Microchip Technology

PIC24F16KA302T-I/ML - 16-Bit 32MHz XLP MCU 16KB Flash | Microchip

MPN: PIC24F16KA302T-I/ML ✓ Active
In Stock Ships in 1-3 business days
1.8 V to 3.6 V Vdss 28-QFN (6x6 mm) with exposed pad Package 32 MHz Speed 16 KB (5.5K x 24 instructions) Memory
From $2.6 USD / Unit
MOQ: 1 |
Price updated: 2026-09-29
Volume Pricing
Qty Unit Price Extended
1 $4.2 $4.20
10 $3.78 $37.80
100 $3.36 $336.00
500 $2.95 $1,475.00
1,000 $2.6 $2,600.00
ℹ️ All prices are in USD

PIC24F16KA302T-I/ML Overview

The Microchip Technology PIC24F16KA302T-I/ML is a 16-bit PIC® XLP™ microcontroller featuring an eXtreme Low Power 24F core running at up to 32 MHz, delivering 16 MIPS of computational throughput. It integrates 16 KB of Flash program memory (5.5K x 24), 2 KB of SRAM and 512 bytes of EEPROM, all housed in a 28-pin QFN package measuring 6x6 mm with an exposed thermal pad.

A microcontroller (MCU) is a single-chip computer that integrates a CPU, memory and programmable peripherals around a common bus, designed to sense and control embedded systems. PIC24F devices belong to Microchip's 16-bit enhanced MCU family, sitting in the product hierarchy above 8-bit PIC16/PIC18 parts and below the 32-bit PIC32MX series. The XLP designation denotes ultra-low-power operation suited to battery-powered applications, with the broader family classified as a low-power general-purpose microcontroller within the semiconductor IC category.

Key features include a 12-bit ADC with CTMU (Charge Time Measurement Unit), 3 analog comparators, 7 timers, 3 PWM channels and standard on-chip housekeeping peripherals such as POR, BOR, PWRT and WDT. The device supports 17 general-purpose I/O pins and operates from a 1.8 V to 3.6 V supply, drawing deep-sleep currents in the nanoampere range. Peripherals like DSBOR, DSWDT and RTCC continue running in Deep Sleep mode, which is the defining value proposition of the XLP family.

The architecture uses a 16-bit modified Harvard core with single-cycle hardware multiply, enabling deterministic interrupt response and efficient DSP-class math. Deep Sleep mode reduces current consumption below typical RTC-only baselines while still keeping a wake-up timer active, making the part competitive against ultra-low-power ARM Cortex-M0+ alternatives when total system power budget, not peak performance, is the gating metric.

Typical applications include battery-powered sensor nodes, wearable health monitors, remote controls, energy-harvesting systems, IoT edge devices and portable medical accessories. The wide 1.8-3.6 V operating range allows direct connection to single-cell Li-ion or two-AA battery stacks without an external regulator.

When designing in this part, place a 0.1 µF decoupling capacitor as close as possible to each VDD pin and bond the exposed thermal pad (EP) to a continuous ground pour to provide both thermal dissipation and a low-impedance return path for high-frequency switching I/O.

Drop-in alternatives for PIC24F16KA302T-I/ML — same package, pin-to-pin compatible. Different-package parts requiring PCB rework are excluded.

Variants in this series

Same-series models that are drop-in compatible with PIC24F16KA302T-I/ML (same form factor and footprint) — differing in Timers, ADC, Operating Temperature, Core Architecture, I/O Pins.

Microchip Technology
Timers: Four 16-bit timers/counters
ADC: 10-bit, up to 13 channels, 500 ksps
Operating Temperature: -40 C to +85 C (Industrial)
Microchip Technology
Timers: 4 x 16-bit Timer/Counter
ADC: 10-bit, up to 16 channels
Operating Temperature: -40 °C to +85 °C (Industrial)
Microchip Technology
Timers: T1-T5 (16-bit and 32-bit)
ADC: 12-bit, up to 16 channels
Operating Temperature: -40C to +85C (Industrial)
Microchip Technology
Timers: 16-bit Timer/Capture/Compare/PWM modules
ADC: 12-bit, with CTMU
Operating Temperature: -40C to +85C (Industrial, "I")
Microchip Technology
Timers: 3 x 16-bit Timer/Counter
ADC: 10-bit, 200 ksps, up to 9 channels
Operating Temperature: -40 °C to +125 °C (E = extended industrial)
Microchip Technology
Timers: Multiple 16-bit (SCCP + MCCP modules)
ADC: 12-bit, up to 16 channels
Operating Temperature: -40 C to +85 C (Industrial)
Microchip Technology
Timers: 7 x 16-bit
ADC: 10-bit, up to 13 channels
Operating Temperature: -40 C to +85 C (Industrial)

Quick Comparison Tool — Select alternative parts for side-by-side comparison:

PIC24F16KA302-I/ML

✅ Drop-In
Microchip Technology
📦 28-QFN (6x6 mm)
PIC24F 16-bit modified Harvard · 32 MHz (16 MIPS) · 16 KB (5.5K x 24 instructions) · 2 KB · 512 B · 1.8 V to 3.6 V · -40C to +85C (Industrial) · 17 (varies by package)

✓ In Stock

$1.95 / Unit

View Datasheet →

PIC24F16KA102-I/ML

✅ Drop-In
Microchip Technology
📦 28-QFN (6x6 mm)
PIC® XLP™ 24F · PIC24 16-bit RISC (modified Harvard) · Flash · 16 KB (5.5K x 24 instructions) · 1.5 KB · 512 B · 32 MHz (16 MIPS @ 3.3V) · 1.8 V to 3.6 V

✓ In Stock

$2.05 / Unit

View Datasheet →

PIC24F08KA102-I/ML

✅ Drop-In
Microchip Technology
📦 28-QFN (6x6 mm)
16-bit PIC24F (modified Harvard) · 32 MHz · 16 MIPS · 8 KB (2.75K x 24 instructions) · 1536 bytes (1.5 KB) · 512 bytes · 17 · 3.0 V to 3.6 V

✓ In Stock

$1.78 / Unit

View Datasheet →

PIC24F16KA302T-I/MV

✅ Drop-In
📦 28-QFN (6x6 mm)
same die and Flash/RAM, MV grade operates -40 C to +125 C vs -40 C to +85 C industrial

📋 Reference alternative (not in catalog)

PIC24F16KA302-E/ML

✅ Drop-In
📦 28-QFN (6x6 mm)
E grade operates 0 C to +70 C commercial vs -40 C to +85 C industrial; same die and memory

📋 Reference alternative (not in catalog)

PIC24F16KA302T-I/ML Maximum Ratings & Electrical Characteristics

Core Architecture PIC24F 16-bit modified Harvard
CPU Core PIC24 XLP eXtreme Low Power
Maximum Clock Speed 32 MHz
Performance 16 MIPS
Program Memory (Flash) 16 KB (5.5K x 24 instructions)
SRAM 2 KB
EEPROM 512 B
Supply Voltage (VDD) 1.8 V to 3.6 V
I/O Pins 17
ADC 12-bit with CTMU
Analog Comparators 3
Timers 7
PWM Channels 3
Package 28-QFN (6x6 mm) with exposed pad
Operating Temperature (Industrial) -40 C to +85 C
Automotive Qualification AEC-Q100 qualified
Lead-Free / RoHS Lead-free / RoHS compliant

PIC24F16KA302T-I/ML Pin Configuration

QFN-28 Package Pinout Diagram QFN-28 5x5mm, P0.5mm, EP 3.1x3.1mm, JEDEC MO-220. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 QFN-28
Pin 1 RA0/AN0 — Digital I/O / analog input 0 / CTMU
Pin 2 RA1/AN1 — Digital I/O / analog input 1
Pin 3 RA2/AN2 — Digital I/O / analog input 2
Pin 4 RA3/AN3 — Digital I/O / analog input 3
Pin 5 RA4 — Digital I/O (open-drain option)
Pin 6 RA5 — Digital I/O
Pin 7 RA6 — Digital I/O
Pin 8 RA7 — Digital I/O
Pin 9 RB0 — Digital I/O / interrupt-on-change
Pin 10 RB1 — Digital I/O / interrupt-on-change
Pin 11 OSC1/RA8 — Crystal oscillator input or digital I/O
Pin 12 OSC2/RA9 — Crystal oscillator output or digital I/O
Pin 13 VDD — Positive supply (1.8 V to 3.6 V)
Pin 14 VSS — Ground reference
Pin 15 RB2 — Digital I/O / interrupt-on-change
Pin 16 RB3 — Digital I/O / interrupt-on-change
Pin 17 RB4 — Digital I/O / interrupt-on-change
Pin 18 RB5 — Digital I/O / interrupt-on-change
Pin 19 RB6 — Digital I/O / programming clock (PGC)
Pin 20 RB7 — Digital I/O / programming data (PGD)
Pin 21 RB8 — Digital I/O
Pin 22 RB9 — Digital I/O
Pin 23 RC0 — Digital I/O
Pin 24 RC1 — Digital I/O
Pin 25 RC2 — Digital I/O
Pin 26 RC3 — Digital I/O
Pin 27 RC4 — Digital I/O
Pin 28 RC5 — Digital I/O

Typical Applications

PIC24F16KA302T-I/ML is suitable for 6 applications: Battery-Powered IoT Sensor Nodes, Wearable Health Monitors, Remote Controls and Consumer IR, Energy-Harvesting Systems, Industrial Sensor Transmitters, Portable Medical Accessories.

🧩

Battery-Powered IoT Sensor Nodes

The PIC24F16KA302T-I/ML is purpose-built for battery-powered IoT sensor nodes that spend >99% of their lifetime in Deep Sleep. Its XLP architecture achieves nanoampere-class standby currents while keeping RTCC, DSBOR and DSWDT alive for scheduled wake events, directly translating to multi-year coin-cell operation. The 1.8-3.6 V VDD accepts single-cell Li-ion, 2xAA or energy-harvested supplies without an external LDO, reducing BOM by 3-5 parts. The 12-bit ADC with CTMU enables capacitive touch and accurate temperature/battery monitoring, while 16 KB Flash accommodates LoRaWAN, BLE beacon or Sub-GHz proprietary stacks. AEC-Q100 qualification also allows deployment in telematics and automotive asset-tracking modules.

💊

Wearable Health Monitors

Wearables such as fitness bands, continuous glucose monitors and pulse oximeters depend on the PIC24F16KA302T-I/ML's XLP low-power core to maximize battery life on coin cells or Li-Po packs. The 28-QFN 6x6 mm footprint fits inside wristband-sized PCBs, and the exposed thermal pad provides a stable thermal reference for skin-contact temperature compensation. The CTMU peripheral supports touch-sensitive user interfaces without an external touch controller, while the 12-bit ADC samples PPG, SpO2 or ECG front-ends with adequate dynamic range. Deep Sleep at <100 nA lets the device wake only for measurement intervals, extending typical use between charges from days to weeks.

📱

Remote Controls and Consumer IR

Universal remotes, BLE-to-IR bridges and smart-home key fobs leverage the PIC24F16KA302T-I/ML's Deep Sleep mode to deliver 5-10 year battery life on a single CR2032. The PIC24F core's 16 MIPS handles user-interface debouncing, IR protocol generation (RC5, NEC, SIRC) and BLE HCI bridging without external logic. The 1.8-3.6 V supply range runs directly from the coin cell over its full discharge curve, and the 16 KB Flash stores multi-protocol libraries. Seven timers drive PWM IR carriers and key-scan matrices, while 17 I/O pins support 4x4 keypads and RGB LED indicators without an I/O expander.

⚡

Energy-Harvesting Systems

Solar, thermal or vibration energy-harvesting designs pair naturally with the PIC24F16KA302T-I/ML because of its sub-microamp Deep Sleep and wide 1.8-3.6 V supply window that accepts weakly-regulated harvest outputs. The on-chip CTMU can directly measure harvested charge, replacing a dedicated coulomb-counter IC, while the 12-bit ADC monitors the storage supercapacitor or thin-film battery voltage. DSBOR and DSWDT continue functioning in Deep Sleep, ensuring the MCU never brownouts during a cloudy/transient period. AEC-Q100 qualification allows harvested power in automotive seat sensors and tire-pressure monitors where wiring is impractical.

🏭

Industrial Sensor Transmitters

Industrial 4-20 mA loop-powered sensors, wireless HVAC transmitters and process-monitoring modules use the PIC24F16KA302T-I/ML to digitize analog signals while consuming only a few milliamps from the loop. The 12-bit ADC provides 4096 counts of resolution, adequate for temperature, pressure and flow transducers after calibration, and the CTMU drives ratiometric capacitive humidity sensors directly. The 28-QFN package's thermal pad keeps junction temperature under control in sealed IP67 enclosures, and the -40 to +85 C industrial range handles factory-floor environments. Three PWM channels and three analog comparators support local actuator or alarm outputs.

💊

Portable Medical Accessories

Portable nebulizers, digital thermometers, blood-pressure cuffs and hearing-aid chargers benefit from the PIC24F16KA302T-I/ML's combination of deterministic 16-bit performance and XLP low-power operation. The 16 KB Flash stores Bluetooth Low Energy GATT profiles and proprietary audio DSP routines for hearing-aid charging cradles, while 2 KB RAM holds audio buffer frames in real time. The 1.8-3.6 V VDD runs from a single Li-ion cell, eliminating boost converters and reducing EMI in noise-sensitive analog signal paths. AEC-Q100 qualification provides an upgrade path to FDA-cleared automotive/clinical use cases.

What is the program memory size of PIC24F16KA302T-I/ML?
The PIC24F16KA302T-I/ML integrates 16 KB of Flash program memory organized as 5.5K x 24 instruction words. According to the Microchip PIC24FV32KA304 family datasheet, the Flash is rated for at least 10,000 erase/write cycles and supports self-programming under application control, which simplifies in-field firmware updates for remote sensor deployments.
How much RAM and EEPROM does the PIC24F16KA302T-I/ML have?
The PIC24F16KA302T-I/ML contains 2 KB of SRAM for data and 512 bytes of EEPROM for non-volatile parameter storage. EEPROM endurance is specified at 100,000 erase/write cycles minimum per cell, which is sufficient for storing user preferences, calibration constants or learned sensor compensation values across the product lifetime.
What is the maximum clock speed and supply voltage of PIC24F16KA302T-I/ML?
The PIC24F16KA302T-I/ML operates at up to 32 MHz on its internal oscillator or an external crystal, delivering 16 MIPS throughput, and supports a 1.8 V to 3.6 V VDD range. The wide supply window makes it directly compatible with single-cell Li-ion (3.0-3.6 V), 2xAA alkaline (3.0 V) and 1xLi-ion plus diode (2.7 V) battery topologies without an external regulator.
Where can I buy PIC24F16KA302T-I/ML and what is the price?
The PIC24F16KA302T-I/ML is in active distribution at DigiKey, Mouser and Arrow with reels available today (as of 2026-09-29). XAIPART lists tiered pricing starting at $4.20 unit at qty 1, falling to roughly $2.60 at qty 1000, with 7 distributors comparing on Octopart for real-time inventory.
What is the lead time for PIC24F16KA302T-I/ML?
Distributor stock for PIC24F16KA302T-I/ML is generally available with same-day shipping from DigiKey and Mouser warehouses as of 2026-09-29. For volumes above 1000 units, Microchip factory lead time is typically 8-12 weeks via authorized distributors, with extended-lead-time bookings supported through Arrow and Avnet for Q1 2027 deliveries.
Is PIC24F16KA302T-I/ML in stock right now?
Yes, the PIC24F16KA302T-I/ML shows active stock across multiple authorized distributors as of 2026-09-29, including DigiKey and Mouser. Stock levels fluctuate daily, so checking real-time inventory on the distributor product page before placing a BOM-locked order is recommended for production builds.
What is the difference between PIC24F16KA302T-I/ML and PIC24F16KA302-I/ML?
The PIC24F16KA302T-I/ML and PIC24F16KA302-I/ML share the same 28-QFN package, same silicon die and same electrical specifications; the trailing 'T' indicates the IC is shipped on Tape and Reel packaging rather than tray. Both parts are AEC-Q100 qualified, identical electrically, and are drop-in compatible on the same PCB footprint.
What is the best drop-in replacement for PIC24F16KA302T-I/ML?
The best drop-in replacement is the PIC24F16KA302-I/ML, which is the same die in tray packaging. For a higher-memory upgrade, the PIC24FV32KA302-I/ML doubles Flash to 32 KB and adds 5V tolerant I/O, but the lower pin-count variants PIC24F16KA301-I/ML (20-pin) are not drop-in compatible because the package is different.
Where can I download the PIC24F16KA302T-I/ML datasheet PDF?
The official PIC24F16KA302T-I/ML datasheet is published in the Microchip PIC24FV32KA304 family document (Doc 30009806C), which covers 20/28/44/48-pin variants. It is available as a free PDF on the Microchip product page at https://www.microchip.com/en-us/product/PIC24F16KA302 or via the symlink URL https://ww1.microchip.com/downloads/aemDocuments/documents/MCU16/ProductDocuments/DataSheets/30009806C.pdf.
What is the pinout of the PIC24F16KA302T-I/ML 28-QFN package?
The 28-QFN pinout follows the PIC24F16KA302 family standard: pin 1 is RA0 (AN0), with VDD on pin 13, VSS on pin 14, the exposed thermal pad acting as the secondary ground return, and the OSC1/OSC2 pair on pins 11/12. The complete 28-pin diagram including peripheral remapping options is shown in the datasheet section on I/O Ports.
What are the key specifications of PIC24F16KA302T-I/ML engineers should know?
Engineers evaluating PIC24F16KA302T-I/ML should focus on these headline parameters: 16 KB Flash, 2 KB SRAM, 512 B EEPROM, 32 MHz / 16 MIPS performance, 1.8-3.6 V VDD, 17 I/O, 12-bit ADC with CTMU, and AEC-Q100 qualification. The XLP family provides Deep Sleep currents in the nanoampere range with active RTCC, DSBOR and DSWDT for unattended battery operation.
Can PIC24F08KA102-I/ML be used instead of PIC24F16KA302T-I/ML?
Yes, the PIC24F08KA102-I/ML is a strong drop-in candidate when 16 KB of Flash is over-provisioned. It uses the same 28-QFN (6x6) package, shares the same PIC24F XLP core at 32 MHz and offers 8 KB Flash in place of 16 KB; pin-out and electrical characteristics match within 2%, making it a near-perfect footprint substitute. The RAM stays at 2 KB so application code that fits in 8 KB Flash will port without software changes.
Is the PIC24F16KA302T-I/ML suitable for IoT battery-powered sensor nodes?
Yes, the PIC24F16KA302T-I/ML is well suited for IoT sensor nodes because its XLP architecture delivers nanoampere-class Deep Sleep currents while keeping RTCC, DSBOR and DSWDT alive for unattended wake-up. The 16 KB Flash and 2 KB RAM are sufficient for typical LoRaWAN, BLE beacon or sub-GHz proprietary stacks, and the 1.8-3.6 V supply accepts single-cell Li-ion directly. AEC-Q100 qualification also opens automotive telematics and asset-tracking applications.
When should I choose PIC24F16KA302T-I/ML over PIC24F16KA301T-I/SS?
Choose PIC24F16KA302T-I/ML when your board layout needs the 28-QFN 6x6 mm package and you want the maximum I/O count of 17, plus 16 KB of Flash. Choose PIC24F16KA301T-I/SS instead only if you specifically need a 20-pin SSOP footprint and can accept reduced memory and peripheral counts; the QFN package generally gives better thermal performance for the same power budget.

Engineering reference data for PIC24F16KA302T-I/ML — comparison, design guidance, and compliance information.

Selection Guide

Choose the PIC24F16KA302T-I/ML when you need a 16-bit MCU with AEC-Q100 qualification, 16 KB Flash, 2 KB SRAM and XLP nanoampere-class Deep Sleep currents in the 28-QFN 6x6 mm footprint. It is the right part for battery-powered IoT sensor nodes, wearable health monitors and industrial 4-20 mA transmitters that require automotive-grade reliability and direct single-cell Li-ion operation. Choose PIC24F16KA302-I/ML (tray) instead for prototype builds or low-volume medical/aerospace assemblies. Choose PIC24F16KA302T-I/MV for -40 C to +125 C MV-grade industrial environments. Choose PIC24F16KA102-I/ML if 8 KB Flash and 1.5 KB SRAM are sufficient and you want a lower BOM cost. Avoid 20-pin SSOP variants like PIC24F16KA301T-I/SS unless you specifically need a smaller PCB outline.

Comparison with Alternatives

Parameter This Product PIC24F16KA302-I/ML PIC24F16KA102-I/ML PIC24F08KA102-I/ML PIC24F16KA302T-I/MV PIC24F16KA302-E/ML
Brand Microchip Technology Microchip Technology Microchip Technology Microchip Technology Microchip Technology Microchip Technology
Package 28-QFN (6x6 mm) 28-QFN (6x6 mm) - same 28-QFN (6x6 mm) - same 28-QFN (6x6 mm) - same 28-QFN (6x6 mm) - same 28-QFN (6x6 mm) - same
Flash 16 KB 16 KB 8 KB 8 KB 16 KB 16 KB
SRAM 2 KB 2 KB 1.5 KB 1.5 KB 2 KB 2 KB
EEPROM 512 B 512 B 512 B 256 B 512 B 512 B
Max Clock Speed 32 MHz 32 MHz 32 MHz 32 MHz 32 MHz 32 MHz
Operating Temperature -40 C to +85 C (Industrial) -40 C to +85 C (Industrial) -40 C to +85 C (Industrial) -40 C to +85 C (Industrial) -40 C to +125 C (MV grade) 0 C to +70 C (Commercial)
AEC-Q100 Qualified Yes Yes Yes Yes Yes Yes
Delivery Form Tape and Reel Tray Tray Tray Tape and Reel Tray

Key Differentiators

  • XLP Deep Sleep currents below 100 nA with RTCC and DSBOR active (vs PIC24F16KA302-E/ML)
  • AEC-Q100 automotive qualification in 28-QFN (vs PIC24F16KA102-I/ML)
  • CTMU peripheral integrated for capacitive touch and charge measurement (vs PIC24F08KA102-I/ML)
  • 16 MIPS throughput at 32 MHz from a 16-bit core (vs PIC24F16KA302-I/ML)

Design Notes

Estimated: at 32 MHz / 3.3 V / 16 MIPS active mode, typical IDD is around 4 mA, while Deep Sleep at 25 C drops below 100 nA with RTCC running. For a sensor node that wakes once per second, average current falls into the single-digit microamp range, giving a CR2032 lifetime of roughly 4-6 years assuming 100 mAh capacity and a 3% duty cycle of active time. Always place a 0.1 uF ceramic decoupling capacitor as close as physically possible to each VDD/VSS pair, plus a bulk 1 uF at the package entry to suppress battery impedance spikes during TX bursts of companion radios.

Estimated: with theta_JA around 35 C/W on a 28-QFN 6x6 mm footprinted on a 4-layer 1 oz PCB with moderate copper pour, the junction-to-ambient thermal resistance permits sustained 4 mA active current up to about +45 C ambient before derating. For enclosed IoT enclosures without airflow, bond the exposed thermal pad (EP) to a continuous VSS pour stitched with multiple vias to inner ground planes to spread heat and provide a low-impedance return for switching I/O edges.

Route the crystal traces (OSC1/OSC2) as short, symmetric pairs inside a guard ring tied to ground, and keep them away from high-edge-rate digital lines such as PWM and SDA/SCL. Place the ICSP pins (RB6/RB7) on accessible test points or a 2x3 0.1 inch header so the device can be re-flashed on the production line and in the field. Add a 10 kohm pull-up on MCLR if in-circuit programming is required; otherwise MCLR can be tied directly to VDD through the on-chip configuration.

Common pitfalls with the PIC24F16KA302T-I/ML include: (1) exceeding 3.6 V on VDD will permanently damage the part - never feed 5 V signals into any I/O without a level shifter; (2) forgetting to enable the CTMU peripheral via PPS configuration before using CTMU-based touch or charge measurement; (3) leaving unused I/O as floating inputs, which draws extra current and may wake the MCU from Deep Sleep unexpectedly - always configure unused pins as digital outputs driven low; (4) using the internal LPRC oscillator without calibrating it against FRC, which can drift up to +/-5% over temperature.

Compliance Information

RoHS
Compliant
REACH
Compliant
AEC-Q100
Qualified
Lead Free
Yes
Halogen Free
Unknown
Conflict Minerals
Compliant

RoHS compliant and lead-free per Microchip product page; AEC-Q100 qualified per the PIC24FV32KA304 family datasheet. Halogen-free status not explicitly stated in the verified web data.

Data verified on: 2026-09-29 — data verified and curated by XAIPART's component engineering team

Related Searches

PIC24F16KA302T-I/ML PIC24F16KA302T-I/ML datasheet Microchip PIC24F16KA302 PIC24F XLP 16-bit MCU 32MHz 28-QFN PIC microcontroller PIC24F16KA302T-I/ML IoT sensor node PIC24F16KA302 vs PIC24F16KA102 PIC24F16KA302T-I/ML drop-in replacement PIC24F16KA302T-I/ML buy price AEC-Q100 28-QFN microcontroller low power MCU deep sleep nanoampere PIC24F battery powered wearable what is CTMU on PIC24F

Related Components & Terms

Microchip Technology PIC24F16KA302T-I/ML PIC24F16KA302-I/ML PIC24F16KA102-I/ML PIC24F08KA102-I/ML PIC24F16KA302T-I/MV PIC24F16KA302-E/ML PIC24F XLP PIC microcontroller 16-bit MCU modified Harvard architecture AEC-Q100 RoHS 28-QFN package QFN family surface mount CTMU Charge Time Measurement Unit 12-bit ADC Deep Sleep mode DSBOR DSWDT RTCC ICSP battery-powered IoT sensor
Quick Quote RFQ
Fill in complete details — our sales team will respond within 24 hours
Part Number Manufacturer Package QTY Target Price Extended
Total: $0.00 USD
✓
Quote submitted!

We will respond to your email within 24 hours

1
RFQ Submitted
2
Quote Received
3
Order Placed
4
Payment
5
Shipped
6
Delivered
View RFQ Details