You could learn to make a perfect solder joint and still make a mess of your first board.
Not because your technique failed, but because you put a part in backwards, or soldered the tall components before the flat ones and could no longer lay the board down, or hit one particular pad that simply refused to get hot while every other pad on the board behaved perfectly.
None of those are soldering problems. They are reading problems, and they are the reason this lesson sits before the one where you pick up the iron.
The last section is the one that matters most, and it is the one almost nobody is told: some pads on your board will drink heat so fast that correct technique at a correct temperature still will not work. If that happens and you do not know why, you will assume your iron is broken. It is not.
By the end of this lesson you will be able to decode the printing on a board, get every polarized part the right way round, work through a build in an order that keeps everything in place, and recognize a pad that needs a different approach before you fight with it.
7.1 What is actually on the board
The aim of this section is to name the layers you are looking at, because "the green bit" and "the shiny bit" are doing very different jobs.
A circuit board is a sandwich, and from your point of view there are four things worth naming.
The substrate. The rigid base, usually fiberglass, usually the thing people mean by "the board". It is an insulator and it does nothing electrically. It is a chassis.
The copper. Thin sheets of copper bonded to the substrate, etched into the pattern of your circuit. Where it runs in thin lines it is called a track or trace. Where it opens into a large area, it is a pour or plane. This is the circuit.
The solder mask. The colored lacquer over the top, usually green but often black, blue, white or red. It insulates and protects the copper, and crucially it stops solder sticking where it should not. Solder will not wet the mask, which is why a solder bridge is a fault rather than the normal state of affairs. Where the board wants you to solder, the mask is deliberately left off, and that opening is a pad.
The silkscreen. The white printing on top of everything: outlines, letters, numbers, symbols. It is purely information, printed for you, and it has no electrical function whatsoever. Section 7.2 is about reading it.
So a pad is simply a window in the mask where bare metal is exposed for you to solder to. On a through-hole board there is a hole through the middle, and the wall of that hole is plated with copper too, connecting the pad on one face to the pad on the other. That is the plated through hole from Lesson 1.
Figure 7.1: The layers of a board. Substrate, copper, solder mask, and the silkscreen printed on top for you to read.
7.2 Reading the silkscreen
The aim of this section is to make the printing legible, because it is a compressed language that is never explained and is completely consistent once you know it.
Every part on a board has a reference designator, a letter and a number printed beside its outline. The letters are near-universal:
| Letter | Component |
|---|---|
| R | Resistor |
| C | Capacitor |
| D | Diode, and LEDs are diodes, so LEDs too |
| LED | Sometimes used explicitly instead of D |
| Q | Transistor, including MOSFETs |
| U or IC | Integrated circuit, a chip |
| SW | Switch |
| J or CN | Connector or jack |
| Y or X | Crystal or oscillator |
| F | Fuse |
| TP | Test point |
| JP | Jumper |
So R7 is the seventh resistor, C4 the fourth capacitor. Your kit's instructions will refer to parts by these names, and they are printed on the board next to where each one goes. That is the whole system.
Two of those are worth knowing because they are not components at all, and they turn up often on educational boards:
-
TP, a test point. A bare pad with nothing to fit in it, put there deliberately so you can touch a meter probe or a scope lead to a particular signal. ⛔ You are not meant to solder anything to it. Seeing an empty pad and assuming you have lost a part is a common and unnecessary worry. -
JP, a jumper. A pair of pads, sometimes with a thin track already joining them, that select an option: a voltage, an address, a mode. Some are meant to be bridged with solder, some are meant to be cut, and some take a removable link. The board's documentation will tell you which, and this is one of the few places where a deliberate solder bridge is correct rather than a fault.
You will also see:
- Outlines showing the footprint and, usually, the orientation of the body.
-
Values, sometimes printed too:
220R,10 uF,100 nF. - Polarity marks, which section 7.3 covers, and which are the ones that matter.
- A pin 1 indicator on chips: a dot, a notch in the outline, or a squared-off pad among round ones.
⚠️ One genuine trap. Silkscreen is printed on the component side. When you flip the board over to solder, you are looking at the back and everything is mirrored. It is very easy to count three pads along on the solder side and be at the wrong one. Check orientation from the component side, before you turn the board over.
7.3 Polarity, and the parts that care
The aim of this section is to stop the single most annoying beginner mistake, which is not a bad joint but a good joint holding a part in backwards.
Some components work either way round. Some are destroyed, or destroy something else, if you get them wrong. And because you solder them in place, getting it wrong means desoldering, which is Lesson 10 and is much harder than getting it right first time.
Parts that do not care
Resistors, most ceramic capacitors, most switches, and simple two-pin connectors. Fit them any way round.
For resistors it is worth turning them so the color bands all read the same direction anyway. It looks better and it makes checking your work far easier.
Parts that absolutely do care
LEDs. The longer leg is the anode, the positive side. The body also has a flat spot on the rim on the cathode, negative, side. On the board, look for a + marking, a squared-off pad, or a flattened edge in the printed outline matching the flat on the LED. ⚠️ Once you trim the legs, the length clue is gone forever, so check before you cut.
Electrolytic capacitors. The cylindrical ones. Usually a stripe down one side of the can marked with - symbols, and the shorter leg is negative. The board will normally mark +. Fitted backwards, these can fail, and larger ones can fail energetically. Take the extra second.
Diodes. A band printed round one end marks the cathode. The board outline will have a matching bar.
Transistors and MOSFETs. Usually a flat face on one side of the body and a matching flat edge in the printed outline. The amomii Script insert says it well: match the curved face of the component to the curved line on the board.
Integrated circuits. A notch or dot marking pin 1, matching a notch or dot in the silkscreen.
Anything with a plug on it. Connectors, battery holders, barrel jacks, headers with a keyed shroud.
The habit worth building: before you solder anything with a leg length difference, a stripe, a flat, a notch or a dot, look at the board and confirm the match. Say it out loud if it helps. It costs two seconds and it saves twenty minutes of desoldering.
Figure 7.2: The polarity marks worth knowing. A long leg, a flat rim, a printed stripe, a notch, and what each one matches on the board.
7.4 The order to work in
The aim of this section is to give you a sequence that keeps parts where you put them, because gravity is about to become your main opponent.
Remember from Lesson 1: you push parts in from the top, then turn the board over to solder. Which means every part you have fitted and not yet soldered is being held in only by friction, and is about to be upside down.
Shortest to tallest
Work through the board from the flattest components to the tallest.
Resistors and diodes lie almost flat against the board, so when you flip it over, the board rests on the bench and holds them in place for you. Do those first, and gravity is on your side.
If you fit a tall part early, say a big electrolytic capacitor or a connector, the board can no longer lie flat. It rocks on that one tall part, and every flat component you fit afterwards falls straight back out when you turn it over.
So the usual order runs something like: resistors and diodes, then small capacitors, then transistors and chip sockets, then switches and headers, then tall capacitors and connectors, then anything mechanical.
⚠️ If your kit gives you a specific order, follow the kit. Its designers know things about that board that a general rule cannot. amomii's own kit inserts say exactly this, and give a numbered sequence.
Work in batches, and check before you commit
Fit a handful of the same component, bend their legs slightly outward on the underside to hold them, flip, solder them all, then trim.
Bending the legs is the trick that makes this pleasant. A leg splayed out at even a small angle turns a part that falls out into a part that stays put. Do not bend them flat against the board, which makes them hard to desolder later; a few degrees is plenty.
And check orientation before you flip, not after. Once it is soldered, fixing it is Lesson 10.
Trim as you go, or trim at the end
Either. Trimming as you go keeps the board tidy and lets you inspect each joint clearly. Trimming at the end is faster. What you should not do is leave them untrimmed permanently: long legs short against each other, catch on things, and make the board impossible to seat in a case.
Figure 7.3: Shortest to tallest. Flat parts let the board lie down and hold themselves in. One tall part fitted early and everything else falls out.
7.5 Why some pads refuse to get hot
The aim of this section is the one that earns this lesson its place before you pick up the iron. It explains a failure that looks exactly like a broken tool.
You will be soldering happily. Twenty joints, all fine. Then you hit one pad and nothing happens. The solder will not melt properly, or it melts and sits there refusing to flow out. Same iron, same temperature, same technique, same solder, thirty seconds ago it was working.
Your iron is fine. That pad is connected to a lot of copper.
What is going on
Lesson 2 said the iron's job is to move heat into the joint faster than the joint can carry it away. Most pads carry heat away slowly, because they are only joined to thin tracks.
But boards also have ground planes: large open areas of copper, often covering most of a layer, used as a common electrical reference and to spread heat. A pad connected to one of those is thermally attached to an enormous slab of metal, and copper is one of the best heat conductors there is.
When your tip lands on that pad, the heat does not stay there. It runs straight out into the plane and spreads across the whole board. You are not heating a pad any more, you are trying to heat the board, and your iron cannot win that race.
This is exactly the thermal recovery problem from Section 2.1, meeting the worst case it will ever meet.
How to spot one before it fights you
- Look at the underside in a good light. A large uninterrupted expanse of copper under the mask is usually visible as a subtle change in color and reflectivity.
- Ground pins are the usual suspects: the negative side of a battery connector, the barrel of a jack, the shield of a USB port, the mounting pads of a metal connector.
- Look for a thermal relief. Board designers know about this, so a well-designed pad connected to a plane is usually joined by four thin spokes rather than merged into the copper completely. It looks like a small wheel. That is deliberate, and it is there to slow the heat escape so you can solder it. If you see the spokes, someone has already helped you.
- The absence of spokes on a big connector is the warning sign.
What to do about it
In order:
- Give it longer. Genuinely. Some pads need four or five seconds rather than one. Watch for the solder to change from sitting to flowing, and be patient in a way you do not need to be elsewhere.
- Use a bigger tip if you have one. More metal at the tip means more stored heat and a bigger contact area. This is more effective than more degrees.
- Turn the temperature up, but only for that joint. Up to 370 or 380 °C is reasonable here. ⛔ Turn it back down afterwards, for the tip-life reasons in Section 6.4.
- Add a little fresh flux. It helps the solder flow when it finally does get hot enough.
- Preheat, if you have many of them. Warming the whole board gently, even with a hot air source held well back, means the plane is no longer starting from cold. This is more than a beginner needs, but it is the professional answer.
⛔ What not to do: sit on the pad for thirty seconds hoping. If it has not taken in five or six seconds, lift off, let it be, and come back with a bigger tip or more heat. Prolonged heat is how pads lift off boards, which is the one damage in this course that is genuinely difficult to repair.
Figure 7.4: A pad on a thin track heats easily. A pad merged into a ground plane pours your heat into the whole board. The four-spoke thermal relief is the designer helping you.
What this sets up
Lesson 8, your first joint, is the technique, and you are now ready to apply it to a real board rather than to a practice pad.
Lesson 9, good joint bad joint, includes the specific look of a joint that never quite got hot enough, which after this lesson you will be able to explain.
Lesson 10, desoldering and rework, is where you go when you fit something backwards despite section 7.3.
Frequently asked questions
What does the writing on a circuit board mean? The white printing is the silkscreen, and it is purely information. Letters and numbers like R7 or C4 are reference designators naming each part: R for resistor, C for capacitor, D for diode, Q for transistor, U for chip, SW for switch, J for connector.
Which way round does an LED go? The longer leg is positive, the anode. The body has a flat spot on the negative side. Match those to the board's plus mark, squared pad or flattened outline. Check before you trim the legs, because the length clue disappears.
What order should I solder components in? Shortest to tallest. Flat parts like resistors let the board lie flat and hold themselves in when you turn it over. Fitting a tall part early makes the board rock and everything else falls out. If your kit specifies an order, follow it.
Why won't one pad on my board melt solder? It is probably connected to a ground plane, a large area of copper that conducts your heat away as fast as you supply it. Give it longer, use a bigger tip, raise the temperature briefly, and add fresh flux. Your iron is not broken.
What is a thermal relief on a PCB? A pad joined to a large copper area by four thin spokes instead of merging into it. It looks like a small wheel and it exists to slow heat escaping, so that the pad can actually be soldered.
How long should I heat a stubborn pad? Five or six seconds at most in one go. If it has not taken, lift off and come back with more heat or a bigger tip. Prolonged heat lifts pads off the board, which is difficult to repair.
Do I bend the component legs before soldering? A few degrees outward on the underside, yes. It holds the part in place when you turn the board over. Do not fold them flat against the board, which makes later desoldering much harder.
Next
Lesson 8: Your first joint, step by step. The actual technique, one beat at a time, with the timings and what to look for at each moment.
Soldering Fundamentals is free and always will be. No sign-up, no gate, no catch.
