Ultra-Low Power Sleep Mode Comparison: ATtiny402 vs PIC10F202

Project Overview

This project demonstrates and compares the ultra-low power sleep capabilities of two modern microcontrollers - the ATtiny402 and PIC10F202. The test program verifies RAM retention during sleep mode and measures actual power consumption using a capacitor discharge method.

Test Program Functionality

Both microcontrollers run identical test programs that:

1. Blink Pattern: LED blinks N times (0.2s on, 0.2s off), followed by 0.5s pause
2. Sleep Mode: Press button to enter ultra-low power sleep
3. RAM Retention Test: Wake from sleep increments blink count (N+1 blinks)

This pattern proves that RAM contents are preserved even at nanoamp-level sleep currents.

Circuit Configuration

ATtiny402 Circuit (Top)

Recently bought from Farnell, ATTINY402-SSNR, Date code: 2305.

PIC10F202 Circuit (Bottom)

Recently bought from Farnell, PIC10F202T-I/OT, Date code: 2336.

Capacitor

10 µF, 0805 SMD, Digikey 1276-6456-1-ND, measured: 9.1 µF.

Test method

The ATtiny was programmed with avrdude and the serialupdi explained here. After programming, the FTDI adapter was removed.

The 6 pin PIC was soldered to an adapter board for the 8 pin version, and programmed with a MiniPro TL866CS outside of the circuit, and then the adapter board inserted in a breadboard.

For both tests, the following sequence was done:

• connect 3 V from a power supply, the counter was then 1, LED blinks
• the button was pressed, which put it in sleep mode
• pressed again, the counter was 2 and the LED blinks twice in a loop
• button pressed again for sleep mode
• power supply was disconnected
• time measured until voltage reached 2.5 V with the 20 V range of the Fluke 8842A (> 10 gigohm input resistance, measured about 0.65 µF input capacitance)
• connected power supply again
• button pressed to wake it up
• verify that the counter value was retained and now increased to 3 for the LED blink sequence

The Fluke was connected all the time during the tests. To avoid any influence of it, I removed everything completely from the breadboard as well, and measured it again after 10 minutes for the PIC, which resulted in 6 nA instead of 1.6 nA. Probably influenced by the cable or multimeter.

Power Consumption Results

For the ATtiny402, 185 nA sleep current was measured, and for the PIC10F202, 1.6 nA, but with the Fluke connected.

Capacitor Discharge Measurement Method

Theory

When a capacitor powers a circuit, it discharges according to the fundamental equation:

$$I = C \cdot \frac{dV}{dt}$$

Where:

• $I$ = Current (Ampere)
• $C$ = Capacitance (Farad)
• $dV$ = Voltage change (Volt)
• $dt$ = Time interval (seconds)

ATtiny402 Measurement with 10 µF Capacitor

• Voltage drop: 3.0V → 2.5V (ΔV = 0.5V)
• Time: 27 seconds
• Calculation: $$I = 10 \times 10^{-6} \cdot \frac{0.5}{27} = 185.2 \text{ nA}$$

Result: 185 nA (0.185 µA) - Within datasheet limits, typical 100 nA.

With a uCurrent in the 1 mV/nA setting: 148 nA.

PIC10F202 Measurement with 10 µF Capacitor

Test setup for the PIC with the adapter:

• Voltage drop: 3.0V → 2.5V (ΔV = 0.5V)
• Time: 52 minutes (3,120 seconds)
• Calculation: $$I = 10 \times 10^{-6} \cdot \frac{0.5}{3120} = 1.60 \text{ nA}$$

Result: 1.6 nA, but without the Fluke: 6 nA - Still way below datasheet limits, typical 100 nA at 2 V.

I compared it then with a uCurrent in the 1 mV/nA setting, and could measure 7 nA.

Software

Project Structure

The test programs are organized in two MPLAB X projects:

• ATtiny402: attiny402-test.X/
  • Source: main.c
  • Hex file: attiny402-test.X.production.hex
• PIC10F202: pic10f202-test.X/
  • Source: main.c
  • Hex file: pic10f202-test.X.production.hex