INA118U
Product Specifications
INA118U
Texas Instruments — Power Management & Signal Chain
Product Overview
The INA118U 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
INA118U: the active surface mount member of the low power precision family
The INA118U is the eight pin SOIC member of the INA118 family, and it is the string in this family that is still active while the plastic DIP rows move to last time buy. The die is the three amplifier instrumentation amplifier with 50 µV of maximum offset, 0.5 µV per degree of drift, 5 nA of maximum bias current, 110 dB of minimum common mode rejection at a gain of 1000, 40 V of input protection, 350 µA of quiescent current and a gain range from 1 to 10,000 on one resistor.
The carrier is a tube of 75 pieces, the body is the eight pin SOIC, and the moisture level is level 3 with a 168 hour floor life. The reel version of the same body ships 2500 pieces. The temperature range on the addendum is minus 40 to plus 85, and the current revision of the data sheet widened the operating range of the die to minus 40 to plus 125, which is a recommended operating condition rather than a specification limit.
The INA118U rows as the addendum publishes them
| Part class | Three amplifier instrumentation amplifier, precision, low power, surface mount |
| Vendor status | ACTIVE, production data |
| Data sheet | SBOS027C, revised September 2022 |
| Input offset voltage | 50 µV maximum |
| Offset drift | 0.5 µV per degree maximum |
| Input bias current | 5 nA maximum |
| Common mode rejection at G = 1000 | 110 dB minimum |
| Input overvoltage protection | 40 V without damage |
| Supply range | 2.25 V to 18 V total |
| Quiescent current | 350 µA |
| Gain equation | G = 1 + 50 kΩ divided by RG |
| Gain range | 1 to 10,000 V/V |
| Body | SOIC, D outline, 8 pins |
| Carrier, tube row | 75 pieces in a tube |
| Carrier, reel row | 2500 pieces on a large tape and reel |
| RoHS | Yes |
| Lead finish | The addendum directs the buyer to the vendor for the finish on this row |
| Moisture level | Level 3 at 260 degrees, 168 hour floor life |
| Temperature range | Minus 40 to plus 85 degrees on the addendum |
| Part marking | 118U with the vendor prefix |
| Revision row | INA118U.A and INA118U.B, both active |
| Successor named by the vendor | INA818, offered for new work |
| Alternative body | Plastic DIP, last time buy, same die |
The lead finish column is the odd one on this row: it points the buyer to the vendor rather than stating a finish, while the neighbouring rows state nickel palladium gold. On a build that has a finish restriction, that is a question to settle before the order rather than after, and the answer comes from the vendor's material declaration rather than from the addendum.
Why the SOIC row survived when the DIP did not
The through hole package is the one that loses volume first, and the surface mount package carries the same die at a lower assembly cost. The vendor's decision to keep the D row active and the P row last time buy follows the volume rather than the technology, and it leaves a customer with a socketed board in the awkward position of needing a mechanical adapter or a board revision.
The adapter is not a good answer for a precision amplifier. An eight pin DIP to SOIC adapter adds a pair of contacts, a length of trace and a thermal junction, and on a 50 µV part the contact EMF is a measurable offset. A board revision is the honest route, and the revision is small because the pin out is the same function in a smaller body with a different footprint.
Level 3 on a 75 piece tube
A tube of 75 pieces that is opened for a build of twenty boards leaves fifty five parts on the shelf, and level 3 gives a 168 hour window from the moment the bag is opened. A low volume instrument that builds twenty boards a quarter will use the tube over four quarters, which is far outside the window, and the parts have to go back into a dry pack between builds or be baked before each use.
The practical arrangement is a dry cabinet with a log, and the log is what an auditor asks for. Without a log the floor life is an assumption, and a level 3 part that has been through three bake cycles is a part whose solderability and whose moisture performance are both unknown quantities.
The 350 µA family and its noise
A low power amplifier trades current for noise, and this family's noise is higher than the 3 mA precision parts in the same catalogue. That matters on a channel with a low level signal and a wide bandwidth, and it does not matter on a channel that is already limited by the source impedance or by the converter. The decision is made by writing the noise budget rather than by comparing headline figures.
The other term that low power makes worse is the output drive. A 350 µA amplifier drives a limited load current, and a channel that has to drive a 1 kΩ load to 4 V is asking for 4 mA, which the part will not deliver over temperature. A buffer after the amplifier is the answer, and the buffer adds its own offset and drift to the budget.
Buying INA118U
The tube is 75 and the reel is 2500, both active, both level 3. A build of a few hundred boards takes the tube and a build of a few thousand takes the reel. The part is the active string in the family, which makes it the one to quote for anything new, and the plastic DIP is the one to quote only for a lifetime buy.
The successor named by the vendor is a different part with a different pin out, and it should be evaluated on a new board rather than substituted. A customer who is revising the board anyway should compare the two, and a customer who is not should stay with this string while it is active.
Questions that arrive with this part number
The first is whether the U is pin compatible with the plastic DIP part. It is not, and the function is the same in a different footprint. The second is whether the reel row and the tube row carry the same die, and they do.
The third is whether the part can drive a 16 bit converter directly. It can on a short trace with a small load capacitance, and it needs a series resistor when the converter has a switched capacitor input, because the charge injection from the converter appears as a step at the amplifier output.
For a customer comparing this part with its named successor, our pin to pin cross reference page sets out the parameters that have to match, and the amplifier cross reference page lists the families we compare across the catalogue.
The charge injection point is worth a number. A 16 bit converter with a 10 pF sampling capacitor switching a 4 V reference moves 40 pC into the amplifier output on each conversion, and an amplifier with a 1 MHz loop and no series resistor shows that as a step of a few millivolts that takes microseconds to settle. A 100 Ω series resistor and a 1 nF capacitor at the converter pin turn the step into a small glitch that the converter's own sample window absorbs, and the resistor costs nothing in accuracy because the converter input is high impedance when it is not sampling.
The other end of the same problem is the reference. A converter that samples against a reference which moves during the conversion produces a gain error that looks like noise, and the reference needs a capacitor at the pin as well as the amplifier. On a low power board where the reference is shared with the amplifier, the two capacitors should be separate parts rather than one shared part, because the converter current and the amplifier current have different spectral shapes.