HMC-AUH318-Die
Product Specifications
HMC-AUH318-Die
Analog Devices — Analog & Mixed-Signal ICs
Product Overview
The HMC-AUH318-Die from Analog Devices is a high-performance analog / mixed-signal IC providing precision signal acquisition, conversion, and conditioning. Analog Devices (ADI) is a world leader in high-performance analog, mixed-signal, and DSP ICs for instrumentation, healthcare, automotive, and communications.
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-resolution data conversion with excellent SNR and THD
- Wide dynamic range and high linearity across operating conditions
- Integrated voltage reference with low temperature drift
- Flexible digital interface — SPI, I2C, or parallel options
- Industrial/automotive temperature range with guaranteed specs
- Low power consumption with power-down modes
Key Specifications
| Resolution | bits |
| Channels | — |
| Sample Rate | kSPS |
| Interface | — |
| Supply Voltage | V |
| Operating Temp | °C |
| Package | — |
| INL | LSB |
| SNR | dB |
| THD | dB |
Applications
Request Quote — In Stock, 100% New & Original, Full Warranty
HMC-AUH318 die: 24 dB at 71 GHz to 76 GHz and a 4 dB output match
The HMC-AUH318 is a three stage gallium arsenide HEMT medium power amplifier die covering 71 GHz to 76 GHz with 24 dB of gain, 17.5 dBm at one decibel compression, 20 dBm saturated, 7 dB of input return loss and 4 dB of output return loss, from two 4 V drains at 80 mA each. The die is 2.65 mm by 1.60 mm by 0.05 mm with titanium gold pads and backside, and the published table is probed die data. Compared with its sibling at 81 GHz to 86 GHz it offers two more decibels of gain and half a decibel more saturated power at a lower frequency, and it gives up three decibels of input match to do it.
Published values
| Parameter | Value |
|---|---|
| Frequency range | 71 GHz to 76 GHz |
| Gain | 21 dB minimum, 24 dB typical |
| Input return loss | 7 dB typical |
| Output return loss | 4 dB typical |
| Output power at 1 dB compression | 17.5 dBm typical |
| Saturated output power | 20 dBm typical |
| Supply | 4 V on two drains, 80 mA each, 160 mA total |
| Gate bias | VGG1 and VGG2 adjustable minus 0.8 V to plus 0.3 V, minus 0.1 V typical |
| Die size | 2.65 mm by 1.60 mm by 0.05 mm |
| Backside and pads | titanium gold |
| Measurement basis | probed die at 25 degrees |
Four decibels of output return loss and what it costs downstream
Four decibels is a voltage standing wave ratio of about 4.4 to 1, which is a severe mismatch for a stage that is meant to drive an antenna or a mixer. What it means in practice is that the load impedance presents a large reactive component, and any change in the load moves the gain and the compression point of the amplifier. A 71 to 76 GHz link that measures well into a laboratory load can lose a decibel or more when connected to a real antenna with its own reflection.
The mitigation is a matched buffer or a short taper into an isolator, and at these frequencies both are lossy. The cheaper route is to accept the mismatch and characterise the amplifier with the actual load attached, then set the link budget on the measured result rather than on the 20 dBm saturated figure from a 50 ohm fixture. That measurement is also the only way to see whether the load pulls the gain flatness, which matters more than the absolute gain in a wideband link.
Why this die runs at half the dissipation of a power amplifier and still needs a carrier
4 V at 160 mA is 0.64 W of DC, and at 20 dBm saturated the output is 100 mW, so 0.54 W becomes heat. That is a modest number in absolute terms, but the die face is 4.24 square millimetres of 50 micrometre gallium arsenide, which cannot spread heat laterally, so the flux under the active area is much higher than the average and the backside attach carries all of it.
At 75 GHz the backside is also the radio frequency ground. A carrier of copper tungsten and an attach with a controlled bond line are therefore both electrical and thermal parts, and a void under the output stage shows up as a gain notch at the top of the band before it shows up as a temperature rise. For a module that has to work over temperature the attach specification belongs on the assembly drawing with a void percentage, not in a note.
Two gates and one current measurement
Both gates are set independently to reach 80 mA per drain, and the two drains are separate pads so each can be measured. In a module the two currents are usually tied together and measured once, which hides a 20 mA imbalance between stages. That imbalance is worth finding, because the second stage carries the output and its current is the one that sets the compression point.
A bias board that brings each drain out on its own test point costs two extra pads and turns a part that either works or does not into one whose per stage current can be recorded. Given that a die at 75 GHz is expensive and the assembly is not reworkable, the extra two pads are cheap.
The pages that answer this most often are the RF component sourcing page for the family alternatives, how to choose an RF amplifier for the acceptance test, and one stop RF supply for the case where a replacement has to be fitted before the next build.
Two questions buyers ask
Is the AUH318 a drop in for the AUH317? No. The frequency ranges do not overlap, the gain differs by two decibels and the port matches differ, so a substitution needs a new radio frequency check even though the die outline is the same.
Is 20 dBm available over the whole band? That is the saturated figure at 25 degrees on probed die. Into a mismatched load the same part delivers less, and the 4 dB output return loss means the load has to be measured with it.