HXC40028
Product Specifications
HXC40028
Hisiwell — Discrete & Power Semiconductors
Product Overview
The HXC40028 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
HXC40028: eight watts at 35 per cent on 6.4 to 7.2 GHz
The HXC40028 is the second member of the 6.4 to 7.2 GHz group in the Hisiwell gallium arsenide internally matched family, and the step up from the 36 dBm part is where the thermal question starts to matter. The window is 799 MHz, 11.76 per cent fractional at a 6.8 GHz centre. The vendor states in-plane matching synthesis and thin film hybrid integration, with 50 ohm at both ports. The index row reads 6.4 to 7.2 GHz, saturation output not printed, output at 1 dB compression at least 39 dBm, gain at least 9 dB, drain efficiency at least 35 per cent, 10 V drain rail, plus or minus 0.8 dB flatness, continuous wave.
39 dBm at compression is 7.94 W. On 10 V the DC input is 22.70 W, the drain current 2.27 A and the heat 14.75 W, which is two and a half times the dissipation of the 36 dBm member for twice the output. The drive is 1.00 W at 30 dBm, one decibel more power than the smaller part needs and a different driver class in practice.
Nothing in the match changes between the two members beyond the die size and the current. The flatness figure and the 50 ohm port impedance are family figures, so a board built for the 36 dBm part carries this one with a change of feed width and thermal path.
The parameter rows Hisiwell publishes for HXC40028
| Part class | Gallium arsenide internally matched power transistor |
| Matching technique, as stated by the vendor | In-plane matching synthesis, thin film hybrid integration |
| Frequency range | 6.4 to 7.2 GHz |
| Bandwidth, calculated | 799 MHz, 11.76 per cent fractional at a 6.8 GHz centre |
| Output at 1 dB compression as published | at least 39 dBm |
| Output at 1 dB compression in watts, calculated | 7.94 W |
| Saturation output power as published | not published, the index prints a slash in that column |
| Gain as published | at least 9 dB |
| Drain efficiency as published | at least 35 per cent |
| Drain voltage | 10 V |
| Input and output impedance, as stated by the vendor | 50 ohm at both ports |
| Flatness across the band | plus or minus 0.8 dB |
| Operating mode | continuous wave |
| Drive power for the compression point, calculated | 30 dBm, 1.00 W |
| DC input at compression, calculated | 22.70 W |
| Drain current at compression, calculated | 2.27 A at 10 V |
| Dissipation at compression, calculated | 14.75 W |
| Dissipation if efficiency reaches 50 per cent, calculated | 7.94 W |
| Output per 100 MHz of the band, calculated | 0.99 W |
| Load voltage at compression into 50 ohm, calculated | 19.93 V RMS, 28.18 V peak |
| Thermal resistance, junction to case | not published |
| Package outline and dimensions | not published in text, the product page carries one outline image |
| Scattering parameters | not published |
| Gate bias point and quiescent current | not published |
| Operating and storage temperature | not published |
| Orderable addendum, marking and packing | not published |
| Neighbour one step below at 36 dBm | HXC40027, 40 per cent, 1.00 A |
| Neighbour one step above at 41 dBm | HXC40029, 38 per cent, 3.31 A |
Five points of efficiency cost two and a half times the heat
The step from the 36 dBm member to this one is 3 dB of output and five points of efficiency lost, and the heat goes from 5.97 W to 14.75 W. That is the shape of the whole family: the larger dies run at lower efficiency because the matching network and the current crowding cost more than the extra die area returns. A system that needs 8 W at this band should plan the 14.75 W as a floor, not as a comfortable operating point.
At 35 per cent, 22.70 W of DC input for 7.94 W of output leaves 14.75 W. A 1.5 C per watt interface and plate puts the case about 22 degrees above ambient, which is workable in a rack mounted unit with airflow and marginal in a sealed enclosure. If the enclosure is sealed, the answer is a cold plate or a lower power member, not a thicker thermal pad.
The 2.27 A drain current is still inside a single wide track with vias, and the choke has to carry that current without saturating under modulation. Above 2 A the physical arrangement of the bias network starts to matter for stability as well as for loss, because the inductance in the feed becomes a feedback path at 7 GHz.
Seven gigahertz in a real system
The 6.4 to 7.2 GHz window covers upper C band fixed service links and the government allocation above them. In a point to point link the linearity requirement drives the back off, and at 8 W the back off has a thermal cost that the smaller member does not carry. A link specified at 32 dBm average out of this part runs 7 dB below compression and dissipates nearly the same heat as at compression, because the current only falls slowly.
In a radar or a data link the operating point is nearer compression and the published figures apply directly. The 9 dB of gain is then the limiting number for the drive chain: a variable gain stage ahead of this part has to deliver 30 dBm cleanly, and any loss in the interstage filter comes straight off the available headroom.
Linearity, again, has no published data. Nothing in the vendor table describes two tone behaviour, adjacent channel power or the compression behaviour under a modulated envelope, so the first article has to be measured. On a part at this power the measurement is worth doing before the thermal stack is frozen, because the result often moves the operating point by a decibel or two.
The ladder on the 6.4 to 7.2 GHz window
The four members are 36, 39, 41 and 44 dBm. This one sits in the middle of the lower half, one decibel below the 41 dBm part and one step above the entry device, and the choice usually comes down to whether the driver can supply 1.00 W or 1.58 W. In a two stage line up the difference is often invisible, because the driver is chosen for the final stage anyway.
The 44 dBm member on the same window is three decibels above this part and needs 3.16 W of drive instead of 1.00 W, so a line up that starts here and later moves up has to change the driver, the feed width and the thermal plate together. It is usually cheaper to choose the final member first and size the driver for it than to grow the chain one step at a time. A product family that ships two power options from one board should leave the driver footprint large enough for the 3.16 W case from the beginning.
Our RF power transistor cross reference lists the pin compatible alternatives on this band, and the component lead time page covers what happens to a schedule when the chosen member is on allocation.
Questions buyers ask before ordering HXC40028
Is 39 dBm the minimum or the typical? The vendor prints at least, so 7.94 W is a floor at the stated conditions. A part that measures higher is inside the specification, which matters when the link budget has no margin.
How much does the case temperature move the output? There is no curve in the vendor material. Gallium arsenide compression typically falls by 0.02 to 0.03 dB per degree at this power, so a 40 degree rise costs about a decibel. That is an expectation, not a specification.
Can it be used at 12 V with a series resistor? No. The rail is 10 V and the current is 2.27 A, so a resistor would dissipate 4.5 W and give a rail that moves with modulation. A regulated 10 V source is the only practical answer.
How do we start? Send the output required, the modulation, the ambient at the base plate and the rail available. We will confirm the family member that fits, quote from the current factory schedule and ship a sample with measured output, gain and efficiency across the band.