Contact Us

HXN40057

Brand: Hisiwell
Hisiwell HXN40057 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

HXN40057 — Discrete & Power Semiconductors | Superb Automation

HXN40057

Hisiwell — Discrete & Power Semiconductors

Product Overview

The HXN40057 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

HXN40057: fifty watts at 5.0 to 5.3 GHz on a 28 V rail

The HXN40057 is the continuous wave member of the Hisiwell group that covers 5.0 to 5.3 GHz. The vendor index row reads 5.0 to 5.3 GHz, saturated output at least 47 dBm, gain at least 11 dB, drain efficiency at least 45 per cent, 28 V drain rail, plus or minus 0.8 dB of flatness, continuous wave. That is 50.12 W of radio frequency from a single packaged transistor, on a rail that most microwave systems already carry.

The numbers behind the headline are the ones that decide whether the part fits an enclosure. At 45 per cent efficiency 50.12 W of output draws 111.4 W from the rail, which is 3.98 A at 28 V, and leaves 61.3 W of heat to leave the flange. Those three figures are calculated from the published rows and are labelled as calculated wherever they appear, because the vendor prints efficiency and output and nothing else.

The drive requirement is unusual for a part this size. With an 11 dB gain floor the drive for rated output is 36 dBm, which is 3.98 W. Four watts of drive is not a small signal input, and it changes the shape of the amplifier chain: this part is not the last stage after a 20 dBm driver, it is the last stage after another power stage. Systems that expect to drive a 50 W device from a 1 W source will need one more stage in front of it.

The 300 MHz window is the other figure to read carefully. At a 5.15 GHz centre it is 5.83 per cent fractional bandwidth, which is narrow for an internally matched part and tells you it was built for a channel plan rather than for a laboratory sweep. Three channels of 100 MHz, or a pair of 5.15 to 5.25 GHz allocations, sit inside it comfortably.

The parameter rows Hisiwell publishes for HXN40057

Part classGallium nitride power transistor, internally matched
Frequency range5.0 to 5.3 GHz
Bandwidth, calculated300 MHz, 5.83 per cent fractional at a 5.15 GHz centre
Saturation output power as publishedat least 47 dBm
Saturation output power in watts, calculated50.12 W
Gain as publishedat least 11 dB
Drain efficiency as publishedat least 45 per cent
Drain voltage28 V
Flatness across the bandplus or minus 0.8 dB
Operating modecontinuous wave
Output at 1 dB compressionnot published, the index prints a slash in that column
DC input at saturation, calculated111.4 W
Drain current at saturation, calculated3.98 A at 28 V
Dissipation at saturation, calculated61.3 W
Dissipation if efficiency falls to 40 per cent, calculated75.2 W
Drive power for the rated output, calculated36 dBm, 3.98 W
Output per 100 MHz of the band, calculated16.71 W
Load voltage at saturation into 50 ohm, calculated50.1 V RMS, 70.8 V peak
Thermal resistance, junction to casenot published
Package outline and dimensionsnot published in text
Scattering parametersnot published
Pinch off voltage and gate bias pointnot published
Operating and storage temperaturenot published
Moisture level and floor lifenot published
Orderable addendum and packingnot published
Neighbour on the same band at 53 dBm, pulsedHXN40060, same rail, 10 dB gain

Where the 61 W of heat goes

61.3 W on a carrier offering 3 C per watt rises 184 C over the case, which no flange will hold. On a 1 C per watt carrier the rise is 61 C, and only then does the junction sit inside a working margin. The arithmetic is blunt: at this power level the carrier is not a heatsink choice, it is the part that decides whether the transistor survives.

The vendor publishes no junction to case figure, so the only path to a defensible number is measurement. The first article should be run at saturation for thirty minutes with a thermocouple on the flange and one on the sink, and the gap between them recorded. A hand built assembly commonly shows a fifteen degree difference across the bolts, and that fifteen degrees is worth 15 per cent of the drain current budget.

Air cooling stops being adequate somewhere near 200 W of dissipation on a conventional fin stack, and this part sits at a third of that, so forced air with a copper spreader is a workable answer for a rack product. A sealed enclosure with no airflow needs a cold plate instead. Either way, the thermal path is settled first and the board is fitted around it.

The efficiency row is a minimum. If the real part runs at 50 per cent instead of 45, the dissipation falls from 61.3 W to 50.1 W, which is a fifth less heat for the same output. Some of that margin should be kept rather than spent on a smaller sink, because the same part at the top of its temperature range loses efficiency rather than gaining it.

Drive, load and the 5.0 to 5.3 GHz channel plan

Four watts of drive at 5 GHz is a real stage. A packaged driver at 36 dBm is a common building block, and the pair of them forms a two stage line that delivers 47 dBm with about 22 dB of total small signal gain. The driver has to hold its own flatness across the band, or the plus or minus 0.8 dB of the output stage is spent before the antenna sees it.

The load voltage is the item that catches people out. 50.12 W into 50 ohm is 50.1 V RMS and 70.8 V peak, which is above the 28 V rail and above the rating of a good many SMA connectors once temperature and altitude are counted. The output network, the circulator if there is one, and every cable used in test has to be rated for 70 V peak at 5 GHz with a margin.

5.0 to 5.3 GHz covers the upper part of the 5 GHz WLAN allocation, the 5.15 to 5.25 GHz and 5.25 to 5.35 GHz windows used for fixed links, and the lower edge of the C band satellite downlink. A part with 300 MHz of matched bandwidth makes one amplifier serve the whole allocation rather than one channel of it, which is worth more to a product line than the last half decibel of gain.

Where the same chassis has to serve a wider span, the family steps out rather than sideways. HXN43004 covers 4.4 to 6.7 GHz at 51 dBm with 8.5 dB gain, and HXN43005 covers 5.0 to 5.9 GHz at 47 dBm with 10.5 dB gain and 52.4 per cent efficiency. Both trade gain for span, and both are the right answer when the channel plan is not fixed at the time the board is built.

Picking between the 5.0 to 5.3 GHz members

Hisiwell lists two parts on this exact band, this one at 47 dBm continuous wave and HXN40060 at 53 dBm pulsed. The difference is not only six decibels of output. The pulsed member runs at 100 microseconds and 10 per cent duty, so its average dissipation is a tenth of the peak, and its gain floor is 10 dB rather than 11. A continuous wave system cannot use it as a drop in replacement even though the band, the rail and the flatness match.

Two generic parts also cover this window from above. HXN42024 spans 0.6 to 6.0 GHz at 44 dBm with 9 dB gain, and HXN42023 spans 0.7 to 6.2 GHz at 49 dBm with 7 dB gain. Both are broadband parts, and both are the cheaper answer when one amplifier has to serve several bands in one box. Against them this part offers 3 dB more gain and 1 dB more output inside a much narrower window.

For a cross reference the sequence matters more than the part number: rail, then mode, then band, then power, then gain. A replacement that matches the power but runs on 32 V or on a pulsed duty cycle is not a replacement. Our RF power transistor cross reference page sets out that sequence and the questions we ask before a candidate goes on a bench.

Where the original part is out of production rather than simply out of stock, the same band is often served by a part from a second vendor with a different die and the same footprint class. The obsolete RF part replacement route covers that case, including the bench work needed to prove the substitute before a production order is placed.

What to settle before the part goes on a board

The gate bias point is not published. The vendor gives 28 V on the drain, 45 per cent efficiency and 11 dB of gain, and stops there. There is no pinch off voltage, no quiescent current recommendation and no gate voltage window on the page, so the first article is a bias sweep against drain current, not a plug and play assembly.

The second item is pulse capability under continuous wave drive. A continuous wave part run into a mismatched load can see standing wave voltages well above the 70.8 V peak of a matched load, and the vendor publishes no load mismatch tolerance. A circulator or an isolator at the output is cheap insurance on a 50 W stage, and it also protects the part from a disconnected antenna during test.

The third item is the feed. 3.98 A at 28 V through a board trace is a copper question, and the decoupling bank has to hold the rail through the envelope of the modulation. A bank sized for a 10 W part will sag on this one at the start of every burst.

Nothing on the vendor page covers packing, marking or moisture level, so the purchase order has to name the band, the rail, the mode and the quantity, and the reel or tray is confirmed with the factory before a date is quoted. Lead time for the family comes through the RF procurement route, where this part sits on the same line as the smaller members of the band.

Questions buyers ask about HXN40057

Is 47 dBm the saturation figure or the compression figure. The index prints 47 dBm in the saturation column and a slash in the 1 dB compression column, so the published number is saturation. A system that needs linear operation has to back off from it, and the vendor gives no compression point to budget against.

Can it run at 32 V or 48 V. The published rail is 28 V. The family has 32 V and 50 V members on other bands but none on 5.0 to 5.3 GHz, so a higher bus needs a step down feed rather than a different part in the same footprint.

What is the gain flatness across the 300 MHz. The vendor prints plus or minus 0.8 dB for the band. That is a peak to peak envelope across the window, so a link budget built on the 11 dB floor and the flatness figure together is the conservative reading.

Does the part ship with a data sheet. The vendor publishes an index row and a product page, and no data sheet appears on either, so there are no scattering parameters, no thermal derating curves and no load pull data. Everything on this page that is not in those two sources is marked calculated. Where a customer needs measured data before a build freeze, our guide to choosing an RF amplifier sets out what can be established on a bench and what has to come from the vendor.