HXC50097
Product Specifications
HXC50097
Hisiwell — Discrete & Power Semiconductors
Product Overview
The HXC50097 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
HXC50097: the 80 W entry rung of the 50 V HF to 1 GHz singles
The HXC50097 is an 80 W single ended LDMOS RF power transistor on a 50 V rail, filed on the HF to 1.0 GHz line and marked CW on the index. The vendor page calls it highly rugged and unmatched, states that it can run pulsed, CW or any modulated signal in linear or saturated operation, and lists ten application bands from 2 MHz to 1 GHz. It is the bottom rung of the single ended ladder on this shelf and it is the row most often bought for a first article.
Eighty watts on 50 V is a modest ask of the device and a demanding ask of everything around it. 80 W into 50 ohm is 63.2 V rms and 89.4 V peak, which is only 1.79 times the rail. That ratio is the lowest of any row above 50 W on this shelf, and it means the drain never has to swing far above its supply: the load line is comfortable, the breakdown margin is wide, and the part is far from the edge of its ratings at rated power. What it also means is that the part will not fail first. The bias network, the driver and the heat sink will.
No gain, no efficiency, no thermal resistance and no breakdown voltage are published for this row. That is the normal state of this shelf and not a fault of this part, but it moves the burden of proof to the buyer: the amplifier has to be measured, and the measurement has to include the case temperature at full power, because there is no published number to fall back on.
The HXC50097 rows as the vendor publishes them
| Part class | LDMOS RF power transistor, unmatched, single ended |
| Frequency range, index row | HF to 1.0 GHz |
| Frequency range, vendor page | HF to 1.0 GHz |
| Output, index row | 80 W |
| Output in dBm, calculated | 49.03 dBm |
| Drain rail, index row | 50 V |
| Mode, index row | CW |
| Signal flexibility, vendor page | Pulsed, CW or any modulated signal, linear or saturated |
| Load voltage at 80 W into 50 ohm, calculated | 63.2 V rms, 89.4 V peak |
| Peak against the 50 V rail, calculated | 1.79 times |
| Load voltage at a 2 to 1 mismatch, calculated | 179 V peak |
| Load voltage at a 3 to 1 mismatch, calculated | 268 V peak |
| DC input at 60 per cent efficiency, calculated | 133 W, 2.67 A at 50 V |
| Heat to remove at 60 per cent, calculated | 53 W |
| DC input at 55 per cent efficiency, calculated | 145 W, 2.91 A at 50 V |
| Heat to remove at 55 per cent, calculated | 65 W |
| DC input at 70 per cent efficiency, calculated | 114 W, 2.29 A at 50 V |
| Heat to remove at 70 per cent, calculated | 34 W |
| Gain | Not published |
| Efficiency | Not published |
| Thermal resistance, junction to case | Not published |
| Breakdown voltage and maximum ratings | Not published |
| Package | Not published on the row or the page |
| Orderable addendum | None published |
| Moisture level and packing quantity | Not published |
| Application bands listed, page | 2 to 30 MHz, 30 to 88 MHz, 54 to 88 MHz, 88 to 108 MHz, 118 to 140 MHz, 136 to 174 MHz, 160 to 230 MHz, 30 to 512 MHz, 470 to 860 MHz, 100 kHz to 1000 MHz |
What the application list actually tells a buyer
The ten bands on the page are not a marketing list, they are a map of where a 50 V LDMOS part with an unmatched input is normally used. The FM band at 88 to 108 MHz and the two television bands at 54 to 88 and 470 to 860 MHz are the broadcast use cases. The avionics band at 118 to 140 MHz and the commercial ground band at 136 to 174 MHz are the air traffic and two way radio cases. The 30 to 512 MHz span is the one that covers jammers and multiband military radios, and the 100 kHz to 1000 MHz span is the instrumentation and industrial heating case.
What the list does not say is how the part behaves at the ends of each of those spans. A single network cannot hold gain flat from 100 kHz to 1 GHz, so the part is used in a sub band with the matching cut for the 20 or 50 MHz the system occupies. The published band is a capability envelope, not a flatness guarantee, and the difference matters when a buyer reads 1 GHz and assumes 80 W at 1 GHz with the same network that gives 80 W at 100 MHz.
The list is also useful in the other direction. A buyer whose requirement sits at 120 MHz avionics or at 470 MHz television is inside two of the listed bands and can reasonably ask for a reference fixture at that frequency. A buyer whose requirement sits at 2.4 GHz is outside every band on the page and should be looking at a different shelf, not at this row with a hopeful note about a wideband match.
The rail, the load and the mismatch case at 80 W
Eighty watts is where the numbers stop being dramatic and start being useful. The load voltage is 63.2 V rms, the peak is 89.4 V, and the ratio to the rail is 1.79. Nothing in that set of numbers threatens a 50 V part with a normal breakdown rating, and the ruggedness claim on the page is credible because the part is not being pushed.
The mismatch case is where the network gets tested rather than the die. At 2 to 1 the load sees 179 V peak and at 3 to 1 it sees 268 V. A 250 V capacitor in the output match survives the 2 to 1 case and fails the 3 to 1 case, and a failure at 3 to 1 happens during an antenna fault, which is exactly when nobody is watching the instrument. Choose the output network components on the mismatch case, not on the matched case, and put the margin in the capacitor voltage rating rather than in the connector.
The supply side is easy at this power. At 60 per cent efficiency the pair draws 2.67 A from the 50 V bus, which any reasonable bus can supply without droop. The interesting case is the modulated signal: an 80 W CW rating does not mean 80 W of average power with a 10 dB peak to average ratio, because the peak envelope at that ratio would be 800 W and the part would be running 10 dB above its rating on peaks. On a modulated signal the average power has to come down by the peak to average ratio, and that derating is the buyer's calculation, not the vendor's.
Thermal at 80 W: small heat, real consequences
At 60 per cent efficiency the heat is 53 W and at 55 per cent it is 65 W. Those are small numbers next to the 260 and 400 W rows, and they are the reason this part is often built on a modest heat sink with a natural convection path rather than a forced air tunnel. The trap is that small heat loads are easy to underestimate because the temperature rise looks unimportant until the part is in a sealed enclosure at 60 degrees ambient.
With no published thermal resistance, the calculation has to start from a measured case temperature. A practical sequence for a first article is to run the amplifier at rated output for thirty minutes, read the flange temperature with a contact probe, read the ambient inside the enclosure, and then compute the sink to ambient resistance from the difference. That gives a number for the actual assembly, which is more useful than any figure the vendor could publish for a reference fixture.
The junction figure then follows from the vendor once the thermal resistance is obtained. Until it is, a conservative assumption is 1 degree per watt junction to case, which on 65 W of heat gives a 65 degree rise and puts the junction at 150 degrees with an 85 degree case. That leaves room against a 225 degree limit and it is honest about being an assumption rather than a published figure.
Drive, bias and the parts of the chain that are not on the data sheet
With no gain published, the driver has to be sized on an assumption. On this shelf, single ended rows between 80 and 200 W publish nothing, while the 28 V rows of similar power publish 15 to 23 dB. A reasonable planning figure for a 50 V unmatched LDMOS at this power is 15 to 18 dB, which puts the drive for 80 W between 1.3 W and 2.5 W. A 3 W driver with a pad covers both cases and gives room for the gain loss that comes with age and temperature.
The bias side is where most first articles fail. An LDMOS part needs a gate voltage set for a quiescent current, typically 100 to 200 mA on this class of device, and the threshold voltage varies from part to part by a few hundred millivolts. The bias network must be adjustable per device, must be sequenced so the gate is negative before the drain is applied, and must not be able to apply the full drain voltage with the gate floating. A single fixed divider is cheaper and fails on the second build.
The third item is the input match. Unmatched means the input impedance is whatever the die presents, and it changes with frequency. A driver that is stable into a mismatch and a pad that isolates it from the variation are worth more than a decibel of gain. Measure S11 at the input of the assembled stage across the band and put the result in the build file, because the next build will use a different driver and the same question will come back.
Where the HXC50097 sits on the ladder
Above it on the same 50 V rail and the same HF to 1 GHz band are the 100 W and 140 W singles and the 170 W and 200 W rows, then the pairs at 260, 400 and 500 W. The steps are small: 80 to 100 W is 0.97 dB, 100 to 140 W is 1.46 dB, 140 to 200 W is 1.55 dB. A requirement that lands between two rungs is normal on this shelf, and the answer is to choose the rung above and run it backed off rather than the rung below and run it flat out.
The choice against the 100 W row is the clearest example. They differ by 0.97 dB, which is inside the measurement uncertainty of most bench setups, so the decision comes down to availability, price and thermal headroom rather than to electrical performance. A buyer who needs 70 W of headroom across temperature takes the 100 W row and runs it at 70 per cent; a buyer who needs 70 W at 55 degrees ambient with a passive sink takes the 140 W row and gains 2.4 dB of margin for one rung of cost.
The full ladder with drive, current and heat at each step is on the RF power transistor cross reference page. The matching and bias components that go around this part, including the capacitors chosen on the mismatch case rather than the matched case, are covered on the RF component sourcing page, and the second source question for a row with no published gain is set out on the one stop RF supply page.