I have a switching power supply (RC Esc). I am measuring the input voltage from battery using a voltage divider. The Divider is attached to a differential DAQ. I have very large transient spikes in both the voltage and current (current shunt). I can suppress a lot of the spike by placing a capacitor across Vlo and Vhi right before going into DAQ. Is this common practice and what is the math involved in the spike reduction / cutoff frequency.

  • 1
    \$\begingroup\$ Show the circuit, find the impedance (resistance of the DAQ input) \$\endgroup\$ – Voltage Spike Sep 19 '19 at 19:38

Yes some low pass filter is common and essential.

Technically it's called a transient suppression filter for : Conducted load transients V=LdI/dt, radiated crosstalk and anti-alias suppression (where the rise time interferes with an error of the sampled value) .

The details of the filter need to be defined in terms of your needs for timing and accurate voltage measurement or SNR. i.e. if you are looking for a slow error recovery like low battery , fly home or slow down immediately, that affects how much you can filter the input with a low ESR cap for special low pass filter.

The most common method for measuring DC voltage accurately in the presence of a high transient spike noise is to choose the BW of noise to be suppressed and design a filter to suppress that to an acceptable SNR or the signal.

Since you have some sampling rate for data, your signal BW is likely to be DC to 1/3 of the sampling rate. You must have some anti-aliasing filter to prevent additional errors from noise spectrum above the sampling BW appearing falsely as a signal.

This is not a complete answer and depends on your specs for SNR, BW , load spike noise levels and signal integrity or acceptable error.

|improve this answer|||||

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.