Chips & Components

Everything that isn't a breadboard strip lives in the Parts palette on the left: 74LS and CD4000 logic chips, CPUs, 65xx interface chips, memory chips, switches, LEDs, displays, resistors, capacitors, inductors, diodes, transistors, oscillators, the 555 timer, and power/clock bricks. This page covers finding a part, seating it on a board, and the (surprisingly varied) ways different parts rotate and flip once they're down.

The parts palette

The palette opens with every section collapsed, grouped by function:

• BOARDS — the breadboard kits and loose strips, pinned at the top (see The Desk & Breadboards).
• CHIPS — the logic chips, folder-grouped by function, ending with the Timer group (the 555 timer), the Op-amps group (the LM358), the Interface group (the 65xx PIA/VIA, the ULN2003A Darlington array and the 4N35 and PC817 optocouplers) and the PROCESSOR group (the W65C02 and Z80A CPUs). Which logic family it shows — 74LS, CD4000, or both, each in a folder of its own — is Settings → Data Sheets → Chip family (see Choosing a logic family). The 555 belongs to neither family, so it shows in every mode; showing CD4000 alone, it shares the Timer group with the CD4000 timers.
• COMPONENTS — Switches, Resistors, Capacitors, Inductors, Diodes, Transistors, Relays, LEDs, Displays, Audio, Oscillators, Regulators and Power, in that shelf order. Capacitors, Inductors, Diodes and Transistors carry a red (i) beside their names: in the digital engine these parts do only part of what the real ones do, and are there for the export and a complete design (see Inductors, diodes and transistors). With Spice Lite on they are simulated for real, and the (i) goes. Five parts are there only while Spice Lite is on, because the standard engine can't simulate them at all: the speaker (Audio), and the electronic load, mains, the AC adapter and the transformer (Power). One already on a desk stays there either way, and searching for one with Spice Lite off says where to turn it on.
• Memory — the ROM/RAM chips, pulled out of CHIPS into a top-level group of their own.
• ANNOTATIONS — labels and notes (see Probing & Net Names).
• SIGNALS — bench stimulus buttons, pinned at the bottom (see External signals).

With the tray shut, the strip it leaves on the desk's edge carries one icon per section in the same order — click one to open the tray on that section alone. To keep the tray down to one section all the time, turn on Auto-close tray folders in Settings: opening any folder then closes the rest.

The tray's right edge is a drag handle: pull it out to give long part names more room, and the width you leave it at is remembered.

Type in the Filter parts… box to search by id, title, or description — matching, it forces every group open so results aren't hidden behind a collapsed folder. Click any entry to arm placement — a ghost of the part follows your pointer until you click it down on a board, or press Esc to cancel.

Placing a DIP chip

A DIP chip always straddles the trench: half its pins seat in row e, the other half in row f, running the standard counterclockwise DIP numbering — pin 1 at the anchor hole in row e, pins continuing left to right along e, then wrapping back right to left along f. The notch end of the chip (or the dot beside pin 1) marks pin 1 and always faces left. Move the ghost over a pin-board and it snaps to the nearest legal seat; it turns red if the seat is already occupied or falls off the edge of the board.

The wide chips — the 24-, 28-, 32- and 40-pin packages: the memories, the 6502 and its peripherals, the Z80 and the 74LS181 — are 0.6 inch across, twice a small chip, and seat at that true width, their two rows of pins six holes apart across the trench. There are four ways to do that, and the chip takes whichever is nearest the pointer as you place or drag it:

Every one makes the same connections — each half-column is one connected strip — so pick the one that leaves room for your wiring. The chip's body covers the three rows between its pins, so those holes can't take a wire or another part's lead; wire to each pin from the rows outside it instead. A dragged chip moves from one seat to the next as the pointer travels, wherever you took hold of it. A desk saved before wide chips seated this way keeps them in rows e and f just as they were; move one on its own and it takes its true width, bringing the wires plugged into it out from under its body.

Because a chip's footprint is the same two rows no matter how it's turned, placing one is a matter of picking the column — there's no click-to-rotate step while placing a chip the way there is for a rail or a resistor; instead, rotation happens afterward (see below).

Placing a discrete

Most discretes — slide switches, push buttons, toggle buttons, LEDs (in their default horizontal form) and LED bars (bar8) — are linear: they seat along a run of adjacent holes in any single grid row (any of a–j), not just rows e/f. Drop one anywhere its footprint fits and every free hole underneath it is available.

A few parts don't fit that linear model:

• 7-segment displays (seg7cc, common cathode; seg7ca, common anode) are the real single-digit 0.56-in part: ten legs, five along the top and five along the bottom, 0.6 in apart, so each straddles the trench like a wide (600-mil) chip — five legs in row d, five in row h (or b/f, c/g, e/i), its body over the rows between. Pin 1 is bottom-left, counting round anticlockwise: 1 e, 2 d, 3 COM, 4 c, 5 dp, 6 b, 7 a, 8 COM, 9 f, 10 g — the common comes out twice, joined inside, so wire either. Common cathode: COM to ground, drive a segment HIGH (a CD4511B's way); common anode: COM to +V, pull a segment LOW (a 74LS47's). Give each segment its own resistor. R turns it end for end. The older single-row 8-segment digits (seg8cc, seg8ca) are retired: they are gone from the tray, but a desk that has one still loads and runs it.
• bar8iso — the isolated 8-segment LED bar — is packaged as a 16-pin DIP, so it straddles the trench exactly like a chip: anodes A1–A8 in row e, cathodes K1–K8 in row f.
• DIP switch banks (sw-dip1, sw-dip2, sw-dip4, sw-dip8) are likewise DIP-packaged (2/4/8/16 pins for 1/2/4/8 switch positions), straddling the trench the same way: each position's two facing pins — one in row e, one in row f — are its own independent SPST switch.
• The 6 mm tactile button (sw-tact) is the push button in every breadboard tutorial. Its four legs straddle the trench two columns apart: pins 1 and 2 in row e, 4 and 3 facing them in row f. The two legs that face each other across the trench are joined inside the switch, so both halves of each column are one contact. Holding the button joins the left column to the right. The column between the legs is under the body, so nothing can be plugged into those two holes.
• The DPDT push switch (sw-dpdt) is the 8 mm self-locking switch with the blue cap. It is a 6-pin DIP across the trench: pole 1 is pins 1·2·3 in row e and pole 2 is pins 6·5·4 in row f, with each pole's common (COM1, COM2) in the middle. Released, each common is joined to its A pin (1A, 2A, the pins at the notch end). Click it to latch it down, and each common moves to its B pin (1B, 2B). Click again to release it.
• Oscillator cans (osc-full, osc-half) are rigid four-cornered shapes rather than a line of pins — a full can is 7 holes by 4, a half can 4 holes square, with legs only at the four corners. A can can seat anywhere on the grid, including straddling the trench, since its shape (not a row) determines its footprint.
• Character LCD modules (lcd16x2, lcd20x4) are linear — a 16-way header along 16 adjacent holes in one row — but the module itself is much bigger than that row. See below.

Component values

A resistor has a Resistance, a capacitor a Capacitance, an inductor an Inductance, a Zener diode a Zener voltage and a transistor a Part number. Right-click the part, choose Properties…, and either pick from the field's list — the ▾ beside it, or ↓ — or type any value at all, the way it is printed on a parts drawer or a schematic. Typing narrows the list to the entries holding what you have typed (4 leaves 47Ω, 470Ω, 4.7kΩ, 47kΩ and 470kΩ; u finds µ).

A plain number is in the field's own unit — ohms, farads, henries or volts — so 100 in a Capacitance is a hundred farads, which is out of range and says so. The prefixes are p n u/µ m k M G, or spelled out (pico … giga, and meg as SPICE writes it). A lowercase m is always milli and an uppercase M always mega: 1m on a resistor is a milliohm, refused as out of range rather than guessed at. A prefix letter — or R for ohms, V for volts — can stand in for the decimal point (4k7, 2R2, 4n7, 5V1). A unit belonging to something else is refused by name: 10uF in a Resistance is "a capacitance, not a resistance".

Press Enter or move out of the field to apply it. A value that reads is shown back in its tidy form — 100 kilo ohms becomes 100kΩ — to three figures, or to as many as you typed (4753 is 4.753kΩ; nothing is rounded away). One that doesn't stays as you typed it, in red with the reason under it ("Not a resistance value", "Out of range: 0.1Ω to 100MΩ"), and the part keeps the value it had. Any value in range is taken, on the list or not.

When the value a part holds is not a standard one — one it is actually made in — a small amber ⓘ appears to the right of the field. Click it for the standard values either side (3k offers 2.7kΩ and 3.3kΩ; a value past the end of the series, the one end it is past), and click one to take it. It is advice, never a fault: a value that is not standard is still taken exactly as typed, and a click anywhere else, or Escape, puts the card away. Resistors and capacitors are measured against the E12 series across their whole range (so 4.7n is standard though the list does not offer it), a potentiometer against the track values pots are sold in, and an inductor or a Zener against its own list.

A Zener's list pairs each voltage with its part — 5.1V (1N4733A) — and picking one sets both. Typing a voltage on the list (5.1, 5V1) brings its part number with it, typing a part number on it (1n4742) picks its entry, and any other voltage stands alone, with no part number.

A capacitor's Type (Ceramic or Electrolytic) and a transistor's Type (NPN, PNP, N- or P-channel MOSFET) swap the part where it stands and reopen its card as the new one; like any change to the wiring, the Type is greyed while the circuit runs. The value comes across untouched, so a capacitance outside the new type's range shows red rather than being changed. A transistor's list holds the parts of its own type, and a part number from another type's list is cleared by the swap (one you typed yourself is kept). Typing another type's part number is allowed, with a warning — 2N3906 is a PNP transistor.

An older project's values load as they were. One that cannot be read is kept exactly as written, and shows red when you open the card.

On the desk a resistor wears its colour code — four bands for a value two figures say (4.7k: yellow, violet, red, gold), five for one that needs three (4.99k) — and a capacitor has its value printed on it. Both values appear in the Schematic View, the BOM, and the KiCad export.

A resistor's value changes nothing in the simulation: a resistor is a weak pull whatever it measures, and a lamp is lit or burnt by whether one is in the loop. Only the timing chips read it (below).

Potentiometers

COMPONENTS ▸ Resistors ▸ Potentiometer is a three-pin trimmer: a resistive track between the two outer pins and a wiper on the middle pin that taps it. It seats along one row like a switch, pins 1, W and 3 in three holes side by side. Press R while placing it, or with it selected, to turn it end-for-end: pins 1 and 3 swap ends and its screw swings to the other side.

Its Properties… card has the track's Resistance — picked from the values pots are sold in (100 Ω, 200 Ω, 500 Ω, 1 kΩ … 22 kΩ, 25 kΩ, 47 kΩ, 50 kΩ … 1 MΩ, 2 MΩ) or typed like any resistor's — and the wiper's Position, a slider from 0 % to 100 %. The resistance from the wiper to pin 1 is Position × Resistance, and to pin 3 is what is left of the track:

(for a 100k part). The slider shows exactly that as you move it — the resistance to pin 1 at its left end and to pin 3 at its right, 15k ━●━━━ 85k — and changing the Resistance updates both. It applies as you drag it, while the circuit runs as well, and the brass screw on the part turns to show where the wiper is.

Turning it while the circuit runs

The Properties card dims the desk behind it, so to watch what the knob does, float the slider instead: right-click the pot and choose Float Position Slider, or click the pop-out icon at the end of the Position row in its Properties card. The slider opens in a small window of its own, floating above Chip Hippo, placed off the circuit near the pot. While you point at the window or it has the focus, the pot it turns glows on the desk.

• Drag its title bar to put it anywhere on the screen — beside the app, or on another display. Float as many pots as you like, one window each; floating one that is already open brings its window to the front.
• Turn it by dragging the thumb, with the mouse wheel over the window (1 % a notch), or from the keyboard: the arrow keys step 1 %, Shift+arrow and Page Up/Page Down step 10 %, and Home/End go to either end. Space runs and stops the circuit from the slider's window too.
• Each move is the same change the Properties slider makes, so it applies live and can be undone the same way. The window follows the part: an undo, a new Resistance or a new Name show up in it.
• It stays open through Run and Stop. It closes with its own close button, when the pot is deleted, or when you switch to another desktop or open another project.

Each side with track left behaves exactly like a resistor: a weak pull, and a value the timing chips read — a pot as a 555's RB tunes its rate. A side with none left is the wiper resting on that pin: a plain wire, so the two nets join, and an LED fed through it with nothing else in the loop burns.

Capacitors

COMPONENTS ▸ Capacitors holds two:

• Capacitor (ceramic) — a disc, not polarised, leads one hole apart.
• Capacitor (electrolytic) — a can with a stripe down its negative side, leads one hole apart. Pin 1 is +, pin 2 is −.

Both place, rotate and bend exactly like a resistor (see Rotating & flipping), and they take holes the way any part does, so a wire or a pin can't share one.

A capacitor carries a value, not charge. It does not filter, smooth, decouple or store anything, and it connects nothing: the simulation treats it as open, even when its legs sit across the two rails. What it is for is the timing chips — the 555, the CD4047B, the CD4060B, the CD4098B and the CD4538B read the capacitor (and the resistors) wired to their timing pins and run at the rate the datasheet gives for those values; see Timers. A capacitor still counts as connected when the checks ask whether a net has anything on it.

A reversed electrolytic makes no difference to the simulation. Watch the stripe anyway if you mean to build it.

The first time you place a capacitor a note says all this; Don't show again puts it away for good.

Inductors, diodes and transistors

On the standard engine these parts do as much as a logic simulator can honestly do, and no more: no forward drop, no gain, no thresholds, no Zener breakdown, no inductive kick. Spice Lite simulates all of those. What they always do is export to KiCad as the real part (see Exporting). Every one of them — and the capacitors — has an optional Part number in Properties… (1N4148, 2N2222, 2N7000…). It is printed on the part, listed in the BOM and written to the export, and it changes nothing about how the part behaves.

Inductors (COMPONENTS ▸ Inductors) take an optional Inductance (see Component values); leave it blank for a bare inductor. Its Style is the part it is, seen from above like everything on the desk: a Coil — a toroid standing on edge over its leads, copper wound round a dark ferrite ring — or a Can, a drum in a black sleeve, its value printed on top. Holes between leads is 1 (leads 0.2 in apart, the smallest part), 2 (0.3 in, the default) or 3 (0.4 in, the biggest). On a part lying along a row it moves the second lead, so a size whose lead would land on a taken hole, or off the end of the board, is refused — with the reason under the choice — and the choice is greyed while the circuit runs, like any other change to the wiring. On the standard engine an inductor conducts like a wire: its two leads are one net, whatever it looks like. So one across the rails is a short, exactly as a wire would be. Under Spice Lite one with an Inductance is a real inductor, and its card gains a Winding choice.

Diodes (COMPONENTS ▸ Diodes) are the Diode and the Zener diode. Pin 1 is the anode; pin 2, the cathode, is the end the band marks. A diode is placed across four holes, like a resistor, but its body is shorter, so its leads bend down to three (a10 to a12) — drag a leg in with Option. A diode is one-way: a HIGH on its anode passes to its cathode, and nothing ever passes back. A LOW or undriven anode leaves the cathode to whatever else is on it — so two diodes into one net, with a pull-down resistor there, make a diode-OR. The HIGH passes at the strength it arrived with: from a rail or a chip output it drives the cathode, through a resistor it only pulls it. Like an LED, a diode wired forward straight across two strongly driven nets (rail to rail, or an output into ground) burns. A Zener takes an optional Zener voltage, picked with its part number or typed (see Component values), printed and exported, but in the simulation it is exactly a diode: reverse breakdown is analog, so it regulates nothing here.

The Bridge rectifier (also under COMPONENTS ▸ Diodes) is MCC's DB105: four diodes in a 4-pin DIP, rated 1 A and 600 V. Like the tactile button it straddles the channel on legs two columns apart: + (pin 1, by the notched corner) and − (pin 2) in row e, the two AC pins (~) facing them in row f. Its body sits between the rows, so the hole between each side's legs stays free. Its part number is printed on it and defaults to DB105; set another of the DB101–DB107 family in Properties… (they differ only in the reverse voltage they stand). Inside, each ~ pin conducts to + and − conducts to each ~ pin, so whichever way an AC input swings, + is the top of it and − the bottom. On the standard engine those are four one-way diodes: a HIGH on either ~ pin reaches +, and nothing is rectified, since nothing alternates. Under Spice Lite it rectifies an AC adapter or a transformer. Press R with it selected to turn it half a lap.

Transistors (COMPONENTS ▸ Transistors) are an NPN and a PNP bipolar transistor and an N-channel and a P-channel MOSFET, each standing over three holes in a row with its pin letters printed on it: E B C for the BJTs, S G D for the MOSFETs. A BJT is a TO-92 unless you pick TO-220 (a TIP120 or a TIP31C) under Package in its Properties…; a MOSFET is a TO-220 — the power part, its metal tab behind it — unless you pick TO-92. The package is how it is drawn, listed in the BOM and exported; on the standard engine it is never how it behaves (under Spice Lite it picks the default Grade). Its Type turns it into any of the four where it stands. Picking a part number from the list sets its package (and, under Spice Lite, its grade) to match. Real pinouts differ by part number (a 2N2222 is E·B·C, a BC547 C·B·E, an IRLZ44N G·D·S), so select one and press R to turn it end-for-end.

A TO-92's legs bend, as the real ones do: hold Option (Alt) and drag one leg to any free hole — the other two stay where they are, and the body stands over the middle of its legs. Drag the transistor by its body and it moves whole, every leg keeping its bend; R still turns it end-for-end (pins 1 and 3 trade holes); a leg plugged in beside a chip you drag bends after it, as a resistor's does. Put the three legs back in three holes in a row, in order, and it is the straight part again. A TO-220's legs are a rigid row, so they never bend — and while a TO-92's legs are bent, TO-220 is greyed under Package (hover it for why) until they stand in a row again.

In the simulation each transistor is a switch:

When it is on, the two switched pins are joined as a closed switch joins them; when it is off, they are apart. What an undriven or undefined control does is where the two kinds differ, as they do on a bench:

• A BJT follows its base and remembers nothing: with its base floating or undefined it is off.
• A MOSFET's gate holds charge: it stays in the last state its gate was driven to — off until the gate has been driven at all. So a gate left floating (or driven by a floating CMOS output) keeps the MOSFET on or off rather than passing the confusion down the channel. While it runs on that held charge, the lamp on its face is ringed in amber, its hover says so, and its Properties… card warns you: a real gate leaks that charge away, so tie it to a defined level with a resistor.

While the circuit runs, a transistor's lamp lights while it conducts. Put a resistor in an LED's leg as you would on a bench — with none, the LED burns — and a transistor switched on straight across the rails is a short.

A pin wired only to one of these parts counts as connected, conducting or not: a reversed diode or an off transistor is a valid board, not a floating input. Supplies don't pass through them, though — a chip powered through a diode or a transistor sees no supply and stays unpowered.

Regulators, relays and the electronic load

These are the parts a bench uses to power and switch things rather than to compute. Each does what a logic simulator honestly can on the standard engine, and what it really does under Spice Lite.

Regulators (COMPONENTS ▸ Regulators) are the LM7805, LM7809, LM7812 and LM7815 fixed regulators and the LM317 adjustable one, each a TO-220 standing over three holes in a row. A 78xx's legs are IN · GND · OUT from the left; an LM317's are ADJ · OUT · IN, and two resistors set its output — R1 from OUT to ADJ (240 Ω is usual) and R2 from ADJ to ground, for Vout = 1.25 V × (1 + R2 / R1): 240 Ω and 390 Ω make 3.3 V. Press R while placing one, or with it selected, to turn it end-for-end: its tab swings to the front and its legs read OUT · GND · IN (IN · OUT · ADJ for an LM317) from the left. On the standard engine a regulator's output is a supply — chips on it are powered, like chips on a rail — whenever its input is a supply at least its dropout above what it holds (2 V for a 78xx, 1.7 V for the LM317); below that its output is nothing. Regulators chain: a 7805 fed from a 7812's output works.

The relay (COMPONENTS ▸ Relays) is a 5 V SPDT relay — the blue cube of a Songle SRD — over five holes in a row: COIL+, COIL−, COM, NO, NC (C+ C− COM NO NC on its face). Its Coil voltage in Properties… is 5 V or 12 V. The coil has no polarity. On the standard engine COM is joined to NO while one coil leg is HIGH and the other LOW, and to NC otherwise; its lamp lights while it is pulled in. A real coil wants about 70 mA at 5 V — more than a logic output gives — so drive it from a transistor or a ULN2003A, with a diode across the coil. Press R while placing it, or with it selected, to turn it end-for-end: the cube swings to the other side of its row, NC now at the left.

The electronic load (COMPONENTS ▸ Power) is a bench brick like the power supply, with a red + and a black −: wire them across a supply, a rail or a regulator's output to draw a current from it on purpose. Its Mode is CC — constant current, its Current from 1 mA to 5 A — CR, constant resistance, its Resistance from 0.1 Ω to 100 kΩ — CV, constant voltage, its Voltage from 0.1 V to 100 V — or CP, constant power, its Power from 1 mW to 100 W. The card shows only the mode's own field (the others keep their values), and the badge shows the setting (CV 3.3V, CP 2W). In any mode it draws at most 5 A. It draws nothing on the standard engine (a logic simulator has no currents); under Spice Lite it shows the volts, amps and watts it is taking. It is left out of a KiCad export, since it is an instrument rather than part of the circuit.

The speaker

The speaker (COMPONENTS ▸ Audio, shown while Spice Lite is on) is a small 12 mm PCB speaker on two legs 0.3 in apart, its round body standing over the rows behind them. Press R (while placing it, or with it selected) to turn it a quarter turn about its + leg: 0.3 in is the channel's own width, so stood on end it straddles the trench — one leg in row e, the other in row f, the body over the channel. Put a tone across it and you hear it: a 555 astable, a counter's output, a clock brick. Its coil is only 8 Ω, so feed it through a resistor and a capacitor (47 Ω and 100 µF from a 555's OUT), or from a transistor. Its Impedance is 8, 16 or 32 Ω, and its Volume (0–100 %) sets how loud it plays on top of the power the circuit puts into it. While it plays, the ring round its cone lights and the cone shows the pitch (the circuit's own, "456 Hz"), as does its hover. Its coil is rated 0.5 W: kept past that it warns, and heated to twice that it burns open in smoke and goes silent for the rest of the run (see Speakers).

It plays the circuit's pitch at the run's speed: at ×¼ a 440 Hz tone sounds at 110 Hz, as everything else on the desk runs at a quarter of its speed. Pause, Step or Stop and it falls silent. Its pin-assignments window has an example with four desktops: a fixed Tone, Pitch on a pot, a Key that sounds only while a button is held, and a two-tone Siren.

Character LCD modules

The Displays group holds two HD44780 character-LCD modules: a 16×2 (the standard 1602A) and a 20×4 (the 2004A). They're drawn to their real sizes — 80 × 36 mm and 98 × 60 mm of PCB — and their screens are live: run the circuit, drive the module, and the characters appear on the glass.

Both plug in through a 16-way header along one row, and the pin assignment is identical between them, so what you learn wiring one applies to the other. The header runs along the module's top edge, which means the body hangs below the row it plugs into — so seat one on a bottom row (a) and it clears the board it's plugged into rather than covering it. The placement ghost shows you the whole module, so you can see this before you click.

Driving one is the ordinary HD44780 parallel bus: put a command or character code on DB0–DB7, set RS (0 = instruction, 1 = data) and R/W (0 = write), and pulse E — the byte latches on E's falling edge. Wire VDD/VSS to a 5 V rail. V0 (contrast) and A/K (backlight) are cosmetic in the standard engine; under Spice Lite the backlight is an LED and V0 sets the contrast. During a read the module drives DB0–DB7 itself, so tri-state anything else sharing that bus.

Both modules show the same controller datasheet in their pin-assignments window — it's one document, because RS/R/W/E, the bus and the address maps are the controller's and are identical across the two sizes.

Rotating & flipping

Rotation behavior is not one rule for every part — it depends on what kind of part it is. This is the part worth reading carefully.

Chips (R, mid-drag or while selected). A DIP chip's footprint maps onto itself when flipped — same two rows, same columns — so flipping only reverses which physical pin sits where; the chip never has to move. Select a placed chip and press R to flip it 180° in place, or press R while mid-drag to flip it before you drop it. Either way its pin-assignments window updates to show the new numbering.

bar8iso, the 7-segment displays, DIP switch banks, the tactile button and the DPDT push switch (R while selected or while placing). These are DIP-packaged, so they flip exactly like a chip: R turns the part 180° in place, the same holes, only the pin numbering reverses (a switch bank's own position states don't move — position 1 is still position 1, just wired to the opposite pins now). The drawing turns with it: bar8iso's bar 1 and its pin-1 dot move to the far corner, a digit stands on its head. None has the turn-in-hand behavior of a plain LED — treat them as chips for rotation purposes.

Resistor, capacitor, inductor, diode and LED (R, both while placing and once placed). These two-lead parts start in a horizontal footprint form (pin 1 and pin 2 a fixed span apart along one row). Press R while the ghost is armed to turn it into a vertical two-free-ends form instead: pin 1 stays at the anchor hole, and pin 2 becomes a free lead that can land on any other free hole — including a hole on a different strip entirely, such as reaching up to a power rail. Each R press while placing steps the ghost a further quarter turn, cycling through all four compass directions before repeating. Once the part is placed, select it and press R to rotate it 90° at a time — pin 1 stays put and pin 2's lead swings around it, hunting for the next free hole to land in. Where its own length has no hole that way (a part a row or two from the channel or the board's edge), the lead takes the nearest length that does — never shorter than the part's body allows (a diode's three holes, a resistor's four) — and goes back to its own length on the next turn that has room.

Transistors, regulators, potentiometers, relays and the bar graph (bar8) (R while selected, or while placing). These turn end-for-end in place, as the bussed resistor array does: the same holes, with the pin order reversed (E B C becomes C B E). The body turns with them — a TO-92's flat face, a TO-220's tab, a trimmer's screw, a relay's cube and the bar graph's block all swing to the other side of the row — while the printing on it stays the right way up.

Oscillator cans (R, but the exact step differs by state). A can spins around its own centre, not around one pin, and the step size changes depending on whether you're still placing it:

• While placing, R steps the ghost a full 90° quarter-turn each press, so you can hunt through every orientation for one that fits.
• Once seated and selected, R behaves differently for the two sizes: the square osc-half can still steps 90° at a time, but the rectangular osc-full can jumps straight from 0° to 180° (and 90° to 270°) — because a non-square footprint only has two genuinely distinct orientations once it's down (rotating it a further 90° would just retrace the same two footprints it already swept through while placing).

LED polarity (F, while placing only). Independent of rotation, press F while an LED's placement ghost is armed to flip which lead is the anode and which is the cathode, before you click it down.

Moving a seated part — and its wiring

Drag a placed part and it moves with everything plugged into it. Every wire end sitting in a column-half one of the part's pins occupies comes along, keeping its own row and its offset from the part, so the circuit after the move is the circuit before it. A wire with its other end somewhere else stretches — that end stays put — so nothing is quietly disconnected. (Moving the part alone would leave the wires in the holes you laid them in, and what was pin 1's input would now be pin 3's.)

A resistor or LED plugged into the part comes too — by the leg that is connected to it. A leg in one of the part's column-halves is attached to it exactly as a jumper in the next hole along is, so it travels; the other leg stays exactly where it is and the part simply bends around it, the way a resistor's legs bend on a real bench. Plugged in at both legs, it travels whole instead. (A chip's pins can't bend, so a chip next door is never dragged along — it isn't a lead, it's a body.)

Only what is actually connected to the part comes — a wire or a leg in a column the part merely spans without having a pin there (a push button reaches two holes three columns apart, not the one between them) is left alone, because it was never connected to it. And nothing chases further than one step: what travels lands in the same column-half its pin lands in, so a rider never leaves anything of its own stranded behind it.

The drop is all or nothing. If any of those has nowhere to go — its hole is taken by something that isn't moving, it would run off the end of the strip, or the bend would squeeze a resistor's legs closer than its body is long — the part and everything it would have carried turn red, and releasing puts everything back. Half a move would silently cut the connections it left behind. The holes the part and its wiring are leaving count as free, so sliding a chip one column along is fine even though its new pins overlap its old ones. And where the part is drawn when you let go is where it lands: a seat it showed you in green is the seat it takes.

A resistor or LED moves by its body whatever part of it you press — a leg included — both legs together, carrying its wiring like any other part. To move one leg on its own, hold Option (Alt) and drag that leg (see Selecting things): it's a re-bend, and carries nothing — that leg can land in any hole, at any angle, on any strip, so there is nothing for a wire to follow.

Moving a part bare — Shift

Sometimes leaving the wiring behind is the point — you want to shuffle a chip along and rewire it yourself. Hold Shift as you press, then drag: the part moves on its own and every wire end and resistor or LED leg plugged into its columns stays in its hole. Whatever the part's pins land on is checked as usual, so a pin can't land in a hole a wire is still sitting in. It's still one undo step.

Shift works on a whole selection too (see below): only the selected parts move. A selected wire still moves — it was selected — and a selected board still carries everything seated on it. A wire plugged into a power or clock brick always travels with the brick, because it is plugged into the brick itself.

Two details worth knowing:

• What comes along is decided when you press, not when you let go. Once the part is in hand, the set is fixed.
• Moving across the trench takes the wiring over with it, keeping the arrangement. Rows a–e and f–j are separate nodes, so a wire that stayed in its row would be left in the half its pin had just left — it travels the same number of holes the part did instead, so a wire two holes from the part is still two holes from it, on the same side. Drag far enough that the wiring would run off the edge of the board and the drop is refused; a row nearer the trench fits.
• Moving to another board works the same way — the wiring lands in the matching holes over there, and anything the wires reach back to stays where it is. While the part is over the gap between two boards there is nowhere for it to land, so the wiring sits where it really is and turns red until the part is over holes again.

Moving several parts at once

A selection drags as one unit. Draw a box around a group of parts, or add them one at a time with Cmd/Ctrl-click, then drag any one of them: everything selected travels by the same amount, keeping the arrangement exactly as you built it. Power and clock bricks count as parts here, and come too; so do selected wires (moved rigidly, both ends) and selected boards (with everything seated on them).

Everything above applies unchanged, just to the whole group:

• The wiring comes with it — every wire, and every resistor or LED leg, riding any member. A wire between two selected parts travels at both ends; one that leaves the group stretches, keeping its far end where it is; a resistor plugged into a member bends after it exactly as above.
• The thing under your pointer snaps to the hole grid, and everything else moves by exactly the same step.
• The drop is all or nothing. If one part, one wire end, one leg or one board has nowhere to land, everything turns red and releasing puts the lot back — a group that dropped some of its members would be a rearrangement you never asked for.
• The whole move is one undo step.
• Hold Shift as you press to move the selection bare: only what you selected moves, and the wiring plugged into it stays where it is.

A plain click inside a selection narrows it to what you clicked, exactly as clicking a part outside one does.

Occupancy — one hole, one lead

Every hole on a breadboard — and every terminal on a power/clock brick — holds at most one lead, whether that's a chip pin or a wire end. Placing a part checks every one of its derived pin positions against every other part and every wire already on the desk; if any pin would land on an occupied hole, or off the edge of a board entirely, the ghost turns red and the drop is refused. This is the same rule a wire's endpoints follow (see Wiring, Nets & Buses) — chips, discretes, and wires all compete for the same holes, with no separate bookkeeping for any of them.

A rotated resistor or LED's free lead is the one case where a pin can legally resolve to nothing — if you later move or delete the strip under that lead, the part stays exactly where it is and that leg simply floats, unconnected, just as it would on a real bench.

The pin-assignments window

Right-click any placed part — chip, discrete, or brick — and choose Pin Assignment, at the top of its context menu, to open its floating pin-assignments window: a diagram of every pin/terminal and, for most chips, a cropped datasheet excerpt below it. A real chip's diagram stays fixed at its canonical layout no matter how you've flipped it on the desk (it matches the physical part, not the placement); a DIP-packaged discrete — bar8iso, any DIP switch bank, the tactile button or the DPDT push switch — is the exception: its diagram reflects its current R flip, since it has no real notch of its own. See The Chip Library for the full detail on what the window shows and how it sources its datasheet crops.

See also: The Desk & Breadboards for how boards and strips work, and Wiring, Nets & Buses for connecting components together once they're placed.