HXN40008
Product Specifications
HXN40008
Hisiwell — Discrete & Power Semiconductors
Product Overview
The HXN40008 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
HXN40008: 48 dBm from the same 32 V rail over 1.2 to 1.4 GHz
The HXN40008 is the second rung of a ladder that Hisiwell publishes as one table. It covers 1.2 to 1.4 GHz, delivers at least 48 dBm of saturation power, at least 13 dB of gain and at least 50 per cent efficiency, runs from a 32 V drain and holds plus or minus 0.8 dB of flatness in continuous wave. The part above it in the same band, the HXN40007, is the 40 dBm member, and the two share a rail and a band, so the step from one to the other is 8 dB of output for one dB less gain.
Forty eight dBm is 63.1 W. At 50 per cent efficiency that is 126.2 W of DC input, which at 32 V is 3.94 A, and 63.1 W of heat to remove from a device whose thermal resistance the vendor does not publish. Those four numbers are the reason this part is a different engineering problem from the ten watt member: the current is six times higher, the heat is six times higher, and the drive needed is 35 dBm, or 3.2 W, which is itself a power amplifier stage rather than a driver from a signal generator.
The drive requirement is the first thing a system budget has to survive. A 35 dBm source into a 32 V rail stage is normally the 40 dBm member of the same family running 5 dB back, and that in turn needs 26 dBm, which is 0.4 W. The chain is therefore three stages deep for a 63 W transmitter, and any filter or coupler between them subtracts from the margin directly, because the 13 dB of gain is printed as a floor and not as a typical.
The band is the same as the 40 dBm part, from the aeronautical distance measuring band below 1215 MHz to the GNSS L2 signal at 1227.6 MHz and the 1.3 GHz video links, and a part with plus or minus 0.8 dB of flatness over 200 MHz is a wideband saturated stage. What it is not, on the published rows, is characterised for linear operation: there is no figure for output at 1 dB compression anywhere on the sheet, so the only safe reading of 48 dBm is as a saturated transmitter output.
The parameter rows the vendor publishes
| Part class | Gallium nitride power transistor with internal matching |
| Frequency range | 1.2 to 1.4 GHz |
| Saturation power as published | at least 48 dBm |
| Saturation power in watts, calculated | 63.1 W |
| Gain as published | at least 13 dB |
| Efficiency as published | at least 50 per cent |
| Output at 1 dB compression | not published |
| Input impedance | 50 ohm |
| Output impedance | 50 ohm |
| Drain voltage | 32 V |
| Flatness across the band | plus or minus 0.8 dB |
| Operating mode | continuous wave |
| DC input at saturation, calculated | 126.2 W |
| Dissipation at saturation, calculated | 63.1 W |
| Drain current at saturation, calculated | 3.94 A |
| Drive power for rated output, calculated | 35 dBm, 3.2 W |
| Matching technology | in-plane matching synthesis on a film hybrid carrier |
| Stated application | RF and microwave systems |
| Package | not published on the parameter page |
| Thermal resistance | not published |
| Scattering parameters | not published |
| Operating and storage temperature | not published |
| Moisture level and floor life | not published |
| Orderable addendum | not published |
| Lower power member, same band and rail | HXN40007 at 40 dBm and 14 dB gain |
| Next step up, same band and rail | HXN40009 at 51 dBm and 13 dB gain |
| Higher voltage member, same band | HXN40011 at 56 dBm from 50 V in pulsed operation |
| Lower band member | HXN40002 at 48 dBm from 32 V over 0.9 to 1.2 GHz |
| Same output, wider band | HXN43001 at 50 dBm over 1 to 4 GHz from 30 V |
What 63 W of heat does to a carrier in continuous wave
Continuous wave at 48 dBm means 63 W of heat for as long as the transmitter is keyed, and the vendor publishes no junction to case figure, so the carrier has to be sized from experience and then measured. A copper tungsten carrier with a thin sintered interface and a forced air heatsink can be worth 2 to 3 C per watt in total. At 3 C per watt, 63 W is 189 C of rise, which no junction survives. The same arithmetic at 1 C per watt gives 63 C, which a 60 C cabinet turns into a 123 C junction and a workable build.
That single comparison is the useful thing to take from this page. A 63 W continuous wave device is not a part that can be bolted to a small extrusion, and the difference between one C per watt and three is the difference between a product and a burn in test. Liquid cooling, a vapour chamber or a heat pipe is normal at this level, and a pulsed duty cycle is the other route.
Efficiency moves the number more than any thermal trick. The published 50 per cent is a floor; at 55 per cent the dissipation at the same 63 W output falls to 51.6 W, and at 60 per cent it falls to 42 W. The vendor page prints no typical efficiency and no curve, so the only way to know which one a delivered part gives is to measure it, and the answer decides the heatsink before it decides anything else.
Where the datasheet stops and the bench starts
The first gap is compression. Saturation power is measured with the input driven past the point where output stops following input, so 48 dBm says nothing about a modulated signal with a peak to average ratio. A system that needs 48 dBm of average power needs a device whose saturation point is 6 or 8 dB higher, and this sheet offers no row that would let a buyer set that margin except by measuring it.
The second gap is the small signal data. At 1.2 to 1.4 GHz, with 50 ohm ports presented by the internal match, the return loss at the band edges and the gain slope across the band are the two numbers a cascade needs. The flatness row of plus or minus 0.8 dB describes the saturated condition and does not substitute for a sweep of gain against frequency at a backed off drive level.
The third gap is the DC behaviour. No quiescent current, no gate bias voltage and no recommended bias sequence is published. A GaN device of this size at 32 V needs a negative gate supply that is present before the drain, and a buyer has to know the pinch off window to set it. That is the vendor's data to give, and it is not on the page.
The fourth gap is the mechanical data. No outline drawing, no base material and no mounting torque is given, and at 63 W the base material is not a detail: the thermal expansion of the carrier has to match the device flange or the solder joint fails in thermal cycling. Ask for the drawing and the base material with the quotation.
Where a 63 W L band stage still fits
The fit is a transmitter that can carry a large heatsink: a ground based data link, a telemetry uplink, a test amplifier for a laboratory, or a driver for a 56 dBm pulsed stage where the 1 ms duty cycle means the 63 W of heat is only 6.3 W on average. That last use is the most economical, because the same device in a pulsed chain averages a tenth of the dissipation without any change to the part.
It is a poor fit for a compact airborne unit without a thermal path, and a poor fit for any linear multicarrier application, for the compression reason above. It is also a poor fit as a first stage: 13 dB of gain with 35 dBm of drive required means the preceding stage is itself a power amplifier, so this part is a middle or final stage and never the front of a chain.
Our RF power transistor cross reference work keeps the drive level, the rail and the gain floor next to the output power for this whole family, because the parts that look interchangeable on output alone differ by 10 V of rail and by 3 dB of gain, and both of those decide the bias supply and the stage count.
Questions we answer before an order is placed
The first is the duty cycle. The same part at 48 dBm in continuous wave and in a 1 ms pulse at 10 per cent differ by a factor of ten in average dissipation, and the vendor's continuous wave rating is the harder of the two. Ask what the keying looks like before the heatsink is chosen, because a build that assumes pulsed operation and then runs continuous wave will fail in the first hour.
The second is the bias supply. A 32 V rail at 3.94 A is 126 W of DC, which needs a supply with current limit and a drain feed that does not ring, plus a negative gate rail sequenced ahead of the drain. The published page gives no quiescent current, so the gate bias point has to come from the vendor or from a measured sweep on the first article.
The third is the measurement route. Our production test covers DC parameters, functionality and customer fixtures, and the RF power sweep at 1.3 GHz is done on the customer bench or at a partner RF laboratory. That division is worth stating in the quotation so that nobody expects a 48 dBm sweep in the incoming inspection report.
The fourth is the documentation. Seven parameter rows and no full document means the purchase order should ask for the outline drawing, the measured efficiency at the customer frequency, the pinch off voltage and the moisture classification. Our incoming inspection plan records what arrives against that list, and the RF procurement process is where the band specific requests are raised with the vendor before the first lot ships.