Rigid-Flex PMT Base Evaluation

This project develops and characterizes a flex-rigid PMT base (FRP) for LXe detectors, integrating voltage division and long-distance signal readout into a single structure to replace traditional cables and hub boards.

My Role and Project Status

My primary responsibility was testing and performance evaluation of rigid-flex PMT base circuits, with measurements informing layout improvements. This page summarizes prototype-level mechanical, thermal, and electrical characterization, including issues that guide the next design iteration.

Motivation

• Future LXe detectors require increasingly dense photosensor layouts with strict space and radiopurity constraints.
• Conventional rigid PCB bases can introduce mechanical interference, routing complexity, and assembly limitations in compact PMT arrays.
• A flexible base design provides a route toward improved mechanical compatibility while preserving stable voltage division and signal readout.

System Architecture

• 3×3 PMT module (9 HV channels, 36 signal channels).
• Flex–rigid structure:
  • Rigid base board: ~0.2 m × 0.2 m
  • Flexible readout tail: ~0.9 m (total length ~1.1 m)
• Direct connection to flange via Micro-D connectors (no external cable bundle).
• Cut-out regions introduced to reduce material and background.

Design Evolution

• V1–V2: FR4 rigid boards + cable/hub → wiring complexity, failure risk.
• V3: PI rigid base → lower background, validated fabrication.
• V4: Flex–rigid PCB → integrated base + signal routing, cable-free design.
• V5 (ongoing): optimized layout and stack-up based on test feedback.

Mechanical & Thermal Validation

• Repeated bending test: >100 cycles with stable electrical continuity.
• Thermal cycling: down to approximately -37°C and back to room temperature. This test does not establish reliability at liquid xenon operating temperature.
• Observed issues:
  • Partial delamination after thermal cycles
  • Local PI swelling during soldering → Indicates need for improved stack-up and material control.

Results from Subsequent Layout Improvements

Thermal-stability, connectivity, noise, attenuation, and crosstalk measurements informed changes to signal-trace spacing and ground isolation. Relative to the earlier baseline:

These values summarize prototype comparisons; they do not establish full-detector qualification.

Earlier Prototype Electrical Performance

The values below summarize prototype bench measurements. They should not be interpreted as detector-wide performance or qualification results.

• High-voltage stability:
  • Tested at 1000–1200 V for 3 hours
  • One discharge observed at 1200 V
• Noise:
  • ~10 mVpp baseline noise (no shielding)
  • Reduced to ~5 mVpp with shielding (~50% improvement)
• Signal attenuation:
  • Strong frequency dependence; improves at low temperature
• Crosstalk:
  • ~10–15% between adjacent traces
  • Strongly dependent on trace spacing, not region location
• Waveform quality:
  • Noticeable distortion observed → identified as a critical issue for next iteration

Key Engineering Insights

• The measurements suggest that routing geometry (spacing, layer separation) is an important contributor to signal integrity.
• Shielding layers are essential for noise suppression.
• HV and signal lines must be strictly separated across layers.
• Flex–rigid integration introduces new failure modes (delamination, mechanical stress coupling).

Optimization Strategy (Next Iteration)

• Introduce multi-layer stack-up:
  • Signal / Ground / HV / Ground / Signal
  • Top/bottom shielding layers
• Increase trace spacing to suppress crosstalk.
• Improve impedance matching and routing continuity.
• Evaluate high-frequency materials (e.g., LCP) for reduced attenuation.
• Strengthen mechanical robustness (thickness, bonding process).

Outcome

• Built and characterized a prototype integrated PMT base and readout architecture. High-voltage discharge, crosstalk, waveform distortion, and delamination remain limitations before detector deployment.
• Established a unified framework linking:
  • mechanical design
  • electrical performance
  • detector integration constraints
• Identified concrete upgrade directions toward deployable systems in future detectors (e.g., PandaX-20T / xT).