HMC907APM5E
Product Specifications
HMC907APM5E
Analog Devices — Analog & Mixed-Signal ICs
Product Overview
The HMC907APM5E 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
HMC907APM5E: same amplifier, and the two transitions the package adds
The HMC907APM5E is the 0.2 GHz to 22 GHz pHEMT distributed amplifier supplied in a surface mount package instead of as bare die. The electrical values that get quoted for it come from the die document, which is taken with two 0.025 mm bonds of at least 0.31 mm on a 50 ohm fixture, so the packaged part carries those figures minus whatever the package transitions and the longer bond loops cost at each port. For a buyer comparing the two order codes the question is not which is faster, it is where the loss appears and what it does to the budget.
Published values
| Parameter | Value |
|---|---|
| Frequency range | 0.2 GHz to 22 GHz, die document |
| Gain | 14 dB typical, die document |
| Output power at 1 dB compression | 28 dBm typical, die document |
| Output third order intercept | 41 dBm typical, die document |
| Supply | 10 V at 350 mA |
| Matching | 50 ohm input and output, no external matching components |
| Assembly | standard surface mount, no wire bonding on site |
| Reference figures | die measured with two 0.025 mm bonds of 0.31 mm minimum length |
Where the package loss shows up and where it does not
A package adds a transition at the input, a transition at the output and a longer bond loop on each side. At 2 GHz the three together are a few tenths of a decibel and nobody notices. At 22 GHz a 0.4 mm loop in place of a 0.31 mm bond is enough to move the input match by several ohms, and the visible symptom is not a lower gain reading at mid band but a gain slope that tilts down at the top of the range.
The other half of the cost is thermal rather than electrical. On a carrier the die sits on a metal block, and the attach is the whole thermal path. In a package the path runs die to attach to pad to solder to board, and each joint is in series. The 2.87 W of heat at the compression point has three interfaces to cross instead of one, so the case temperature that the 350 mA was set at is no longer the die temperature.
Why buyers still take the package for a first build
The package removes wire bonding, die attach and the visual inspection that goes with them from the assembly flow. For a programme that is still at ten boards a month, that is the difference between an internal process nobody owns and a reflow profile the contract assembler already runs. The die becomes attractive once the volume justifies the assembly step or once the insertion loss at the top of the band has to come back.
The trade has a procurement side that is easy to miss. Die and package are separate order codes with separate lead times, separate moisture sensitivity handling and separate last time buy behaviour. Die usually goes to end of life earlier and without a published replacement, because the wafer bank runs out rather than because the part is withdrawn. A build that has qualified on die should either hold a die bank or keep the packaged order code documented as the fallback with the extra loss accounted for.
What to put on the purchase order either way
Both order codes are the same amplifier, so the acceptance test is the same: gain against frequency from 0.2 GHz to 22 GHz, output power at the compression point, and the drain current at 10 V. The difference is that the packaged part can be tested on a connectorised board while the die has to be tested in a fixture or on a carrier, and a fixture measurement of a die is only as good as the bond length that was used.
For a second source or a replacement study the three figures to compare are the gain at 1 GHz, the gain at 20 GHz and the current at 10 V. A part that matches at 1 GHz and reads two decibels low at 20 GHz is not a substitute however close the headline gain looks, because the application that needed 22 GHz will lose its top octave.
Two pages worth reading before a purchase order goes out are the RF power transistor cross reference and how to verify a drop in replacement; the first lists what else fits the socket, the second is the test sequence for a drop in, and RF procurement covers the case where the shortage is already on the line.
Two questions buyers ask
Is the packaged part a drop in for the die on an existing carrier? No. One is attached and bonded, the other is reflowed. The impedance presented at each port also differs because the bond loop and the package transition differ, so the match has to be rechecked rather than assumed.
Do the die figures apply to the packaged order code? They are the reference values, taken on die with 0.025 mm bonds of 0.31 mm. Expect the packaged part to be slightly below them at the top of the band, and treat the difference as part of the link budget rather than as a defect.