Low-cost cross-correlation noise setup for measuring the Boltzmann constant and the elementary charge

This paper presents a low-cost, robust, and compact experimental setup using dual-channel cross-correlation sampling with standard electronic components to accurately measure the Boltzmann constant and elementary charge via thermal and shot noise, making it an ideal educational tool for university and high school laboratories.

The Gist

Imagine you are trying to hear a single person whispering in a crowded, noisy room. If you stand there with just one ear, the chatter of the crowd (the background noise) might drown out the whisper. But what if you had two friends standing right next to you, both listening to the same whisper? If you compare what Friend A hears with what Friend B hears, the random chatter they don't agree on cancels out, leaving you with a crystal-clear version of the whisper.

That is exactly the clever trick scientists Zitong Peng, Jie Zheng, and Xiaokai Yue used in this paper. They built a low-cost, "DIY-friendly" machine to listen to the tiniest whispers of electricity in order to measure two of the universe's most fundamental secrets: the Boltzmann constant (which relates heat to energy) and the elementary charge (the size of a single electron's electric charge).

Here is a breakdown of how they did it, using simple analogies:

1. The Problem: The "Whisper" is Too Quiet

For decades, physics students have tried to measure these constants by listening to "electrical noise."

• Thermal Noise: Imagine a resistor (a tiny component that resists electricity) as a crowded dance floor. Even if the room is still, the dancers (electrons) are jiggling around because of heat. This jiggling creates a tiny, random electrical "buzz."
• Shot Noise: Imagine electrons flowing through a wire like raindrops hitting a tin roof. They don't fall in a smooth stream; they hit individually and randomly. This creates a "pitter-patter" sound.

The problem is that these sounds are incredibly faint—like trying to hear a pin drop in a hurricane. Traditional tools (like expensive oscilloscopes) are often too "noisy" themselves to hear the pin drop. They add their own static, making the measurement impossible for a regular university lab.

2. The Solution: The "Two-Ear" Trick (Cross-Correlation)

Instead of using one expensive, high-tech ear, the team built a setup with two ears (two amplifiers) listening to the same signal simultaneously.

• The Setup: They took a tiny, cheap audio amplifier (the kind used in high-end headphones) and connected it to a cheap computer data card (an ADC).
• The Magic: They ran the signal through two separate channels.
  • Channel 1 hears: The Whisper + The Static of Channel 1.
  • Channel 2 hears: The Whisper + The Static of Channel 2.
• The Math: When they compared the two, the "Whisper" (which is the same in both) stayed loud. But the "Static" (which is random and different in each channel) canceled itself out.

This is called Cross-Correlation. It's like asking two people to describe a song they both heard. If they both say "there was a drum beat," you know it's real. If one says "I heard a cat meow" and the other says "I heard a dog bark," you know those were just their own imaginations (noise) and you ignore them.

3. The Results: Measuring the Universe with a Backpack Budget

Because they used this "two-ear" trick, they didn't need a million-dollar lab. They built the whole thing for about $83 USD (roughly the cost of a nice dinner for two).

• Measuring Heat (Boltzmann Constant): They measured the "jiggling" of electrons in a resistor. By comparing the noise to the temperature, they calculated the Boltzmann constant. Their result was within 1.6% of the official value.
• Measuring Electrons (Elementary Charge): They measured the "raindrop" effect of electrons flowing through a light sensor. By counting the noise, they calculated the size of an electron's charge. Their result was within 2% of the official value.

4. Why This Matters

• Democratizing Science: Before this, you needed a very expensive lab to do these experiments. Now, a high school or a small university can build this in a weekend for less than $100.
• Teaching Tool: It lets students see the invisible. Instead of just reading about "randomness" in a textbook, they can watch the noise on a computer screen, see how the "two-ear" trick cleans it up, and actually calculate the fundamental rules of the universe.
• Simplicity: The design is so straightforward that students can build it themselves (DIY), learning about circuits, noise, and data analysis all at once.

In a Nutshell

The authors took a complex, expensive physics experiment and turned it into a cheap, accessible, and clever "noise-canceling" gadget. By using two cheap ears instead of one expensive one, they managed to hear the universe's tiniest whispers and measure the fundamental building blocks of reality, proving that you don't need a fortune to discover the secrets of nature.

Technical Summary

1. Problem Statement

The measurement of fundamental physical constants, specifically the Boltzmann constant ($k_B$) and the elementary charge ($e$), via electrical noise (thermal and shot noise) is a staple of advanced undergraduate physics education. However, existing experimental setups face significant barriers to widespread adoption:

• High Cost and Complexity: Traditional methods often require high-precision oscilloscopes or expensive RMS-DC converters, making them inaccessible for many university and high school laboratories.
• Signal-to-Noise Ratio (SNR) Issues: Thermal and shot noise signals are extremely weak (on the order of nV/$\sqrt{\text{Hz}}$). Direct measurement is often obscured by the intrinsic background noise of the measurement instruments (amplifiers and ADCs).
• Lack of Intuitive Data: Some low-cost alternatives (using RMS-DC converters) fail to provide students with direct access to the noise power spectral density (PSD), hindering the ability to analyze frequency-domain characteristics and identify interference.
• Component Limitations: Previous low-cost designs sometimes utilized amplifiers with high input current noise, which introduces significant errors in thermal noise measurements due to the interaction with the source resistance.

2. Methodology

The authors developed a low-cost, DIY-friendly experimental setup that integrates a dual-channel cross-correlation technique with standard electronic components.

• Hardware Architecture:

  • Amplification: The system uses Burr-Brown™ OPA1656 FET-input operational amplifiers. These were selected for their exceptionally low input voltage noise (2.9 nV/$\sqrt{\text{Hz}}$) and ultra-low input current noise (6 fA/$\sqrt{\text{Hz}}$), which is critical for minimizing errors in thermal noise measurements. The circuit employs a two-stage non-inverting amplifier design with a total gain of 60.4 dB.
  • Data Acquisition: A cost-effective AD7606 16-bit ADC module (200 kHz sampling rate) replaces expensive oscilloscopes. It is controlled via a Python wrapper around a C-driver, allowing for flexible data acquisition and processing.
  • Power Supply: The system is powered by standard 9V batteries to ensure low noise and stability, though AC-DC regulated modules are suggested for long-term use.
  • Shielding: All noise-sensitive components are housed in aluminum shielding boxes to minimize electromagnetic interference (EMI).

• Cross-Correlation Technique:

  • The setup utilizes two independent, identical measurement channels.
  • Principle: The target noise signal ($\delta V$) is correlated between both channels, while the intrinsic background noise of the amplifiers and ADCs ($V_{\text{osc1}}$ and $V_{\text{osc2}}$) is uncorrelated.
  • Mathematical Approach: By calculating the Cross Power Spectral Density (CPSD) ($S_{\text{cross}}$), the uncorrelated background noise terms average to zero over time, leaving only the correlated signal of interest. This effectively suppresses the instrument's noise floor without requiring ultra-low-noise hardware for every component.

• Experimental Procedures:

  • Thermal Noise ($k_B$): Measured using thin-film resistors (500 $\Omega$ to 4 k$\Omega$) at room temperature. The PSD is measured and averaged over 2,000 repetitions to improve SNR.
  • Shot Noise ($e$): Generated using a photodiode driven by an LED. The photocurrent is varied to observe noise scaling with current. The circuit is designed to produce opposite polarity noise voltages in the two channels, resulting in a negative cross-spectral density.

3. Key Contributions

• Cost Reduction: The total cost of the experimental hardware (excluding the computer) is approximately $83 USD, a fraction of the cost of traditional setups.
• Educational Accessibility: The design is modular and uses standard components, making it suitable for "Do-It-Yourself" (DIY) implementation in high school and undergraduate labs.
• Direct Spectral Observation: Unlike RMS-based methods, this setup allows students to directly visualize and analyze the noise power spectrum, facilitating better error analysis (e.g., identifying 50/60 Hz mains interference).
• Optimized Component Selection: The paper explicitly demonstrates why the OPA1656 is superior to other low-cost amplifiers (like the ADA4898) for thermal noise experiments due to its negligible input current noise, preventing quadratic noise errors.
• Validation of Cross-Correlation: The study provides a practical, low-cost validation of the cross-correlation method for suppressing instrumental noise in fundamental constant measurements.

4. Results

• Boltzmann Constant ($k_B$):

  • The experiment yielded a value of $k_B = (1.3931 \pm 0.0098) \times 10^{-23}$ J/K.
  • This compares to the accepted value of $1.3806 \times 10^{-23}$ J/K, resulting in a relative uncertainty of approximately 1.6%.
  • The linear fit of noise power vs. resistance passed almost through the origin, with a y-intercept ($S_0$) an order of magnitude smaller than the measured noise, confirming effective background noise suppression.

• Elementary Charge ($e$):

  • The experiment yielded a value of $e = (1.6249 \pm 0.0109) \times 10^{-19}$ C.
  • Compared to the accepted value of $1.6022 \times 10^{-19}$ C, the relative uncertainty is approximately 2%.
  • The linear relationship between shot noise and photocurrent was highly consistent, with standard deviations of data points around 1%.

• Noise Suppression:

  • The cross-correlation method successfully reduced the noise floor. While single-channel measurements showed significant 1/f noise and instrument background, the cross-correlation spectrum was flat (white) and stable across the 500 Hz to 10 kHz bandwidth.
  • Standard deviation analysis confirmed that noise reduction followed the theoretical $1/\sqrt{N}$ scaling law as the number of averaging repetitions increased.

5. Significance

This work bridges the gap between high-precision physics research and accessible education. By demonstrating that fundamental constants can be measured with ~1-2% accuracy using a sub-$100 setup, the authors provide a robust framework for:

1. Democratizing Advanced Physics: Enabling institutions with limited budgets to perform high-level experiments on quantum fluctuations and statistical mechanics.
2. Pedagogical Enhancement: Offering students a hands-on understanding of signal processing, Fourier transforms, and the statistical nature of noise, while teaching advanced techniques like cross-correlation.
3. Future Research: The setup serves as a platform for exploring more complex topics such as quantum transport and two-photon statistics, as the methodology is scalable and adaptable.

In summary, the paper presents a highly effective, low-cost solution that transforms the measurement of $k_B$ and $e$ from a specialized research task into a feasible and insightful educational experiment.