Via in Pad: PCB Design Rules, Filling and Capping Guide
Via in pad explained: when to use it, filled and capped vias (VIPPO), solder wicking, cost, IPC-4761 types, and design rules for BGA and QFN layouts.
Via in pad explained: when to use it, filled and capped vias (VIPPO), solder wicking, cost, IPC-4761 types, and design rules for BGA and QFN layouts.

Short answer: Via in pad means drilling a via directly inside a component’s copper pad instead of routing a short trace from the pad to a via beside it. It saves space and shortens the electrical path, which makes it the standard way to fan out fine-pitch BGAs and to pull heat out of QFN thermal pads. The catch is solder: an open via in a pad wicks solder away from the joint, so for SMT pads the via is normally filled with resin or copper and plated over flat, a process often called VIPPO (via in pad plated over). Use it when pitch, signal integrity or heat demand it, and spec the filling clearly in your fab notes.

In a conventional layout, every SMT pad connects to a via through a short trace called a dog-bone. Via in pad (also written via on pad or pad on via) removes the dog-bone and places the via inside the pad itself. The via barrel connects the pad straight down to inner layers or to the other side of the board.
That sounds like a small change, but it changes how the board is built. A plain through-hole via is an open, plated hole. If you put an open hole in an SMT pad, the paste melts during reflow and capillary action pulls it down the barrel. To prevent that, the fab fills the via, planarizes it and plates copper over the top so the pad is flat and solid again. That extra fill-and-cap step is the main cost of via in pad, and it is why you should use it on purpose rather than everywhere. It sits in the same family of choices as blind and buried vias, and the two are often combined in HDI boards.
Via in pad earns its cost in four common situations:
If none of those apply, a dog-bone with a standard via beside the pad is cheaper and easier to build.

| Option | What the fab does | Good for | Risk |
|---|---|---|---|
| Open (unfilled) via in pad | Nothing extra; plated hole left open | Rarely acceptable on SMT pads; sometimes tolerated in large thermal pads with tiny drills | Solder wicking, voids, starved joints, paste on the far side |
| Tented or plugged via | Solder mask or plug covers one or both ends | Vias next to pads, not inside them | Trapped chemistry, outgassing, uneven pad surface |
| Resin-filled, copper-capped (VIPPO) | Fill with non-conductive epoxy, planarize, plate copper over | BGA pads, QFN thermal pads, most via in pad work | Higher cost and lead time; needs clear fab notes |
| Copper-filled via | Fill the barrel with plated copper | Microvias and high-current or high-thermal vias | Highest cost; usually limited to small drills |
The key point: for any via inside an SMT pad that gets solder paste, ask for filled and capped vias. Open vias in pads are the source of most via in pad assembly failures, from voided BGA balls to QFNs that float on uneven solder.
Fabs describe via protection using IPC-4761, which lists types from simple tenting to full fill and cap. In practice you will see a few names:
Write the type into your fab notes rather than writing “via in pad” and hoping. A clear note such as “All vias in SMT pads: non-conductive epoxy fill, planarized, copper capped per IPC-4761 Type VII” removes ambiguity at quote time. The solder mask layer and the surface finish both sit on top of the cap, so flat ENIG or ENEPIG pads pair well with VIPPO under fine-pitch BGAs.
These rules keep via in pad buildable and reliable:
Via in pad adds process steps: filling, curing, planarizing and a second plating cycle. Expect a higher price per board and a longer lead time than a dog-bone design with standard vias. On dense boards that is often still the cheaper choice overall, because it can save routing layers, shrink the board or avoid HDI build-ups. The decision is a trade-off between process cost and the layers and area it saves, which is why it belongs in your DFM review and not only in routing.
Add a via in pad line to your PCB design review checklist so these get caught before Gerbers go out.
Via in pad errors are usually mismatches between the layout and the notes: a via inside a pasted SMT pad with no fill and cap specified, a footprint that mixes open and filled vias, a drill that breaks the annular ring, or an aspect ratio the chosen fab cannot fill. All of these can be checked automatically. An AI-assisted DFM pass reads footprints, drill data, stackup and fab notes together, finds every via that lands inside a pad, and flags the ones that will wick solder or cannot be built as specified, before the board is quoted.
That is the IntelCAD angle on via in pad: not just a definition, but a check tied to the actual pads, parts and fab notes on your board.
Via in pad places a via directly inside a component’s copper pad instead of connecting the pad to a nearby via with a short trace. It saves space and shortens the electrical path, and is common for fine-pitch BGAs and thermal pads.
For SMT pads that receive solder paste, yes in most cases. An open via wicks solder away from the joint, so fabs fill the via with resin or copper and plate it over to make a flat pad.
VIPPO stands for via in pad plated over. The via is filled, planarized and capped with copper so the pad is flat and solderable, which is the usual way to build via in pad for BGAs.
Yes. Filling, planarizing and extra plating add cost and lead time. On dense boards it can still lower total cost by saving routing layers or board area.
Yes. A grid of filled and capped vias in the thermal pad moves heat into internal planes. Segment the paste stencil so the part does not float during reflow.
Want every via that lands in a pad found and checked against your fab notes before you send Gerbers? Start Your Design with IntelCAD for AI assisted PCB design and manufacturability checks.
Design your next board by describing it.
Start Your Design
Gerber files for PCB explained: which layers to export, RS-274X vs X2, file extensions, drill files, a fab package checklist and how to check files first.

Short answer: PCB copper thickness is usually specified as copper weight in ounces per square foot. 1 oz copper is about 1.37 mil (roughly 35 µm) thick, 0.5 oz is about half that, and 2 oz is about 2.74 mil (roughly 70 µm). Most boards use 1 oz finished copper on outer layers and 0.5 …

Short answer: Clearance is the shortest distance between two conductors through air. Creepage is the shortest distance between them along the surface of the insulating material. On a PCB, clearance protects against arcing from voltage peaks and transients, while creepage protects against tracking, the slow build-up of conductive paths across a contaminated board surface. Both …