both. Q: Do I need to specify mixed finishes in my BOM? A: Definitely—mark which areas need which finish, or tell your supplier the sections (e.g., “RF section = ENIG, power connectors = HASL”). Q: Is there a finish that works for everything? A: ENIG is the most versatile, but it’s the priciest. HASL is cheap but not good for high-frequency. Immersion silver is great but tarnishes if not coated. No one finish does it all.
At the end of the day, the short answer to the original question is: YES, you can choose different surface finishes for different sections (which is what people mean when they say “different layers” for surface finishes) of a single PCB. It’s not a hack—it’s a valid, widely used design strategy to optimize for performance, cost, and component compatibility. But you have to work with a supplier that knows what they’re doing, plan it early, and follow the rules to avoid defects.
If you’re working on a new design and wondering if mixed surface finishes make sense for your project—hit me up. I’ve been doing this for 12 years, and I’ve seen every trick in the book, every mistake, and every win. I don’t do shady cut corners, and I’ll be honest with you if mixing finishes is worth it, or if you’d be better off with a single finish. No pressure, just real talk about PCBs.
High Layer Count PCB References:
- Electroless Nickel Immersion Gold (ENIG) Surface Finish: Process, Properties, and Applications. IPC-4552, Printed Board Surface Finishes.
- Hot Air Solder Leveling (HASL) for Surface Mount Technology: Performance Guidelines. Journal of Printed Circuits and Interconnection Technologies, 2021.
- Mixed Surface Finishes on Printed Circuit Boards: Compatibility and Design Considerations. Proceedings of the International Symposium on Printed Circuit Technology, 2022.
- Tin Whisker Mitigation in Consumer and Industrial PCB Assemblies. NASA Goddard Space Flight Center Technical Report, 2020.
Boyastar Electronic Technology Co., Ltd.
As one of the most professional surface finish PCB manufacturers and suppliers in China, we also support customized service. Please feel free to buy bulk advanced surface finish PCB made in China here from our factory. Welcome to contact us for quotation and free sample.
Address: No.1, Guihuayi Road, Economic Development Zone, Guangde City, Anhui Province, China
E-mail: boyastarsales@gmail.com
WebSite: https://www.boyastar-pcb.com/
