HX1020120P2
Product Specifications
HX1020120P2
Hisiwell — Discrete & Power Semiconductors
Product Overview
The HX1020120P2 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
HX1020120P2: three power figures for one part and a one gigahertz band
The HX1020120P2 is a gallium nitride RF power transistor from Hisiwell for the 1 to 2 GHz window on a 28 V rail. Three different power figures appear in the vendor text. The family index prints 140 W of saturation output with 20 dB of gain and 80 per cent drain efficiency. The product overview sentence calls it a 120 W part. The parameter block on the same product page prints a maximum saturated power of 155 W and a best drain efficiency of 81.1 per cent.
That is three figures spread over 1.1 dB, and a buyer is entitled to ask which one is the specification. The answer from the documents is that none of them is labelled as a guaranteed minimum. The page therefore quotes 140 W as the index figure, carries the 155 W parameter block and the 120 W sentence, and treats the lot data as the number a purchase order is written against.
The band is one gigahertz wide, 1 to 2 GHz, which is narrow for a part at this power and is the reason the efficiency can be 80 per cent. A narrow band match at 140 W is a practical lumped network rather than a transmission line transformer, which is a different board, a different bill of materials and a different assembly process from the wideband members of the same shelf.
One hundred and forty watts is 51.46 dBm. With 20 dB of gain the drive at saturation is 31.46 dBm, which is 1.40 W. At the 155 W figure the drive is 31.90 dBm, 1.55 W. At 80 per cent drain efficiency the DC input is 175.0 W, the drain current is 6.25 A at 28 V and the heat is 35.0 W. At 81.1 per cent and 155 W the DC input is 191.1 W, 6.83 A and 36.1 W of heat.
The parameter rows behind the HX1020120P2
| Part class | Gallium nitride RF power transistor, unmatched |
| Frequency range | 1 to 2 GHz |
| Saturation output, index row | 140 W |
| Maximum saturated power, product parameter block | 155 W |
| Power in the overview sentence | 120 W |
| Power in dBm, calculated | 51.46 dBm at 140 W, 51.90 dBm at 155 W, 50.79 dBm at 120 W |
| Spread between the three figures | 1.11 dB from 120 W to 155 W |
| Power gain | 20 dB |
| Drive power at 140 W, calculated | 31.46 dBm, 1.40 W |
| Drive power at 155 W, calculated | 31.90 dBm, 1.55 W |
| Drain efficiency, index row | 80 per cent |
| Best drain efficiency, parameter block | 81.1 per cent |
| DC input at 80 per cent, calculated | 175.0 W, 6.25 A, 35.0 W of heat |
| DC input at 81.1 per cent and 155 W, calculated | 191.1 W, 6.83 A, 36.1 W of heat |
| Drain rail | 28 V |
| Operating mode | Continuous wave |
| Load voltage at 50 ohm, calculated | 83.7 V RMS, 118.3 V peak at 140 W |
| Load voltage at 155 W, calculated | 88.0 V RMS, 124.5 V peak |
| Load voltage under a two to one mismatch, calculated | about 167 V peak at 140 W |
| Output per GHz of band, calculated | 140 W per GHz |
| Matching | Unmatched, external network on the customer board |
| Body and package | Not published |
| Thermal resistance, junction to case | Not published |
| Input and output impedance | Not published in text |
| Gate bias point and quiescent current | Not published |
| Compression point | Not published |
| Orderable addendum, marking and packing | Not published |
Six amps, thirty five watts of heat and one gigahertz to match
At 80 per cent the part leaves 35 W of heat for 140 W of output, which is the best ratio on this shelf and the reason a narrow band part is worth having. Six point three amps at 28 V through a board is a 175 mil trace at one ounce, and the drain choke carries the full current with a saturation margin above seven amps.
Thirty five watts is a bolted spreader and a finned sink, or a metal backed board with a liquid plate. It is not a DFN on a laminate with a via field. The package is not published, but at this power the body is a flanged ceramic or a large plastic over moulded part, and the interface is a torque specification rather than a solder joint.
The output match is easier than on the wideband members for two reasons. One gigahertz of band allows a two or three section lumped match with a low pass structure, and the capacitors in that match see 118.3 V peak matched and about 167 V peak under a two to one mismatch, which is inside the range of 250 V parts but well outside 50 V ones.
The drive is the part that catches people. Twenty decibels of gain is a lot for a 140 W device, and it means the stage in front can be a small one. It also means the input is sensitive: a 1.4 W drive that varies by a decibel across the band is a 0.5 dB output variation, so the driver has to be flat rather than merely adequate, and the input match has to hold across the band or the gain curve tilts.
HX1020120P2 against the one gigahertz shelf
| HX1020120P2 | 1 to 2 GHz | 140 W | 20 dB | 80 per cent | Reference point |
| HX1020240P2 | 1 to 2 GHz | 280 W | 20 dB | 80 per cent | 3.01 dB above, 28 or 48 V |
| HXN50256 | 0.7 to 2.7 GHz | 120 W peak | Not published | Not published | 0.67 dB below, 1.7 GHz wider |
| HXN50254 | 0.7 to 2.7 GHz | 60 W peak | Not published | Not published | 3.68 dB below |
| HX2040185F | 2 to 4 GHz | 185 W | 18 dB | 70 per cent | 1.21 dB above, different band |
| HX446050F2 | 4.4 to 6 GHz | 50 W | 15 dB | 55 per cent | 4.47 dB below, different band |
The HX1020240P2 is the same band and the same gain, 3.01 dB higher, on 28 or 48 V. A 48 V rail halves the current for the same power, which for a 280 W part is the difference between 12.5 A and 7.3 A, and that difference decides whether the board is a thick copper laminate or a conventional one.
Against the wideband group the trade is band for gain and efficiency. The HXN50256 covers 0.7 to 2.7 GHz at 0.67 dB less output with no published gain and no published efficiency, and this part covers one gigahertz with both published. For a product that only needs the 1 to 2 GHz window, this part is the easier build.
The comparison across the shelf is where the RF power transistor cross reference earns its place, because the three power figures printed by the vendor are not comparable between members without a note of which figure is which. The one stop RF supply page covers buying the driver stage and this part as one kit rather than from two schedules.
Board notes for a narrow band hundred and forty watt stage
The match is the first item and it is where the money goes. One gigahertz of band at 140 W is a low pass ladder on a laminate with a defined dielectric, and the series element carries 6 amps of RF current. The shunt capacitors see 118 V peak, so the choice is a high Q porcelain or a thin film part rather than an ordinary ceramic.
The bias is the second item. No gate voltage and no quiescent current is published. Class AB at this power draws a substantial quiescent current and the drain feed has to hold the rail through the envelope of the signal, which means the bulk bank is sized for the pulse current rather than the average.
The sequencer is the third item. Negative gate present and clamped before the drain rises, drain removed before the gate falls, bleeder on the gate. A 28 V part with 6 amps of drain current will destroy itself in microseconds if the gate floats while the drain is live.
The thermal interface is the fourth. No junction to case resistance is published, so the case temperature at rated output in the final enclosure is the measurement that closes the loop, and the first article is where it is taken.
Questions buyers ask about HX1020120P2
Is it 120 W, 140 W or 155 W? All three appear in the vendor documents, the 140 W in the index, the 155 W as a maximum, the 120 W in the overview sentence. We quote 140 W as the working figure and ask for the lot data before a schedule is set.
What is the drive power? At 20 dB of gain and 140 W of output it is 1.40 W, and at 155 W it is 1.55 W. That is derived from the gain row and it holds as long as the gain is 20 dB at the band edges, which the vendor does not state separately.
Is it linear enough for a modulated signal? The vendor publishes no compression point, so the back off has to be measured. The 80 per cent efficiency is a saturation figure and the efficiency at a backed off operating point is lower.
Can it run on 48 V? The index row and the parameter block both state 28 V for this member. The 28 or 48 V option belongs to the HX1020240P2 in the same family, so a 48 V rail is a different part number rather than a different bias point.
How do we start? Send the band edges, the output at the connector, the signal, the ambient and the rail. We will confirm which power figure the current lot is screened to, quote from the schedule and ship a sample with gain and efficiency measured at 1.0, 1.5 and 2.0 GHz.