Monday, November 7, 2011

Comprehensive Look at Sources of Error in DC Voltage Measurements Part 1

In this two part post we will (or at least attempt to) take a comprehensive look at all of the factors that can lead to errors in a DC voltage measurement with a DMM and how to eliminate them so you can achieve the highest accuracy possible in your measurement. In part one we will cover radio frequency interference, thermal EMF errors, noise caused by magnetic fields, and common mode rejection. If you are a seasoned DMM measurement veteran and you feel I missed something in the following sections please add it as a comment.


Radio Frequency Interference -- Most voltage-measuring instruments can generate false readings in the presence of large, high-frequency signal sources such as nearby radio and television transmitters, computer monitors, and cellular telephones. Especially when the high frequency energy is coupled to the multimeter on the system cabling. This effect can be severe when the cabling is 1/4, 1/2, or any integer multiple of the high frequency wavelength. You probably have experienced this type of effect first hand if you ever placed a mobile phone near speaker wiring and heard bursts of noise from the speaker that were certainly not part of the intended audio experience. To reduce interference, try to minimize the exposure of the system cabling to high-frequency RF sources. You can add shielding to the cabling or use shielded cabling.  If the measurement is extremely sensitive to RFI radiating from the DMM or your DUT, use a common mode choke in the system cabling, as shown in the figure below, to attenuate DMM emissions. Often you can see this same EMI reducing method being used on the data cable for your computer monitor.


Thermal EMF Errors -- Thermoelectric voltages, the most common source of error in low level voltage measurements, are generated when circuit connections are made with dissimilar metals at different temperatures. Each metal-to-metal junction forms a thermocouple, which generates a voltage proportional to the junction temperature. It is a good idea to take the necessary precautions to minimize thermocouple voltages and temperature variations in low level voltage measurements. The best connections are formed using copper-to-copper crimped connections. The figure below shows common thermoelectric voltages for connections between dissimilar metals.

Agilent benchtop DMMs use copper alloy for their input connectors
Noise Caused by Magnetic Fields -- When you make measurements near magnetic fields, take precautionary steps to avoid inducing voltages in the measurement connections. Voltage can be induced by either movement of the input connection wiring in a fixed magnetic field, or by a varying magnetic field. An unshielded, poorly dressed input wire moving in the earth’s magnetic field can generate several millivolts. The varying magnetic field around the ac power line can also induce voltages up to several hundred millivolts. Be especially careful when working near conductors carrying large currents. Where possible, route cabling away from magnetic fields, which are commonly present around electric motors, generators, televisions and computer monitors. In addition, when you are operating near magnetic fields, be certain that the input wiring has proper strain relief and is tied down securely. Use twisted-pair connections to the multimeter to reduce the noise pickup loop area, or dress the wires as closely together as possible.

For more on magnetic coupling and other spurious coupling issues in measurements check out the post Ground Loops and Other Spurious Coupling Mechanisms and How to Prevent Them

Common Mode Rejection (CMR) -- Ideally, a DMM is completely isolated from earth-referenced circuits. However, there is finite resistance between the DMM’s input LO terminal and earth ground. This can cause errors when measuring low voltages that are floating relative to earth ground. Check out the post Understanding Common Mode DMM Specifications for more information on CMR.

Stay tuned for part 2 next week!


9 comments:

  1. Nicely written--This post is equally informative as well as interesting . Thank you for information you been putting on making your site such an interesting If you have time please visit my RTCA-Do160 page.

    ReplyDelete
  2. I want someone to help me with my chemistry homework because that is too tricky and complicated so i needed help

    ReplyDelete
  3. Current clamps are electrical devices with two jaws that open to allow clamping around an electric conductor.www.themultimeterguide.com

    ReplyDelete
  4. Multimeter is good tool for testing. Read more best multimeter reviews here http://www.zoniv.com/best-multimeter/

    ReplyDelete
  5. Java is an object oriented programming language. It is used in a variety of computing platforms, you can see it nearly everywhere nowadays, from embedded devices and mobile phones to enterprise servers and supercomputers. See more c language cheat sheet

    ReplyDelete