← All memosMemo 01 · Sense · 4 min read
How does a sensor become a decision?
Daniel McNair · Sample memo · September 2026
A sensor does not hand a computer the truth. It produces a physical signal—often a voltage—that must be measured, translated, and interpreted.
Start with a magnet
A Hall-effect sensor changes its output voltage when a magnetic field changes. Imagine its output is 2.5 volts with no field and rises by 5 millivolts for each additional gauss. A 200-gauss field would produce about 3.5 volts.
That voltage is still analog: it can take any value within the sensor’s range. A microcontroller’s analog-to-digital converter, or ADC, measures it and assigns a digital number. Software can then compare that number to a threshold and turn on an LED.
Magnetic field → voltage → ADC reading → software decision.
The engineering question
How small a change can the system detect? A 10-bit ADC spanning 0 to 5 volts has 1,024 possible codes, so one step is about 5 V / 1024 = 4.88 mV. With a sensor sensitivity of 5 mV per gauss, that is roughly one gauss per ADC step.
Resolution is only part of the answer. Electrical noise, sensor error, reference-voltage drift, and how often you sample can all limit a real measurement. If the magnetic field changes rapidly, too few samples may miss the change or make it appear slower than it is.
Why defense engineers care
Radar receivers, navigation sensors, and aircraft health monitors all convert physical signals into digital information. The software decision is only as trustworthy as the measurement chain beneath it. Engineers therefore specify the sensor, ADC, filtering, timing, and tests as one system.
Interview prompt
Question: A 12-bit ADC spans 0 to 5 volts. What voltage does one code represent?
Answer: Approximately 5 / 4096 = 0.00122 V, or 1.22 millivolts. More bits improve ideal voltage resolution, but do not remove sensor noise.
A little about me
Curious about people.
Drawn to systems.
I’m Daniel, an electrical and computer engineering student at Duke. I’m learning how circuits and code become systems people can depend on. Each week, I’ll take one engineering question I’m curious about, learn the fundamentals, and speak with a Duke professor whose work touches on it. Then I’ll share what I learned in plain language: how the technology works, where it could be useful, and what I’m still trying to understand.
Where it started
A different way to serve.
I grew up in Annapolis, Maryland with an Iranian mother and a family history of Navy service. Those parts of my life taught me to pay attention to people’s experiences, while also making me curious about the technological systems that they rely on. I want to understand how technology works down to its bare bones and what it takes for an idea to become useful in someone’s life.
What I’m following now
From a signal to a decision.
Studying engineering has given that curiosity a practical direction. I’m especially interested in embedded systems, sensors, semiconductors, and defense technology: how they work, how they fail, and how to make them reliable.
Read the memos ↗Building & learning
Experience.
Engineering, investing, and research have each taught me a different way to ask better questions.
Current focus
Electrical & Computer Engineering
Duke University · Student
Building a foundation in circuits, computing, mathematics, and physics. In the lab, I’ve worked with sensors and microcontrollers to turn physical measurements into decisions in code.
Past experience
Research, analysis & people
Duke Impact Investing Group
Investment Analyst/Project Manager
Researched early-stage companies, spoke with founders, and helped evaluate a potential investment in BluShift Aerospace.