Data Acquisition Measurement Techniques

In order to make the best measurements with a data acquisition, DAQ system and equipment, a number of techniques and technologies are used to give the best performance.

Data Acquisition Sensors Includes:
Thermocouple temperature sensor Flow meters

More about data acquisition: Understanding data acquisition Data logging systems DAQ systems DAQ cards & modules DAQ measurements

Data acquisition utilises samples that have been collected by various sensors and returned to the controlling computer or controller of whatever format.

These samples are generally analogue in format and they are returned to the controller on a periodic basis so that the various parameters and measurements that need to be made can be collected.

In this way the controller or computer can see how the various measurements change over time the effect changes in one parameter may reflect into another.

Accordingly a variety of processes, techniques and technologies need to be used to provide the accurate data in a digital format at the required rate.

Basic DAQ measurement system

Whatever the format for the DAQ system, whether it is a data logging system, a data acquisition device with a single node, a compact DAQ rack system or some form of larger system, the basics techniques and technologies do not change.

Concept of a data acquisition system, DAQ showing how the phenomenon is measured or sensed, conditioned and converted from an analogue to a digital format and then processed by a computer. — Concept of a data acquisition system, DAQ
*Note that for a data acquisition system many inputs are likely to be received into the computer simultaneously*

The overall process involves first taking the measurement with some form of sensor. This may be remote or it may be local.

Once this has been done the signal is conditioned. The sensors or transducers need to have their output converted into a format that can be accepted by the next stage in the process which is an analogue to digital converter. This signal conditioning can take many forms.

Once the transducer output has been suitably conditioned it is converted from its analogue format into a digital format. This is accomplished using an analogue to digital converter, A2D. There are several choices to be made at this stage in terms of the resolution of the A2D and also the frequency of the samples, i.e. sample rate

Once the signal is in the digital format, i.e. in the digital domain, the data can be processed by the computer or controller and then stored and presented as required.

This is a rather simplified description of what happens, but it serves to provide a summary of the overall flow of the measurements in the DAQ system.

More detailed descriptions of the different functional blocks are given in the sections below.

DAQ sensors of transducers

Sensors are the interface between the physical world and the electrical or measurement world. It means that parameters like temperature, force, and movement are converted to voltage or current signals that can be used as inputs to an analogue to digital converter in the data acquisition system.

Common sensors used for data acquisition and data logging measurements include items such as thermocouples, and thermistors to measure temperature, accelerometers to measure movement, and strain gauges to measure force.

There are also flow meters, pressure sensors and others that are widely used in industrial processes.

When selecting a sensor for a particular DAQ measurement system, it is important to consider factors such as the accuracy of the sensor and any the signal conditioning required to produce an accurate and readable signal.

Some of the more commonly available transducers are included in the table below:

List of Some of the More Popular Data Acquisition Transducers
Thermocouple	Strain Gauge
Flow meter

It should also not be forgotten that triggers and logic states as well as voltages themselves can be measured and these are often required and can be fed directly into the DAQ, device, card or module.

Signal conditioning

The electrical signals that come from the sensors often need some processing or 'conditioning' before they can be converted from their analogue format into a digital one. This can be key to the performance of the overall data acquisition measurements and their accuracy and validity.

As a result there is normally some circuitry required to enable the signal from the sensor to be in the right format for the analogue to digital converter. This circuitry is generally referred to as 'Signal Conditioning" and this may take several forms dependent upon the type of sensor, the environment and various other factors.

The most common types of signal conditioning include amplifiers, attenuators, filtering and the use of a Wheatstone bridge.

It is also often necessary to linearise the input and add compensation in other ways as well. These capabilities can all be added to the analogue signal conditioning.

This signal conditioning is placed ahead of the analogue to digital conversion, and it is often included in the DAQ card or module, or directly within the DAQ device circuitry if a single unit data acquisition unit is used.

In most cases the signal conditioning circuitry is contained within a data acquisition device, module or card, but signal conditioning may also be included as part of the transducer itself. For example, load cells contain the bridge completion, calibration circuitry, and amplification. Many MEM (micro-electro-mechanical) sensors also contain signal conditioning. By including it as part of the transducer, the overall transducer is able to provide a signal ready for presentation to any further limited signal conditioning ready for passing to the A2D.

Digitisation

One of the most important elements of a data acquisition system and the resultant measurements is the digitisation or analogue to digital conversion.

It is here that a circuit called an analogue to digital converter is used as the basis for the circuit. It takes in the analogue signal and using a set of electronic circuitry, it converts the analogue signal into a digital format. This is in a format that can be presented to the computer circuitry for storage and processing as required.

There are several aspects to the operation of the digitisation or analogue to digital conversion process of the data acquisition or data logging system.

Sample rate: One of the key elements of the digitisation process is the rate at which samples are taken. Samples are generally taken at a set repetition rate and this is known as the sample rate. Sometimes, samples may also be taken when a trigger signal is received.
Resolution: Any sample that is taken is represented in a digital format. The number of bits of the output data is known as the resolution.

Basic data acquisition measurements

Whatever the application used for the data acquisition system, a huge number of measurements can be made depending upon the sensors being used, but the basic measurements that are made in any DAQ card or DAQ module can be simplified down to some of the basic elements:

Voltage
Digital signals
Frequency or time interval

The sensors that are used in data acquisition card measurements often return values of voltage in particular that can then be converted to the values of displacement, temperature, or whatever is being measured.

These data acquisition card or module measurements that may be termed primary measurements will be looked at in turn, and then their applications in making other measurements in data acquisition systems will be covered.

Voltage

Voltage is one of the most commonly made measurements in any data acquisition system. Not only is it used as a data acquisition measurement in its own right, but it is also used for measurements with many other sensors including thermocouples, strain gauges, gas concentration probes and many more.

Voltages that are presented to a voltage measuring device is essentially an analogue quantity. However as data acquisition uses computer techniques, the analogue voltage needs to be converted into a digital format. The voltage measurement will use a form of digital to analogue converter to convert the analogue voltage into a digital representation of its value. The greater the number of bits, the greater the resolution of the measurement.

Typically the voltages that are required to be measured in data acquisition systems range from a few millivolts up to a few volts. Voltages that are much higher than this need to be reduced before they are measured.

Some sensors used for data acquisition require voltage measurements to be made with a very high impedance device because their source impedance is high. Any current drawn from the sensor will apply a load which will distort the reading. The types of sensor that fall into this category are the glass electrodes types that are used for measuring pH or gas concentration. Special high impedance voltmeters are available for these measurements.

Digital signals

Often in any data acquisition system it is necessary to measure the status of an indicator. This will typically be one of two states. Under these circumstances it is not necessary to measure the actual voltage, although this is one way to do it, but rather measure whether the line is above or below a given voltage level. This can be done with a simple comparator.

In many data acquisition applications the particular sensor may not be a logic circuit, but a simple switch. To generate a logic signal a voltage (often 5 V so that it is TTL compatible) is applied through a current limiting resistor and the resulting voltage or absence of it is measured. It is worth noting that the circuit with the switch should be configured so that when the switch is open circuit, the sense line is not left open circuit otherwise stray pick-up may be a problem. This can be done simply by providing the voltage through a suitable resistance and then having the switch taken to ground. Thus when the switch is closed the sense line will be shorted to ground and when it is open it will see the supply voltage through the current limiting resistor.

When a the number of instances a digital signal changes state in a given time, or the interval between them needs to be measured, then frequency and time interval measurement techniques are needed.

Frequency and time interval measurements

In many data acquisition applications, frequency or time interval measurements are required. The two measurements are very similar. They are simply the reciprocal of each other. Time interval is 1 / f.

Data acquisition modules that can count time intervals or measure frequency are widely available. The time interval measurement measures the time taken between a logic state crossing from one state to another. A frequency measurement counts the number of state changes in a given time.. Some simple switching in the circuitry enables both time interval and frequency to be measured.

One of the key elements with any frequency or time interval measurement is to ensure that the counter timer sees all the required pulses and no more. This can sometimes be more difficult than it may appear at first sight. Long lines can alter the shape of the pulse, slowing it down so that the counter timer does not trigger. Alternatively for frequency measurements, stray pick-up can introduce additional pulses. To overcome these problems, lines should be kept short and the line impedances should be low to reduce the effect of pick-up. By using these and other basic precautions counts should be accurate and repeatable.

Data acquisition derived measurements

Using the basic voltage, digital and frequency / timer measurements, it is possible to perform an enormous variety of other measurements in a data acquisition system. Many sensors, for example, provide an analogue voltage proportional to the quantity they are measuring. This means that the voltage measurement in a data acquisition system is one of the most important techniques to master.

Current measurements Current is a measurement that is commonly made within the wider range of electronics test. It is also widely used for data acquisition measurement applications as well. Although there may not be many applications where it is necessary to monitor the current being consumed by a unit, current is often used in data acquisition applications to transmit signals in noisy environments. The reason for this is that current measurements are generally much lower impedance and they are less affected by pick-up of any electrical noise in the environment.

In order to sense or measure the current, the signal is converted into a voltage by placing a small resistor in the circuit and then measuring the voltage across it. This resistor should be a high precision type as its accuracy will have a direct bearing on the accuracy of the measurement. The actual tolerance or precision requirement for the resistor is determined by the level of accuracy needed for the overall measurement.

The values of current that are normally used may range up to 20 milliamps, and the resistors used may be of the order of 100 ohms, but these figures will depend upon the given application. Tolerances on the resistors used are often as tight as 0.01%.
Temperature measurement with thermocouples Temperature measurement is one that is often needed in data acquisition systems. Thermocouples consist of a junction of two metals that produce a small voltage dependent upon the temperature difference between the thermocouple junction itself and the point where the thermocouple wires terminate. This is known as the cold junction. The way in which this occurs is beyond this page on the website, but it is worth noting that the voltages produced are small. As a result care is required to ensure that there is no stray pick-up from the environment and that there are no DC offsets on the system. Either of these could produce significantly erroneous results.
Resistance measurements Another measurement that is widely used in data acquisition systems is the resistance measurement. This again uses a voltage as the basis of the measurement. Essentially it is done by using a current source with an accurately defined level of current, and then measuring the resultant voltage across the resistance under test. This technique can then be used not just for determining the resistance of particular elements but as the basic for other parameters that are required within data acquisition systems.

To prevent errors when the values of resistance are small, the leads connecting the element to be measured to the system must have a much lower level of resistance. Accordingly these need to be kept short and sufficiently low resistance.
Strain gauges and strain measurement Strain measurement is used in many data acquisition applications. These data acquisition systems may be required for geographical applications as well as monitoring strain on vessels in engineering manufacturing applications.

Strain measurement can be considered as a special case of resistance measurement. As the changes in value are very small, a Wheatstone bridge arrangement is normally used, and in view of the variations that may be present, the changes in strain are measured as deviations from initial values. A voltage is applied to the bridge circuit and the voltage across the required element is measured, being converted from a voltage to a digital value in the normal way. Changes are generally small and therefore reasonably high levels of resolution are needed.
As a result of the fact that only the changes are measured it is necessary to determine the initial values by some other means. Additionally as the values obtained from the strain gauge are dependent upon the supplied voltage, this needs to be measured and accurately maintained, especially if the system is to be used over a period of time where voltages, etc may drift.

The measurement technology and techniques for data acquisition have been refined over many years to enable highly accurate data to be collected. The overall data acquisition measurement system involves several stages and several techniques. Understanding the techniques used ensures that the optimum system can be developed and the best accuracy obtained for all the data collected.