TMS320F28033PNT
Product Specifications
TMS320F28033PNT
Texas Instruments — Power Management & Signal Chain
Product Overview
The TMS320F28033PNT from Texas Instruments is a power management / signal chain IC delivering high efficiency, precision regulation, and robust protection. Texas Instruments (TI) is a global leader in analog ICs, power management solutions, and embedded processors.
Superb Automation partners with leading brands including Qorvo, TI, ADI, XILINX, NVIDIA, Intel, ST, MACOM, Renesas, and Coilcraft to provide complete BOM and supply-chain solutions.
Key Features
- High-efficiency power conversion with low quiescent current
- Wide input voltage range with precision output regulation
- Integrated over-current, over-temperature, and short-circuit protection
- Small footprint packaging for space-constrained designs
- AEC-Q100 qualified options available for automotive applications
- Flexible sequencing and soft-start control
Key Specifications
| Input Voltage | V |
| Output Voltage | V |
| Output Current | A |
| Switching Frequency | kHz |
| Efficiency | % |
| Quiescent Current | µA |
| Topology | — |
| Package | — |
| Operating Temp | °C |
| Protection | — |
Applications
Request Quote — In Stock, 100% New & Original, Full Warranty
TMS320F28033PNT: 60 MHz Piccolo control microcontroller with CLA in an 80-pin LQFP
TMS320F28033PNT is a 32-bit real time control microcontroller with a floating point accelerator in a LQFP body of 80 pins. It runs motor control, digital power and sensing loops at 60 MHz with a control law accelerator that executes maths in parallel with the main core. The PNT order code is the 80-pin LQFP in the T grade, and it is the body with the full 16 channel ADC rather than the reduced channel count of the 56-pin package.
Order code and package
| Order code | TMS320F28033PNT |
| Package | 80-pin LQFP, PN body, 0.5 mm pitch |
| Packing | tray |
| Core | 32-bit C28x at 60 MHz, 16.67 ns instruction cycle |
| Accelerator | control law accelerator present on this device number |
| Memory | 64K by 16 flash, 10K by 16 SARAM, 1K by 16 OTP, 8K by 16 boot ROM |
| Temperature | minus 40 to plus 105 degrees Celsius on the T suffix |
Electrical parameters that decide the application
| ADC | 12-bit, 16 channels on the 80-pin body, 4.6 MSPS, 216.67 ns conversion |
| PWM | 14 ePWM channels with high resolution mode |
| Capture and encoder | one eCAP, one eQEP, high resolution capture |
| Serial | one SCI, two SPI, one I2C, one LIN, one eCAN |
| Supply | single 3.3 V rail with an internal regulator, no power sequencing |
| Code security | 128-bit key and lock over flash, SARAM and OTP |
| Boot | boot ROM with SCI, SPI, I2C, CAN and parallel port loaders |
Family variants and how they differ
| F28030 | 16K flash and 6K SARAM, no accelerator |
| F28031 | 32K flash and 8K SARAM, no accelerator |
| F28033 | 64K flash and 10K SARAM with the accelerator, the part on this page |
| F28035 | 64K flash and 10K SARAM with the accelerator and a second ePWM set |
Points to check before this part goes on a BOM
The order code carries three separate pieces of information: PN is the 80-pin LQFP body, T is the minus 40 to plus 105 degree grade, and the trailing letter group identifies the packing. The same silicon in the PAG 64-pin body has 14 ADC channels and 12 ePWM channels, so a board that needs every analogue input has to stay on the 80-pin body.
Flash is 64K by 16 bits, which is 128 kilobytes, and SARAM is 10K by 16 bits. A motor control product with a field bus stack and a data logger can fill 64K words faster than expected, and the F28035 in the same family has the same memory but more PWM channels rather than more program space. Memory is the first limit to check before choosing between them.
The control law accelerator is a separate 32-bit floating point unit that runs its own code from a dedicated task and triggers on an ADC interrupt. It shares memory with the main core but not the pipeline, so a current loop can run entirely in the accelerator while the main core handles communications. A part from the same family without the accelerator cannot be substituted where that split exists.
The ADC converts in 216.67 ns at 4.6 MSPS, which is 12 samples per 20 kHz PWM period on one channel. At 16 channels shared across three sample and hold groups, the number of conversions per period is what sets the practical control bandwidth, and a product that wants three phase currents plus a DC bus voltage has to plan the trigger order rather than convert everything in one burst.
The single 3.3 V supply with an internal regulator and no power sequencing requirement is a real advantage over older C2000 parts that needed 1.8 V and 3.3 V in a defined order. The consequence is that the internal regulator has to be bypassed correctly: the VREG pin needs its capacitor close to the pin, and a board that copies a 1.8 V external supply arrangement onto this part will not boot reliably.
The 128-bit code security module locks flash, SARAM and OTP. Once the password locations are programmed the part can only be erased through the password, so a production line that programs the wrong image into a locked device loses that device. Many C2000 products ship with the module unlocked and rely on other means, which is a decision that has to be made before the first production batch.
The boot ROM loads from SCI, SPI, I2C, CAN or the parallel port, and the boot mode pins are sampled at reset. A board that straps the boot pins for flash and then wants to reprogram over CAN during end of line test needs a way to change those straps, because the ROM loader is only entered from the pin state and not from a command.
Longevity is the practical question with this family: it has been in production since 2009 and the revision Q data sheet is the current one. Silicon revisions change errata, and the errata list for a given revision is what a safety review asks for. A buyer should record the revision letter marked on the device rather than assume the reel matches an older qualification report.
The converter reference is brought in on a dedicated pin, and its value sets the full scale of every channel. An external reference with a published drift is the normal route for a product that measures current to a specification; the internal reference is adequate for a control loop that only needs repeatability within a run.
The ePWM trip zones are hardware paths that turn the outputs off without firmware. In a motor drive the comparator output and the trip input form a cycle by cycle current limit that protects the bridge while the loop catches up. That path has to be tested by forcing a fault, because a trip that is configured but not wired looks identical to a working one until the first short.
Programming goes through JTAG or through a loader in ROM, and the part is factory fresh with the security module unlocked. A production fixture that locks the device should do it after the functional test, because a locked device with a bad image cannot be erased without the password.
The 80-pin body also brings out the external reference pin and the comparator inputs, which the smaller bodies do not. A product that measures an analogue signal against a threshold in hardware uses those comparators to trip the PWM without a firmware decision, and the trip path is what protects the power stage during a fault.
Where this order code fits
It is used in motor drives, solar inverters, digital power supplies and appliance controllers where a floating point loop has to run beside a field bus. Teams comparing it with a neighbouring part start from our MCU Cross Reference page, and teams replacing a part that has gone end of life start from the Component Lead Time Solutions page.
Questions buyers ask
Is the 64-pin body a substitute? Only if the product gives up two ADC channels and two ePWM channels. The silicon is the same, but the pins are not brought out, so a 16 channel board cannot move to the 64-pin part without a board change.
Does it need an external 1.8 V rail? No. The part runs from a single 3.3 V supply and generates its core voltage internally. That is the main difference from the older F281x generation.
Can the accelerator be used with the main core idle? The accelerator runs tasks triggered by peripheral events and executes independently, so the main core can sleep or handle other work. The code still has to be written for the split; it is not automatic.
What temperature does the T suffix cover? Minus 40 to plus 105 degrees ambient. The same device number in other suffixes covers other ranges, so the suffix is part of the orderable description and not a packing detail.
SUPERB holds and sources this line for production and repair demand, quoted against the exact ordering string with date codes, packing and country of origin stated on the offer. When a part is moving, the BOM Shortage Recovery page sets out the checks that keep a substitute from failing on the bench.