Design, waterproofing, and mass production of the 3-inch PMT frontend system of JUNO

This paper details the design, mass production, and acceptance testing of the high-voltage dividers, waterproof cables, and connectors for the over 25,600 3-inch photomultiplier tubes used in the JUNO central detector.

The Gist

Imagine you are building a giant, underwater camera to take pictures of ghostly particles called neutrinos. These particles are so shy they pass through everything, including the Earth, without stopping. To catch them, scientists built the JUNO experiment, a massive tank filled with a special glowing liquid (scintillator) deep underground.

Inside this tank, they needed to install over 25,000 tiny, super-sensitive eyes (called 3-inch Photomultiplier Tubes, or PMTs) to see the faint flashes of light when a neutrino hits the liquid.

This paper is the "instruction manual" and "quality control report" for how they built, waterproofed, and tested these eyes so they could survive deep underwater for decades. Here is the story of how they did it, explained simply.

1. The Eyes and Their Glasses (The PMTs and Dividers)

Think of each PMT as a tiny camera eye. But an eye needs electricity to work.

• The Problem: These eyes need a high-voltage battery (like a very strong flashlight) to function, but they are tiny. You can't just tape a giant battery to them.
• The Solution: They built a tiny "circuit board" (called an HV divider) that acts like a voltage ladder. It takes the high voltage and carefully steps it down to the right levels for the different parts of the eye.
• The Challenge: They had to fit all these electronic parts onto a board smaller than a hockey puck. They used tiny, high-quality components (like surface-mounted resistors) to make sure the "ladder" didn't break under pressure or heat.

2. The Umbilical Cords (Cables and Connectors)

Once the eyes are ready, they need to talk to the computer on the surface.

• The Cables: They used special underwater cables (like umbilical cords) that carry both electricity and the picture signal. These cables are made of a special plastic that doesn't glow in the dark (so it doesn't confuse the sensors) and has a "self-healing" powder inside. If the outer skin gets a tiny scratch, this powder swells up like a super-absorbent diaper to stop water from leaking in.
• The Connectors: You can't have 25,000 individual wires going to the computer. So, they grouped 16 eyes together and plugged them into a single "power strip" (a 16-channel connector).
• The Seal: These connectors are like high-tech diving suit zippers. They use rubber rings (O-rings) and special glue to ensure that even if the ocean pressure tries to crush them, no water gets inside.

3. The "Potting" Process (The Waterproof Cocoon)

This is the most critical part. The place where the eye meets the wire is the weakest point. If water gets there, the whole system shorts out.

• The Analogy: Imagine dipping a delicate electronic device into a jar of jelly.
• The Process: They took the eye, the circuit board, and the wire, soldered them together, and then poured a special two-part polyurethane "jelly" (resin) around them inside a plastic shell.
• Why Jelly? This jelly hardens into a solid, waterproof block. It protects the electronics from the crushing water pressure and stops any tiny electrical sparks (which could look like fake signals) from being seen by the sensors.
• The Safety Net: They didn't just rely on one layer. They used tape, shrink-wrap, and the jelly itself, creating multiple layers of defense, like an onion made of waterproof armor.

4. Sorting the Eyes (Grouping by Weight and Voltage)

Not all eyes are exactly the same. Some are slightly heavier (thicker glass) and some need slightly more electricity to work.

• The Weight Sort: Deep underwater, water pressure is huge. If a glass eye is too thin, it might implode (crush like a soda can).
  • They weighed every single eye.
  • The lightest (thinnest) eyes were placed near the top of the tank where the water is shallow.
  • The heaviest (thickest) eyes were placed at the bottom where the pressure is highest.
• The Voltage Sort: They also grouped the eyes by how much electricity they needed, so that 16 eyes plugged into one "power strip" would all work perfectly together.

5. The "Stress Test" (Acceptance Testing)

Before sending 25,000 eyes to the bottom of the ocean, they had to prove they wouldn't fail.

• The Dark Room: They put the eyes in a pitch-black room (like a camera darkroom) and turned them on.
• The Test: They checked three things:
  1. Brightness: Is the eye seeing clearly? (Gain)
  2. Clarity: Is the picture sharp, or is it blurry? (Resolution)
  3. Silence: Is the eye quiet, or is it making noise in the dark? (Dark Count Rate)
• The Result: They tested thousands of eyes. Only about 0.7% failed (like finding one bad apple in a barrel of 140). They swapped the bad ones with spare eyes and re-tested them.

6. The "Air Bubble" Problem

During the early days of filling the tank with water, they had a scare. The water was so pure that air bubbles started forming inside the connectors, causing electrical sparks (like static electricity).

• The Fix: They realized the air inside the connector was "hungrier" than the water. So, they started pumping Nitrogen gas into the water. This made the water "fuller" of gas than the connector, forcing the gas to flow out of the connector and into the water, raising the pressure inside and stopping the sparks.

The Big Picture

This paper is a triumph of engineering. It describes how a massive international team took 25,000 delicate, high-tech eyes, wrapped them in layers of waterproof armor, sorted them by weight, tested them in the dark, and successfully installed them deep underwater.

Thanks to this work, the JUNO detector is now "seeing" the universe's most elusive particles, helping us understand the fundamental building blocks of our reality. It's like building a submarine that can see ghosts, and this paper explains how they made sure the submarine didn't leak!

Technical Summary

Here is a detailed technical summary of the paper "Design, waterproofing, and mass production of the 3-inch PMT frontend system of JUNO."

1. Problem Statement

The Jiangmen Underground Neutrino Observatory (JUNO) requires a massive, highly reliable photodetection system to detect neutrinos. The central detector utilizes 25,600 3-inch Photomultiplier Tubes (PMTs) submerged in ultra-pure water at depths up to 43.5 meters.

• The Challenge: These PMTs must operate in a high-pressure, ultra-pure water environment for over 20 years. They require a custom frontend system that integrates high-voltage (HV) distribution, signal transmission, and waterproofing.
• Specific Constraints:
  • Reliability: The system must achieve a channel failure rate of less than 10% over a 6-year period (MTBF > 120 years).
  • Waterproofing: The electronics (HV dividers, cables, connectors) must be hermetically sealed to prevent water ingress, which would cause short circuits or corrosion.
  • Signal Integrity: The system must transmit single-photon signals with minimal noise, crosstalk, and attenuation over cable lengths of 5m and 10m.
  • Safety: PMTs are fragile glass spheres; they must be grouped by weight to prevent implosion due to hydrostatic pressure differences.

2. Methodology

The paper details the end-to-end engineering process from component design to mass production and acceptance testing.

A. Component Design

• HV Divider:
  • Designed with a positive HV scheme for 11 dynodes.
  • Evolution: Initially considered wire-wound resistors for high voltage tolerance, but switched to surface-mounted 1206 resistors (15 MΩ) and ceramic capacitors to fit the tight space (48 mm diameter) on the PMT base.
  • Specs: Total resistance 210 MΩ, max working voltage 2,000 V, DC current < 6.5 µA.
  • Protection: The divider PCB is coated with black polyurethane to prevent flashover (micro-discharges) in air.
• Cable and Connector:
  • Cable: Custom RG178 coaxial cable (50 Ω) with an HDPE jacket and a superabsorbent polymer (sodium polyacrylate) water-blocking powder layer.
  • Connector: A novel 16-channel waterproof connector developed with AXON'. It features a plug (attached to 16 PMTs) and a receptacle (mounted on the underwater box).
  • Sealing: Uses redundant axial/radial O-rings, injection molding, and adhesive. The design ensures signal ground is isolated from the mechanical ground.
  • Gas Issue Mitigation: Addressed HV breakdown caused by air diffusion from connectors into degassed water by introducing a continuous nitrogen flow into the water circulation system to equalize pressure and prevent Paschen breakdown.

B. Mass Production and Integration

• PMT Grouping:
  • By HV: PMTs were grouped into 16 categories based on the voltage required to achieve a gain of $3 \times 10^6$ (ranging 900–1300 V).
  • By Weight: To prevent implosion, PMTs were weighed and classified into three groups (W1, W2, W3). Lighter PMTs were installed at shallower depths (<10m, <25m), while heavier ones were placed deeper (<42m).
• Waterproof Potting:
  • The PMT pins, HV divider, and cable were soldered and enclosed in an ABS plastic shell.
  • Sealing Layers:
    1. Inner seal: Epoxy (Loctite ES4412) for the cable entry.
    2. Mechanical support: ABS shell filled with polyurethane.
    3. Outer seal: Butyl tape wrapped around the PMT glass/shell interface, covered by a shrinkable tube.
  • Process: 16 PMTs were assembled into a single unit connected to one 16-channel plug.

C. Testing and Validation

• Long-Term Reliability: Accelerated aging tests were conducted in France and Chile at elevated temperatures (up to 62°C) and pressures (up to 10 bar).
  • Result: Simulated 40+ years of operation with zero leaks. Failure In Time (FIT) values were calculated at <500 (France) and 300 (Chile), well below the requirement of 951.
• Acceptance Testing:
  • Conducted at Guangxi University using a 128-channel test bench.
  • Criteria: Gain ($2\text{--}5 \times 10^6$), Single Photoelectron (SPE) resolution (<45%), and Dark Count Rate (DCR < 3 kHz).
  • Calibration: Established a linear correlation between the manufacturer's HV (HZC) and the potting test HV (GXU) to optimize testing efficiency.

3. Key Contributions

1. Miniaturized HV Divider: Successfully adapted a high-voltage divider using surface-mounted components for the constrained space of a 3-inch PMT base, maintaining high reliability (<0.05% annual failure rate).
2. Novel 16-Channel Connector: Developed a custom waterproof connector with a double-sealing mechanism and a unique water-blocking cable design, solving the challenge of connecting 25,600 PMTs efficiently.
3. Robust Potting Strategy: Implemented a multi-layer sealing technique (Epoxy + Polyurethane + Butyl tape) that passed rigorous underwater pressure testing with no leakage.
4. Mass Production Workflow: Established a scalable process for grouping 26,000 PMTs by weight and HV, integrating them into 1,609 groups, and performing automated acceptance testing.
5. Problem Solving: Identified and resolved the HV breakdown issue caused by gas diffusion in ultra-pure water using a nitrogen circulation solution.

4. Results

• Production Volume: Successfully produced and delivered 25,600 instrumented PMTs plus 144 spares (Total 25,744 tested).
• Reliability: Long-term underwater tests confirmed a FIT value of 300–500, meeting the stringent JUNO reliability requirements.
• Performance Metrics:
  • Gain: Mean gain of $3.002 \times 10^6$ with a spread of ~10%.
  • SPE Resolution: Majority of PMTs achieved <45% resolution.
  • Crosstalk: Analog charge crosstalk between channels was measured at <0.3%.
  • Dark Count Rate (DCR): Average DCR was ~420 Hz (lower than bare PMT tests due to better stabilization time), with a peak shift attributed to packaging foam fluorescence.
• Quality Control:
  • Rejection Rate: Only 0.7% of PMTs failed acceptance tests (177 units out of 25,744).
  • Leakage: In a random sampling of 217 groups (approx. 13% of total) subjected to underwater pressure tests, zero potting leaks were detected.
  • Failures: 4 PMT glass implosions occurred during pressure testing (0.1% failure rate), which were successfully replaced.

5. Significance

This paper documents the successful engineering and mass production of a critical subsystem for the JUNO experiment, one of the world's largest neutrino detectors.

• Enabling Physics: The 3-inch PMT system provides a linear reference for calibrating the larger 20-inch PMTs, improves energy resolution, and aids in cosmic muon tracking and proton decay searches.
• Scalability: The methodologies developed for grouping, potting, and testing thousands of sensitive optical sensors serve as a blueprint for future large-scale underwater or underground physics experiments.
• Reliability Assurance: The rigorous validation process, including accelerated aging and destructive testing, provides high confidence that the detector will operate reliably for its intended 20+ year lifespan in a harsh deep-water environment.