HXN50156x2
Product Specifications
HXN50156x2
Hisiwell — Discrete & Power Semiconductors
Product Overview
The HXN50156x2 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
HXN50156x2: a 1500 W amplifier row built from two HXN50156 transistors
The HXN50156x2 is not a transistor. The product page says it plainly: the row describes an amplifier assembled from two HXN50156 gallium nitride transistors on a 50 V rail, delivering more than 1500 W of pulsed output across 960 to 1225 MHz, and a buyer who wants to build the same amplifier is told to order the single HXN50156 instead. The shelf row beside it prints 1500 W, 13 dB of gain, 57 per cent efficiency and a pulse mode, and it prints those numbers for the pair rather than for one device.
The distinction matters at every step of a quotation. A 1500 W figure is 61.76 dBm, and with 13 dB of gain the drive needed is 48.76 dBm, which is 75.2 W. A buyer who treats the row as a single transistor will plan a 75 W driver for one device, when the same 75 W is what the pair needs as a block. Split the pair and each transistor delivers about 750 W, which is 3 dB down, and the drive for each falls with it.
The band is also worth reading carefully. The index row prints 0.96 to 1.2 GHz and the page prints 960 to 1225 MHz. Those are not the same window at the top: 1.2 GHz is 1200 MHz and the page allows 25 MHz more. Where a programme operates at 1215 MHz, the page is the document that governs and the row is the one that would be quoted in a catalogue.
The HXN50156x2 rows as the vendor publishes them
| Part class | Amplifier assembled from two HXN50156 gallium nitride transistors |
| Frequency range, product page | 960 to 1225 MHz |
| Frequency range, index row | 0.96 to 1.2 GHz |
| Band width, calculated | 265 MHz on the page, 240 MHz on the row |
| Output power, product page | More than 1500 W pulsed |
| Output power, index row | 1500 W |
| Output in dBm, calculated | 61.76 dBm |
| Output per transistor, calculated | About 750 W each, 3 dB below the pair |
| Output per megahertz of band, calculated | 5.66 W per MHz |
| Gain, index row | 13 dB |
| Drive needed for the pair at 13 dB, calculated | 48.76 dBm, 75.2 W |
| Efficiency, index row | 57 per cent |
| DC input at 57 per cent, calculated | 2632 W, 52.6 A on a 50 V rail |
| Heat to move at 57 per cent, calculated | 1132 W, before any duty cycle derating |
| Drain rail, index row | 50 V |
| Load voltage at 1500 W into 50 ohm, calculated | 273.9 V RMS, 387.3 V peak |
| Peak swing against the rail, calculated | 7.75 times |
| Mode, index row | Pulse |
| Pulse width and duty on the page | Not published |
| Package and body | Not published for the pair or for the single transistor |
| Thermal resistance, junction to case | Not published |
| Load impedance at the output | Not published |
| Applied application list | RF and microwave systems |
| Orderable addendum and packing | None published for either part number |
Buying a pair or buying an amplifier
There are two ways to read this row and the choice changes the whole bill of materials. The first is to buy the pair as a block, in which case the vendor supplies a matched combination whose output is already combined and whose input is already split. The second is to buy two HXN50156 transistors and build the splitter and the combiner on the customer board, which is what the page tells a buyer to do when the intent is to reproduce the amplifier.
The first route is faster and the second is cheaper. A combiner built from two 750 W devices has to handle the loss of the splitter, the amplitude and phase balance between the two paths, and the isolation when one device fails. Those three numbers do not appear on the page, and on a pulse part at 1215 MHz they are the numbers that decide whether the pair survives a mismatch.
The second route carries the combiner inside the customer's own scope, and with it the risk. Where the programme has an RF engineer with a load pull bench, the second route is the normal one. Where it does not, the pair as a block is the honest purchase, because a pair that arrives combined has been balanced by the party that made the die.
What 75 W of drive means upstream
The 13 dB of gain on the row is the number that sets the driver chain. A 1500 W output at 13 dB needs 75.2 W of drive, and 75 W at 960 to 1225 MHz is itself a two stage amplifier with its own bias, its own heat and its own control loop. A programme that budgeted a single 10 W driver will be 9 dB short and will discover it after the first module is built.
There is a way to soften this. Where the pair is driven as two separate paths, each device needs half the drive at half the output, and the driver can be split earlier at a lower level where a single stage still has the gain. That arrangement trades a splitter loss for a simpler driver, and it is the usual route when the amplifier has to fit a defined enclosure.
The published gain is a small signal figure at one point on the band, and no gain curve is published. On a 265 MHz window the gain at 960 MHz and at 1225 MHz will differ, and the driver has to be sized for the weaker end. Without a curve the safe assumption is that the top of the band is the weaker end, and the margin to allow is a decibel or two.
Pulse mode and the heat that is not there on average
At 57 per cent efficiency the pair draws 2632 W and puts 1132 W into heat during a pulse. That is more heat than a small room heater, and it is only manageable because the mode is pulse. The page does not state the pulse width or the duty cycle, so the average heat cannot be calculated from the vendor data at all; the row says pulse and stops.
What can be said is that the average dissipation falls in proportion to the duty. At a 10 per cent duty the average heat drops to 113 W plus quiescent, which is a normal heatsink problem rather than a special one. At a 30 per cent duty it is 340 W, which is a liquid cooled problem. The duty is therefore a specification the buyer has to set, not a detail.
The peak voltage does not fall with duty. Every pulse puts 387.3 V peak at the load, which is 7.75 times the 50 V rail, and the output network and the connector have to stand it. That is the number that decides whether a standard N connector is adequate or whether something larger is needed at 1500 W.
Where the pair sits against the single transistor rows
Around it the shelf holds single device rows in the same 0.96 to 1.4 GHz region: the HXN50154 at 800 W on 960 to 1200 MHz, the HXN50152 at 350 W across 960 to 1400 MHz and 400 W in a narrow window, and the 500 W row at 960 to 1400 MHz that appears under two part numbers. Those rows publish a gain and an efficiency and are meant to be built into an amplifier by the buyer.
The pair is the only row on this shelf that describes a finished combination. Where the requirement is a number of watts at the flange of a module and the schedule is short, the pair is the lower risk purchase. Where the requirement is a board with a specific outline, the single devices give the freedom the pair does not.
The comparison across these rows is set out on the RF power transistor cross reference page, including the point where a pair of smaller devices costs less than one larger device and where it does not.
Questions buyers ask about the HXN50156x2
Do I order one HXN50156x2 or two HXN50156? The page says the row describes an amplifier built from two HXN50156, and that a buyer who wants to build the same amplifier should order the single transistor. So the x2 string is a description of a combination and the HXN50156 is the orderable device. The pair as a supplied block has to be confirmed with the factory rather than assumed from the row.
Is 1500 W a peak or a continuous figure? The mode column says pulse and the page says pulsed output. No pulse width and no duty cycle are published, so the figure is a pulse figure without a stated condition, and any continuous use has to be treated as outside the published data.
Can it run at 1215 MHz? The page allows up to 1225 MHz and the index row stops at 1.2 GHz. The page is the more detailed document and permits the higher frequency, but the vendor's standard sentence adds that performance is not guaranteed outside the stated window, so 1215 MHz sits inside the page window and outside the row window, which is a good question to put in writing.
How is a part like this bought and checked? Through a channel that can prove the batch, with an incoming check at the operating band and a pulse test at the stated condition. The RF component sourcing page describes the trace, and where the module needs its driver and its passives against the same date the one stop RF supply page covers the consolidated buy.