8 Tips for Getting More out of Your DMM

Tip 1 Avoid Measurement Errors Caused by Connections, Test Leads, and DMM Wiring
The easiest way to eliminate errors caused by wiring is to make a null measurement. For DC voltage or resistance measurements, select the appropriate measurement range, connect the probes together and wait for a measurement—the closest you can get to a zero input—then press the null button. The readings below will have the null measurement subtracted. Null measurements are great for DC and resistance measurement functions. But this technique is not ideal for AC measurements. AC converters don't work well in the lower parts of the range; the Agilent 34401A DMM's analog converter is not specified below 10% of full scale. The Agilent 34410A and 34411A DMMs use digital technology that can measure all the way down to 1% of full scale, but they can't be used to measure shorts.

Connections
If you make connections between dissimilar metals, you create a thermocouple junction. A thermocouple junction generates a voltage that varies with temperature. This voltage is low, but if you're measuring small voltages or your system has many connections, it's a problem. Think of this thermocouple junction as being at the DUT, the relay (multiplexer), and your DMM. Using a copper-copper junction minimizes this offset.

You can use offset compensation to measure any offset voltage and deduct this error when making resistance measurements. Figure 1 shows two measurements taken in an offset compensated measurement, the first with a current source and the second without. Subtracting the second reading from the first and dividing it by the known current source value gives the actual resistance value. Since two readings are taken in the measurement, the reading speed will be slower, but the measurement accuracy will be improved. Offset compensation can be used for both two-wire and four-wire resistance measurements.

figure 1

Offset compensation uses two measurements. The first measurement is a standard ohm measurement; the second is a measurement of the offset produced by the thermal EMF. The voltmeter reading is the difference between these two measurements divided by the known current source. The

four
-wire ohm method is the most accurate method for measuring small resistances. Test lead resistance and contact resistance are automatically deducted using this method. The four-wire resistance measurement connection is shown in Figure 2. Using a known current source and measuring the voltage produced by the resistor, the unknown resistance value can be calculated. An additional set of test leads is used to carry the current to the unknown resistor, and the voltage produced across it can be measured by the voltage sensing leads. No current flows through the voltage sensing leads, so it does not produce a voltage drop.

No current flows through the voltage-sensitive leads. The DMM divides the measured voltage by the known current to obtain the unknown resistance.

Internal DMM offset
Autozero is used to eliminate error sources within the DMM. When Autozero is enabled, the DMM internally disconnects the input signal after each measurement to obtain a zero reading. This zero reading is then subtracted from subsequent measurements. This prevents the effect of offset voltages present in the DMM input circuitry on measurement accuracy. Autozero is always enabled for four-wire measurements, but you can
disable . When Autozero is disabled, the DMM takes a zero reading once and then subtracts it from all subsequent measurements. A new zero reading is taken each time you change the function, range, or integration time.

Tip 2 Measuring Large Resistors
Settling Time Effects
Capacitance in parallel with a resistor creates settling time errors after initial connection and after a range change. Modern DMMs insert a trigger delay that gives the measurement time to stabilize. The length of the trigger delay depends on the function and range selected. These delays are adequate for resistance measurements when the combined capacitance of the cable and the device is less than a few hundred pF, but if there is shunt capacitance on the resistor or you are measuring resistances above 100 kΩ, the default delays may not be enough. Settling can take quite a while due to the effects of the RC time constant. Some precision resistors and multifunction calibrators use shunt capacitors (1000 pF to 100 μF) that, along with the high value resistors, filter out noise currents injected by the internal circuitry. Dielectric absorption (wetting) effects in cables and other devices can increase the RC time constant and require longer settling times. In this case, you may need to increase the trigger delay before making the test.

Offset Compensation in the Presence of Capacitance
If there is shunt capacitance on the resistor, you may need to turn off the offset compensation. When the offset compensation takes the second reading without the current source, it will measure any voltage offset. But if the device has a long settling time, this can result in an erroneous offset measurement. DMMs use the same trigger delay for offset measurements to try to avoid settling time issues. Adding a trigger delay is another
solution .

Connections in High Resistance Measurements
When you measure large resistances, insulation resistance and surface contamination can cause considerable errors. Take various precautions to keep high resistance systems "clean." Test leads and fixtures are very sensitive to leakage caused by moisture absorption by insulation materials and "dirty" surface films. Nylon and PVC are relatively poor insulators (1013 GΩ) compared to PTFE Teflon insulators (109 Ω). If you are measuring a 1 MΩ resistor in wet conditions, the contribution of nylon or PVC insulator leakage to the error can easily be as high as 0.1%.

Tip 3 Using DC Bias for AC Measurements
Many signals contain both AC and DC components. For example, an asymmetrical square wave contains both. Many audio signals also contain a DC offset created by the DC bias current used to drive the output transistors. Some situations require measuring a DC+AC voltage, while others may only require the AC component. For this audio example, the amplifier gain is the comparison of the input AC voltage to the output AC voltage.

Most modern multimeters use a DC blocking capacitor in front of the AC RMS converter. It isolates the DC voltage and allows the multimeter to measure only the AC value. More importantly, the multimeter can scale the AC signal for the best measurement. For example, when measuring the AC ripple of a power supply, the multimeter isolates the high level DC and amplifies the AC signal according to the range selected for the AC component.

For the most accurate AC+DC measurements, the two components should be measured independently. The multimeter can achieve the best possible DC measurement by using the appropriate range and integration time to suppress the AC component. When making AC measurements, select the appropriate range according to the AC component. You can calculate the AC+DC RMS value using the following formula:

True RMSAC+DC = √ ( AC2 + DC2 )

Agilent's new 34410A and 34411A use DC blocking capacitors when making AC voltage measurements. AC measurements are made digitally, which results in faster settling times and the ability to handle higher crest factors, which is often encountered when measuring pulse trains. When measuring pulses, make sure the pulses do not contain frequencies above the multimeter's bandwidth. The 34410A and 34411A can measure AC signals up to 300 kHz. If there is a significant AC component at frequencies below 8 kHz, the 34410A and 34411A have a DC function with peak detection to accurately measure both DC and AC components. For higher frequency signals, you can measure the AC component alone and use the formula to calculate the AC+DC measurement result.

Tip 4 Measuring Low Frequency AC Signals with a DMM
Most modern multimeters can measure AC signals as low as 20 Hz. However, some applications require measuring even lower frequencies. To make such measurements, you need to select the right multimeter and configure it appropriately. Consider these examples:

The Agilent 34410A and 34411A multimeters use digital sampling techniques to make true RMS measurements down to 3 Hz. They digitally increase the settling time to 2.5 s at the slow filter setting. To make the best measurements, you should note:

1. It is very important to set the correct AC filter. The filter is used to smooth the output of the true RMS converter. At frequencies below 20 Hz, the correct setting is LOW. In the LOW filter setting, a 2.5 s delay is inserted to ensure that the multimeter is stable. Set the low filter with the following command.

VOLTage:AC:BANDwidth MIN

2. If you know the maximum level of the signal being measured, set the manual range to help speed up the measurement. The longer settling time for each low frequency measurement will significantly slow down the auto range.

We recommend that you set the range to manual.

3. The 34401A uses a DC blocking capacitor to block the AC RMS converter from measuring DC signals. This allows the multimeter to measure the AC component using the best range. When measuring sources with high output impedance, allow ample time for the DC blocking capacitor to stabilize. The settling time is not affected by the frequency of the AC signal, but is affected by any changes in the DC signal.

The Agilent 3458A has three methods for measuring AC RMS voltage; its synchronous sampling mode can measure signals as low as 1 Hz. To configure the multimeter for low-frequency measurements:

1. Select the synchronous sampling mode:
SETACV: SYNC

2. When you use the synchronous sampling mode, the input signal is DC coupled for the ACV and ACDCV functions. In the ACV function, the DC component is mathematically subtracted from the reading. This is an important consideration because the combined AC and DC voltage levels can cause an overload condition even if the AC voltage itself is not overloaded.

3. Selecting the appropriate range speeds up measurement because the auto-range feature can cause delays when you measure low-frequency signals.

4. To sample a waveform, the multimeter needs to determine the signal period. Use the ACBAND command to determine the pause value. If you do not use the ACBAND command, the multimeter may pause before the waveform repeats.

5. Synchronous sampling mode uses level triggering to synchronize the signal. However, noise on the input signal can cause false level triggers and inaccurate readings. It is important to select a level that provides a reliable trigger source. For example, avoid the peak of a sine wave because the signal changes slowly and noise can easily cause false triggers.

6. To get accurate readings, make sure your surroundings are electrically "quiet" and use shielded test leads. Enable level filtering, LFILTER ON, to reduce sensitivity to noise.

Configuring the 34401A can be done in the same way as the 34410A and 34411A. The 34401A
converts the RMS voltage using an analog circuit with DC blocking capacitors. It can measure signals as low as 3 Hz. For best results, select a low-frequency filter, use manual ranging, and verify that the various DC biases are stable. When you use a slow filter, you insert a 7-second delay to allow the multimeter to stabilize.

Tip 5 Selecting Sensors for DMM Temperature Measurements
There are four common sensors used for DMM temperature measurements: resistance temperature detectors
(RTDs), thermistors, IC temperature sensing devices, and thermocouples. Each has its own advantages and disadvantages.
Using thermistors for better sensitivity
Thermistors are made of semiconductor materials and offer high sensitivity, but they have a limited temperature range, typically -80°C to 150°C. The relationship between temperature and resistance for a thermistor is nonlinear, so the conversion algorithm is complex. Agilent multimeters use the standard Hart-
Steinhart approximation to provide an accurate conversion with a typical resolution of 0.08°C.

Using RTDs for better accuracy
Resistance temperature detectors (RTDs) offer a very accurate and highly linear relationship between resistance and temperature, with a measurable temperature range of approximately -200°C to 500°C. Modern multimeters, such as the Agilent 34410A, provide temperature measurements of IEC 751 standard RTDs with a sensitivity of 0.0385 Ω/°C.
IC temperature sensing devices produce a voltage that is linear with degrees Celsius
Many manufacturers offer probes that produce a voltage that is proportional to degrees Celsius and Fahrenheit. These probes typically use IC temperature sensing devices, such as the National Semiconductor LM135 series. These IC devices cover a temperature range of -50°C to +150°C. You can easily calculate the temperature from the probe output displayed by the multimeter. For example, 270 mV is equivalent to 27°C.

Thermocouples that provide extreme temperature measurements
Thermocouples measure an extremely wide temperature range of -210°C to 1100°C, and their rugged construction withstands the demands of harsh environments. Unlike other types of temperature sensing devices, thermocouples make relative temperature measurements, so they also require a reference junction for absolute temperature measurements. However, for most applications, adding an external reference junction is not practical. We recommend the Agilent 34970A data logger and the 34901A 20-channel multiplexer with built-in reference junction. The 34970A also has built-in temperature conversion algorithms for commonly used thermocouples.

Summary
To monitor one temperature, a thermistor and a multimeter like the 34410A are simple, low-cost solutions. To get an accurate temperature reading, use an RTD. When monitoring multiple temperatures or high temperatures, a dedicated data logger is the best choice.

Tip 6 Making Grouped Measurements with a Multimeter
Multimeters typically use a two-stage trigger system; two sets of trigger conditions must be met to get one reading. Figure 6 shows the two-stage trigger model used in the 34401A multimeter. Normally, the number of samples and the number of triggers are both set to 1, and one reading is taken when a trigger is received. The number of samples can also be increased, that is, N readings are taken when a trigger is received. If the number of samples remains at 1 and the number of triggers is increased to N, then each reading requires a trigger. In both cases, a trigger delay is inserted between readings. The

default trigger delay is configured by the multimeter to achieve measurement stability and varies depending on the range and function. The trigger delay can be set manually. It is important to note that this delay is implemented in software and will have a time-determined amount of variation. In addition, the measurement time will also vary, making it difficult to sample signals with fixed time intervals using this scheme. Figure 7 shows a series of measurements made using the trigger delay.

Figure 8 shows the second trigger model. This is the model used in the 34410A, 34411A, and 3458A. It allows the trigger delay and the time between samples to be set independently. Alternatively, use a sampling loop (n readings) to get readings faster and with minimal time variation. Most sampling loops are implemented in hardware, with minimal firmware
to ensure consistent timing.

The 34410A, 34411A, and 3458A can be configured to sample readings as quickly as possible, but a timer can also be used.
To configure a burst measurement, set the trigger delay to allow for stability after the trigger and before the first reading. Use a timer to set the precise time interval between readings. The 34410A and 34411A have front panel data logging capabilities to simplify configuration of burst measurements.

Tip 7 Peak Detection with a Multimeter
Multimeters are well suited for sampling low frequency signals using the DC function. Bandwidth is usually limited to 8 kHz or less. Traditionally, analog peak detection circuits are used to capture and hold the peak voltage until the A/D circuit can measure it. This technique provides high bandwidth and can also be used to capture spikes of very short duration. This technique is also used in multichannel systems where an analog-to-digital converter is connected to peak detectors on each channel. This common technique samples the signal very quickly, preserving its maximum and minimum values.

In many applications, the energy contained in the noise spikes displayed by the oscilloscope is relatively small. Noise is usually introduced by EMI and can mask the signal of interest - for example, a car engine generates a lot of EMI. Physical measurements such as temperature and fuel level sensors usually change very slowly. This can inject high frequency noise into the filters used and the slower A/D. Therefore, it is not necessary to sample the output of the filter with a high speed A/D.

For peak determination and measurement, a multimeter is a very suitable tool. The multimeter provides signal conditioning (gain, attenuation, and low-pass filtering) and an appropriate sampling rate (1 kSa/s to 50 kSa/s). Most multimeters have built-in math functions that can be used to determine maximum and minimum values. To get the highest reading rate, you may want to perform post-processing, as math functions may slow down the reading rate. Other ways to increase reading speed include selecting a small time slot and turning off autozero and display.

Characterizing signals and determining peak values ​​are common tasks with Agilent's new 34410A and 34411A multimeters with the peak detect feature. When you monitor a DC signal, you can use the secondary display to show the high peak, low peak, and peak-to-peak values. The peak detect feature always samples at 50 kSa/s regardless of the multimeter's time slot setting and does not require math operations. Figure 9 shows the normal peak reading update after each reading.

An alternative to taking a single reading is to set up a group of readings, which returns only one peak measurement in a group of DC readings, as shown in Figure 10.

A third solution is to change the multimeter's measurement slot time and take a long reading. This third method is shown in Figure 11. A longer measurement returns a peak measurement result.

The 34410A and 34411A peak detection samples the signal every 20 μs. The peak value is held until the next trigger. You can change the measurement slot time to hold the peak value longer. Each peak measurement provides peak-to-peak, high peak, and low peak values.

Tip 8 Get the most out of your multimeter with accessories

Easier to inspect
You often have to double the circuit board inspection: Watch the probe slip caused. 34401A, 34410A and Agilent 34133A precision work easier. This light, and equipped with Agilent patented spot probe can help absorb small slip punctures into the solder joint.

High-voltage and high-current detection
High-voltage probes allow you to safely measure high voltages with your multimeter. The Agilent 34136A high-voltage probe is used by the 34401A, 34410A, and 34411A and has a fixed input impedance mode (input resistance is 10 MΩ). This probe is a 1000:1 voltage divider that extends the voltmeter's measurement capabilities to 40 kV DC.

You can measure DC and low-frequency AC current (30 A, 15 A continuous) with the Agilent 34330A current shunt (shown above), which is a precision 0.001 Ω resistor in a plastic case sealed with epoxy. The shunt outputs 1 mV when it passes 1 A of current. This current can be measured using the binding posts on the shunt. Simply
attach the leads securely to the binding posts.

Do you spend time looking for the shunts or user manuals? If you keep them in the instrument's "backpack," you always know where they are. Agilent offers two sizes of nylon bags that fit on top of our popular multimeters. The 34162A is for shorter instruments, such as the 34410A and 34411A multimeters. The 34161A fits the 34401A and 34420A multimeters.

Want to make a four-wire ohm measurement?
Buy another probe.

If you want to make a four-wire ohm measurement, you need a second set of test leads. The 34138A test lead kit is suitable for the 34410A and 34411A. It includes some very sharp probes and a small grabber. The Agilent 34132A Deluxe Test Lead Kit includes 2 test leads, pluggable spring-loaded probe hooks, alligator clips, contact pins and nylon bag.

Build a neat terminal block to minimize offset errors
The Agilent 34171A digital multimeter input terminal connector block is a set of two connectors that provide a convenient and reliable method for wiring to all five input terminals. The terminals are made of a low-heat copper alloy to minimize the induced voltage of different metal connections. It is suitable for 34401A, 34410A and 34411A. To achieve the minimum thermal offset voltage, use bare copper wire of the same size for all connectors.