MEASURE FOR MEASURE
Qualitative Indications From Quantitative Instruments
by Graeme C. Payne
When we make measurements with inspection, measuring and test equipment (IM&TE), we assume and expect that all of the measurements are quantitative. That is, the result of the measurement represents some physical property—length or mass of an object, amount of electric current, pH of a solution and so on. This expectation is higher if the IM&TE is calibrated.
However, that expectation is not always justified, because a measuring instrument can also assess qualitative properties.
First, remember that quantitative refers to properties that can be expressed as a range of quantities identified by measurements. Qualitative refers to properties that are different in ways not normally subject to physical measurements.
Some multifunction IM&TE includes functions that might appear to be quantitative but are actually qualitative. One example is a pair of functions commonly found on digital multimeters (DMMs): diode test and continuity (see photo). Either of these functions can produce a numeric value on the display, but that number is not a quantitative measurement.
DMMs—handheld, bench top or rack mounted—typically measure current, resistance and voltage. Those are all quantitative measurements. They have performance specifications and can be calibrated. Of these, the direct current (DC) voltage is the most important, because all other parameters are converted to a DC voltage measurement.
For example, resistance is measured by passing a known current through the unknown resistance and measuring the voltage drop. From Ohm’s law, the meter calculates the resistance value (voltage divided by current).
People who use DMMs to troubleshoot and repair electronic equipment sometimes need other tools as well, and they and their employers often have limited budgets. IM&TE manufacturers recognize that, so additional features can be included in a DMM, often at a low cost.
The DMM in the photo, for example, can also measure temperature and capacitance. These are derived by leveraging existing functions from the basic measurements. The output of a thermocouple is a small DC voltage and can be measured directly.
Capacitance is often measured by sending the known current (from the resistance function) through the unknown capacitor and measuring the time for it to charge up to a specific voltage. Temperature and capacitance are quantitative values with performance specifications.
This DMM will also test diodes and small transistors and indicate continuity. These three functions are qualitative, not quantitative. There are several reasons. The most fundamental is that the typical user of these functions is principally interested in a qualitative indication—good or bad—and the qualitative indication happens to be inexpensive to implement.
Continuity is the presence of a continuous electrical path from one point to another. The ideal value would be close to zero ohms. A continuity test function sounds a buzzer when low resistance is present, typically less than 10 ohms.
This speeds up testing because the audible tone means the technician can concentrate on putting the test leads where they need to be instead of watching the meter. If she is testing a multi-conductor cable for continuity, after checking the first pair of pins she can quickly move on to the next pair when she hears the tone. Different action is needed only if the tone is not heard.
Many DMM specification sheets only show the presence of the function, without any performance specifications. If there is a specification, it is often something like “tone below 15 ohms, no tone above 20 ohms (typical).”
The diode test function uses the DMM’s internal current source (the same one used for resistance and capacitance). This is the general schematic symbol for a semiconductor diode:
A diode acts like a one-way turnstile gate. When electrons (electric current) flow from left to right, the diode conducts, and there is a small voltage drop. Current cannot flow in the opposite direction.
A DMM uses the resistance function to show indications for the diode test. That is natural, since that function’s current source is being used. But it is important for everyone—from the technician using the DMM to the metrology engineer writing the calibration procedure—to remember that the parameter being checked is not resistance. What is actually displayed is the voltage across the conducting diode junction, which, for a silicon diode, will be about 0.6 volts. That is considered a good indication.
If the meter indicates close to zero, the diode is shorted and therefore bad. If the meter indicates over-range, it might be connected backward. If the meter still indicates over-range after changing the test lead connections, then the diode is open (bad). All of these are qualitative indications. The only function of the diode that is tested is its ability to conduct when properly connected, and the main interest of the technician is knowing whether that part needs to be changed.
Quantitative measurement of any parameter of a diode requires at least an oscilloscope and a regulated current source, and the resulting set of curves will demonstrate that the relationship between current and voltage is nonlinear. (Resistance is a linear relationship.)
If a DMM’s diode test function has a specification (many do not), it is often something like “3.000 V range, ± 1% of reading (typical).” Sometimes the magnitude of the test current is specified—“test current 1 mA (typical)”—but usually it is not, and if the current is not accurately known, then no meaningful measurement can be made.
DIGITAL MULTIMETER EXAMPLE: The diode test and continuity functions are circled.
Typical vs. Approximate
It is very important to understand the word “typical” when it appears in performance specifications. A specification is something that describes the expected performance of a population of instruments and is part of what the manufacturer guarantees.
“Typical” and similar words indicate cases in which parts of the specifications are excluded from the guaranteed performance. “Typical” means the manufacturer believes most of the instruments will perform as stated but there is no guarantee the one you have actually will.
The word “typical” is a practical notice to the user of several things:
- That part of the specification should not be used to evaluate or compare instruments. Even though your particular instrument might meet the specification, there is no assurance that any other unit will. There is nothing that can be fixed if the specification is not met.
- The values might be approximate or estimated, with an unknown amount of uncertainty.
- That part of the specification is typically not verified during a calibration procedure. Often, the only thing that can be done with any meaning is to verify that the feature appears to do something.
- If your use of the instrument depends on the “typical” specification, you might need to look for something else.
So, it is possible for calibrated IM&TE to have functions that appear to produce numeric results but are in fact only qualitative indications. These functions are there to help you, the user, do your job faster and easier.
Sometimes the only thing of interest is the presence or absence of an aural or visual indicator. Sometimes you have to be aware that the settings of the instrument might not relate to the actual parameter being tested, as when the resistance function is leveraged to test a diode.
Finally, you always have to watch out for “typical” and its cousins when they appear in the specification pages, as it always indicates a parameter that might not be calibrated and might raise other important questions for you.
GRAEME C. PAYNE is president of GK Systems, a consulting firm specializing in measurement science. He is a contributor to The Metrology Handbook (ASQ Qual-ity Press, 2004) and is past chair of the ASQ Measurement Quality Division. Payne is a senior member of ASQ, a certified quality technician, calibration technician and quality engineer.