HXN40063
Product Specifications
HXN40063
Hisiwell — Discrete & Power Semiconductors
Product Overview
The HXN40063 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
HXN40063: 43 dBm in the 5.3 to 5.9 GHz band on a 28 V rail
HXN40063 is the 43 dBm member of the Hisiwell GaN internally matched family for 5.3 GHz to 5.9 GHz, with 12 dB of gain, 45 percent drain efficiency, a 28 V drain rail, plus or minus 0.8 dB of flatness and a continuous wave rating. The package brings the input and output matching to 50 ohm, so a 20 W stage can be built on a two layer laminate with a microstrip launch instead of a tuned network.
43 dBm is 19.95 W, and this page is written for the engineer who has to decide whether that is the right row. The band covers the 5.8 GHz industrial, scientific and medical allocation and a set of licensed links, and the question that comes up first is not gain but what the part does at the band edges, where flatness of plus or minus 0.8 dB is 20.2 percent of the power and a link budget has to carry it.
The numbers here are the vendor table rows for this part, read from the Hisiwell GaN internally matched list. Where a figure is derived, the arithmetic is shown so it can be checked: 19.95 W of output at 45 percent efficiency means 44.3 W of DC input, which on a 28 V rail is 1.58 A and 24.4 W of heat in the flange.
The 5.3 to 5.9 GHz ladder and the two rows that matter
The family covers this band in eight power steps from 37 dBm to 53 dBm, and the step between neighbours is 3 dB, which is a factor of two in power. That spacing is deliberate: a link budget that gains 3 dB by moving up one row gains it without a combiner, and a product that needs 1 or 2 dB has to buy the next row up and run it backed off.
| Part | Band (GHz) | Psat (dBm) | Gain (dB) | Vd (V) | Mode |
| HXN40061 | 5.3 to 5.9 | 37 | 12 | 28 | CW |
| HXN40062 | 5.3 to 5.9 | 40 | 12 | 28 | CW |
| HXN40063 | 5.3 to 5.9 | 43 | 12 | 28 | CW |
| HXN40064 | 5.3 to 5.9 | 47 | 11 | 28 | CW |
| HXN40065 | 5.3 to 5.9 | 50 | 10 | 28 | CW |
| HXN40066 | 5.3 to 5.9 | 51 | 10 | 32 | CW |
| HXN40067 | 5.3 to 5.9 | 53 | 10 | 28 | Pulse 100 us, 10 percent |
Two details in that table decide more projects than the power column. First, gain falls as power rises: 12 dB at 37 to 43 dBm, 11 dB at 47 dBm, 10 dB from 50 dBm up. A chain built around a 20 W part and later moved to a 100 W part loses 2 dB of gain and needs another driver stage or a different one. Second, HXN40066 is the only 32 V row, so a product that has a 28 V rail and no boost converter cannot use it.
The specification rows we quote against a customer requirement
| Function | Internally matched GaN power transistor, 50 ohm in and out |
| Frequency band | 5.3 GHz to 5.9 GHz |
| Saturated output power | 43 dBm, which is 19.95 W |
| Output power at 1 dB compression | Not published as a separate row in the vendor table |
| Gain | 12 dB at saturation |
| Drain efficiency | 45 percent at saturation |
| Drain voltage | 28 V nominal |
| DC input at saturation | 44.3 W |
| DC current at saturation | 1.58 A from a 28 V rail |
| Heat in the flange at saturation | 24.4 W |
| Flatness in band | plus or minus 0.8 dB |
| Operating mode | Continuous wave |
| Input return loss expectation | Set by the internal match, confirm on the data sheet curve |
| Stability | Confirm the vendor curve at the low end of the band with the bias feed in place |
| Harmonic termination | Board decision, outside the internal match |
| Bias sequencing | Gate negative before drain, drain removed before gate |
| Package | Flange mount, screw fixed |
| Interface | 50 ohm microstrip launch at both ports |
| Cooling | Heatsink sized on 24.4 W continuous |
Twenty watts continuous and the heatsink that carries it
At 24.4 W in the flange the heatsink is not optional and it is not small. A flange to heatsink interface with a 0.05 mm thermal pad has about 0.5 degrees per watt, and a modest extruded heatsink in still air has 1.5 to 2.5 degrees per watt. That puts the flange 49 to 73 degrees above ambient, so a 25 degree room gives a flange at 74 to 98 degrees before the junction rise is counted. The vendor curve gives the junction to case number, and the arithmetic, not the room temperature, decides whether the product needs a fan.
The failure mode when the path is short is not a dead part. It is a part that makes its power cold and loses 1 dB after twenty minutes, with the gain drifting as the junction heats. That shows up in a soak test and not in a five minute bench check, so the acceptance test for a 24 W stage belongs at thermal equilibrium.
The flange interface is also where a plating problem shows. A nickel plated heatsink with a bare aluminium base and a thermal pad is fine; a bare copper heat spreader with a pad that has been handled without gloves is not, and the oxide layer that grows in a humid store is enough to add a degree per watt. We ask for the interface material and the torque figure before the parts ship, because both are in the customer process rather than in the part.
Gain, backoff and what 0.8 dB of flatness costs
At 12 dB of gain the driver for this part has to deliver 31 dBm, which is 1.26 W, and most of the small GaAs or GaN driver parts in the catalogue can do that with margin. The driver has to be linear enough that the correction loop, if there is one, keeps the composite signal inside the spectral mask at the power the system actually runs, which for a 43 dBm part is usually 6 to 8 dB of backoff, not at saturation.
Flatness of plus or minus 0.8 dB across the band is 20.2 percent of power at the edges against the centre. In a system with a level control loop that is invisible. In an open loop transmitter it is a 1.6 dB spread across the band that has to be trimmed with a pad or equalised at the driver. We see this most often on test equipment and on satellite up converters, where the band is used end to end and the level at the top of the band is what the customer measures.
How we test and what we do not claim
Our line tests DC parameters and function, and for a flange part that means the gate threshold window, the drain to source insulation, the presence and value of the bias network on the carrier and the assembly workmanship under the flange. We do not publish gain or power figures of our own for this part, and we do not run a 5.3 GHz to 5.9 GHz load pull. Those measurements come from the customer fixture or from a partner RF laboratory, and the report is filed against the unit serial.
What we do control is what reaches that bench. The part number, the date code, the reel or tray it came from and the certificate chain are recorded, and the flange and the launch are inspected before the unit is powered. That process is on our incoming inspection plan, the record that follows the parts is on our component traceability report, and a first article report for a new assembly is written against the checklist on our NPI production launch checklist.
What to send with an enquiry for HXN40063
Three things make a quotation for this part usable rather than indicative: the duty cycle, the band edges that matter, and the rail available on the board. With those we can say whether the 43 dBm row or the 47 dBm row is the right one, and whether the 32 V member is worth a boost converter. Without them the quote is a price for a part number, and the part number may be the wrong row of a seven row ladder.
For a second source or a drop in comparison, our RF power transistor cross reference puts the equivalent parts from other vendors beside this one, with the gain and the drain voltage called out because those are the two figures that break a drop in. The passive side, the laminate, the connectors and the bias parts, is on our one stop RF supply page.