Isolated Voltage Measurement Circuit Design
Budget: €30 – €250 EUR
I have an Arduino-based board that is powered from its own DC rail, yet it has to monitor several other DC lines whose grounds are completely separate. Right now I’m simply feeding each line through a voltage divider, and—unsurprisingly—the readings jump around or drift because the grounds are floating. The symptom is clear: inaccurate voltage readings that make the data unusable.
What I need is a clean, reliable way to sense those isolated DC voltages without tying the reference points together. Whether that means introducing isolated amplifiers, optocouplers, tiny transformers, or another proven technique is up to you; I’m open to whichever approach gives stable, repeatable results and keeps the measurement side safe for the microcontroller.
Here’s what I’d like back:
• An updated schematic that drops into my existing design, showing every connection required for proper isolation
• A concise parts list with recommended component values and footprints I can actually buy today
• A short rationale explaining how the proposed stage cures the isolation problem and what accuracy I can expect
KiCad, Eagle, or even a clearly annotated PDF is fine for the schematic, as long as I can replicate it. If you can include quick SPICE or other simulation plots to illustrate behavior, that’s a bonus but not strictly required.
Once I have a working revision that delivers stable, accurate readings on my bench, I’ll consider the job complete.
What I need is a clean, reliable way to sense those isolated DC voltages without tying the reference points together. Whether that means introducing isolated amplifiers, optocouplers, tiny transformers, or another proven technique is up to you; I’m open to whichever approach gives stable, repeatable results and keeps the measurement side safe for the microcontroller.
Here’s what I’d like back:
• An updated schematic that drops into my existing design, showing every connection required for proper isolation
• A concise parts list with recommended component values and footprints I can actually buy today
• A short rationale explaining how the proposed stage cures the isolation problem and what accuracy I can expect
KiCad, Eagle, or even a clearly annotated PDF is fine for the schematic, as long as I can replicate it. If you can include quick SPICE or other simulation plots to illustrate behavior, that’s a bonus but not strictly required.
Once I have a working revision that delivers stable, accurate readings on my bench, I’ll consider the job complete.