1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

I Electronic Noise General Questions - Laboratory Setting

  1. Dec 7, 2016 #1
    Hello all,

    I'm an undergraduate at a large research university in Massachusetts working in a lab where my group contributes to the Muon g-2 experiment at Fermilab in chicago. Essentially what I'm working on is optimizing an optical rotation (Faraday) magnetometer system for use in magnetic field calculations of timescales on the order microseconds (particle decay) and had a few general questions I was hoping some of you might be able to help me with.

    To make all this short (something antithetical to my nature so please bear with me) I have a system; the magnetometer (an SF-59 crystal rod with a solenoid wrapped around it and a function generator attatched to generate a B field which has a 1mW ~406nm laser passing through the central axis of the rod and into a 90 degree beamsplitter which bifurcates the beam at a right angle and distributes it into two photodetectors). The photodetectors are wired up in a subtractor circuit (the intent and result of which is to subtract away the common background noise experienced by both photodetectors. The circuit is encased in a metal box which is intended on shielding the circuit from ambient electronic noise. The circuit is wired to a DAQ assist data collection box and fed into LabView where I can analyze the voltage and fourier transform it to determine the strength of our signal versus the noise. The objective is to reduce the noise we see at 60 Hz as much as possible in order to see a minute signal at the actual project.

    Given that exposition I was wondering a few things; I have an idea of how electrical noise is generated and enters a system but was seeking validation, clarification or correction in my understanding. Household electronics are powered by electrical currents (typically AC currents -- which oscillate at a particular frequency, say 60 Hz) as a result of the flow of charge photons are radiated away with frequencies corresponding to the number of oscillations a second of the current. For instance the lightbulbs operate at 120 Hz and as a result anytime I take data with the lights on I get an enormous spike at 120 Hz in my power spectrum). Now I understand the generation of noise (I think!) but exactly how it seeps into a system im not as clear on. I imagine the photon gets absorbed by some portion of the system (say the laser beam or the actual circuit (through inductance of the resistor?) or the cables leading from the circuit to the data acquisition terminal) and then contributes that frequency component to the overall electrical signal as its transduced back into electrical current. Any clarification or further understanding (as there always is) is much appreciated. I need to know the origin and process behind the noise before I can think through how to effectively eliminate it.

    In addition to that general question, being that my project emphasizes 60 Hz noise (as that's the frequency our signal will be at) I was wondering if there were any standard conventions for eliminating 60 Hz noise or more broadly speaking if there are conventions for eliminating noise of any arbitrary frequency?

    Furthermore does anyone know of any common sources of 60 Hz noise that might nudge me in the right direction on my investigation? As I mentioned the 120 Hz noise coming from the room lights previously I imagine there are some common appliances which put out noise at 60 Hz as well -- perhaps computers? Monitors? Cellphones? Etc...

    I thank anyone who was willing to read through my entire message. I would appreciate as prompt a message as possible (as I need to address this issue again tomorrow at noon) so i encourage anyone who wants to reply to any portion of my message (the general question about understanding noise, the conventions question or listing any possible sources) to please do so without feeling obligated to respond to everything I've posted, a response to any part of my questions will be greeted with appreciation and gratitude. Thanks to anyone willing to help a frustrated undergrad out!
  2. jcsd
  3. Dec 8, 2016 #2


    User Avatar
    Science Advisor

    For 120Hz, the photosensors may be picking up the room lighting, throw a blanket over them and see if it helps.
    Power supply ripple in the electronics is another common source. Many switching power supplies are worse than a linear power supply.

    For 60, 180, 360, and 540Hz, generally hum pickup is due to the electric field around the nearby power wiring, to a lesser extent due to the magnetic field.
    The electric field can be reduced by enclosing all of your devices and wiring in a conductive enclosure, then connecting each enclosure to a common point of your first stage of amplification/instrumentation. The signal wiring between devices should be done with shielded cables, particularly any low level or high impedance signals, as from the photosensors..

    Magnetic field shielding requires a ferromagnetic enclosure, such as steel. This may be the same enclosure used for the electric field shielding.
    Signal wiring between items should be twisted-pair wiring so that the magnetically induced voltage is balanced in the two conductors. Balanced or differential outputs and inputs should be used If the instrumentation in use has them available.

    Fluorescent lamps are notorious for wide-band interference, predominately at 120Hz for those using a magnetic ballast (the older four-foot overheads and desk lamps), and several kHz for electronic ballasts (the screw-in 'twisties.') Use incandescents when you can.

    Good luck! And let us know how it turns out.
  4. Dec 8, 2016 #3
    Thanks for the prompt response and considerate insight Tom G! I figured shielding via a conductor/ferromagentic metal would most likely be the way to go, I'm currently running through a diagnostic procedure of sorts in order to ascertain where exactly the noise is leaking into my system and I'm keeping the points you laid out in mind, so thank you again :)
  5. Dec 9, 2016 #4


    User Avatar
    Science Advisor
    Gold Member

    If your lab has fluorescent lights, then you may pick up 180 Hz noise (and other odd harmonics) from the transformer ballasts. This arises from saturation of the iron cores. It's a particular problem for magnetic sensors.

    As for electric pickup, I recommend Ott's book Noise Reduction Techniques in Electronic Systems.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted