ISO7321FCDR
Product Specifications
ISO7321FCDR
Texas Instruments — Power Management & Signal Chain
Product Overview
The ISO7321FCDR 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
ISO7321FCDR: A Dual 25 Mbps Isolator With Input Noise Filters and the Low Default
ISO7321FCDR is a dual-channel digital isolator with integrated input filters in a SOIC (D-8) body of 8 pins. It runs 25 Mbps with 33 ns of typical delay, filters short noise pulses at the input pins and holds the outputs low when the input side loses power, in a narrow eight pin body.
Order code and package
| Ordering string | ISO7321FCDR |
| Package | SOIC, 8 pins, D narrow body |
| Carrier | 2500 pieces on a large reel, level 2 260 C, one year floor life |
| Tube sibling | ISO7321FCD, 75 pieces in a tube |
| Part marking | 7321FC |
| Temperature range | Minus 40 to plus 125 degrees Celsius |
| Channel split | Two channels, one each way |
| Default output | Low, the F suffix sets it |
| Input filter | Integrated noise filter on both inputs |
Electrical parameters that decide the application
| Signaling rate | 25 Mbps |
| Isolation rating | 3000 VRMS for one minute per UL 1577 |
| Transient isolation | 4242 VPK per DIN V VDE V 0884-10 |
| Propagation delay | 33 ns typical at a 5 V supply |
| Common mode transient immunity | 65 kV per microsecond typical at 5 V |
| Barrier life | Above 25 years |
| Supply current, ISO7321 at 5 V | 1.7 mA per channel typical at 1 Mbps |
| Supply current, ISO7321 at 3.3 V | 1.2 mA per channel typical at 1 Mbps |
| Supply range | 3 V to 5.5 V |
| Level translation | 3.3 V and 5 V in either direction |
| Input threshold | TTL levels |
| Input filtering | Short pulses at the input pins are rejected |
| ESD and transient | System level ESD, EFT and surge immunity claimed |
| Enable pin | None in the eight pin body |
Family variants and how they differ
| ISO7321C | Same channel split with the default high |
| ISO7320FC | Both channels forward with the low default |
| ISO7320C | Both channels forward with the default high |
| ISO6721FBDR | 50 Mbps dual with the low default when the rate matters |
| ISO6721BDR | 50 Mbps dual with the default high |
| ISO7721 | Wide body dual at 100 Mbps for a reinforced barrier |
Points to check before this part goes on a BOM
The input filter is the reason this part survives a noisy cabinet. A pulse of a few nanoseconds at the input pin is rejected instead of being passed on as an edge, which is what an unfiltered isolator of the same rate would do.
33 ns of typical delay is more than the 11 ns of the ISO672x family. A link that is timing sensitive has to be budgeted against the slower figure even though both parts are called 25 and 50 Mbps parts.
The F suffix gives a low output on input power loss. A receiver that treats a low as a start bit will see a false start during power up, which is the trade the suffix makes.
The eight pin body has no enable pin. Parking the channel means holding the input at a defined level rather than disabling the output.
The 25 Mbps rate and the 3000 VRMS barrier put this part between the older 1 Mbps isolators and the newer 50 Mbps duals. Where the rate is enough, the lower price and the input filter are the reasons to stay on this family.
The tube row is 75 pieces and the reel is 2500, both at level 2 with a one year floor life. A prototype build can be served from a tube without committing to a reel.
The supply floor of 3 V is above a 2.5 V rail, so this part cannot be used directly on a 2.5 V logic side. The 50 Mbps family covers 2.25 V and above, which is the route for a 2.5 V controller.
The part translates between 3.3 V and 5 V in either direction, and the supply range is the same on both sides. A product that mixes a 3.3 V sensor with a 5 V controller can therefore use one part for the whole interface rather than a shifter and an isolator.
Filtered inputs in a drive cabinet
Motor cables inside a cabinet couple short pulses into signal wiring, and an isolator without an input filter passes them as real edges. The filter on this part is a fixed function, not a programmable one, so the pulse width it rejects is the figure to compare against the noise the cabinet actually produces.
The input filter against a real cabinet
A variable frequency drive couples short pulses into signal wiring through the cable capacitance, and an isolator without an input filter treats each pulse as an edge. The filter on this part is fixed and integrated, so the noise the cabinet produces has to be compared with the pulse width the part rejects rather than with a number taken from a competitor page.
The practical test is a bench run with the drive switching at the working frequency and the signal cable dressed the way it will be in the cabinet. A filter that works on a clean bench proves nothing about the installation, and the part that survives the dressed cable is the one to qualify.
Where the 33 ns delay goes in a budget
An isolated digital input that is read by a controller every millisecond does not care about 33 ns. A closed loop that closes through the barrier does care, because the delay sits inside the loop and shifts the phase margin. The same part can be invisible in one product and marginal in another.
The delay also drifts with temperature and supply, and the document gives a typical figure at 5 V. A loop that is stable on the bench at 25 degrees has to be checked at the top of the ambient range with the supply at its low limit, which is where the delay is longest.
When the input filter is not enough
An integrated filter rejects short pulses, and a pulse that is wide enough to pass it is treated as data by any isolator. Where the noise in a cabinet is that wide, the answer is shielding, twisted pair and a common mode choke rather than a different isolator, because no digital barrier can tell a genuine edge from a noise pulse of the same width.
That is why the filter is described as a filter and not as immunity. It raises the noise level a product survives in a real cabinet, and it does not replace the wiring practice that keeps the noise out of the signal pair in the first place.
What the filter buys in a drive cabinet
Short pulses coupled into signal wiring are rejected before they reach the barrier, which keeps a controller from counting noise as edges. That is the difference between a product that passes an immunity test in a cabinet and one that only passes it on a bench.
Where this order code fits
It is chosen where a cabinet generates short noise pulses on signal wiring and the isolated side must not see them as data. Teams comparing it with a neighbouring part start from our Pin to Pin Cross Reference page, and teams replacing a part that has gone end of life start from the Supplier Quality Audit page.
Questions buyers ask
How much noise does the input filter reject? The document describes the filter as rejecting short pulses at the input pins without publishing a single pulse width limit in the features list, so the rejection has to be verified against the noise the equipment produces rather than assumed from a number.
Why take this part instead of the 50 Mbps dual? Cost and the input filter. When 25 Mbps is enough, the slower family is the cheaper route and the filter is not offered on the faster one.
Does it work at 3.3 V on the input and 5 V on the output? Yes. The supply range is 3 V to 5.5 V on each side and the part translates between the two rails.
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 How to Verify a Drop in Replacement page sets out the checks that keep a substitute from failing on the bench.