HX339180F3
Product Specifications
HX339180F3
Hisiwell — Amplifiers & Comparators
Product Overview
The HX339180F3 from Hisiwell is a precision amplifier / comparator delivering high gain, wide bandwidth, and low distortion for signal conditioning. 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
- High gain-bandwidth product for wideband signal amplification
- Low input offset voltage and drift for precision DC applications
- EMI-hardened design for robust operation in noisy environments
- Rail-to-rail input/output for maximum dynamic range
- Low quiescent current for power-sensitive designs
- Unity-gain stable with capacitive load drive capability
Key Specifications
| Channels | — |
| GBWP | MHz |
| Slew Rate | V/µs |
| Vos | µV |
| Iq per Channel | µA |
| Input Type | — |
| Supply Voltage | V |
| Operating Temp | °C |
| Package | — |
Applications
HX339180F3: the same 2 to 10 GHz amplifier in a 3 x 3 mm QFN, and what the package changes
The F3 suffix on this series means the die has been assembled into a 3 x 3 mm QFN. The electrical core does not change: 18 dB of gain, 19 dBm at the 1 dB compression point, 80 mA from 5 V, 2 to 10 GHz. What changes is everything around the numbers, and those are the parts that decide whether a pick and place line can use it.
Published values
| Parameter | Packaged F3 | Bare die sibling |
|---|---|---|
| Frequency range | 2 to 10 GHz | 2 to 10 GHz |
| Small signal gain | 18 dB | 18 dB |
| Gain flatness | plus or minus 1.5 dB | not published |
| Output P-1dB | 19 dBm | 19 dBm |
| Input VSWR | 1.7:1 | 1.4:1 |
| Output VSWR | 1.3:1 | 1.2:1 |
| Supply | +5 V at 80 mA | +5 V at 80 mA |
What the package costs at the input
The input match moves from 1.4:1 to 1.7:1, which is 15.6 dB of return loss becoming 11.7 dB. In a 50 ohm system that is the difference between 2.8 percent and 5 percent of incident power reflected. For a stage fed by a filter or a switch with its own mismatch, the two reflections add in phase at some frequencies, and 1.7:1 combined with a 1.5:1 load can push the effective input match past 2.5:1. If your source is not well matched, add a pad or plan for a tuning stub at the input.
What the package buys
A published flatness figure is the first gain: plus or minus 1.5 dB across 2 to 10 GHz is the number you can put in a budget, where the die version leaves you measuring it yourself. The second is assembly. A QFN part goes through a standard reflow line with standard inspection, and it does not need die attach, wire bonding or a bare die handling area. For volumes above a few hundred units, that usually costs less than the die even at a higher unit price.
Power and heat
0.40 W of DC input and 0.079 W of RF at compression, so about 0.32 W is dissipated. In a 3 x 3 mm package with an exposed pad soldered to a ground plane, that is a comfortable number and does not normally need thermal vias beyond the standard footprint.
Two questions buyers ask
Can I use the F3 where the die version was qualified? The RF core is the same, but the input match and the flatness figures are different, so a like for like qualification is not automatic. Re run the gain and return loss sweep at your band edges.
Which one should I stock? If your line does reflow and your annual volume is in the thousands, the F3 removes a process step. If you build modules and already have a die attach area, the die version gives you a better input match and a smaller footprint.
SUPERB supplies both forms of the HX339180 so the choice can be made on process cost rather than availability. Lead time and price break for the packaged part are confirmed per order, and the component lead time solutions page covers scheduled demand. To compare this part against other 2 to 10 GHz drivers, see the amplifier cross reference hub.
The series index, including the 6 to 18 GHz members, is on the amplifiers and comparators page.
HX339180F3: the reflow line sees this part before the RF bench does
Eighteen dB, 19 dBm and 80 mA from 5 V across 2 to 10 GHz in a 3 by 3 mm QFN. The electrical case for the packaged version over the bare die is already made by the published flatness figure. The other half of the case is process: what a QFN in this band needs from the assembly line, and what happens to the RF result when those requirements are treated as mechanical details.
The exposed pad is the RF ground, not a heat slug
In a 3 by 3 mm QFN at 10 GHz, the ground path from the die to the board runs through the exposed pad and a via field. The 0.32 W dissipated in this part is small enough that the pad does not need to be a thermal feature. It does need to be an electrical one: a via under or immediately beside the pad, on the same net as the ground plane the input and output launches reference, or the return current from the output port has to find its way around the via field and the 1.3 to 1 output match published in the row will not appear on your board. A common pattern is a grid of 0.3 mm vias at 0.6 mm pitch under the pad, connected to the ground plane on layer two with no thermal relief.
| Assembly item | What to control | Why it matters at 10 GHz |
|---|---|---|
| Exposed pad solder | void rate and reflow profile | the pad carries the RF return, not just the heat |
| Via field under the pad | count, pitch, no relief | source inductance sets the gain at the top of the band |
| Input and output traces | 50 ohm, referenced to the same plane | a launch that is not 50 ohm appears as extra mismatch loss |
| Moisture handling | the level is not in the row, so ask | a plastic package that absorbs moisture can delaminate in reflow |
| Cleaning | no aqueous wash into an unsealed package | residue under the part changes the match and cannot be seen |
What the package gives back in repeatability
The published input match is 1.7 to 1, which is 11.7 dB of return loss and 5 percent of the incident power reflected, against 1.4 to 1 and 2.8 percent for the die. That is the price. What the package sells is that every unit has the same launch: the bond wires inside the package are fixed, so the part that arrives on tape and reel is the part that was measured, and a 500 unit build does not need a die attach qualification, a wire bond pull test or a bare die handling area. For a build where the RF section is a small part of a larger assembly, that repeatability is worth more than the 0.1 dB of extra mismatch loss.
How to read the 1.5 dB flatness figure against a filter
Plus or minus 1.5 dB across 2 to 10 GHz is what the packaged part publishes and the die does not. In a chain with a filter after it, the flatness adds directly to the filter's own passband ripple, and 1.5 dB plus 0.5 dB is 2 dB of total variation that has to be inside the dynamic range of whatever follows. If the following stage is a 19 dBm compression point amplifier and the output is planned at 12 dBm, there is 7 dB of headroom and 2 dB of ripple is harmless. If the output is planned at 17 dBm, the ripple takes the peaks to compression and the flatness figure stops being a gain question and becomes a distortion question.
Two questions buyers ask
Can the same board take the die and the package? Not without a different footprint and a different input match. The package route is 3 by 3 mm with a pad, the die route is attach and bond, and the two need different launch geometry. Pick one before the board is released rather than after.
Is the packaged part qualified for the same environment? The row publishes no temperature range, so the honest answer is that it is not stated. For a 2 to 10 GHz part used in an instrument the question is usually about the solder joints rather than the die, and the assembly checks that matter are the same ones that apply to any QFN in this band, which our one stop RF supply page covers together with the die alternative on amplifier cross reference.
For a build that has to keep the same band when the part is unavailable, the route on obsolete RF part replacement starts from the gain and the match rather than from the part number.
Request Quote — In Stock, 100% New & Original, Full Warranty