Summary of "AI Assisted Troubleshooting"

Overview / Goal

Test whether an AI (Google Gemini) can guide hands-on troubleshooting of a bench power supply using a camera/screen share. The experiment is framed as a step toward AR/remote-assisted hardware repair.

Device under test

Model: VLT130-1103
Manufacturer: EOS (as referenced)
Noted specs on unit: ~2.9 A max
Expected outputs: 24 V on pin 2 (other pins may provide 12 V, etc.)

Tools used

Digital multimeter (visible on camera)
Screen share so the AI could see the meter/readings

Step-by-step troubleshooting flow (as guided by the AI)

AI requests model number and symptoms.

User reported: “don’t know how to test it.”
AI instructs to set the multimeter to DC voltage and measure the 24 V output (pin 2) with the black lead on ground.
- User identifies pins: pins 3 and 4 reported as ground; pin 2 expected 24 V.
- Result: 0 V measured on the output.
AI suggests checking for a power switch (none on the unit) and then verifying input power.
AI instructs setting the multimeter to AC voltage (V with wavy line) and measuring the AC input.
- Result: ~119.2 V AC present on the input.
AI conclusion: Input AC is present but there is no DC output → the power supply is likely defective. AI offers next-step suggestions (user declined).

Key takeaways / technological points

Demonstrates a practical workflow for remote/AI-guided troubleshooting:
- Identify device and expected outputs.
- Use camera to show meter/readings.
- Follow standard electrical checks (measure DC output, confirm AC input).
Useful checklist for power-supply diagnosis:
- Confirm expected output pins.
- Check DC output with a multimeter.
- Verify AC input.
- Inspect power switch/settings.
Potential future: combining live camera/AR with AI for step-by-step repair guidance.