Heavy Copper PCB Surface Finish: Why One Board-Level Entry Is Not Enough

Open the PCB168 manufacturing capabilities page and the standard-PCB block lists four surface finishes: HASL, ENIG, OSP, Immersion Tin/Silver. Now open the heavy copper PCB product page and read its specification table. The surface finish row says HASL/ENIG/OSP.

Immersion tin and immersion silver are on one list and not the other.

That difference is the whole reason this article exists. A surface finish is not a preference you carry over from your last board. It is a plating process applied to a specific copper geometry, and heavy copper changes that geometry enough that the available set of finishes changes with it. Your RFQ has one field labelled "surface finish." A heavy copper board frequently needs more than one answer in it.

What actually changes when copper gets thick

A standard board runs 1-2oz copper. The heavy copper page defines its range as 3oz (105µm) and up, through 20oz, on 1-16 layers, at 0.8mm to 6.0mm board thickness, with 6mil minimum line width — versus 2mil/2mil on the standard capability list. That minimum line width gap is the visible symptom of something that matters for plating: thick copper etches with significant undercut, which is why the product page states that heavier copper needs wider spacing and etch compensation.

Three consequences follow, and every one of them lands on the finish decision.

Surface area and topography. Heavy copper boards are dominated by large uninterrupted copper pours, not by small isolated pads. The heavy copper page's own application list — power converters, automotive electronics, industrial control, welding equipment, UPS, solar inverters — describes boards where the copper is the current path, so the exposed metal is measured in square centimetres, not square millimetres. Plating a large continuous surface uniformly is a different problem from plating a pad field.

Vertical relief. At 3oz the copper stands roughly 105µm above the substrate; at 6oz roughly 210µm. Soldermask and any wet chemistry now have to negotiate a genuine step, not a film-thickness bump. Anything that relies on even drainage or even flow across the surface has to cope with that relief.

Thermal mass. The reason you specified heavy copper — copper conducts heat far better than FR-4, as the product page puts it — is also the reason the board behaves differently in any thermal process. A board that is engineered to spread heat will spread it during finishing and during assembly too.

None of that makes a finish impossible. It makes the finish a per-geometry decision rather than a board-level one.

The four pad groups a heavy copper board actually has

Before comparing finishes, separate the surfaces. Most heavy copper boards carry four functionally distinct groups, and they have different requirements:

1. Signal and control pads. Regular SMT footprints, often on the same board as the power section — gate drivers, sense resistors, microcontrollers, connectors. These behave like pads on any board: coplanarity and fine-pitch solderability matter.
2. High-current solder terminations. Wide pads for power devices, bus connections, ring lugs soldered directly to copper, large through-hole terminals. Here the concern is wetting a large thermal mass, not coplanarity.
3. Pressure-contact and screw-mount areas. Bolted lugs, busbar landings, clamped joints, screw-down terminal blocks. These are never soldered. Their requirement is a stable, low-resistance, mechanically durable contact surface that survives clamping force and time. A finish optimised for solderability is answering a question these areas never ask.
4. Test, probe and edge features. Points that get contacted rather than joined, including whatever your test strategy needs. The capabilities page commits to 100% electrical testing (flying probe / fixture) and AOI inspection on every layer, so these surfaces will be touched.

Groups 2 and 3 are the ones a standard board does not have, and they are exactly the groups that a single board-level finish entry silently mishandles.

HASL on heavy copper: the thermal mass question

HASL appears on both the capabilities list and the heavy copper product page, so it is available across the range. Its characteristics interact with heavy copper in a specific way.

HASL is a molten-solder process: the board is coated and the excess is levelled off. The resulting surface is solder, which is why it wets readily and why it is the default for through-hole and high-current terminations. On heavy copper, two things deserve attention.

First, the board is a heat spreader. Immersing a high-thermal-mass board in molten solder and then levelling it is a thermal event, and the board's heat distribution is not uniform when copper distribution is not uniform. The heavy copper page notes that high-Tg FR-4 (TG170/TG180) is the specified material precisely because these boards see thermal stress; that material choice is part of why the process is viable, and it is also a reminder that the thermal path matters. If your stackup mixes very heavy outer copper with thin dielectrics, say so — that is an input your fabricator needs, not a detail to leave implicit.

Second, HASL's surface is not flat by nature. Levelling reduces the solder crown but does not eliminate it. On the wide terminations of group 2 that is irrelevant and arguably helpful. On group 1's fine-pitch footprints, coplanarity across the pad field is what determines whether paste deposits and placement behave, and a mixed-geometry board gives you both surfaces at once. If the same board carries 8oz power pours and a fine-pitch signal section, "HASL" as the single answer means you accepted a coplanarity outcome for the signal section that you never explicitly discussed.

Third, and most commonly missed: HASL is a solder coating, and solder coating on a pressure-contact surface is a different proposition from solder coating on a joint you are about to reflow. Bolted and clamped joints depend on the contact surface staying dimensionally and mechanically stable under load. If group 3 exists on your board, it needs its own decision, not inheritance from group 2.

ENIG on heavy copper: uniformity over large areas

ENIG is on both lists as well. It is an immersion process: nickel is deposited over the copper and a thin gold layer over the nickel. The result is flat, which is why it is the standard answer for fine-pitch assembly, and it is a plated metallic surface rather than a solder coating, which is why it is the usual candidate for contact and probe areas.

On heavy copper, the relevant question shifts from "is it flat" to "is it uniform everywhere." ENIG deposits from solution, so coverage across a very large continuous copper area, and into the corners of a 210µm step, depends on process control over that geometry — not on the nominal specification. This is a legitimate engineering-review item and it is why the product page invites you to send your stackup for a manufacturability review before committing to production.

ENIG's flatness makes it the natural fit for group 1 on a mixed board. Its metallic, non-solder surface makes it the natural candidate for group 3's bolted and clamped areas and for group 4's probe points. What it is not automatically better at is group 2: wetting a large thermal mass is HASL's strength, and the thin gold on ENIG is a wetting layer, not a solder reservoir.

So on a genuinely mixed heavy copper board, the technically coherent answer is frequently not one finish. It is ENIG on the signal and contact areas, HASL on the large solder terminations — or the reverse, depending on which group dominates your assembly risk. Whether a given board can be selectively finished, and what that costs, is a fabricator answer. The point here is that you cannot get that answer if your RFQ never asked the question.

OSP on heavy copper: the shelf-life and rework trap

OSP is on the capabilities list and on the heavy copper product page. It is an organic coating over bare copper: extremely flat, no metallic overlay, and by construction a temporary protection for the copper underneath.

Two properties matter here and neither is about the copper thickness itself.

It is single-purpose. OSP protects copper until soldering, then is consumed. It is not a contact surface. Group 3's bolted and clamped joints and group 4's repeatedly probed points are not what OSP is for.

It is time-and-handling sensitive. Because the protection is organic and consumed, boards have a working window and tolerate a limited number of thermal excursions. Heavy copper boards are frequently exactly the boards that see extended assembly sequences, multiple solder operations, and rework on large terminations — because reworking a high-thermal-mass joint takes longer and more heat than reworking a small one. If your assembly plan includes selective soldering of power terminals after SMT, or field rework on bus connections, OSP's consumed-coating behaviour is a planning input, not a footnote. I am deliberately not quoting a shelf-life figure: PCB168's published pages do not state one, and the real number depends on storage and packaging. Ask for it in writing rather than assuming a number from another supplier's datasheet.

OSP's genuine strength on heavy copper is cost and flatness on boards where the whole assembly happens in one controlled pass and nothing on the board is a pressure contact.

Immersion tin and immersion silver: the list difference you must resolve

Here is the one that sends people into a quote round they did not plan for. The capabilities page lists Immersion Tin/Silver among available finishes. The heavy copper product page's specification table lists HASL/ENIG/OSP.

Read that difference correctly. It tells you the published heavy-copper specification set is narrower than the published general capability set. It does not tell you that immersion tin is impossible on heavy copper — the heavy copper page is a product specification page, not an exhaustive process exclusion list, and the "Ordering" section explicitly names HASL, ENIG and OSP as the available finishes for these boards.

What it does tell you is actionable: if your design requires immersion tin or immersion silver on a 3oz-or-heavier board, that is an engineering confirmation item, not a checkbox. Raise it in the RFQ, ask for written confirmation against your specific copper weight, layer count and board thickness, and get a named alternative if the answer is no. Discovering this after a purchase order is issued is how a schedule slips.

The same discipline applies whenever a finish you need is absent from the specification table of the product family you are ordering. The narrower list is the one that governs your order until engineering says otherwise in writing.

What "surface finish: ENIG" costs you when it is the only thing you wrote

A single board-level entry is not neutral. It hands four decisions to whoever processes the order, and they will be resolved by default:

• Which pad groups get which treatment becomes "all of them get the same one," including the bolted joints that never see solder.
• Coplanarity expectations for the fine-pitch section become whatever the chosen process delivers on a board dominated by heavy copper geometry.
• Contact-surface durability on screw-mount areas becomes unspecified, so it is verified against nothing.
• Coverage uniformity across the large pours becomes an internal process question rather than a documented acceptance criterion between you and the supplier.

None of those defaults is necessarily wrong. All of them are unexamined, and three of them surface at assembly or in the field rather than at incoming inspection.

How to declare finish per pad group in the RFQ

Put this in the RFQ body or in a note layer in your data package. It is short, and it changes the conversation from a checkbox to an engineering review.

1. Declare the copper weight the finish applies to. Outer-layer finished copper weight and, if the stackup is mixed, which layers carry which weight. The finish question cannot be answered without it.

2. Identify the pad groups on the board. Reference designators, net names or a marked-up drawing. Four lines is usually enough:

• signal/SMT pads: [where]
• high-current solder terminations: [where]
• pressure-contact / screw-mount areas: [where, plus fastener size and whether the joint is clamped in service]
• test and probe points: [where]

3. State the finish requested for each group, and flag where you are open to a recommendation. If you want one finish everywhere, say that explicitly so it reads as a decision rather than an omission. If a group has no soldering requirement, say so — that is the single most useful sentence you can add, because it tells engineering that group 3 needs a contact surface, not a solderable one.

4. State the assembly sequence that follows. Number of reflow passes, whether selective or wave soldering hits the power terminations after SMT, whether rework on large joints is anticipated, and expected storage time between fabrication and assembly. This is what makes OSP viable or non-viable, and it is the input most often left out.

5. Raise any finish not on the product-family specification table as an explicit question. For heavy copper: if you need immersion tin or immersion silver, ask for written confirmation against your copper weight rather than assuming the general capabilities list applies.

6. State how the finish will be accepted. The capabilities page commits to AOI inspection on every layer and 100% electrical testing (flying probe / fixture), and lists cross-section analysis for quality validation and ionic contamination / cleanliness testing per IPC standards. Reference the standard you are accepting against by name and ask which of these applies to the finish on your board. Do not invent a plating-thickness number or a porosity limit and put it in the PO — ask what is measured, how, and on what sample. An acceptance criterion you made up is worse than none, because it looks agreed.

7. Keep it consistent with the rest of your declarations. Finish interacts with soldermask coverage over thick copper, with finished copper weight versus base copper, and with trace width and spacing. Those are separate declarations and they are covered separately — see below. What matters is that the finish note does not contradict them.

Where this sits among the other heavy copper declarations

Surface finish is one field. These are the adjacent ones, each with its own decision and its own failure mode:

• Soldermask over thick copper — coverage, registration and edge behaviour on a 105-210µm step — is the heavy copper soldermask coverage declaration.
• The copper weight itself, driven by current and temperature rise, is the current and thermal budget decision.
• Geometry limits at a given copper weight are in trace width and spacing versus copper weight.
• Whether your PO says base copper or finished copper is finished copper thickness versus base copper.
• General finish behaviour independent of copper weight is the PCB surface finish comparison.
• Acceptance language for the whole order is the heavy copper purchase specification and acceptance.

When the pad groups and the assembly sequence are documented, send the package through heavy copper PCB and request a quote. Full published limits are on the capabilities page.

FAQ

Can I use one surface finish across an entire heavy copper board?

Often yes, and it is the simplest path when the board is either all power or all signal. It becomes a problem on mixed boards and on boards with pressure-contact areas, because one finish then has to serve solderability on a large thermal mass, coplanarity on a fine-pitch field, and mechanical contact stability at a bolted joint. Those are different requirements. Declare it as a deliberate choice rather than leaving the field to be interpreted.

Why does the heavy copper page list HASL, ENIG and OSP while the capabilities page also lists immersion tin and silver?

Because the product specification table for heavy copper is narrower than the general capability list. Treat the narrower list as governing your order. If you need immersion tin or immersion silver on 3oz-or-heavier copper, raise it in the RFQ and get written confirmation against your specific copper weight, layer count and board thickness before the PO.

Which finish should go on screw-mount and bolted busbar areas?

That depends on the joint, and it is worth an engineering conversation rather than a rule of thumb. The useful step on your side is to declare that those areas exist, mark where they are, state the fastener size, and state that they are not soldered in service. A pressure contact needs a stable low-resistance surface under clamping load, which is a different specification from a solderable one — and it is the distinction that a single board-level finish entry erases.

Does heavy copper rule out OSP?

No. OSP appears on the heavy copper product page. What constrains it is your assembly plan, not the copper weight: OSP is a consumed organic coating, so extended assembly sequences, multiple solder passes, anticipated rework on large terminations, and long storage between fabrication and assembly all work against it. Boards assembled in one controlled pass with no pressure contacts are where it fits. Ask PCB168 for the storage window in writing rather than assuming a figure.

What does my RFQ need to include for the finish question to be answerable?

Outer-layer copper weight, a marked-up identification of the pad groups, the finish requested per group or an explicit "same everywhere," the assembly sequence including reflow passes and any selective soldering or rework, and the acceptance method you expect. Without the copper weight and the assembly sequence, any finish recommendation is a guess.

Should I specify a plating thickness in the purchase order?

Only if you have a requirement you can defend and a confirmed number from the supplier. Inventing a thickness or porosity limit and putting it in the PO creates an acceptance criterion that looks agreed but was never engineered. Ask what is measured, by what method, and on what sample. PCB168's published quality controls include AOI on every layer, 100% electrical test by flying probe or fixture, cross-section analysis, and cleanliness testing per IPC standards; ask which of these covers the finish on your board.

Related links

• Manufacturing Capabilities
• Heavy Copper PCB