Open almost any modern device and the board inside will look nothing like the one you have been learning on.
No legs poking through. No holes. Instead, hundreds of tiny rectangles sitting flat on the surface, some smaller than a grain of rice, some smaller than the period at the end of this sentence. It can be a slightly deflating sight right after you have finally gotten good at soldering a resistor.
This lesson exists so that sight does not deflate you. It is deliberately not a surface-mount tutorial. It is an explanation of what you are looking at, why it is like that, why this course started you somewhere else on purpose, and what it would actually take to work on it.
By the end of this lesson you will be able to recognize surface-mount construction and its common package types, explain why the industry moved to it, and make an informed decision about whether to go there yourself.
12.1 What surface mount actually is
The aim of this section is the one structural difference, from which everything else follows.
Through-hole, which you have been learning, puts a component's legs through holes in the board and solders them on the other face. The board is drilled, the holes are plated, and the joint passes through the material.
Surface mount, usually shortened to SMD or SMT, puts the component on top of a pad and solders it there. Nothing goes through. There are no holes and no legs, just small metal terminations sitting on small metal pads.
That single change has consequences everywhere:
- No drilling, which is a slow and expensive manufacturing step.
- Both sides of the board are usable, because nothing has to pass through.
- Components can be far smaller, because they no longer need legs long enough to reach through a board.
- Boards can be much denser, so devices get smaller and cheaper.
- Assembly can be fully automated. A machine places thousands of parts an hour, and the whole board goes through an oven that melts pre-applied solder paste in one pass.
That last point is the actual reason for everything else. Surface mount exists because it can be assembled by machine, and the entire economics of modern electronics rests on that.
Figure 12.1: Through-hole passes a leg through the board and solders the far side. Surface mount sits on top of a pad. Everything else follows from that.
12.2 The sizes, and what they mean
The aim of this section is to decode the numbers, because they look like part codes and are actually dimensions.
Small passive parts, resistors and capacitors, are named by a four-digit code that is simply their size:
| Code | Size | What it is like |
|---|---|---|
| 1206 | 3.2 × 1.6 mm | Comfortable by hand. A good place to start. |
| 0805 | 2.0 × 1.25 mm | Still very manageable by hand. |
| 0603 | 1.6 × 0.8 mm | Doable, and where most people stop enjoying it. |
| 0402 | 1.0 × 0.5 mm | Grain-of-sand territory. Tweezers and magnification. |
| 0201 | 0.6 × 0.3 mm | Not a hand-assembly proposition. |
The code is the size in hundredths of an inch: 1206 is 0.12 by 0.06 inches. Once you know that, the numbers stop being mysterious.
The reassuring part: 1206 and 0805 are genuinely hand-solderable with an ordinary iron and a steady-ish hand. If you ever want to try surface mount, start at 1206 and you may be surprised how approachable it is.
Figure 12.3: The same sizes at true relative scale. 1206 is a good deal larger than the numbers suggest.
Chip packages
Chips are named by their lead style rather than by size:
- SOIC, SSOP, TSSOP: legs out of the sides, bent down, visibly spaced. Hand-solderable, and the wider ones are quite forgiving.
- QFP: legs on all four sides. Hand-solderable with care and a technique called drag soldering.
-
QFN, DFN: pads underneath the body with no legs sticking out. Difficult by hand, because you cannot see or reach most of the joints.
- ⚠️ With one growing exception. Many modern QFNs are made with side-wettable flanks: the pads wrap a little way up the side of the package, leaving a small vertical face exposed at each corner of the body. Where a part has them, solder forms a visible fillet up that face, which means it can be hand-soldered with a fine tip and plenty of flux, and inspected by eye afterwards. They were introduced largely so automated optical inspection could see the joints, and hand assembly gets the benefit for nothing. Worth looking for on the datasheet before assuming a QFN is out of reach.
- BGA: a grid of solder balls entirely underneath. ⛔ Not hand-solderable in any meaningful sense, and there is no side-wettable equivalent to rescue it. This needs an oven with a controlled temperature profile, and inspection needs X-ray.
⚠️ The rule of thumb, refined: if you can see the joint, you can probably solder it. That is why side-wettable flanks matter, because they turn an invisible joint into a visible one.
12.3 Why this course started you on through-hole
The aim of this section is to justify a choice, because it was deliberate and the reasons are good.
It was not because through-hole is easier in some vague way. It was for four specific reasons.
You can see what is happening. Every idea in this course, wetting, the fillet shape, the hollow flank, is visible on a through-hole joint at ordinary reading distance. On an 0603 part, the same process is happening in a space smaller than a grain of rice, and you would be learning to recognize something you cannot actually see. You cannot learn to read joints you cannot see.
The board holds the part for you. A through-hole component is mechanically located by the board before you solder anything. A surface-mount part sits loose on top and will slide, tilt or stick to your tip. That is a coordination problem stacked on top of a technique problem, and learning both at once is unnecessarily hard.
Mistakes are recoverable. Lesson 10 showed you can remove almost anything from a through-hole board with braid and a pump. Surface-mount rework, particularly on a package with hidden joints, often needs hot air.
The skills transfer completely. Everything you have learned still applies: heat the joint not the solder, flux is what makes it flow, wetting is the test, shape not shine. Surface mount is not a different craft. It is the same craft at a smaller scale with different handling.
12.4 What it takes, honestly
The aim of this section is a straight answer about equipment, without either gatekeeping or pretending.
If you want to try surface mount, here is what genuinely changes.
What you already have that still works
Your temperature-controlled iron is fine. A fine chisel tip is still the right choice, and this surprises people who assume small parts need a needle point. Chisel wins for the same reason as Section 2.3: contact area.
What you will actually need to add
- Tweezers. Fine, straight, non-magnetic. Genuinely essential rather than a nice-to-have, since your fingers cannot hold an 0805.
- Magnification. Reading glasses at 1206, a headband magnifier or a lamp below that. Fix your lighting first, as Lesson 4 said.
- Flux, separately. This becomes non-negotiable rather than optional. Surface-mount work uses far more flux than through-hole, and most of the technique is really flux management.
- Fine solder. 0.4 mm or thereabouts.
- Braid, thin, for the drag-soldering technique and for clearing bridges, which happen constantly.
What you need only for the harder packages
- Hot air. For QFN, DFN, and any rework where you must melt every joint at once. This is the real dividing line.
- Solder paste and a stencil, if you build whole boards rather than repairing them.
- A hotplate or a reflow oven.
- A microscope, at the small end. This is the one item that is genuinely necessary rather than a convenience, and it is why Lesson 4 put it in the "do not rush" list rather than the "never" list.
The two techniques worth knowing the names of
Tack and solder. For a two-terminal part: put a small amount of solder on one pad only, hold the part in place with tweezers, melt that one pad so the part is tacked down, then solder the other end properly, then go back and redo the first. Nearly all hand SMD work is this.
Drag soldering. For a chip with many fine legs: flux the whole row generously, put a small bead of solder on the tip, and drag it steadily along the legs. Surface tension pulls solder onto each joint and leaves the gaps clear. It looks like it should produce a solid bridge and it does not, provided there is enough flux. Any bridges that do form come off with braid.
Figure 12.2: Tack one end, then solder the other properly, then go back and redo the first. Almost all hand surface-mount work is this one move.
12.5 Should you bother
The aim of this section is a straight recommendation rather than a shrug.
If you want to repair modern devices: yes, eventually. Phones, laptops, appliances and games consoles are surface mount throughout, and no amount of through-hole skill will let you replace a failed capacitor on a laptop board. Repair is the strongest reason to go there.
If you want to build kits and projects: not urgently. There is an enormous amount of genuinely rewarding through-hole work available, and plenty of well-designed boards deliberately do the difficult surface-mount work at the factory and leave the through-hole for you.
That is worth knowing as a general pattern rather than as a product point. A well-designed kit will often have its fine surface-mount parts pre-soldered, precisely because those are the parts a beginner cannot reasonably do and would not learn much from. It is a sign of a thoughtfully designed kit rather than a limitation.
If you are curious: buy an SMD practice board. They cost very little, they come with a bag of parts and a board full of footprints in descending sizes, and an hour with one will tell you more than any article. Start at 1206 and work down until it stops being fun. That point is different for everyone and it is worth finding yours.
What this sets up
Lesson 13, what to build next, is the last lesson, and it is about turning all of this into something you own rather than something you know.
Frequently asked questions
What is the difference between through-hole and SMD? Through-hole passes component legs through drilled holes and solders them on the far side. Surface mount sits components on top of pads with nothing passing through. Surface mount allows smaller, denser, double-sided boards and, crucially, machine assembly.
Can you solder SMD components by hand? Yes, up to a point. 1206 and 0805 parts are genuinely comfortable with an ordinary iron. 0603 is doable. Chips with visible legs like SOIC and QFP are hand-solderable. QFN packages are difficult, though many modern ones have side-wettable flanks that expose a small vertical pad at each corner and can be hand-soldered and inspected. BGA is not realistically hand work at all.
What size SMD can a beginner solder? Start at 1206, which is about 3.2 by 1.6 mm and much more approachable than it looks. Work down from there until it stops being enjoyable.
Do I need hot air for surface mount? Not for parts with visible joints, which an ordinary iron handles. Hot air becomes necessary for packages where the joints are underneath the component, and for most surface-mount rework.
What do the numbers like 0805 mean? They are dimensions in hundredths of an inch. 0805 is 0.08 by 0.05 inches, roughly 2.0 by 1.25 mm.
Why did this course teach through-hole first? Because you can see what is happening. Every idea here, wetting, the fillet shape, the hollow flank, is visible at reading distance on a through-hole joint. You cannot learn to read joints you cannot see. The board also holds the part for you, and mistakes are recoverable.
Is SMD soldering a completely different skill? No. Same physics, same rules: heat the joint not the solder, flux makes it flow, wetting is the test. What changes is handling, magnification and how much flux you use.
Next
Lesson 13: What to build next. Where to practice, what makes a good first real project, and how to keep going once the course ends.
Soldering Fundamentals is free and always will be. No sign-up, no gate, no catch.
