INTERPRETING THE READINGS:
The Lies Multimeters Tell
 

There is no doubt as to the impact digital multimeters (DMMs) have had on the electrical and electronic world. However, when it comes to even simple electrical testing, DMMs have been known to provide erroneous readings thus leading the user up the garden path - and then slamming the door neatly in the face!

There are two issues with DMMs, and each has unique problems.

The first is the more expensive types tend to be very electrician friendly in that they average out the readings. The result is slight variations in the input voltage cause very little change in the reading and give the user confidence the supply is stable.

At first this appears to be a good thing. Not so. Should there be a bad joint somewhere, such meters cause the user to miss this. Cheaper models, those without the averaging mechanism, will have the reading change dramatically thus indicating something is wrong.

The second issue is these meters have a very high input impedance. Should a cable being tested run together with a live cable there may be sufficient coupling between the two (please note coupled, not induced) for the digital meter to read a voltage close to the working voltage of the installation thus making the user believe the circuit is still live.

Although not in itself a bad thing, it can cause the user to waste a significant amount of time tracing why the circuit is live, when in actual fact it is fully isolated and perfectly safe to work on.

Cheaper DMMs have a lower input impedance and are therefore more suited to testing for 'dead' circuits (but it takes a lot of convincing to get people to purchase more economical makes!).

The alternative is the good ol' fashioned analogue moving coil meter as these typically have a much lower input impedance; In the order of 4kW/V i.e. 500V scale is 2MW vs. the typical 200MW of a DMM.

Even with the correct kit, there are times one just has to accept what they say about bad workmen! Good training is the only way out of this hole.

  >>


| | Ask a Question |

© 07.10.04