CD4017BE
Product Specifications
CD4017BE
Texas Instruments — Power Management & Signal Chain
Product Overview
The CD4017BE 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
CD4017BE: the decade counter in a 16-lead plastic DIP
CD4017BE is the through-hole member of the CD4017B family, a five stage Johnson counter with ten decoded outputs in a 16-lead plastic DIP, 25 pieces to a tube. The vendor descriptor reads CMOS Decade Counter with 10 Decoded Outputs, the document is the CD4017B and CD4022B data sheet revision C, and the feature line gives a 10 MHz typical counting rate at 10 V, fully static operation, symmetrical output characteristics, a quiescent current tested to maximum at 20 V, and 5 V, 10 V and 15 V parametric ratings. The Schmitt trigger on the clock input is what makes the part usable with a switch, an RC oscillator or a long cable rather than only with a clean logic edge.
The addendum lists CD4017BE as Active in production, PDIP (N) with 16 leads, 25 pieces to a tube, RoHS yes with a nickel-palladium-gold finish, no moisture level because the body is through-hole, minus 55 to plus 125 degrees, and the marking CD4017BE. The same document carries the SOIC rows CD4017BM, which is Obsolete, and CD4017BM96, which is Active at 2500 pieces, plus the SOP row at 2000 pieces and the TSSOP rows, so the DIP is the only format that arrives in a tube of 25.
The CD4017BE row as the vendor addendum publishes it
| Part class | Decade counter, five stage Johnson, ten decoded outputs |
| Vendor descriptor | CMOS Decade Counter with 10 Decoded Outputs |
| Vendor document | CD4017B and CD4022B data sheet revision C |
| Counting rate, vendor feature line | 10 MHz typical at 10 V |
| Operation | Fully static, clock can be stopped indefinitely |
| Clock input | Schmitt trigger, unlimited rise and fall times |
| Inputs | Clock, clock inhibit, reset |
| Outputs | Ten decoded outputs, one high per clock period |
| Carry out | One pulse per ten clock cycles |
| Supply range | 3 V to 18 V |
| Parametric ratings | 5 V, 10 V and 15 V tables |
| Output type | Push pull, balanced outputs |
| Input type | Standard CMOS |
| Quiescent current | Tested to maximum at 20 V on every unit |
| Package for this code | PDIP (N), 16 leads |
| Packing | 25 pieces in a tube |
| Lead finish | NIPDAU |
| Moisture level | Not applicable for this package type |
| Operating range | Minus 55 to plus 125 degrees |
| Status | Active, production |
| Part marking | CD4017BE |
| Same string in the addendum | CD4017BE.A and CD4017BEE4, both 25 piece tube |
| SOIC rows | CD4017BM Obsolete, CD4017BM96 Active at 2500 pieces |
| Ceramic rows | CD4017BF and CD4017BF3A, 25 piece tube |
What happens at power up, and the reset pulse every board needs
The counter is static, so it holds its count with the clock stopped, and it starts from an unknown state when the supply comes up. That is the single most common integration fault with this part: the board powers on, the counter sits on an arbitrary one of the ten outputs, and the sequence starts from the wrong step. The fix is a reset pulse at power up, either from an RC network on the reset pin or from a microcontroller pin, and the reset input is active high and asynchronous, so a short pulse is enough.
The reset pin also needs a defined level when the pulse is not applied. A 100 k resistor to ground holds it low across the temperature range, and the input leakage of 1 microamp at 18 V leaves 0.1 V across that resistor, which is well inside the low threshold at any rail in the range. Leaving the pin open is the second most common fault, because the pin then follows whatever couples into it, which on a board with a relay or a motor is a great deal.
Driving ten outputs from one package
Each decoded output is a push pull CMOS stage with the family drive, so an LED with a series resistor works and a small transistor base works. What does not work is a load that needs more current than the output can give, because the output level then drops and the internal logic sees a disturbed supply. On a 5 V rail a red LED at 3 mA with a 1 k resistor is inside the drive; a 20 mA LED is not, and the usual answer is a transistor per output or a driver array.
Because only one output is high at a time, the total current drawn by the load is the current of one output rather than ten, which is the reason a sequencer built this way can drive ten loads from one package without a thermal problem. The dissipation that does occur sits in the one conducting output and in the package, and a 5 V rail with 5 mA into a load is 25 mW, which no package in this family notices.
Chaining, inhibiting and the count that must not run away
The carry out gives one pulse every ten clocks, so two packages chain into a count of one hundred and three into a thousand. The carry out is a clean CMOS output rather than an open drain, which means it can drive the next clock input directly, and because the clock input has a Schmitt trigger the slower edges of a long chain do not produce extra counts. The clock inhibit input stops the count and holds the current output active, which is the way to freeze a sequence without losing the position.
One behaviour to plan for is what happens when the clock runs faster than the decade can follow. The anti lock gating keeps the sequence valid, but the decoded outputs then carry pulses shorter than a clock period, and a load that needs a stable level sees a flicker rather than a step. A sequencer that must not flicker is clocked well below the 10 MHz typical figure, and the practical ceiling for a visible sequence is set by the load, not by the counter.
Where the DIP body still earns its place
A 16-lead DIP in a 25 piece tube is a repair part and a prototype part. Test fixtures, lamp sequencers and teaching boards built in the 1980s and 1990s are still in service, and a socketed DIP is the only practical way to keep them running. The same is true of a prototype that is reworked weekly: a socket saves the board from repeated soldering, and the tube format suits a drawer of spares better than a reel does.
For a new reflow build the reel SOIC is the right format and the tube DIP is the wrong one. The die is identical, so the electrical numbers above hold for either, and the decision is about assembly and service rather than about performance.
Incoming checks on a counter
The functional check is a clock, a reset, and a walk through all ten outputs, with a check that no two outputs are high together and that the carry out pulses once per cycle. The reset check is separate, because a counter that steps correctly can still ignore reset, and that fault only appears on the board at power up. The body marking distinguishes the DIP from the SOIC, and the obsolete SOIC row CD4017BM makes the marking check worth doing rather than assuming.
The reference pages are the incoming inspection plan, date code management, and the pin to pin cross reference when a legacy board needs a second source for the counter.
Questions that come up on the DIP counter
Whether the counter can be clocked by a mechanical switch. It can, because the clock input has a Schmitt trigger, and a 100 nF capacitor plus a 100 k resistor in front of it removes the worst of the contact bounce.
Whether the outputs can drive a seven segment display. They cannot, because the ten outputs are one of ten rather than a segment pattern, and the display needs a decoder or the CD4026B.
Whether the part runs at 3.3 V. It does, with a lower counting rate than the 10 MHz printed for 10 V, and the Schmitt thresholds scale with the supply so a 3.3 V clock swing is inside the specification.
Whether two packages can share one clock and one reset. They can, and that is the normal way to build a two digit counter, with the carry out of the first feeding the clock of the second.