HX0645
Product Specifications
HX0645
Hisiwell — Discrete & Power Semiconductors
Product Overview
The HX0645 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
HX0645: 45 W on the index row, 55 W on the product page, one 6 GHz window
The HX0645 is a gallium nitride RF power transistor filed on the Hisiwell 50 V shelf listing at 45 W, DC to 6 GHz, 18 dB of gain, 75 per cent efficiency, a 48 V rail and continuous wave mode. The product page behind the row states 55 W of maximum saturated power, 68 per cent of best saturated drain efficiency, and a supply that can be either 48 V or 28 V. The two documents agree on the band and disagree on the power, the efficiency and the rail.
The power gap is 0.87 dB, from 46.53 dBm to 47.40 dBm. That is smaller than the 2.30 dB gap on the 20 W member of the same shelf, and it is inside the range where a saturation measurement on different benches can differ, so the two numbers may describe two points on one curve rather than two different parts. The efficiency gap runs the other way, from 75 per cent to 68 per cent, and at 55 W the DC input comes to 80.9 W rather than the 60.0 W implied by 45 W at 75 per cent.
The rail statement on the product page is the more useful one. A supply that works at either 48 V or 28 V is rare on this shelf and it tells the buyer that the die has margin at the lower rail, which is what a mobile or vehicle mounted amplifier needs. It does not say what the output is at 28 V, and that is the number a 28 V programme has to ask for.
The HX0645 rows as the vendor publishes them
| Part class | Gallium nitride RF power amplifier transistor, flanged and unflanged body |
| Frequency range, both documents | DC to 6 GHz |
| Saturation output, index row | 45 W |
| Saturation output, product page | 55 W maximum saturated |
| Difference between the two, calculated | 0.87 dB |
| 45 W in dBm, calculated | 46.53 dBm |
| 55 W in dBm, calculated | 47.40 dBm |
| Gain, index row | 18 dB |
| Drain efficiency, index row | 75 per cent |
| Drain efficiency, product page | 68 per cent best saturated |
| Drain rail, index row | 48 V |
| Drain rail, product page | 48 V or 28 V |
| Mode, index row | Continuous wave |
| Mode, product page | Continuous wave and pulsed, no pulse rating given |
| Load voltage at 45 W into 50 ohm, calculated | 47.4 V RMS, 67.1 V peak |
| Load voltage at 55 W into 50 ohm, calculated | 52.4 V RMS, 74.2 V peak |
| Peak swing against a 48 V rail at 55 W, calculated | 1.55 times |
| Peak swing against a 28 V rail at 55 W, calculated | 2.65 times |
| DC input at 45 W and 75 per cent, calculated | 60.0 W, 1.25 A on a 48 V rail, 15.0 W of heat |
| DC input at 55 W and 68 per cent, calculated | 80.9 W, 1.69 A on a 48 V rail, 25.9 W of heat |
| Package and body | Flanged and unflanged versions, sizes not published |
| Thermal resistance, junction to case | Not published |
| Load impedance at the output | Not published, no load pull table |
| Applied application list | RF and microwave systems |
| Data sheet link on the product page | None, the parameter block is two bullet lines |
| Orderable addendum and packing | None published |
The 28 V question, and why it decides the driver
At 28 V a 55 W output needs 74.2 V peak at the load, which is 2.65 times the rail. That ratio is high but not impossible for a tuned class AB stage on a GaN die with a low knee voltage, and it is exactly the kind of operating point that GaN is chosen for. The consequence is that the drain current at 28 V doubles compared with 48 V for the same DC input, from 1.69 A to 2.89 A, and the feed, the choke and the decoupling all have to be sized for the higher current.
The gain also moves with the rail. A part biased at 28 V usually shows slightly less gain than the same part at 48 V, because the transconductance and the load line both change. The index row publishes 18 dB and does not say at which rail. A 28 V programme that budgets 18 dB of gain from the catalogue may find 16 or 17 dB on the bench, and that one or two decibels comes straight out of the driver requirement.
The practical route is to ask the factory for the 28 V row before the driver is chosen. If the answer is that the part is characterised only at 48 V, the 28 V option is not a specification and the amplifier should be built at 48 V with a down converter in front of it. If the answer carries a 28 V output figure, the driver can be sized once and not twice.
Comparing 45 W and 55 W on the same board
The 0.87 dB between the two published powers is small enough that the same board can usually carry both, which is not true of the 2.30 dB gap on the 20 W member. The load voltage rises from 67.1 V peak to 74.2 V peak, a 10.6 per cent increase, and the DC current rises from 1.25 A to 1.69 A at 48 V because the efficiency also falls. The heat load rises from 15.0 W to 25.9 W, which is a 73 per cent increase on the same footprint.
That last number is the one that decides whether the board carries both. A heat sink sized for 15 W will run a 55 W part at a junction temperature that may be outside the sheet, and the failure mode is not a sudden one; it is a gain that sags with temperature and a part that fails at the end of a hot week. Where both figures are in play, the thermal path has to be sized on the higher heat load and the lower efficiency.
Where the choice is open, the 45 W figure at 75 per cent is the more attractive operating point on paper, and the 55 W figure at 68 per cent is the more output for the same die. A programme that needs the last decibel will take the 55 W number and pay for it in heat; a programme that needs a long life at a moderate output will run the same part below its saturation point and take the efficiency back.
What an order for the HX0645 has to state, and what it cannot yet
No orderable addendum, packing quantity, marking or moisture level is published. The purchase order has to name the part, the body variant, the quantity, the band, the mode, the rail and the bias condition. The body variant is the field most often left open, and on this row the page confirms that a flanged and an unflanged version both exist while mapping neither to a suffix.
The rail is the second field that has to be stated, because the page offers two and the price may differ between them. A quotation that does not name the rail is not a quotation for a part; it is a quotation for a family. The factory reply should confirm the rail, the saturated output at that rail, the gain at the band of use and the efficiency at the same point.
Nothing on the page supports a lead time or a minimum quantity. Both have to come from a factory reply, and until then the honest line to a customer is that the date is unconfirmed rather than that it is short.
Questions buyers ask about the HX0645
Is the output 45 W or 55 W? The index row says 45 W and the product page says 55 W maximum saturated. The two are 0.87 dB apart, which is within the spread of saturation measurements across benches, so the likely reading is one curve and two points on it. The number to plan with is the one the factory confirms for the rail in use.
Can it really run at 28 V? The page states that the part operates from either a 48 V or a 28 V supply. It publishes no output figure at 28 V, and the load voltage at 55 W against a 28 V rail is 2.65 times the rail, which is a demanding operating point. The rail is possible; the output at that rail is unconfirmed.
What is the difference between the F and the P suffix? The page names the part as HX0645F and HX0645P and describes the two as flanged and unflanged bodies. It does not print a dimensional drawing, a flange thickness or a mounting hole pattern, so the suffix can be read as a body choice and cannot be read as a mechanical specification.
How is a batch like this bought? Through a channel with an unbroken trace and an incoming check at the intended band. The RF component sourcing page sets out the trace, and where the same bill carries the driver, the circulator and the passives the one stop RF supply page covers the consolidation.