HX07141KPM
Product Specifications
HX07141KPM
Hisiwell — Discrete & Power Semiconductors
Product Overview
The HX07141KPM from Hisiwell is a discrete power semiconductor delivering efficient switching and power conversion. Hisiwell is a recognized semiconductor supplier providing reliable components for global electronics manufacturing.
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
- Low Rds(on) for minimal conduction losses in power switching
- Fast switching speed with low gate charge for high-frequency operation
- Avalanche-rated for robust operation under inductive load conditions
- High junction temperature rating for demanding environments
- Logic-level gate drive compatible for simplified design
- RoHS and AEC-Q101 qualified options available
Key Specifications
| Vds / Vce | V |
| Id / Ic | A |
| Rds(on) / Vce(sat) | mΩ |
| Qg | nC |
| Technology | — |
| Package | — |
| Configuration | — |
| Mounting | — |
Applications
Request Quote — In Stock, 100% New & Original, Full Warranty
HX07141KPM: 1000 W pulsed across 700 to 1400 MHz in a sealed cermet body
The HX07141KPM is a pre-matched gallium nitride RF power transistor rated at 1000 W across 700 to 1400 MHz, sealed in a cermet package and operated from a 50 V supply in pulse mode. The product page is more specific than the shelf row beside it: 15 dB of power gain, 65 per cent of typical added efficiency, a static bias of 50 V at 200 mA, and a pulse condition of 100 microseconds at a 10 per cent duty cycle.
The index row tells a different story on two of those four numbers. It prints 18 dB of gain, 70 per cent of efficiency and a 48 V rail. So one document says 15 dB at 50 V and the other 18 dB at 48 V, and the difference in gain is 3 dB, which is a factor of two in drive power: 1000 W is 60.00 dBm, so 15 dB of gain needs 45.00 dBm, or 31.6 W, while 18 dB needs 42.00 dBm, or 15.8 W.
That choice of driver is the single most expensive decision around this part. A 32 W driver is a two or three stage chain with its own bias and its own heat, while a 16 W driver can often be a single stage. A quotation that mixes the 15 dB figure from the product page with the 48 V rail from the index row describes a part that does not exist on either document.
The HX07141KPM rows as the vendor publishes them
| Part class | Gallium nitride RF power transistor, pre-matched, hermetically sealed cermet body |
| Frequency range, product page | 700 to 1400 MHz |
| Frequency range, index row | 0.7 to 1.4 GHz |
| Band width, calculated | 700 MHz |
| Band ratio, calculated | 2 to 1 |
| Output power, product page | 1000 W |
| Output power, index row | 1000 W |
| Output in dBm, calculated | 60.00 dBm |
| Power gain, product page | 15 dB |
| Gain, index row | 18 dB |
| Drive needed at 15 dB, calculated | 45.00 dBm, 31.6 W |
| Drive needed at 18 dB, calculated | 42.00 dBm, 15.8 W |
| Efficiency, product page | 65 per cent typical added |
| Efficiency, index row | 70 per cent |
| Drain rail, product page | 50 V |
| Drain rail, index row | 48 V |
| Static bias, product page | 50 V at 200 mA |
| Mode and pulse condition, product page | Pulse, 100 microseconds, 10 per cent duty |
| Load voltage at 1000 W into 50 ohm, calculated | 223.6 V RMS, 316.2 V peak |
| Peak swing against a 50 V rail, calculated | 6.32 times |
| DC input at 65 per cent, calculated | 1538 W, 30.8 A on a 50 V rail, 538 W of heat |
| Average heat at 10 per cent duty, calculated | 53.8 W plus 10 W of quiescent |
| Applied application list | L-band radar, navigation systems |
| Package and body | Cermet, hermetic, dimensions not published |
| Thermal resistance, junction to case | Not published |
| Load impedance at the output | Not published, no load pull table |
| Orderable addendum and packing | None published |
Pulse mode changes the thermal arithmetic, not the electrical one
At 65 per cent efficiency the part draws 1538 W from the rail during the pulse and dissipates 538 W in the die. On a steady carrier that would need a heatsink the size of a small chassis. At 100 microseconds and 10 per cent duty the average dissipation falls to 53.8 W, and with 10 W of quiescent bias the amplifier only has to move about 64 W on average. The die still sees the full 538 W for the length of the pulse, so what matters is the transient thermal impedance over 100 microseconds and not the steady state figure.
That is why the vendor publishes no thermal resistance on the page. A junction to case number measured in steady state says almost nothing about a hundred microsecond pulse, because the heat has not had time to cross the die and the solder. What a buyer needs is the transient curve at the stated duty, and until that curve arrives the heat sink can only be sized from measurement.
The electrical side does not relax with duty. The full 316.2 V peak appears at the load inside every pulse, and the drain has to stand that plus the margin the load mismatch adds. On a 50 V rail this is 6.32 times the supply, which is a reminder that the rail sets the current and not the voltage here; a part in this class is chosen for its breakdown and its thermal path as much as for its watts.
Pre-matched input, unmatched output, and what that leaves on the board
The page calls the part pre-matched, which on this shelf means the input match sits inside the package. For a 1000 W part this is a large convenience: the input impedance of a die this size is a fraction of an ohm, so matching it on the board would need a network with loss the driver cannot spare. With the match inside, the board sees something near 50 ohm at the gate and the driver can be placed further away.
The output is a different matter. No load pull table is published, so the impedance the die wants at 700 to 1400 MHz is unknown from the page, and the output network has to be built from a measurement or from a data sheet that does not yet exist. A buyer who plans a broad 2 to 1 network will usually find that the same network cannot serve both ends of the band at full power, and that the practical solution is a narrower network per sub band.
The cermet body is the third constraint. A hermetic package is chosen where the environment demands it, and it comes with a mounting rule, a torque and a flange that are all unpublished here. A radar or navigation programme that needs the hermetic rating will have to get the drawing before the mechanical interface is released.
Where the HX07141KPM sits on the shelf
At 1000 W this is the highest power row in the 50 V listing, and the only one that reaches four figures on a single ended part. The nearest neighbours are the two 900 W per side members in pulse mode, the 800 W rows at 0.96 to 1.2 GHz and 1.2 to 1.6 GHz, and the 1500 W amplifier rows that are built from two transistors. All of them are pulse parts, and none of them publishes a load pull table.
Below 700 MHz the shelf changes character. The DC to 1.4 GHz 48 V rows publish 18 dB and 82 per cent at 470 W and 240 W, and the DC to 1 GHz parts reach 900 W per side. Those parts have published gain and efficiency and a lower rail, so a programme that can live at 48 V and below 1 GHz has an easier choice than one that needs the full 700 to 1400 MHz span at a kilowatt.
The trade between this row and its neighbours is set out on the RF power transistor cross reference page. Where the application is L-band radar, the deciding column is not the watts but the pulse condition, because a part rated for a hundred microseconds at 10 per cent duty cannot be pushed to a longer pulse without a thermal answer.
Questions buyers ask about the HX07141KPM
Is the gain 15 dB or 18 dB? The product page says 15 dB and the index row says 18 dB, and the rails differ too, 50 V against 48 V. The gain difference is a factor of two in drive, so it has to be settled before a driver is bought. The product page is the more detailed document and is the safer planning figure at 15 dB.
Can the pulse be longer than 100 microseconds? Nothing published allows it. The stated condition is 100 microseconds at 10 per cent duty, which sets both the pulse length and the average dissipation. A longer pulse raises the transient junction temperature for the same heat sink, and the vendor would have to confirm a derated output for it.
What is the output impedance? Not published. No load pull table, no scattering parameters and no impedance locus appear on the page. On a kilowatt part this is the single most valuable document a buyer can request, because it decides the output network and the bandwidth that network can hold.
How is a part like this supplied? Through a channel that can prove the batch and hold the hermetic documentation, with the incoming check covering the marking, the flange and a pulse test at the intended band. The RF component sourcing page sets out what that trace covers, and the one stop RF supply page covers how the driver chain and the passives are bought against the same schedule.