Supplier photo of an assembled infinity mirror light cube illuminated on a dark background.
Supplier photo
Temu affiliate offer · Project idea

Build an infinity light cube

Listing specifies 5-volt direct current. Soldering required; battery and glue not included.

View kit on Temu

Tricky Circuit may earn a commission through these links. Product ideas have not been bench-tested by Tricky Circuit and are not verified lesson parts lists.

MCP6002 Sensor Buffer: Measure Input Loading and Output Error

A useful buffer should stop the next stage from disturbing a sensor. It should not make us believe every voltage in the circuit is now exact. This experiment first produces a large, predictable loading error, then uses a buffer to reduce it and measures the smaller errors that remain.

The “sensor” is deliberately a pair of fixed resistors. That removes changing light, temperature and moisture from the first test. Once you can explain every reading, a real sensor becomes easier to diagnose. This is a steady-voltage experiment requiring no firmware.

MCP6002-I/P buffer schematic: 100 kΩ sensor divider, channel A follower, 100 Ω output resistor, 10 kΩ test load and correctly terminated channel B.
Original Tricky Circuit schematic. Both amplifiers are inside one MCP6002; repeated SENSE and supply labels denote connected nets. Nominal predictions are not bench measurements. Open full-size schematic

What the MCP6002 contributes

An operational amplifier, or op-amp, responds to the difference between its two inputs. A voltage follower feeds its output back to its inverting, minus-marked input. This negative feedback lets the output settle close to the voltage on the plus-marked input. The load current mostly comes from the amplifier’s supply, rather than through our high-resistance sensor divider.

Use Microchip MCP6002-I/P: two amplifier channels in an eight-pin PDIP, or plastic dual in-line package. I denotes its industrial temperature grade; /P selects this through-hole package. Viewed from above, notch upward, pin 1 is top left; count down that side to 4 and up the other side from 5 to 8. An MCP6001 or MCP6004 has a different pin arrangement.

The documented operating supply range is 1.8–6 V; this project uses 3.3 V. The 7 V absolute maximum is not a working supply target. “Rail-to-rail input/output” means inputs and outputs can approach the supply rails under specified conditions; it does not mean zero error or unlimited output current.

Parts and equipment

  • One Microchip MCP6002-I/P in an eight-pin plastic dual in-line package, breadboard and short wires.
  • Regulated 3.3 V supply with current limiting; R1 and R2: 100 kΩ, 1%; R3: 100 Ω, 1%; removable load R4: 10 kΩ, 1%.
  • C1: 100 nF ceramic and C2: 1 µF ceramic, both rated at least 10 V.
  • A DC voltmeter with known input resistance, preferably 10 MΩ. DC means direct current; here it describes steady voltage measurements.

Every connection

Pin or partConnection and purpose
8 — VDD; 4 — VSSPin 8 to 3.3 V; pin 4 to ground. C1 100 nF and C2 1 µF across these rails, C1 directly beside the chip.
R1 and R23.3 V → R1 100 kΩ → SENSE → R2 100 kΩ → ground.
3 — VINA+SENSE to channel A non-inverting input.
2 — VINA−; 1 — VOUTAJoin pins 2 and 1 directly. This is channel A negative feedback.
Output and loadPin 1 → R3 100 Ω → LOAD node; removable R4 10 kΩ from LOAD to ground.
5 — VINB+SENSE to the second channel’s non-inverting input.
6 — VINB−; 7 — VOUTBJoin 6 and 7 directly; leave pin 7 otherwise unloaded. Never tie it directly to ground.

VDD labels the positive supply and VSS the ground-side supply in this circuit. VINA and VOUTA label channel A input and output; B identifies channel B. Both channels share pins 8 and 4. A floating input has no defined voltage. The second channel is not floating: it follows SENSE with no external load. Short the output to the minus input, not to the plus input. Keeping the feedback wire short reduces opportunities for noise pickup. Supply bypass capacitors support local current demand; they do not belong across the signal output.

Predict the unbuffered result

With nothing connected to SENSE, the ideal divider gives 3.3 × 100/(100 + 100) = 1.65 V. The divider draws 3.3/200000 = 16.5 µA. Its equivalent source resistance is the parallel combination of the two 100 kΩ resistors: 50 kΩ. That number tells us why a modest load causes a large change.

Temporarily connect R4, 10 kΩ, directly from SENSE to ground. The lower resistance becomes 100 kΩ in parallel with 10 kΩ, or 9.091 kΩ. The new voltage is only 0.275 V. The same answer follows from 1.65 × 10/(50 + 10). This is a wiring consequence, not evidence that the supply suddenly failed.

A voltmeter with 10 MΩ input resistance is not an open circuit. It reduces an otherwise unloaded SENSE to about 1.642 V. A 1 MΩ meter reads about 1.571 V. Know your instrument before blaming an 8 mV difference on the amplifier. Use the same meter and range for the comparisons, and record where it was connected.

Move the load after the buffer

Remove R4 from SENSE. Attach it instead between LOAD and ground, after the 100 Ω output resistor. The ideal amplifier pin now stays at 1.65 V, while LOAD becomes 1.65 × 10000/(10000 + 100) = 1.634 V. Load current is about 0.163 mA. The approximately 16.3 mV drop across R3 is expected; it is not input loading returning through the sensor.

R3 is an output isolation resistor. It reduces direct exposure of the feedback loop to capacitance in leads or the next stage. Our feedback senses pin 1 before R3, so it does not correct R3’s voltage drop. A 100 Ω starting value does not certify stability with any cable or capacitor; this exercise uses short breadboard wiring and a resistive load. Check the exact datasheet and the real response before driving a long cable or a converter’s sampling input.

These numbers ignore amplifier offset and meter loading unless stated. When the meter measures SENSE, its load changes the voltage the follower receives; when it measures LOAD, its 10 MΩ resistance is effectively in parallel with R4 and has much less influence. Simultaneous readings require a model with both meters included.

A useful error budget

Even before the amplifier, two independent 1% divider resistors can give approximately 1.6335–1.6665 V with an exact 3.3 V supply. That ±16.5 mV range is larger than many small errors a beginner may try to “calibrate out.” Measure the resistors and supply if you want a tighter prediction. The buffer cannot correct their ratio.

Input offset is the small input-voltage difference needed for balance; in a follower it appears roughly as an output error. Input bias current is the tiny current entering or leaving each input. Do not treat typical room-temperature values as worst-case guarantees. Temperature, contamination on a breadboard and leakage through a sensor connector can matter in a high-resistance circuit.

Output headroom is the voltage kept between the working signal and a supply rail so the amplifier can still regulate. Start around 1.65 V, comfortably between 0 and 3.3 V. The datasheet’s 25 mV output-swing figures use a 5.5 V supply and a particular load to mid-supply; they are not a promise that this 3.3 V breadboard will track within 25 mV of both rails. Keep both inputs within our 0–3.3 V circuit rails; do not use absolute-maximum input limits as normal signal levels.

Run a comparison you can reproduce

  1. Build the resistor divider first. With a current-limited 3.3 V supply, measure SENSE unloaded by R4, then with R4 directly to ground. Write down both readings and the meter input resistance.
  2. Disconnect power, remove R4 from SENSE and wire the MCP6002. Verify the notch, all eight pins, feedback, and bypass capacitor. Set a modest supply current limit, for example 10 mA.
  3. Power up without R4. Measure SENSE and pin 1 separately. They should be close, but moving one meter can change the test conditions slightly.
  4. Power off, connect R4 at LOAD, then power on. Record SENSE, pin 1 and LOAD. Expect a large improvement compared with the directly loaded divider, plus a small pin-1-to-LOAD drop.
  5. Measure the voltage across R3, keeping the meter in voltage mode. Dividing that voltage by 100 Ω estimates load current without moving a lead into a current socket.
  6. Power off and replace R4 with 100 kΩ if available. The ideal LOAD becomes about 1.648 V. Explain why a lighter load reduces the isolation resistor’s drop.

Check faults without damaging the amplifier

If pin 1 sits near a rail, disconnect power and inspect the feedback path, plus/minus input assignment and shared ground. A follower with missing feedback is no longer a follower. If SENSE falls sharply only when R4 is fitted, check that R4 actually reaches LOAD rather than SENSE. If the supply current limit operates or the chip warms, stop and inspect orientation and shorts.

Do not remove the amplifier supply while leaving an independently powered sensor connected: this schematic is not an overvoltage or power-sequencing protection circuit. Do not attach a motor, loudspeaker, unresisted LED or large capacitor to the output. Stable DC readings alone do not prove fast-waveform fidelity or the absence of high-frequency oscillation; those require appropriate instruments.

Challenge: why not replace the 100 kΩ divider resistors with 1 kΩ parts instead?

That would make the divider much harder to load, but its standing current rises from 16.5 µA to 1.65 mA. The buffer also consumes supply current, so a battery design needs the actual total-current comparison. A buffer is useful for separating a high-resistance source from its load, not automatically the lowest-power solution.

Connect the idea to real sensors

A resistive temperature or light sensor can replace one divider resistor in a later project. First establish its voltage range, resistance range and self-heating limits. Then check whether the amplifier, supply and following input suit those extremes. This experiment’s distinct result is the separation of input loading from output-path error, not a claim that any sensor can connect directly.

Record the chip marking, supply, measured resistances, meter model/input resistance and all comparison readings. Source review and local calculations are complete; this circuit has not been physically tested. It is a useful foundation for a future low-voltage measurement instrument, not a finished protected multimeter input.

Sources and related lessons

Continue with why a voltage divider drops under load, measuring current through a resistor with a voltmeter, and the interactive circuit lessons.

📡Broadcast the signal — amplify the connection.

Leave a Reply

Temu affiliate offers

Find your next electronics project

Explore these do-it-yourself project kits. Check the selected option, included parts and instructions on each offer.

Tricky Circuit may earn a commission if you buy through these Temu affiliate links. These are product ideas; they have not been bench-tested by Tricky Circuit and are not verified lesson parts lists.

Supplier photo of a breadboard, jumper wires and assorted electronic components in a plastic storage case.

Supplier product photo

Try breadboard experiments

Breadboard, wires and assorted parts. Arduino controller board not included.

View kit on Temu

Supplier photo of an assembled blue surface-mount soldering practice board with a circular ring of small lights.

Supplier product photo

Solder a chasing-light board

Surface-mount soldering practice. Battery and battery holder not included.

View kit on Temu

Supplier photo of an assembled two-wheeled line-following car with a blue circuit board, two motors and an empty battery holder.

Supplier product photo

Build a line-following car

Soldering and assembly required. Check battery and track requirements before buying.

View kit on Temu