160nH Broadband Conical Inductors 100MHz - 22GHz Ultra Wideband RF Choke 22GHz Self Resonant Frequency

160nH HALT25008 Broadband Conical Inductor 100MHz-22GHz Ultra-Wideband RF Choke 22GHz Self-Resonant Frequency

Solving the Broadband Choke Dilemma with Conical Topology

Every RF engineer who designs active microwave circuits faces the same trade-off: how to block DC from leaking into the RF path without destroying wideband signal transmission. The physics of conventional wirewound inductors creates an unavoidable conflict—more turns provide better low-frequency isolation but introduce distributed capacitance that self-resonates in the microwave region, creating a narrowband notch right where you need clean throughput.

This is not a manufacturing tolerance issue. It is a fundamental electromagnetic limitation of uniform solenoid geometry. The only way to decouple these competing requirements is to abandon the uniform winding altogether.

The HALT25008 employs a continuously tapered cone winding that resolves this physics-level constraint. By graduating the turn diameter from a 0.08 mm fine-wire apex to a wider base, the structure distributes the parasitic turn-to-turn capacitance along a mechanical gradient rather than concentrating it at a single electrical node. The narrow tip, connected directly to the 50Ω microstrip, presents vanishingly small shunt capacitance to ground—preserving K-band transmission. The broader base packs sufficient turns to generate 160 nH of blocking inductance, active down to 100 MHz.

The SRF Story: Why 22 GHz Matters in Practice

Self-Resonant Frequency is the single number that separates usable chokes from datasheet fiction. A component may carry an impressive inductance specification at 10 MHz, but if that same part self-resonates at 5 GHz, it becomes a capacitor—not an inductor—above that frequency. Your bias tee has now become a high-pass filter you never designed.

The HALT25008 pushes SRF beyond 22.0 GHz. This is not a typical figure for a wirewound component carrying 160 nH. The physics behind this number is worth understanding:

• Wire gauge selection: 0.08 mm (80 µm) diameter. The reduced conductor surface area between adjacent turns cuts the dominant parasitic capacitance term by roughly half compared to a 0.15 mm winding. Since SRF scales as 1/√(LC), halving the parasitic C pushes the resonance upward by approximately 40% in frequency.
• Polyimide insulation: The enamel dielectric has a low relative permittivity (ε_r ≈ 3.5) and is applied in a thin, uniform coating, further suppressing the turn-to-turn capacitive coupling that limits conventional enameled-wire chokes.
• Air-core topology: No ferrite. No core loss tangent increasing with frequency. No permeability collapse at elevated temperature or DC bias. The SRF is governed solely by the winding geometry—a deterministic, repeatable relationship.

Operational Bandwidth Characterization

Each HALT25008 is verified on a calibrated vector network analyzer with microstrip fixturing and TRL de-embedding. Three characteristic bands define the operational envelope:

Technical Specifications Summary

Model Comparison: Understanding the Wire-Gauge Trade-Off

Hoan produces two 160 nH conical inductors differentiated by conductor diameter. Selecting the correct variant avoids over-specifying (and over-paying) or under-specifying (and risking field failures):

Integration Guidelines for Maximum Bandwidth

Mounting a conical inductor is not equivalent to soldering an 0805 chip component. These practices are essential to achieve the rated bandwidth:

1. Orientation: The apex (narrow tip) must face the RF microstrip and sit perpendicular to the transmission line. Any angular offset introduces asymmetric coupling that manifests as degraded S11 above 15 GHz.
2. Lead trim: The tip-side flying lead must be kept below 0.5 mm from solder pad to winding body. Each additional 0.5 mm of lead length adds roughly 0.3 nH of series parasitic inductance—enough to shift the apparent SRF downward by several hundred MHz at K-band.
3. Adhesive stabilization: After soldering, apply a single micro-dot of non-conductive, low-outgassing epoxy (Epotek H70E recommended) to the side of the winding. This arrests microphonic modulation of turn spacing under vibration. Do not encapsulate the full coil—excess dielectric adds unwanted shunt capacitance.
4. Thermal relief: At 500 mA continuous in high-temperature environments, allow at least 0.5 mm of air clearance around the coil body to support convective cooling.
5. Soldering thermal budget: 280–320°C tip, ≤3 seconds dwell per joint. The polyimide enamel tolerates brief thermal excursions but prolonged heating can anneal the fine copper and alter its mechanical properties.

Where This Component Fits

• Wideband bias tees requiring single-inductor coverage from VHF through K-band
• DC feed networks in 100G/400G/800G fiber-optic transceiver optical subassemblies (TOSA/ROSA)
• GaAs and GaN MMIC amplifier drain/gate bias injection
• PIN diode switch and step attenuator DC return paths
• Phased-array antenna element bias distribution
• Millimeter-wave VNA extender and test-fixture bias networks
• X/Ku/K-band radar and electronic warfare receiver front-ends

Common Questions from RF Design Teams

Q: We need flat group delay through our bias tee. Does the HALT25008 introduce dispersion?
A: Because the HALT25008 has no ferrite core, there is no frequency-dependent permeability contributing to phase nonlinearity. The air-core structure combined with the distributed-capacitance topology yields essentially constant group delay across the rated band. Measured S21 phase deviation is less than ±5° from 100 MHz to 18 GHz on a properly de-embedded fixture. The .s2p file includes full phase data for your own dispersion analysis.

Q: How repeatable is the SRF from lot to lot?
A: The SRF of an air-core conical inductor is determined by winding geometry, not material properties. Since Hoan controls the taper profile, turn count, and wire tension through automated winding, lot-to-lot SRF variation is typically within ±1.5 GHz. Each production batch is sampled on a VNA, and the statistical process control data is available under NDA for high-volume programs.

Q: Can we get a 3D EM simulation model instead of just .s2p data?
A: Yes. Hoan provides a parameterized HFSS 3D model upon request. The model captures the exact tapered geometry, wire diameter, and lead configuration needed for full-wave EM co-simulation of your PCB layout. This is particularly valuable above 20 GHz where microstrip pad geometry and launch discontinuity dominate the measured response.

Q: What is the failure mechanism if we exceed the 500 mA rating?
A: The immediate effect is increased I²R heating within the 0.08 mm copper winding. The polyimide insulation is rated for continuous operation at 200°C, so insulation breakdown is not the primary concern. Rather, sustained over-current raises the copper resistivity (positive TCR of +0.39%/°C), which elevates DCR, which generates more heat—a slow thermal runaway. The 500 mA rating includes margin to prevent this positive-feedback loop from initiating at the maximum rated ambient of +125°C.

Q: Do you provide fixtured S-parameter data, or just probe-station measurements?
A: All .s2p data is acquired with the HALT25008 soldered onto a characterized microstrip calibration fixture using TRL de-embedding to move the reference plane to the component leads. This captures the real-world assembly environment, including solder joint parasitics, rather than idealized probe-landed data that does not represent the installed condition.