DRV8884PWPR
Product Specifications
DRV8884PWPR
Texas Instruments — Power Management & Signal Chain
Product Overview
The DRV8884PWPR 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
DRV8884PWPR: a 1 A stepper bridge with the sense resistors removed
The DRV8884PWPR is a two bridge stepper driver with an integrated indexer and integrated current sense, published ACTIVE on a 2000 piece reel. It drives one bipolar stepper at up to 1.0 A full scale or 0.7 A rms per bridge, from an 8 to 37 V rail, with microstepping up to 1/16 and a simple step and direction interface. The two sense resistors that a discrete driver needs are inside the package, replaced by a reference resistor on the RREF pin that sets the full scale current for both bridges.
The consequence is a smaller board and one less source of mismatch between the two phases. The cost is that the current accuracy depends on the internal sense rather than on a pair of 1 per cent resistors the customer chooses: the data sheet quotes 6.25 per cent full scale accuracy, and the trip accuracy degrades at low current settings, reaching 25 per cent when the setting is between 10 and 20 per cent of full scale. A mechanism that needs accurate holding current at a tenth of full scale should know that before it specifies this part.
The DRV8884PWPR rows as the addendum publishes them
| Part class | Stepper motor driver, two H bridges, integrated indexer |
| Vendor status | ACTIVE, production data |
| Data sheet | SLVSDA5E, January 2016, revised March 2020 |
| Operating voltage | 8 V to 37 V |
| Full scale current | 1.0 A per bridge |
| RMS current | 0.7 A per bridge at 24 V and 25 degrees ambient |
| Microstepping | Full step to 1/16, non circular and standard modes |
| Current accuracy | 6.25 per cent of full scale |
| Trip accuracy, 10 to 20 per cent | 25 per cent |
| RDS(on), high side | 716 mΩ typical, 798 mΩ maximum at 24 V and 1 A |
| RDS(on), low side | 684 mΩ typical, 749 mΩ maximum |
| RDS(on), both FETs | 1.4 Ω at 24 V and 25 degrees |
| Reference transimpedance | 30 kΩ, 28.1 minimum, 31.9 maximum |
| Reference pin voltage | 1.232 V at RREF between 27 kΩ and 132 kΩ |
| PWM off time | 20 µs |
| Blanking time | 1.5 µs |
| Supply current | 5 mA typical, 8 mA maximum, no motor load |
| Sleep current | 20 µA at 25 degrees, 40 µA at 125 degrees |
| Wake time | 0.85 ms typical, 1.5 ms maximum |
| Internal regulator | 3.3 V at 1 mA, 5 V analog rail |
| Charge pump | 5.5 V above VM |
| Step propagation delay | 1.2 µs |
| Body, HTSSOP | 24 pins, 7.80 by 4.40 millimetres |
| Body, WQFN | 28 pins, 5.50 by 3.50 millimetres |
| Reel row | DRV8884PWPR, 2000 pieces, large tape and reel, Active |
| Quad flat row | DRV8884RHRR, 3000 pieces, level 1 |
| Obsolete rows | DRV8884PWP tube and DRV8884RHRT tray, no quantity |
| Marking | DRV8884 |
The two live reels and two obsolete alternatives tell a story about how this part is bought. The tube and the tray were closed, which means every live order ships on tape or in a reel. The WQFN row at 3000 pieces is level 1, so it carries no floor life limit, while the HTSSOP reel is level 3 with the 168 hour window.
Choosing the reference resistor
The full scale current is set by the voltage forced across the internal sense path, divided by the transimpedance of 30 kΩ. With the 1.232 V reference pin, a 27 kΩ resistor sets the largest current the part allows, and a 132 kΩ resistor sets the smallest. Between those two the relationship is linear in the resistor value, and a 1 per cent resistor holds the setting; a 5 per cent part moves the full scale current by the same 5 per cent, which is on top of the 6.25 per cent internal accuracy.
Because both bridges share one reference, the two phases cannot be trimmed separately. That is fine for a stepper, where the two phases are meant to match, and awkward for a mechanism that uses the second bridge for a different load. In that case the second bridge can be disabled and the load driven differently, or the part can be replaced by two single bridge drivers.
Microstepping and the torque question
Microstepping up to 1/16 divides the full scale current into sixteen sine steps per electrical cycle, which is what makes a stepper quiet at low speed. The catch is that the current accuracy applies to the sum of the steps: a 6.25 per cent error at full scale becomes a larger fraction of a single microstep, so the position error at 1/16 is worse than the same motor at full step. For a printer carriage that is acceptable; for a mechanism that counts microsteps to hold a position, the error budget has to include it.
The decay mode matters as much as the step size. Slow decay holds current better at low speed and produces more audible noise; mixed decay reduces the distortion at the zero crossing, where the current has to change direction. The DECAY pin is quad level, set by a resistor, so one board can serve two mechanisms with a different resistor fitted.
Thermal behaviour at 1.4 Ω
Two bridges at 0.7 A rms each dissipate about 1.4 W in the FETs at 25 degrees, and about 2 W at the hot resistance. The HTSSOP pad is 6.4 by 4.4 millimetres and the WQFN pad is smaller but closer to the plane, so the two bodies reach similar junction temperatures on comparable copper. A stepper that holds position continuously at full current is the worst case, because the current never falls to zero; a stepper that moves and then releases the winding runs much cooler.
The 20 µs off time and the 1.5 µs blanking set the chopping frequency near 50 kHz at the typical setting, which is high enough to keep the current ripple small on a 10 mH winding and low enough to keep switching losses modest. It is also high enough to be audible on some motors, which is why the decay mode setting changes the noise more than the step setting does.
Buying DRV8884PWPR
Two thousand pieces on a reel, level 3, and the WQFN alternative at 3000 pieces with no floor life limit. For a printer or a 3D printer main board that uses four of these, a reel is a month of production. For an instrument that uses one, the reel is a two year supply and the date code becomes the oldest part on the shelf.
The obsolete tube and tray rows still appear in drawings from the first years of the part. Any customer still buying tubes has to move to a reel, and the quantity step is 2000 rather than 25, which is a purchasing change rather than an engineering one.
Questions that arrive with this part number
Whether the integrated sense is accurate enough to replace two external sense resistors is the first question. At full scale it is within 6.25 per cent, which is enough for torque and not enough for a current measurement used in a calibration. The second is whether the part can drive two independent DC motors instead of a stepper. It can, in full step mode with the indexer bypassed, but the two bridges then share one current setting.
The third is the difference between this part and the DRV8889, which has the same footprint at 24 pins but adds SPI, 1/256 microstepping and stall detection. The 8884 is the lower cost part for a mechanism that does not need diagnostics, and the two are not pin compatible in function even where the package matches.
One practical note on the reference resistor: it carries a small current and its value sets the full scale, so it should be placed close to the pin and away from the charge pump capacitors, which are the noisiest nodes on the board.
For a build that is choosing between the stepper rows, the analogue cross reference page includes the driver families, and the obsolete part hub covers what to do when a tube row closes.