I'm working with a nano scale motion stage with high precision. To move the stage I supply Sinusoidal wave using (Agilent Keysight) function generator.

This F-gen, goes as low as \$20mV_{pp}\$ in HighZ mode. \$20mV_{pp}\$, corresponds to \$30nm\$ of motion of the stage. For my experiment I need it to move the stage about \$6-7nm\$, so I need to supply small input voltages. My initial attempt was to build a voltage divider circuit (shown below). The output voltage \$V_{out}\$ according to equation (1) should be \$3.7mV\$ when \$V_{in}=20mV\$. Now the problem $is that the circuit produces a huge noise and I cannot read the correct output voltage with oscilloscope. The \$V_{in}\$ (yellow) and \$V_{out}\$ (Red) are shown in the snapshot of the oscilloscope.

\$V_{out}=V_{in}\frac{R_{2}}{R_{1}+R_{2}}\$ ........ (1)

I'm not an expert in field of electronics, I would like to hear your suggestions on how to improve the overall noise issue.

Thank you very much in advance.


simulate this circuit – Schematic created using CircuitLab

enter image description here

  • \$\begingroup\$ Is this how it's actually hooked up? The potentiometer isn't going to be doing anything without the wiper connected. It'll just be a fixed resistor. \$\endgroup\$ – Hearth Nov 5 '18 at 22:26
  • \$\begingroup\$ yes at this point, I just fixed it at \$1K\Omega\$. \$\endgroup\$ – Racaio Cmoto Nov 5 '18 at 22:30
  • \$\begingroup\$ Put a capacitor on the output to set bandwidth to maybe 50Hz. That needs 1/(2*pi*50) timeconstant, or 3 milliSeconds. Using 1Kohm, you need 3uF across the pot. \$\endgroup\$ – analogsystemsrf Nov 6 '18 at 5:33

You can calculate the actual noise of the circuit, as it is basically just an 800Ω resistor in series with the equivalent source (a bit higher, as the signal generator surely has a 50Ω output impedance). At ambient temperature and 1kHz bandwidth, you would have less than 0.2µV of RMS voltage noise due to that circuit (visually <0.6µV p-p). So that is unlikely to be the actual source of your noise.

The possible sources:

  • An oscilloscope is normally not designed to work at inputs <10mV.
  • Grounding problems. Noise being injected between your oscilloscope ground and circuit ground.
  • External interference. Signals coupling into your circuit.

How to solve it:

  1. To actually measure your signal, you would probably have to add a small amplifier (e.g., an instrumentation amplifier that allows you to do a differential measurement).
  2. Make sure you reduced the measurement bandwidth of the oscilloscope.
  3. Adding a filter capacitor would help, and you probably should add it anyway.
| improve this answer | |
  • 2
    \$\begingroup\$ 4. Use a probe with little or no attenuation. At the frequency he's driving it, he can just use a hunk of 50\$\Omega\$ coax with clip-leads on the end. \$\endgroup\$ – TimWescott Nov 6 '18 at 0:55

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.