Hi everyone! I’m a KOSEN student building a DIY EFNMR Quantum Computer.

In this latest experiment, we finally succeeded in capturing the FID signal (damped sine wave)!

Observing the FID Signal!

First, the FID signal of hydrogen atoms (protons) in the Earth's magnetic field occurs at a frequency right around 2000Hz to 2100Hz.

If you look at the bottom graph in the photo, right around the purple dot between 2000Hz and 2100Hz, there is a distinct, sharp peak. This is the exact FID signal we’ve been hunting for!

However, looking at the top time-domain graph, you can't really see a clear damped sine wave. We believe this is because a massive dead time is occurring in the green section due to residual surge current from the coil switching.

This also explains why the peak around 2000Hz–2100Hz is relatively small. Because of the dead time, we are losing the initial head of the damped sine wave where the signal amplitude is at its strongest. What the FFT graph is actually showing is the tail end of the microscopic FID signal—those tiny oscillations right after the green dead-time line drops back to zero in the top graph.

Future Challenges & Countermeasures

This experiment brought two critical hardware challenges to light:

• Due to dead time, we are missing the most crucial data right after sequence switching.
• The signal is still nearly buried in analog noise.

Personal Thoughts: A 6-Month Journey & The 0.25x Speed Discovery!

I am beyond thrilled that we finally observed the FID signal! It took us about half a year to get to this point. For six months, spending hours every single day researching, brainstorming, building, and testing has finally paid off!

This experiment was the very first time we ran all the circuits together and performed real data analysis. Funny enough, while we were running the experiment, I didn't even realize the FID signal had appeared. Because the signal flashes by in a fraction of a second, we were recording a video of the screen and playing it back at 0.5x speed—and we couldn't find anything. I thought we had failed.

However, a little while after returning to my dorm room, I uploaded the video to Google Photos and rewatched it at 0.25x speed. Suddenly, I thought, "Wait... could this be it?!" I paused the video and scrubbed through it frame-by-frame manually—and there it was: the exact frame showing the FID peak on the graph! Going from the disappointment of thinking it failed to discovering we actually caught it made me unbelievably excited.

Riding this momentum, we are going to keep tackling the dead-time and noise issues head-on and complete this DIY quantum computer!

Follow Our Journey!

• Real-time updates on X (Twitter): https://x.com/nmrqcproject
• Detailed development logs and source code on GitHub: https://github.com/nmrqcproject/nmr-qc-project/blob/main/README.md
• Detailed development logs in Japanese / 日本語での詳しい開発記録 (Zenn): https://zenn.dev/nmrqcp