HXC50105
Product Specifications
HXC50105
Hisiwell — Discrete & Power Semiconductors
Product Overview
The HXC50105 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
HXC50105: the 60 W pair with a real document, and the 1.4 degree per watt that sets its limit
The HXC50105 is a 60 W LDMOS RF power transistor built as two 30 W devices in a pair on a 50 V rail, filed on the HF to 1.5 GHz line and marked CW. Unlike most of the rows on this shelf, this one links a document with real numbers: nine pages of maximum ratings, thermal data, electrical characteristics, a functional test at 915 MHz, a package outline and a fixture bill of materials. The document is titled HX515060VP rather than HXC50105, which is the same naming pattern seen on other rows of this vendor, and the numbers inside match the row.
The figure that decides how this part is used is the thermal resistance of 1.4 degrees per watt, measured with the case at 85 degrees and 60 W of CW output. That is nearly three times the 0.5 degrees per watt published for the 260 W row on the same shelf, and it changes the thermal arithmetic completely: at 60 per cent efficiency the heat is 40 W, the rise is 56 degrees, and the junction sits at 141 degrees with an 85 degree case. The margin is real but the coefficient is the highest on the shelf, which means the case temperature has to be held rather than assumed.
Everything else in the document is consistent with a modern unmatched LDMOS part: 125 V breakdown, plus or minus 10 V on the gate, a 225 degree junction limit, ESD Class 2, a flanged ceramic body with two mounting holes and four leads, and a functional row of 23 dB gain at 60 W with 60 per cent efficiency and a 10 to 1 ruggedness rating.
The HXC50105 rows as the vendor publishes them
| Part class | LDMOS RF power transistor, thermally enhanced, unmatched, two devices in a pair |
| Frequency range, index row | HF to 1.5 GHz |
| Output, index row | 30 W per device, 60 W for the pair |
| Output in dBm, calculated | 44.77 dBm per device, 47.78 dBm for the pair |
| Drain rail, all sources | 50 V nominal, 55 V maximum operating |
| Drain to source breakdown, document | 125 V typical |
| Gate to source maximum, document | minus 10 V to plus 10 V |
| Junction temperature maximum, document | 225 degrees C |
| Case operating temperature, document | 150 degrees C |
| Thermal resistance, junction to case, document | 1.4 degrees C per watt at 85 degree case and 60 W CW |
| Power gain at 60 W, document | 23 dB typical at 915 MHz, 25 dBm drive |
| Drain efficiency at 60 W, document | 60 per cent typical |
| Input return loss, document | minus 7 dB typical |
| Ruggedness, document | 10 to 1 VSWR at all phase angles |
| Gate threshold voltage, document | 2.65 V typical at 600 microamps |
| Gate quiescent voltage, document | 3.57 V at 200 mA drain current, 50 V rail |
| On state resistance, document | 900 milliohms, each half of the device measured |
| Input capacitance, document | 28.3 pF at 50 V, 1 MHz, per half |
| Output capacitance, document | 11.9 pF at 50 V, 1 MHz, per half |
| Feedback capacitance, document | 0.38 pF at 50 V, 1 MHz, per half |
| ESD class, document | Human body model Class 2 |
| Package, document | Flanged ceramic, two mounting holes, four leads, PKG-LB or LBB, 16.51 mm overall length |
| Application bands listed, page | Includes 960 to 1215 MHz avionics and 1200 to 1400 MHz L band, plus the HF to 1 GHz list |
| Orderable addendum | None published |
| Moisture level and packing quantity | Not published |
Why 23 dB of gain changes the driver decision at 60 W
At 23 dB and 60 W the drive needed is 0.30 W, or 24.78 dBm. That is a quarter of a watt, which is inside the output of a small driver MMIC and inside the reach of a 1 W part with a pad. Compare it with the 260 W row, which needs 3.27 W at 19 dB, and the shelf starts to make sense: the smaller device has 4 dB more gain and needs a tenth of the drive, because gain falls as the die gets larger.
The high gain has a price. A 23 dB stage with an unmatched input is closer to instability than a 15 dB stage, because the feedback capacitance of 0.38 pF couples more of the output back into the input when the surrounding gain is high. The document quotes a typical input return loss of minus 7 dB, which is a match that can be worked with, but the practical build still wants a pad or an isolator in front of the pair and a build that keeps the input and output networks apart.
The 25 dBm drive in the functional row is also a useful cross check. At 25 dBm in and 60 W out the gain works out at 22.8 dB, which is the 23 dB in the table. A buyer who measures 18 dB on a first article has a fixture problem rather than a part problem, and knowing the expected number saves a week of correspondence.
The thermal coefficient is the constraint, not the power
At 1.4 degrees per watt the case has to be held. Forty watts of heat gives a 56 degree rise, so an 85 degree case puts the junction at 141 degrees and a 100 degree case puts it at 156 degrees. Both are inside the 225 degree limit, so the part is not marginal, but the sensitivity is four times that of the 260 W row: a 20 degree error in the case temperature moves the junction by 20 degrees, while on the 0.5 degree per watt part the same error would move it by less than 5 degrees once the heat is scaled.
The practical rule at this coefficient is to measure the flange rather than the sink. A thermocouple on the sink fin reads several degrees below the flange, and at 1.4 degrees per watt a 10 degree error in the assumed case temperature is a 10 degree error in the junction. Put the probe on the flange, next to the ceramic, and let it settle for twenty minutes at full power.
The second rule is that the interface has to be specified. The document gives a flanged ceramic body with two mounting holes, so the interface is a bolted joint with compound, and the flatness of the flange matters. At 40 W of heat a poor interface costs 10 to 15 degrees, which is affordable but eats the margin that the 1.4 coefficient already spent.
The mismatch case and the output network
At 60 W this pair delivers 54.8 V rms and 77.5 V peak into 50 ohm, which is 1.55 times the rail. At a 2 to 1 mismatch the load sees 155 V peak and at 3 to 1 it sees 232 V. A 250 V capacitor is adequate for the 3 to 1 case with a small margin, which puts this row in a much easier component class than the 140 W single or the 400 W pair on the same shelf.
The 10 to 1 ruggedness rating is the vendor's claim for the die and it is measured at all phase angles, which matters for an antenna that can present any impedance. It does not cover the network: a 10 to 1 mismatch at 60 W can put 500 V across the output match, and the components have to be chosen for that case if the amplifier is expected to survive an open antenna. Rated ruggedness and specified network are two different promises and both belong on the drawing.
The fixture in the document is a narrow band fixture at 915 MHz, which is inside the band but not at the top of it. A buyer who needs this part at 1.3 GHz, where the page lists the L band application, should ask for a fixture at that frequency or measure it. The gain at 915 MHz is 23 dB; at 1.3 GHz the same part will show less, and the difference decides whether a 0.3 W driver is still enough.
What the document does not cover
No orderable addendum is published, so the packing quantity, the marking and the moisture level have to be confirmed in writing. The document carries a revision and a package outline but no tape and reel data, which is normal for a flanged ceramic part and still has to be resolved before a production order.
The document is also silent on behaviour above 1.5 GHz and on the pulse capability that the index row does not claim. Both are reasonable questions for a buyer using the part in an avionics band at 1.03 GHz or an L band radar at 1.3 GHz, and both should be answered by the vendor rather than inferred.
The naming difference between the catalogue number and the document title is the last item. It is the same pattern seen on the 260 W row of this shelf, and it means goods inwards has to accept two names for one part without holding the shipment. The inspection route that handles that case is on the incoming inspection plan page, the second source question for a row with no addendum is on the one stop RF supply page, and the comparison with the 260 W and 400 W pairs on the same rail is on the RF power transistor cross reference page.
Questions buyers ask about the HXC50105
Why is the thermal resistance 1.4 degrees per watt when the 260 W row is 0.5?
Because the die is much smaller. A 60 W part has roughly a quarter of the active area of a 260 W part, so the heat per square millimetre is higher and the path from junction to flange is longer in thermal terms. The absolute heat is lower, 40 W against 173 W, which is why the part still works; the coefficient is simply the price of the smaller body.
Can the HXC50105 run at 1.3 GHz?
The page lists the 1200 to 1400 MHz L band among its applications, so the vendor claims the band. The document, however, measures the functional row at 915 MHz only, and gain at 1.3 GHz on an unmatched part is normally 2 to 4 dB below the 915 MHz figure. A 23 dB part at 915 MHz may be a 19 to 21 dB part at 1.3 GHz, which changes the drive from 0.3 W to 0.6 W.
Is 23 dB of gain realistic in a finished amplifier?
The document measures 23 dB in a demo fixture with 25 dBm of drive, and a production amplifier typically shows 1 to 2 dB less because of connector, cable and combining losses. Budget 21 dB and the amplifier will meet its output on the first build; budget 23 dB and it will be 2 dB short.