- What is the most common problem with precision analog circuits?
Probably grounding errors, but there are a number of frequently seen mistakes. They are mostly sins of omission; engineers are not perfect and can forget things.
- Don't forget to read the data sheet. (Application engineers routinely shout "RTFDS"1 as they hang up after an enquiry.) Extracting implicit information from a data sheet, not just the explicit details, is important.
- Don't forget Ohm's Law. The resistance of a wire or PC track is not zero, and leakage in "insulators" matters when measuring low currents.
- Don't forget the bias current. Sometimes greasy fingerprints provide a current path in the prototype, leading to surprises in the clean(er) final version.
- Don't forget the stray resistances, inductances, and capacitances of the final (crowded) PCB; don't assume that all is well because the breadboard (or the SPICE model) worked.
- Don't forget that EMI and RFI occur everywhere; filter your supplies and input/output leads.
- Don't forget to consider the effects of temperature variation on components (including the effects of differing temperature coefficients in nominally identical components).
- Don't forget to verify that the circuit can tolerate having its supplies (and signals) applied in any order (and with any value of dv/dt)or to ensure that it cannot be exposed to unacceptable power/signal sequences and rates.
- Don't forget that switching power supplies are not as noise free as a battery.
- Don't forget that analog circuits, unlike microprocessors, often do not reset on power up and that you may need to ensure correct start-up.
- Don't forget that circuits don't start instantly: capacitors must charge and precision circuits must stabilize.
- Don't forget that some circuits are unstable when driving a reactive load. An output stage that will drive a wide range of resistive loads may oscillate with capacitance, such as that of a cable.
- Don't forget that noise, like death and taxes, is universal. Every ADC has quantization noise, every resistor has Johnson noiseyou can't avoid them.
- Don't forget that IC designers may not be user friendly. Devices may not work as you expect. Again I say, "RTFDS!"
All these issues, and more, have been discussed in earlier columns. Read them all, and pin up this list where you can read it every day!
1 RTFDS = "Read The Friendly Data Sheet."
For More Information
A Dozen Ways to Make a Circuit Fail
The Contaminator of Signals: HF Common-Mode-Generated Errors
Avoid Common Problems When Designing Amplifier Circuits (pdf)
Power Supply ManagementPrinciples, Problems, and Parts (pdf)
ADC Input Noise: The Good, The Bad, and The Ugly. Is No Noise Good Noise? (pdf)
A Practical Guide to High-Speed Printed-Circuit-Board Layout (pdf)
Practical Techniques to Avoid Instability Due to Capacitive Loading
(Ask the Applications Engineer-32) (pdf)
Avoiding Op-Amp Instability Problems in Single-Supply Applications (pdf)
Do Something with that Unused Pin!
Considerations on High-Speed Converter PCB Design, Part 4: Plane Coupling
Considerations on High-Speed Converter PCB Design, Part 3: The E-Pad Low Down
Considerations on High-Speed Converter PCB Design, Part 2: Using Power and Ground Planes to Your Advantage
Considerations on High-Speed Converter PCB Design, Part 1: Power and Ground Planes
Taming A/D Converter Power Supplies
Bring on the Converter Noise! - Part 2
Bring on the Converter Noise! - Part 1
Lock Down That Noise - Don't Let It Escape
Op-Amp Noise can be Deafening Too
Igor, pass me some supply sequencers!
Grounding Converters (or Philosophy to the Rescue!)