By definition, a DMM measures AC and DC voltage and current. Most meters also handle resistance and some form of continuity checking or diode test. This level of basic capabilities is readily available in a 3½-digit hand-held instrument for as little as $8.95. OK, to be fair, that was an online sale price reduced from the normal $12.
Model U1253A True
rms OLED MultimeterCourtesy of Agilent Technologies
Most professionals wouldn’t consider using such a DMM because they require better accuracy, resolution, safety, and durability. Also, many higher performance meters include an extensive list of additional measurements and features that enhance the user experience, as the marketing people say. Nevertheless, even with a large number of extras, an excellent DMM can be bought for only a few hundred dollars.
Hand-held DMMs are tailored to suit the needs of a technician in a field test or troubleshooting application. Fully guaranteed safety characteristics to at least CAT III or IV are needed for any AC-power work although much more than 0.1% accuracy generally isn’t. Some meters are designed to be especially rugged, but protective rubber bumpers also are found on less robust equipment.
By far the largest category of features specific to hand-held DMMs could be classified as convenience items even though that description tends to minimize their importance. If you’re working outdoors, a sunlight-viewable display is a big help. Similarly, auto-ranging and auto-polarity together with peak detect and measurement hold allow you to make a measurement while using both hands to position probes. A strong magnet attached to a hanging strap easily supports a meter when working among steel equipment racks.
In contrast, a benchtop DMM emphasizes extreme accuracy and may have as many as 8½ digits of resolution. These instruments are used as secondary standards in a design lab. They provide 4- and possibly 6-wire ohms measurements, often frequency, and sometimes capacitance. Benchtop DMMs are the heart of many ATE systems, and models are available with extensive switching capabilities. Prices range up to several thousand dollars.
VTI Instruments’ Business Development Manager Tom Sarfi explained that the 6½-digit DMM development for the Model EX1266 LXI Class A switching/DMM was complete. “The DMM is the core of a modular measurement system for a range of data acquisition applications. Much of our work is focused on adding capability that can work in conjunction with the DMM as part of a scanning measurement system,” he said.
Plug-in PCI/PXI/VXI DMMs also are used in test systems, and generally a PC-based modular system has an economic advantage where an integral display and front-panel controls add little value. This kind of approach is more flexible than a proprietary LXI or benchtop system because you’re not limited to the range of modules only available from the DMM manufacturer. Benchtop and LXI system manufacturers counter that their closely coupled instruments may offer better performance.
The trend toward greater functionality isn’t limited to any one DMM format. In all cases, the intention has been to increase the types of applications that the instrument can address.
Tee Sheffer, president of Signametrics, commented in a 2006 EE-Evaluation Engineering article that his company saw “few applications that use more than 30% of the functionality of our DMMs. However, each application uses a different 30% than the previous one. You have to have a broad range of capabilities to address a major fraction of the applications even if no one application uses all of the capabilities.”
Additional Measurements
True rms
Very low-cost meters often measure a signal’s average value but display rms after changing the scale factor. While there’s a fixed relationship between a sine wave’s average and rms values, the ratio is different for every other wave shape. Only a meter that actually measures rms regardless of the waveform will display the correct result for common signals such as a distorted sine wave.
Model EX1266 Modular LXI Class A
Mainframe-Based Switching/DMMCourtesy of VTI Instruments
True rms has been offered on DMMs for many years, but it’s easy to overlook the importance of this feature. Even though these measurements are insensitive to the shape of most signals, all DMMs have a crest-factor limitation. This means that errors begin to creep in when a signal shape starts approaching a pulse. Crest factor is defined as the ratio of peak to rms and typically must be below about 5 to ensure accurate measurements. A sine wave has a crest factor of 1.414.
Inductance
Inductance relates the voltage across a coil’s terminals to the rate of change in current flow: V = L di/dt. Coils wound on nonmagnetic cores are much easier to measure because there are no core nonlinearities to deal with. However, although they commonly are used at RF frequencies, air-core coils have low inductance, so for practical reasons many larger inductors, such as those used in power supplies, are wound on a magnetic core.
You wouldn’t try to make a resistor or capacitor yourself, but prototype inductors are easy to build from readily available cores and copper wire. What appears to be a simple component actually can have very complex behavior.
DC current changes the operating point on the core-material B-H characteristic curve. As the DC level is increased, the slope of the B-H loop gradually becomes less steep. This means that the coil inductance will be smaller. In the extreme, when the core is saturated, the inductance falls to its air-core value. Consequently, a coil intended to be used at a certain DC current level must be biased at that level to measure the correct inductance. An Agilent Technologies’ application note covers DC current-biased inductance measurement in detail.1
Even if a coil operates in a circuit with no DC current flowing, the level of the AC current still may affect the inductance measurement. Magnetic core material requires an amount of energy to activate it—the so-called magnetizing current in power transformers. So, depending on the core size and material, accurate measurements require AC current to be above a minimum level. A coil’s inductance also varies because magnetic core material properties are not constant with frequency.
The electrical model of an inductor includes components that represent the copper loss of the coil, the core loss resulting from energy dissipated to excite the magnetic material, and a shunt interwinding capacitance. In general, these elements are frequency-dependent. For example, the equivalent series resistance (ESR) at DC is approximately equal to the coil resistance, but at high frequencies, skin effect accounts for a much higher value. Also, at some high frequency the inductance will resonate with the interwinding capacitance.
Dedicated LCR meters, also called bridges, are capable of providing a DC bias and measure inductance as well as ESR over a wide range of frequencies. The ratio of the inductive reactance to ESR is the inductor’s quality factor (Q) or Q = 2 ? f L/ESR. In high-frequency designs, it’s common to specify the Q of a coil at a certain frequency. Measuring the coil inductance at a much lower frequency simply isn’t equivalent.
National Instruments’ (NI) Model PXI-4072 6½-digit DMM includes inductance and capacitance measurement based on a technique documented in an NI technical paper and described by Travis White, product manager for precision DC and switches: “We use a very stable, harmonically limited square wave as the current source for the component under test. Then, the digitizer capability of the on-board FlexADC is used to acquire a waveform of the resulting voltage. Finally, an FFT is performed and an algorithm applied to derive the inductance or capacitance.
“The test method offers the added benefit of compensating for losses in the front end, cabling, and the component under test,” he explained, “by extracting the magnitude and phase of the impedance at the excitation source fundamental and third harmonic frequencies and then comparing the differences to calculate the losses in the system and derive a corrected result.”2
Part of the process involves open/short compensation that minimizes the effect of the DMM probe wires without using a Kelvin 4-wire connection. On the other hand, the highest fundamental test frequency is 10 kHz used to measure inductances less than 1 mH. Above 1 mH, the test frequency drops to 1 kHz and again to 91 Hz for inductances greater than 100 mH.
Model EX623 True rms
AC/DC Clamp MeterCourtesy of Extech Instruments
Obviously, you may not get the same result from this approach that an LCR meter would measure at 1 MHz or 10 MHz. The comment is made in the technical paper that “because of the amount of magnetization current required, you may see an increase in sensitivity to frequency changes and other dependency factors in inductors with cores of larger dimensions, such as those used in transformers and power inductors.”
The point of comparing a DMM’s inductance measurement to an LCR meter’s capabilities isn’t to disparage the DMM. Rather, it’s important to understand the limitations of such a measurement. If an inductor is intended for use with a significant DC bias, you can’t measure its performance without the bias. Similarly, if a coil’s characteristics at 20 MHz are important to you, measuring the inductance at 10 kHz won’t give an accurate picture.
Capacitance
Similar to an inductor’s sensitivity to DC current level, many capacitors exhibit sensitivity to voltage levels. A Murata technical paper specifically addresses multilayer ceramic chip (MLCC) capacitor characteristics, which can include large sensitivities to voltage depending on the dielectric material.3 For example, the capacitance of a type X7R 10-µF 10-V size 1206 surface-mount MLCC was reduced by 60% when the DC bias was raised from zero to 10 V. The capacitance of a similar device made with type Y5V material dropped by 90% under the same conditions.
These materials are attractive because their dielectric constants are high. This means that a large capacitance can be manufactured in a small package. Of course, just as the normal DC bias current must be provided when measuring inductance, so too is the operational DC voltage level critical for these and some other types of capacitors.
Like inductors, capacitors also are characterized by their ESR and Q, in this case Q = 1/(2 ? f C)/ESR. Dissipation factor (D) is equal to the inverse of Q. High-capacitance values can be difficult to measure accurately because the impedance is very low at high frequencies. As a result, the effects of ESR and the DMM probe lead resistance are significant. A 4-wire connection minimizes this problem by applying the excitation current on one pair of wires and measuring the resulting voltage on a separate pair.
Although it’s obvious that you can’t measure X7R or Y5V MLCC capacitors without taking into account their voltage dependence, few people would question a DMM’s resistance measurements. Nevertheless, high-value resistors often show a similar but much smaller voltage sensitivity. For example, the 4500 Series of high-voltage planar resistors available from Welwyn Components is rated for up to 20-kV DC and from 1.0-W to 4.5-W power dissipation.
Depending on the power rating, the maximum guaranteed voltage coefficient of resistance ranges from 1.5 to 5 ppm/V. A 100-M? resistor operating at 10 kV dissipates 1 W. Assuming the worst-case 5-ppm/V value, this resistor could change value by as much as 50,000 ppm or 5%. Even the typical 3-ppm/V value for a 1-W or 1.7-W member of the 4500 family gives a 3% change.
Because of this behavior, it’s common to place several lower value, less sensitive resistors in series if it’s necessary to accurately maintain a certain high resistance at high voltage. The point of the example is to demonstrate that even resistance measurement is not necessarily straightforward. When a component is used at signal levels or frequencies that are very different from a DMM’s test conditions, the resulting measurements probably will not be accurate.
Temperature
According to Extech Instruments’ André Rebelo, “Many multimeters offer temperature as an auxiliary measurement by using type K thermocouple bead probes. The new MM570 DMM goes one step further with inputs for monitoring two temperature readings simultaneously and the convenience of one button push to calculate the difference between the two readings. In addition to a number of heating, ventilation, air-conditioning, and refrigeration (HVACR) applications,” he continued, “differential temperature measurement capabilities are invaluable for electronics testing when heat dissipation of components is being tested or when cooling airflow duct paths are being evaluated.”
Agilent’s U1240 Series Hand-Held DMM also supports dual and differential temperature measurements. Jason Saw, a product manager at the company, explained, “Ensuring the safety of a boiler in a heating system requires measurement of the boiler and air temperatures. Making simultaneous measurements is critical in obtaining real-time readings.”
Type K thermocouples are formed by welding together a Chromel (nickel with 10% chromium) wire and an Alumel (nickel with 5% aluminum silicon) wire and are suitable for measuring temperature differences from -200°C to +1,350°C. Like all thermocouples, they generate a voltage proportional to a temperature difference.
Actually, three thermocouples always are involved: the obvious one measuring the unknown temperature and the two unintended thermocouples formed by connecting the thermocouple wires to copper terminals in the DMM. Making accurate temperature measurements with thermocouples involves the following:
•?Establishing intimate contact between the thermocouple and the DUT.
•?Maintaining the two copper connections at the same temperature by attaching them to an isothermal block.
•?Measuring the temperature of the isothermal block.
•?Linearizing the inverse relationship between thermocouple output voltage and temperature.
The temperature that the DMM displays is the difference between the DUT and isothermal block temperatures corrected to account for the isothermal block temperature being other than 0°C. Thermocouple measurements are subject to several error sources including temperature differences across the isothermal block, isothermal block temperature measurement uncertainty, and thermocouple material impurities.
Type K devices show a nearly linear relationship between voltage and temperature of about 41 µV/°C. With many DMMs, only type K thermocouples can be used because of the on-board scaling and linearization. Keithley Instruments’ 2001-TCSCAN Thermocouple/General-Purpose Scanner Card is used with the company’s Model 2001 DMM and handles thermocouple types J, K, T, E, R, S, and B.
Extech’s EX400 Series DMMs also feature patented, built-in infrared thermometer capabilities. Noncontact temperature measurement is useful if you need to measure temperature a safe distance away from dangerous equipment or of hard-to-reach components. More generally, infrared offers point-and-shoot convenience that makes multiple sequential measurements easy to perform compared to repetitive placement and removal of bead, clamp, or Velcro temperature probes.
Additional Functions
Agilent’s Mr. Saw was asked whether he considered DMM development to be complete. He replied, “It appears that trends are moving toward the intelligent DMM model: instruments packed with extra features on top of DMM functions as well as features that incorporate analytical capabilities beyond the collection of raw measurement data.”
Many of the additional functions included in DMMs in some way process or extend basic measurements to solve application problems. The 2001-TCSCAN Scanner Card customizes and automates the voltage measurement capability already provided by several Keithley DMMs. Data logging and trending are supported in Fluke’s Model 287 Multimeter. As the data sheet states, “TrendCapture quickly and graphically displays logged data. You can log up to 15,000 events and zoom the TrendCapture display.”
A more fundamental change took place when DMMs integrated an AC or DC signal source. This is a common theme with many variations. NI’s DMMs measure inductance and capacitance using a built-in harmonic-limited square wave source. Agilent’s U1250 DMM has a programmable square wave output that can range from 0.39% to 99.6% duty cycle, 0.5-Hz to 4,800-Hz frequency, and 0 to 2.8-V amplitude.
One use for this capability is to simulate speed control pulses from a PWM motor drive. Keithley’s Models 2015 and 2016 generate sufficiently pure sine waves that the instruments are used for audio band quality measurements and analysis such as THD, THD+Noise, and SINAD.
The combination of a source and DMM has proven so useful and popular that a new class of source-measure units (SMUs) has developed. Keithley specialized in low-level, accurate measurement for many years and added a four-quadrant source. Agilent manufactures both DMMs and power supplies as well as SMUs. NI recently announced its first PXI SMU, the single-channel PXI-4130 that can source 40 W in quadrants I and III and sink 10-W in quadrants II and IV.
Several of the additional features available in today’s DMMs are listed in Figure 1. Some are sufficiently popular, such as measurement hold or peak detect, that they appear in instruments from many different manufacturers. Others, such as intrinsic safety or totally waterproof, are much more specialized, so perhaps only one DMM model provides them.
manufacturers appear in bold type.
Summary
The amount of design and development money supporting new DMM functionality ensures that you will continue to see innovative improvements for many years. To determine the best DMM for your needs today, however, a few guidelines may help.
First, determine which basic measurement capabilities you need. Do you require inductance and capacitance, or are they just nice-to-have features and not really necessary? This is important because several very good DMMs do not offer inductance and capacitance. If you do need to make these measurements, can you do so with a DMM—any DMM? Perhaps you need an LCR meter if the measurements must be made with the components biased as they would be in operation.
Next, find a model having the extra features you want, such as data logging or dual/differential temperature measurement. And try to find one that also has a graphic display. Quarter VGA displays have become fairly common in DMMs, and they offer much greater flexibility. Some models actually display waveforms or trends in logged data. Even if the graphical capability isn’t exploited, this type of display supports multiple, simultaneous readouts so you can compare this reading and the previous one. Or, you can view voltage, current, and a related parameter such as frequency.
Finally, attempt to future-proof your purchase as much as possible without compromising the present requirements or budget. A DMM that provides a switching capability or is modular so it can be mixed with other functions is a good choice. On the other hand, given the relatively low cost of a DMM and the continuing feature innovation, you may choose to buy what you need now and address new applications as they arise.
References
- “Wide Range DC Current Biased Inductance Measurement,” Application Note 369-8, Agilent Technolgies, 2008.
- “Capacitance/Inductance Measurements,” http://zone.ni.com/devzone/cda/tut/p/id/3078
- “Capacitance and Dissipation Factor Measurement of Chip Multilayer Ceramic Capacitors,” Murata Manufacturing, TD No. C10E, http://www.murata.com/cap/measure.pdf
FOR MORE INFORMATION | Click below | |
Agilent Technologies | U1253A True RMS OLED Multimeter | Click here |
Extech Instruments | EX623 True RMS AC/DC Clamp Meter | Click here |
Fluke | Model 287 Multimeter | Click here |
Keithley Instruments | Model 2001 7½-Digit DMM | Click here |
National Instruments | PXI-4072 6½-Digit DMM | Click here |
Signametrics | SMU2064 7½-Digit USB Multimeter | Click here |
VTI Instruments | Model EX1266 Class A LXI DMM | Click here |
August 2009