HXC50121
Product Specifications
HXC50121
Hisiwell — Discrete & Power Semiconductors
Product Overview
The HXC50121 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
HXC50121: a 175 W pair on HF to 500 MHz and the first row that behaves like UHF
The HXC50121 is an LDMOS RF power transistor from Hisiwell on the HF table with a 0.5 GHz top frequency and a 175 times two output row. Three hundred and fifty watts total from a 50 V rail in CW. It is the smallest pair on that table and the first where 500 MHz is the top of the window rather than a distant edge.
At 500 MHz a quarter wavelength is 150 mm in air and about 100 mm on a laminate, so the matching network can be printed. That changes the cost structure of the amplifier: the transformer that dominates an HF stage is replaced by copper on a board, and the loss moves from the ferrite into the laminate. It also changes the way the part is specified, because a printed network is repeatable in a way a hand wound one is not.
The row publishes no gain and no efficiency, which is normal on this table. It does publish the frequency, the power, the rail and the mode, and from those four the whole supply, cooling and voltage budget can be worked out. That is what this page does, and it also states plainly what cannot be worked out.
The HXC50121 rows as the vendor publishes them
| Part class | LDMOS RF power transistor, push pull pair |
| Frequency range, index row | HF to 0.5 GHz |
| Top frequency in MHz, calculated | 500 MHz |
| Wavelength at the top of the window, calculated | 600 mm in air |
| Quarter wavelength at the top, calculated | 150 mm in air |
| Saturated output, index row | 175 W times two |
| Pair total, calculated | 350 W, 55.44 dBm |
| Per side, calculated | 175 W, 52.43 dBm |
| Step below the 1 kW pair at 300 MHz, calculated | 4.56 dB |
| Step above the 250 W single ended row, calculated | 1.46 dB |
| Drain rail, index row | 50 V |
| Load voltage at 350 W into 50 ohm, calculated | 132.3 V rms, 187.1 V peak |
| Peak against the rail, calculated | 3.74 times |
| Load voltage per side at 175 W, calculated | 93.5 V rms, 132.3 V peak, 2.65 times the rail |
| Peak at a two to one mismatch, calculated | 374.2 V |
| DC input at 55 per cent, calculated | 636 W, 12.7 A at 50 V, 286 W of heat |
| DC input at 65 per cent, calculated | 538 W, 10.8 A at 50 V, 188 W of heat |
| Output per megahertz of window, calculated | 0.7 W |
| Gain | Not published |
| Drain efficiency | Not published |
| Mode, index row | CW |
| Package and flange | Not published |
| Thermal resistance, junction to case | Not published |
| Load impedance | Not published, no load pull table |
| Data sheet | None reachable for this row |
350 W, a 3.74 rail ratio and a board that can be etched
350 W into 50 ohm is 132.3 V rms and 187.1 V peak, which is 3.74 times a 50 V rail. At this ratio a printed output network is viable: the peak voltage is high enough to matter for capacitor rating but low enough that a well chosen high voltage ceramic survives it, and the current is modest enough that track widths of a few millimetres carry it without significant heating.
The peak under a two to one mismatch doubles to 374.2 V, and that is the number that should set the capacitor voltage class. A part rated at 200 V looks adequate against 187 V and is not adequate against 374 V, and a mismatched antenna is a normal condition on a transmitter rather than an exceptional one.
The heat is 286 W at 55 per cent and 188 W at 65 per cent. Compared with the kilowatt rows this is easy to move, and a forced air heatsink with a proper interface can handle it. But the die at 500 MHz is smaller than the die on the 200 MHz rows, so the heat flux is higher than the total suggests, and the thermal path from die to flange is what limits the part rather than the heatsink.
What changes when the network moves onto the board
On the HF rows the output transformer is a wound component with a core, and its behaviour is dominated by the core material. At 500 MHz the same function is performed by printed lines and discrete capacitors, and the dominant term becomes the dielectric constant and the copper loss. The practical consequences are a smaller, cheaper and more repeatable network, at the price of less ability to tune it after the fact.
That has an effect on how the part should be bought. On an HF stage a buyer can accept a device with wide parameter spread and tune the amplifier around it. On a 500 MHz stage the tuning range is narrow, so device spread shows up directly as unit to unit variation in gain and output. A buyer planning volume production at this frequency should ask for the spread, not only for the typical figure.
It also has an effect on the incoming test. A 350 W part is small enough to test on a bench fixture rather than in the final amplifier, and measuring gain and output at 500 MHz on the fixture is fast enough to do on every unit rather than on a sample. That is the cheapest place to catch a batch problem, and it is well inside the scope described on the incoming inspection plan page.
Where the HXC50121 sits on the 500 MHz band
The 500 MHz rows on this table are crowded, and the HXC50121 is the pair at the bottom of the power range. Above it are the 700 W pair, the 450 W pair and the 500 W pair, and the single ended rows at 250, 170, 80 W and 30 W. That spread is the useful part of the table for a buyer: the same band is offered at 1.4 kW, 900 W, 1 kW, 250 W, 170 W and 80 W.
Read against the single ended 250 W row, this pair is 1.46 dB above it, which is not a large step. The difference is that the pair needs a balun and the single ended part does not, so for a 250 W requirement the single ended row is the simpler build and the pair is the one that allows a balanced output. That choice is a system decision and the power difference is too small to make it for the buyer.
Against the 1.4 kW pair built from two 700 W halves, this row gives away 6.02 dB of output on the same band. A buyer who expects to grow a product upward later should consider starting on the larger part and backing the drive off, because the two rows may not share a footprint or a network. The drive and rail comparison for the family is on the RF power transistor cross reference page.
Buying the HXC50121: what the order has to state
State the sub band inside HF to 500 MHz, the rail, the mode, the flange and body, the quantity, the gain and efficiency to be guaranteed, and the parameter spread the buyer will accept. On a 500 MHz part the spread matters more than the typical value, because the network on the board cannot be retuned per unit.
The order should also state whether the part is expected to survive a full mismatch into an open or a short. The row is silent on ruggedness, so the only defensible answer is the one the factory gives in writing, and a protection circulator is the alternative when it does not.
For a module, the devices, the circulator, the couplers and the passives can be bought against one date and one test plan, which is what the one stop RF supply page describes. That matters at 500 MHz because the passives in the output network are the parts most likely to hold a build.
Questions buyers ask about the HXC50121
Is it 350 W from one device? No. The row is 175 W times two, so it is a pair. Each half is 52.43 dBm.
Can it cover 900 MHz? No. The window stops at 500 MHz. The higher bands need a different table.
What is the current draw? 12.7 A at 55 per cent efficiency and 10.8 A at 65 per cent on a 50 V rail, both calculated rather than published.
What capacitor voltage rating should the output network use? 374.2 V is the peak at a two to one mismatch, so the matched figure of 187 V is not the number to size against.
How much gain does it have? Not published. It has to come from a fixture measurement on the first article.