Contact Us

HXN50134

Brand: Hisiwell
Hisiwell HXN50134 is a discrete semiconductors component designed for DC-DC Converters & Power Supplies and related applications. In stock at Superb Automation — request a quote for competitive pricing and fast delivery.
quote now

Product Specifications

HXN50134 — Discrete & Power Semiconductors | Superb Automation

HXN50134

Hisiwell — Discrete & Power Semiconductors

Product Overview

The HXN50134 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 / VceV
Id / IcA
Rds(on) / Vce(sat)mΩ
QgnC
Technology—
Package—
Configuration—
Mounting—

Applications

Switch-Mode Power Supplies Motor Drives & Inverters Battery Management Systems DC-DC Converters Load Switching Automotive Power Distribution

Request Quote — In Stock, 100% New & Original, Full Warranty

HXN50134: a 450 W per side GaN pair for DC to 1 GHz on a 50 V rail

HXN50134 is an unmatched GaN power transistor from Hisiwell printed in the 50 V table at 450 W times two, over DC to 1 GHz, in continuous wave. The times two notation is the one this vendor uses for the dual rows in the same table, where two devices are combined in one flange so that a single part number delivers the pair a push pull stage needs. The rail is 50 V, which is the point of the whole 50 V family: for a given output power the current is close to half of what a 28 V part needs, and at these power levels the current is what decides the bus bar, the connector and the amount of copper on the board.

DC to 1 GHz is the band that carries FM broadcast, VHF television, avionics VOR and ILS, the 13.56 MHz industrial band and a long list of legacy military radios. Those applications share a property that shapes the purchasing decision: they are in service for decades, the RF chain is usually a single ended or push pull stage with a lot of copper, and the part number often has to stay the same for the life of the product. A 50 V part at this band is chosen for headroom and for efficiency, not for bandwidth.

All figures here come from the Hisiwell unmatched GaN 50 V category table read on 4 November 2026, plus arithmetic derived from those figures. The table prints output power, band, rail and mode for this row, and leaves gain and efficiency blank; the page says so rather than filling the gap, and the device document is the source that governs a purchase.

What 450 W times two does to the rest of the amplifier

450 W is 56.53 dBm. Two of them combined coherently give 900 W, which is 59.54 dBm, before the loss of the combiner. In a push pull stage the two devices do not add in power at the output in the same way as a combiner, but the pair is what makes the transformer and the balun practical at this power level, and the number a system engineer works with is the pair figure.

The current is where 50 V earns its place. At 60 percent drain efficiency, 450 W of output draws 750 W from the rail, which at 50 V is 15 A per device and 30 A for the pair. The same output on a 28 V rail would need 53.6 A, and at those currents the difference is not a resistor value, it is the thickness of the copper and the size of the connector. That is the arithmetic behind every 50 V stage in a broadcast transmitter.

Heat follows the same rule. At 60 percent efficiency each device turns 300 W into heat, and the pair turns 600 W into heat that has to leave through the flange and the heatsink. At the DC to 1 GHz band the flange is large and the heatsink is a copper block with a forced air or water path; the thermal resistance of that stack is the number that decides whether the part runs at its rated output or at 80 percent of it.

The 50 V table around this part

Part numberBandPrinted outputGainEfficiencyMode
HXN50336DC to 1 GHz900 times 2 W16 dB57 percentPulse
HXN50134DC to 1 GHz450 times 2 WblankblankCW
HXN501510.96 GHz to 1.4 GHz400 times 2 WblankblankCW
HXN50131DC to 3 GHz150 times 2 WblankblankCW
HXN50156x20.96 GHz to 1.2 GHz1500 W13 dB57 percentPulse
HXN501540.96 GHz to 1.2 GHz800 W13 dB60 percentPulse
HXN50160F1.2 GHz to 1.6 GHz800 W16 dB60 percentPulse
HX13470PDC to 1.4 GHz470 W18 dB82 percentCW
HX13240FDC to 1.4 GHz240 W18 dB82 percentCW
HXN50126DC to 2 GHz240 WblankblankCW

Two comparisons are worth making. HX13470P delivers 470 W single ended at 82 percent efficiency and 18 dB of gain, which is better on both counts than the row this page covers, and it is a 48 V part. The difference is that HXN50134 is a pair in one flange, which is what a push pull stage wants, while HX13470P is a single device for a different topology. Choosing between them is a topology decision before it is a price decision.

The second comparison is on modes. The 800 W and 1500 W rows in the same table are pulsed, and the 900 times 2 row is pulsed as well. This part and the 400 times 2 row are the continuous wave entries at this band. A requirement written for continuous wave cannot be met by a pulsed row with a duty cycle attached to it, however large the peak number looks.

The figures the table gives for this row

FunctionUnmatched GaN power transistor pair, flange mount
Frequency bandDC to 1 GHz
Output as printed450 times 2 W
Output in dBm per device56.53 dBm
Output in dBm for the pair59.54 dBm before combining loss
Drain voltage50 V nominal
Operating modeContinuous wave
GainNot printed in the category table
Drain efficiencyNot printed in the category table
DC input at 60 percent efficiency750 W per device, 15 A from 50 V
DC input for the pair at 60 percent1500 W, 30 A from 50 V
Heat per device at 60 percent300 W continuous
Heat for the pair at 60 percent600 W continuous
Equivalent current on a 28 V rail53.6 A for the same output, which is the reason for 50 V
MatchingExternal, board side
TechnologyGallium nitride on silicon carbide
PackageFlange mount, screw down
Thermal resistanceIn the device document, not in the category table

What a 30 A continuous pair means for the board

The supply is the first constraint. Thirty amps at 50 V through a PCB trace is not a trace, it is a bus bar or a heavy copper plane. The 50 V family assumes that the drain feed is a machined part or a multilayer copper stack with vias in parallel, and a board that tries to feed this pair through two ounces of copper will drop more voltage than the regulator tolerance allows.

The decoupling is the second. At DC to 1 GHz there is no point in a large electrolytic close to the flange, and the parts that matter are the ones that hold the rail during a modulation envelope: a bank of ceramic capacitors with a low inductance path to the flange, and a feed that is not shared with the driver stage. The energy storage for the envelope sits further back, behind a choke that keeps the RF out of the supply.

The third is grounding and mechanical. A flange at this power has to be flat to the heatsink, and the source connection runs through the flange, so the screw pattern and the torque are part of the electrical circuit as well as the thermal path. A flange that lifts at one corner raises the source inductance and the gain drops without anything failing.

Supply reality for a part in service for decades

Parts in the DC to 1 GHz band tend to sit in equipment with a twenty year life: broadcast transmitters, avionics ground stations, industrial heating. The purchasing question is therefore rarely about the next quarter, it is about whether the same part number will be available when a transmitter needs a spare in year twelve. The answer comes from two places, a last time buy position held against the installed base, and a documented alternative that has been tested in the same fixture rather than compared on paper.

What we can do on a part like this is quote the exact orderable number with the date code, hold a stock position against a service plan, and keep the alternative list for the rows in the same table that are electrically close. We do not claim to measure RF power at this level in our own facility; the RF characterisation belongs to the vendor document or to a laboratory, and our incoming check is the visual, dimensional and marking check plus the traceability paperwork.

Related pages: RF power transistor cross reference, obsolete RF part replacement for the service life case, RF procurement for the supply route, and one stop RF supply for building the rest of the chain around it.

Questions buyers ask about the 450 times 2 row

Is 450 times 2 the output of the part or of each device. The vendor writes it as 450 times 2 W, which is the notation the dual rows in this table carry. The safe reading is that the number as printed describes the pair, and the per device figure has to be confirmed in the document before a combiner or a transformer is specified. That is the first line of any quote we send on this row.

Can a 28 V board use this part. No. The rail is part of the device, not a preference. A 28 V supply on a 50 V part will not reach the rated output and will change the load line the matching network was set for. Moving a product from 28 V to 50 V is a new board, not a part substitution.

What has to be checked at goods inwards. Marking, date code, flange flatness and the packing list against the order, plus a sample check of the gate and drain characteristics with a curve tracer if the programme asks for it. RF output at 450 W is not measured at goods inwards by anybody sane; it is measured in the fixture that the product uses.