---
title: "What are 1-, 2- and 4-quadrant power supplies?"
topic: "Power quadrants"
answer: "They differ in which way current can flow through the output, and at which polarity. A 1-quadrant supply only pushes current out at positive voltage, a 2-quadrant one can also take it back, and a 4-quadrant one does both at either polarity. You need more than one quadrant only when your load gives power back or its voltage has to cross zero."
url: "https://nearfieldnotes.com/notes/power-quadrants/"
site: "Near Field Notes"
author: "Michael Walker"
note: 21
family: "Power"
published: "2026-11-06"
revised: "2026-11-06"
revision: 1
video: "https://nearfieldnotes.com/videos/power-quadrants/"
---

Note 21 · [Power](https://nearfieldnotes.com/families/power/) · Rev 1 · [Watch the video · 0:10](https://nearfieldnotes.com/videos/power-quadrants/)

Power quadrants

# What are 1-, 2- and 4-quadrant power supplies?

**Short answer:** They differ in which way current can flow through the output, and at which polarity. A 1-quadrant supply only pushes current out at positive voltage, a 2-quadrant one can also take it back, and a 4-quadrant one does both at either polarity. You need more than one quadrant only when your load gives power back or its voltage has to cross zero.

The figure drives a coil with AC. Its voltage leads its current by 75°, so for part of each cycle the coil hands back the energy it stored, and the supply has to take it.

A 1-quadrant supply reaches the coil’s path for 29% of each cycle, a 2-quadrant one for 50%. Only a 4-quadrant supply follows all of it.

**Note 21’s opening figure: A coil on AC, round the plane.** One cycle at 60 Hz

![Note 21’s opening figure: A coil on AC, round the plane](https://nearfieldnotes.com/notes/power-quadrants/fig-opening.png)

*What the figure shows:* A V-I plane: voltage across the supply's output along the bottom, from −25 to 25 V, current out up the side, from −6 to 6 A. The top right and bottom left quadrants are marked Source, the other two Sink. A coil of 10 mH with 1 ohm of winding, driven with 5 A of 60 Hz current, traces an ellipse through all four, counterclockwise, peaking at 19.5 V. Below: a 1-quadrant supply reaches 29% of each cycle, a 2-quadrant one 50%, a 4-quadrant one all of it.

A 10 mH coil with 1 Ω of winding takes 5 A of 60 Hz current from a supply. Each stretch of its path is colored by the fewest quadrants a supply needs to reach it.

**The numbers**

*A 10 mH coil with 1 Ω of winding on 5 A of 60 Hz current. Its voltage leads its current by 75°.*

| Quadrant | The supply | Share of each cycle | Fewest a supply needs |
|---|---|---|---|
| Quadrant 1 | Source | 29% | 1 quadrant |
| Quadrant 2 | Sink | 21% | 4 quadrants |
| Quadrant 3 | Source | 29% | 4 quadrants |
| Quadrant 4 | Sink | 21% | 2 quadrants |

## Three kinds of supply

*Simplified output stages*

Every supply ends in an output stage, and the stage decides where the output can go. These are linear stages, drawn with only the parts that set the quadrants.

A bench supply has one transistor between its positive rail and the output. It can push current out, and that’s all it can do.

A 2-quadrant supply adds a second transistor that pulls the output down to ground, so current can flow back in.

A 4-quadrant supply returns that transistor to a negative rail. Now the output can swing below zero as well, with current flowing either way.

**Fig. 21: Three output stages.**

![Fig. 21: Three output stages](https://nearfieldnotes.com/notes/power-quadrants/fig-21.png)

*What the figure shows:* Figure 21: three simplified output stages side by side. 1 quadrant: one npn transistor from the +25 V rail to the output, driven at its base. It can only push current out. 2 quadrants: the same, plus a pnp transistor from the output down to ground, so it can also pull current in. 4 quadrants: the pnp returns to a −25 V rail instead, so the output can also go negative. Under each, a small V-I plane with the quadrants it reaches filled.

Each step adds the part drawn in its color. Under each stage, a small plane with the quadrants it reaches filled.

**The numbers**

| Stage | Output transistors | Does | Quadrants |
|---|---|---|---|
| 1 quadrant | +25 V rail, one npn transistor to the output | Sources at positive voltage | 1 |
| 2 quadrants | Adds a pnp transistor from the output to ground | Sources and sinks at positive voltage | 1 and 4 |
| 4 quadrants | Returns that pnp transistor to a −25 V rail | Sources and sinks at either polarity | 1 to 4 |

## The voltage-current plane

*Where an output can be*

Plot the voltage across a supply’s output against the current out of its + terminal. At any moment the output sits at one point on that plane, and the power it gives is the product of the two:

`P = V I`

where P: power out of the supply (W); V: voltage across the output (V); I: current out, from the + terminal (A).

`= 20 V × (−5 A) = −100 W`

That point is 20 V with 5 A flowing back in. P is negative, so the supply takes 100 W. A supply sources where P is positive, giving power to the load. It [sinks](https://nearfieldnotes.com/glossary/#g-sink) where P is negative, taking power from it.

This note puts voltage along the bottom, as Keithley’s 2460 datasheet does, and counts the [quadrants](https://nearfieldnotes.com/glossary/#g-quadrant) counterclockwise from the top right. Keysight’s guides put voltage up the side, so their quadrant 2 is this note’s quadrant 4. Source and sink mean the same in both.

**Fig. 21b: Four quadrants.**

![Fig. 21b: Four quadrants](https://nearfieldnotes.com/notes/power-quadrants/fig-21b.png)

*What the figure shows:* Figure 21b: the V-I plane. Voltage across the output runs along the bottom, current out up the side. Quadrant 1, top right, and quadrant 3, bottom left, are sources: they give power. Quadrant 2, top left, and quadrant 4, bottom right, are sinks: they take power. Beside it, the same plane with voltage up the side, as in Keysight's guides: there the sink at positive voltage is quadrant 2.

The large plane is numbered as in this note. The small one puts voltage up the side, each quadrant colored as in the large one. There the sink at positive voltage is quadrant 2.

**The numbers**

| Quadrant here | Voltage | Current | The supply | With voltage up the side |
|---|---|---|---|---|
| 1 | + | Out | Source, gives power | 1 |
| 2 | − | Out | Sink, takes power | 4 |
| 3 | − | In | Source, gives power | 3 |
| 4 | + | In | Sink, takes power | 2 |

## 1 quadrant: current goes one way

*A braking motor on a 24 V supply*

A bench supply’s transistor can only push current out of the positive rail. That works until the load pushes current back. A braking motor does. It turns its motion back into current.

With nowhere to go, that current charges every capacitor on the output, and the voltage climbs until something gives. Here moteus-c1, the motor drive in [Note 20](https://nearfieldnotes.com/notes/fet-choice/), brakes a motor from 3,000 rpm on a 24 V supply.

**Fig. 21c: A braking motor on three supplies.**

![Fig. 21c: A braking motor on three supplies](https://nearfieldnotes.com/notes/power-quadrants/fig-21c.png)

*What the figure shows:* Figure 21c: a motor braked hard from 3,000 rpm on a 24 V supply, three ways. Top, the voltage across the supply's output. Bottom, the current out of it. The bench supply can't take current back: the voltage climbs past the board's 51 V rating at 9.7 ms and peaks at 58.8 V, held there by the board's TVS diode. A 2-quadrant supply that sinks 1 A slows the climb, to a peak of 50.7 V. One that sinks 6 A takes all 5.2 A and holds 24 V.

The supply’s output is the drive’s bus, set to 24 V. The board is rated to 51 V. Its TVS diode breaks down between 54.1 and 58.1 V, and the model uses 56 V. As the motor slows, it returns less than the braking current loses in its windings, so the current turns positive until the motor stops at 48.7 ms.

**The numbers**

*Braked at 0 ms. The motor stops at 48.7 ms.*

| Supply | Highest voltage | Most current taken | At 60 ms | Into the TVS diode |
|---|---|---|---|---|
| Bench supply | 58.8 V | 0.00 A | 53.4 V | 1.11 J |
| 2-quadrant, sinks 1 A | 50.7 V | 1.00 A | 24.0 V | 0.00 J |
| 2-quadrant, sinks 6 A | 24.1 V | 5.18 A | 24.0 V | 0.00 J |

Current pushed back

### Past 51 V in 9.7 ms

**Field marks.** On the bench supply the voltage leaves 24 V as soon as the brake bites. It passes the board’s 51 V rating at 9.7 ms and peaks at 58.8 V, where the TVS diode holds it.

**Look-alike.** A supply holds what it’s set to. A bench supply holds 24 V only while the load draws current. Its control can turn the transistor off, but it can’t pull the output down.

**Range.** The motor stops at 48.7 ms, and the output still sits at 53.4 V. Nothing drains the capacitors but the board’s own logic.

**Habitat.** Motors slowing down, capacitor banks as the supply turns down, and batteries charged above the supply’s setting.

## 2 quadrants: a way back in

*What a sink takes, and where it goes*

Add a transistor that pulls the output down, and the supply can sink. Fig. 21c’s 6 A supply takes all 5.2 A the motor sends back and holds the bus at 24.1 V.

What a supply sinks still has to go somewhere. A linear stage turns it into heat in its pull-down transistor:

`P_2 = V I_s`

where P_2: heat in a 2-quadrant stage as it sinks (W); V: voltage across the output (V); I_s: current the supply sinks (A).

`= 20 V × 5 A = 100 W`

That’s 5 A taken in at 20 V. A [regenerative supply](https://nearfieldnotes.com/glossary/#g-regenerative-supply) uses switching stages instead and returns the power to the grid.

Sink ratings often run small. Keithley’s 2281S-20-6 battery simulator sources 6 A but sinks about 1 A. In Fig. 21c, a 1 A sink only slows the climb, to a peak of 50.7 V.

**Fig. 21d: One 2-quadrant supply’s reach.**

![Fig. 21d: One 2-quadrant supply’s reach](https://nearfieldnotes.com/notes/power-quadrants/fig-21d.png)

*What the figure shows:* Figure 21d: what a Keithley 2281S-20-6 reaches on the V-I plane, to scale, from its datasheet: 0 to 20 V, sourcing up to 6 A and sinking up to 1.02 A. Nothing at negative voltage.

The 2281S-20-6’s ratings from its datasheet, to scale. Its sink limit is a typical continuous average.

**The numbers**

| 2281S-20-6 | Datasheet |
|---|---|
| Voltage | 0 to 20 V |
| Sources | Up to 6 A, 120 W |
| Sinks | Up to 1.02 A, typical, continuous average |

## 4 quadrants: anywhere on the plane

*A linear stage on ±25 V rails*

Return the pull-down to a negative rail, and the output can swing below zero too. With its transistors on ±25 V rails, this stage reaches ±20 V and ±5 A, in all four quadrants.

The negative rail costs heat. Sinking at positive voltage, the pull-down now has the output voltage plus the rail’s 25 V across it:

`P_4 = I_s (V + V_r)`

where P_4: heat in a 4-quadrant stage as it sinks (W); I_s: current the supply sinks (A); V: voltage across the output (V); V_r: how far the negative rail sits below 0 V (V).

`= 5 A × (20 V + 25 V) = 225 W`

That’s the same 5 A taken in at 20 V. The circuit gives the stage 100 W, and the stage burns 225 W.

**Fig. 21e: Heat while sinking 5 A.**

![Fig. 21e: Heat while sinking 5 A](https://nearfieldnotes.com/notes/power-quadrants/fig-21e.png)

*What the figure shows:* Figure 21e: heat in a linear output stage while it sinks 5 A, against the voltage across its output. A 4-quadrant stage on ±25 V rails burns 125 W at 0 V, rising to 225 W at +20 V and at −20 V. A 2-quadrant stage, its pull-down to ground, burns from 0 W at 0 V to 100 W at +20 V: just what it takes in. At +20 V the gap is 125 W, drawn from the −25 V rail.

Linear stages. Below 0 V the 4-quadrant stage’s top transistor does the sinking. A 2-quadrant stage can’t go below 0 V, so its line starts there.

**The numbers**

*Sinking 5 A. Linear stages. Rails at ±25 V for 4 quadrants, +25 V and ground for 2.*

| Output | 4-quadrant stage | 2-quadrant stage | Taken from the circuit |
|---|---|---|---|
| −20 V | 225 W | can’t go there | 100 W |
| −15 V | 200 W | can’t go there | 75 W |
| −10 V | 175 W | can’t go there | 50 W |
| −5 V | 150 W | can’t go there | 25 W |
| 0 V | 125 W | 0 W | 0 W |
| 5 V | 150 W | 25 W | 25 W |
| 10 V | 175 W | 50 W | 50 W |
| 15 V | 200 W | 75 W | 75 W |
| 20 V | 225 W | 100 W | 100 W |

Linear 4-quadrant stage

### 225 W to take 100 W

**Field marks.** Sinking 5 A at +20 V, the pull-down transistor has 45 V across it. It burns 225 W, 100 W from your circuit and 125 W from its own negative rail.

**Look-alike.** A sink burns what it takes. That holds for the 2-quadrant stage, 100 W here. The 4-quadrant stage burns 125 W more at every voltage, because its pull-down returns to −25 V, not ground.

**Range.** At 0 V, sinking 5 A still burns 125 W. Sourcing costs far less: 5 A out at +20 V burns 25 W. At −20 V the two transistors swap roles.

**Habitat.** Linear [SMUs](https://nearfieldnotes.com/glossary/#g-smu) and bipolar supplies, where the output transistors sit on a heatsink. Fig. 21f shows one inside.

## Why each step costs more

*The rest of the bill*

Each quadrant adds hardware. A 2-quadrant supply needs a second output transistor and its drive. What it sinks has to go into a big enough heatsink or through a switching stage back to the grid. Its current has to pass cleanly from one transistor to the other each time it crosses 0 A.

A 4-quadrant supply adds the negative rail. Its control has to stay stable while the load pushes back. It also watches four limits, ±V and ±I, instead of two.

**Fig. 21f: Inside a 4-quadrant instrument.**

![Fig. 21f: Inside a 4-quadrant instrument](https://nearfieldnotes.com/notes/power-quadrants/fig-21f.png)

*What the figure shows:* Figure 21f: photo of a Keithley 2400 SourceMeter with its lid off. A heatsink runs the width of the case along the bottom, with a row of power transistors clamped to it, outlined.

A Keithley 2400 SourceMeter with its lid off. Photo by Dave Jones of EEVblog, CC BY 2.0.

The prices follow. Three Keithley instruments of about 100 W list at $1,710, $5,170 and $11,500. The 2-quadrant one costs 3.0 times the bench supply, the 4-quadrant one 6.7 times.

Not all of that is quadrants. The 2460 is an SMU, a source-measure unit. That is a 4-quadrant supply with a precision meter built in. The meter is part of its price.

**Fig. 21g: Three Keithleys of about 100 W.**

![Fig. 21g: Three Keithleys of about 100 W](https://nearfieldnotes.com/notes/power-quadrants/fig-21g.png)

*What the figure shows:* Figure 21g: list prices of three Keithley instruments of about 100 W, new, on a linear dollar scale: 1 quadrant, the 2200-20-5, $1,710. 2 quadrants, the 2281S-20-6, $5,170. 4 quadrants, the 2460, $11,500.

New, at US distributors, as listed on 6 November 2026. The scale is linear.

**The numbers**

*New, US distributors, listed on 6 November 2026.*

| Quadrants | Keithley | Rated | Price | Of the 1-quadrant | Listed by |
|---|---|---|---|---|---|
| 1 quadrant | 2200-20-5 | 20 V, 5 A, 100 W | $1,710 | 1.0× | Transcat |
| 2 quadrants | 2281S-20-6 | 20 V, 6 A, 120 W, sinks 1.02 A | $5,170 | 3.0× | Transcat |
| 4 quadrants | 2460 | ±100 V, ±7 A, 100 W | $11,500 | 6.7× | Valuetronics |

## Which one you need

*One solar cell, three supplies*

It depends on where your load goes on the plane. A solar cell under a full sweep visits three quadrants. Below 0 V it takes power. Between 0 and 0.69 V it makes power, and above 0.69 V it takes power again.

Sweep one from −1 V to +1 V, and each supply sees only the part it can reach.

**Fig. 21h: One solar cell, three supplies.**

![Fig. 21h: One solar cell, three supplies](https://nearfieldnotes.com/notes/power-quadrants/fig-21h.png)

*What the figure shows:* Figure 21h: one small solar cell in sunlight, the ANYSOLAR KXOB25-14X1F, swept from −1 V to +1 V by three supplies, each limited to 100 mA. Each panel is a V-I plane with the quadrants that supply reaches tinted, the cell's real curve dashed, and what the supply read drawn over it. The bench supply can't take the cell's current, so its output floats to 0.69 V and reads 0 mA for 85% of the sweep: it traces 4% of the curve. The 2-quadrant supply can't go below 0 V and traces 43%. The SMU traces all of it.

A KXOB25-14X1F cell in sunlight. Each supply is limited to 100 mA, and each plane tints the quadrants that supply reaches.

**The numbers**

*Cell: ANYSOLAR KXOB25-14X1F, 0.69 V open circuit, 58.6 mA short circuit. Every supply limited to 100 mA, reached at 0.763 V.*

| Supply | Reaches | Traces | Elsewhere |
|---|---|---|---|
| Bench supply | 1 quadrant | 4% | Floats to 0.69 V and reads 0 mA whenever the cell pushes current out |
| 2-quadrant supply | 2 quadrants | 43% | Stays at 0 V and reads −58.6 mA for every setting below 0 V |
| SMU | 4 quadrants | 100% | Follows the whole curve |

Solar cell sweep

### 4%, 43% and 100% of the curve

**Field marks.** The bench supply can’t take the cell’s current. Its output floats to 0.69 V and reads 0 mA for 85% of the sweep. The 2-quadrant supply stops at 0 V and reads −58.6 mA for every setting below it.

**Look-alike.** A dead cell. The bench supply reads 0 mA, as it would with nothing connected. The cell is fine. It holds the output at 0.69 V, above the setting, and the supply can’t take its current.

**Range.** The part most tests want, where the cell makes power, lies in quadrant 4. A 2-quadrant supply traces all of it, maximum power included.

**Habitat.** Solar cell tests. Reverse bias, below 0 V, takes a 4-quadrant supply.

## Choosing a supply

*Two questions*

Two questions sort it. Does the load ever give power back? Does its voltage cross zero?

Key to supplies

### Which supply does your load need?

1. 1a The load only ever takes power Go to 2
2. 1b It sometimes gives power back Go to 3
3. 2a Its voltage keeps one polarity 1 quadrant
4. 2b Its voltage crosses zero 4 quadrants
5. 3a Its voltage keeps one polarity 2 quadrants
6. 3b Its voltage crosses zero 4 quadrants

A supply with floating outputs, wired the other way round, covers a negative voltage with the same quadrants. At 3a, check the sink rating against the current the load gives back. If it falls short, an electronic load across the output, in constant-voltage mode just above the supply’s setting, takes the rest.

## Trace a load yourself

*Seven loads*

Pick a load and see the path it asks a supply to follow. Each part of the path is colored by the fewest quadrants that reach it.

**Note 21’s explorer: The path it asks for.** A motor braked hard from 3,000 rpm on a 24 V supply that can sink 6 A. The voltage holds while the current swings to −5.2 A and back.

![Note 21’s explorer: The path it asks for](https://nearfieldnotes.com/notes/power-quadrants/fig-explore.png)

*What the figure shows:* A V-I plane with the path the chosen load asks a supply to follow, each part colored by the fewest quadrants a supply needs to reach it, with arrows for the direction and a dot where it starts.

Voltage across the output along the bottom, current out up the side, each scaled to the load. The arrows show which way the load runs, and the dot is where it starts.

**The numbers**

| Load | What it is | Quadrants it visits | Fewest a supply needs |
|---|---|---|---|
| Resistor | A 4 Ω resistor as the supply turns up from 0 to 20 V | 1 | 1 quadrant |
| Charging a cell | One lithium-ion cell charged at constant current, 1 A up to 4.2 V, then at constant voltage until its current falls to 50 mA | 1 | 1 quadrant |
| Acting as a cell | The supply acts as the cell, taking the same constant-current, constant-voltage charge from a charger under test | 4 | 2 quadrants |
| Braking motor | A motor braked hard from 3,000 rpm on a 24 V supply that can sink 6 A. The voltage holds while the current swings to −5.2 A and back. | 1 and 4 | 2 quadrants |
| Capacitor | 4,700 µF on the output as the supply turns it down from 20 V to 0 V, taking 1 A | 4 | 2 quadrants |
| Solar cell | One small solar cell in sunlight swept from −1 V to +1 V, with a 100 mA limit | 1, 3 and 4 | 4 quadrants |
| Coil on AC | A 10 mH coil with 1 Ω of winding, driven with 5 A of 60 Hz current | 1, 2, 3 and 4 | 4 quadrants |

#### Conditions

- **Plane:** Voltage across the supply’s output along the bottom, current out of its + terminal up the side. Quadrants counted counterclockwise from the top right, as in Keithley’s 2460 datasheet.
- **Coil:** 10 mH in series with 1 Ω, carrying 5 A peak at 60 Hz. Its voltage peaks at 19.5 V.
- **Output stages:** Linear push-pull on ±25 V rails, ±20 V and ±5 A out. Each transistor drops what the rail and the output don’t share, with no bias current. The 2-quadrant stage returns its pull-down to ground.
- **Braking motor:** qsim/regen.py’s model of moteus-c1’s bus: twelve 4.7 µF capacitors and an SMBJ48D TVS diode. The diode starts to conduct at 56 V, mid-range of its 54.1–58.1 V breakdown, and reaches 76.3 V at 7.9 A. An mj5208 motor, 330 rpm/V, braked at 20 A from 3,000 rpm. Its rotor’s inertia, its winding resistance and the running losses are assumed. Each supply has 1,000 µF and 10 mΩ at its output, and sinks at most 0, 1 or 6 A.
- **Solar cell:** A single-diode model of the KXOB25-14X1F, fitted to its listed 0.69 V open circuit, 58.6 mA short circuit and 55 mA at 0.56 V. Its ideality factor, 1.2, and its shunt resistance, 300 Ω, are assumed.
- **Loads:** The explorer’s paths come from the same models. The charge is an ideal [constant-current, constant-voltage](https://nearfieldnotes.com/glossary/#g-cc-cv) one: 1 A from 3.0 to 4.2 V, then 4.2 V until the current falls to 50 mA.
- **Prices:** New, from US distributors’ listings on 6 November 2026.
- **Solver:** Python. The motor’s bus is stepped every 0.2 µs. Everything else is in closed form or read off the models’ curves.

Models only. Nothing on this page has been measured.

#### Limits

- Real supplies use more complex output stages, often behind a switching pre-regulator. The heat in Fig. 21e is for a plain linear stage.
- The three supplies in Fig. 21c differ only in what they sink. Real ones also differ in speed and in output capacitance, which sets how fast the voltage climbs.
- The 2281S-20-6’s 1.02 A is a typical continuous average from its datasheet.
- Prices change, and part of an SMU’s price is its meter.

#### Files

- qsim/site.py Every number on this page
- qsim/regen.py The braking motor and moteus-c1’s bus
- qsim/solar.py The solar cell and the three supplies that sweep it
- post/cards.py The linear stage’s heat

#### Sources

- [Keithley 2460 datasheet: 100 V, 7 A, 100 W, four-quadrant, and its power envelope](https://download.tek.com/datasheet/2460_Datasheet_1KW-60905-0.pdf)
- [Keithley 2281S datasheet: the 2281S-20-6’s 20 V, 6 A, 120 W, and 1.02 A sink](https://download.tek.com/datasheet/1KW-60206-0_2281S_Datasheet_050119.pdf)
- [Keithley Series 2200 specifications: the 2200-20-5’s 20 V, 5 A, 100 W](https://download.tek.com/manual/SPEC-2200_B_May_2020.doc_SPEC-2200B.pdf)
- [Keysight N6700 specifications guide, edition 13: output quadrants with voltage up the side, p. 47](https://www.keysight.com/us/en/assets/9018-04533/technical-specifications/9018-04533.pdf)
- [Ed Brorein, Keysight’s Watt’s Up? blog, 2012: a supply sinking current “is operating in quadrant 2”](https://wattsup-powersupply.blogspot.com/2012/12/power-supply-current-source-to-sink.html)
- [Keysight RP7900 series: two-quadrant, sunk energy returned to the grid](https://www.keysight.com/us/en/products/dc-power-supplies/ate-system-power-supply/rp7900-series-regenerative-programmable-power-supplies-5kw-10kw.html)
- [Transcat, Keithley 2200-20-5 listing, $1,710](https://www.transcat.com/keithley-2200-20-5-power-supply-dc-programmable)
- [Transcat, Keithley 2281S-20-6 listing, $5,170](https://www.transcat.com/keithley-2281s-20-6-dc-supply-and-battery-simulator)
- [Valuetronics, Keithley 2460 listing, new, $11,500](https://valuetronics.com/products/2460-keithley-sourcemeter-new)
- [Dave Jones, Keithley 2400 Source Measure Unit, CC BY 2.0, via Wikimedia Commons](https://commons.wikimedia.org/wiki/File:Keithley_2400_Source_Measure_Unit_(16622193546).jpg)
- [Vishay SMBJ5.0D–SMBJ188D datasheet, document 87606: the SMBJ48D’s 54.1–58.1 V breakdown and 76.3 V at 7.9 A](https://www.vishay.com/docs/87606/smbj5cdthrusmbj120cd.pdf)
- [mjbots, mj5208 product page: 330 Kv](https://mjbots.com/products/mj5208)
- [DigiKey, ANYSOLAR KXOB25-14X1F: 690 mV open circuit, 30.7 mW](https://www.digikey.com/en/products/detail/anysolar-ltd/KXOB25-14X1F-TR/9990479)
- [moteus-c1’s KiCad files, Apache-2.0: its bus capacitors and TVS diode](https://github.com/mjbots/moteus/tree/main/hw/c1)

What should the next note be? [Ask for one](https://nearfieldnotes.com/#ask-for-a-note)

---

Source: https://nearfieldnotes.com/notes/power-quadrants/

Cite as: Michael Walker, “Power quadrants”, Near Field Notes, Note 21, Rev 1, 6 Nov 2026. https://nearfieldnotes.com/notes/power-quadrants/

Figures: free to reuse under CC BY 4.0 (https://nearfieldnotes.com/license/). Each one's image is linked under its name. Credit Michael Walker, Near Field Notes, and link the figure: this address plus #fig- and its number, like https://nearfieldnotes.com/notes/power-quadrants/#fig-21.