Getting a Negative Reading When Making a Resistance Measurement

Updated Nov 17, 2017

Issue Details

I am using a DMM to measure a resistance. For the purpose of testing I had shorted all the input channels and made a resistance measurement using the DMM Soft Front Panel. I read a -12 mΩ. Why did I get this measurement?


This behavior is something that happens with most DMMs in general and not just with National Instrument DMMs. The issue lies in the fact that we are trying to make a low level voltage measurement  with many considerations. 

Firstly, to increase the resolution and decrease the effect of noise, use the measurement range closest to the resistance you are measuring. If you are measuring a short for example, use the 100Ω range on the NI 4070. This will provide a 1mA test current. For this example, consider a short having the resistance of 0.01Ω. With the 1mA test current the voltage appearing across the load will  be 10µV. This is a small voltage measurement, but in this range the NI 4070 has a resolution of 100nV and a calculated accuracy of ±2µV. Even though this is getting to be a small measurement that noise can affect, this short can still be measured accurately. 

However, if a 100kΩ range was used instead of the 100Ω, then a test current of 10µA is generated. By using this test current a 0.01Ω load would produce a voltage of 100nV. Referencing the specifications for the NI 4070, you will see the maximum resolution in the 100mV DC voltage range is 100nV and 1µV is the maximum in the 1V DC voltage range. The 100kΩ range uses the 1V DC voltage range to measure the voltage across the resistor, therefore you will not get a good resistance measurement by choosing the incorrect range. 

Selecting the incorrect range can result in the incorrect selection of the test current being used. This will result in voltages too small for the DMM to measure. These small voltages will also be susceptible to additional sources of errors that will offset your reading and possibly result in a negative measurement. The main sources of error can arise from thermoelectric EMFs, offsets generated by rectification of RFI (radio frequency interference), and offsets in the voltmeter input circuit. To reduce the error caused by these sources you can do the following:
  • Remove any sources of noise or add shielding (RFI)
  • Allow instruments sufficient warm up time to prevent offsets (thermoelectric EMFs & offsets in voltmeter)
  • Use OCO (offset compensated ohms) to cancel out any offsets (offsets in voltmeter)
  • Try to use the same metal to make all of your connections (thermoelectric EMFs)
  • Reduce any temperature gradients though out your circuit (thermoelectric EMFs)

Measuring a signal using a range that is too large can result in the signal being too small to read because of the resolution and the introduction of noise and other sources of error. This is something common to most general digital multimeters, but by using the above steps these effects can be minimized.


Not Helpful