Sub-THz on-chip antennas - what's realistic gain at 140GHz?

Started by Siemens_IA_Victor · 2026-10-26 02:33 UTC · 2 views · 1 replies
S Member
Designing a 140GHz on-chip patch for a radar demo. Simulated gain is -2dBi which feels useless. Is on-chip even viable at D-band, or should I just go with antenna-in-package?
L Admin
At 140 GHz, on-chip patch gain around −2 dBi (or even a few dB) is not unusual in a lossy BEOL stack — the antenna is fighting metal density rules, thin dielectrics, and substrate modes. “Useless” depends on the link: a short radar demo with a lens/reflector or AiP can still close; a free-space chip-only patch rarely matches a package antenna.

Realistic split many groups use:
- On-chip: OK for probe-station bring-up, very short range, or as a feed into an off-chip lens / dielectric resonator — expect low gain and budget the loss.
- Antenna-in-package / interposer: usually the product path at D-band when you need double-digit dBi or a real EIRP number.
- Hybrid: on-chip launcher + package horn/patch array is common when the foundry antenna kit is limited.

Before you abandon on-chip, check whether the −2 dBi includes feed and mismatch, whether a backside lens or cavity was in the sim, and what EIRP your demo actually needs. If the goal is a customer-facing radar demo, AiP (or a small module antenna) is usually the shorter path to a believable range.

— rfforum 编辑部 / Editorial Team
广告 AdNXP Airfast 5G 基础设施射频功率多芯片模块示意图NXP Semiconductors射频功率、汽车雷达与连接芯片 · RF power, radar and connectivity访问官网 →免费展示 · 资料来自公开信息 · 如需撤下请联系 ads@rfforum.com
广告 Ad广告位招商 / Advertise hereads@rfforum.com · 刊例 Rate cardT2 · 728×90
Log in or register to reply.