Digital panel meter design -

Digital panel meter design

Panel meters are commonly mounted in industrial control panels to measure various physical quantities like temperature, pressure, voltage, current, power, and speed.

These meters take an electrical signal that is proportional to the physical quantity being measured and display the value either on a mechanical dial (in the case of analog panel meters) or a display (in case of digital panel meters). Panel meters come in various standard sizes as listed in the Table 1 below.

Table 1: Size standard for panel meters

Analog panel meters (Figure 1a below ) display the measured quantity on a dial using a needle. These analog meters suffer from various drawbacks like non-linearity and parallax errors. They are also hard to calibrate and have a short life due to aging and mechanical wear and tear. Also, they provide fewer features to meet today’s changing market needs.

To overcome these disadvantages, analog panel meters are being widely replaced by digital panel meters (Figure 1b below). Unlike analog panel meters, digital panel meters (DPM) are linear, more accurate, and easier to read as they are equipped with LED or LCD displays.

They also have memory to store various configuration and user parameters. In addition, they can perform functions beyond making simple measurements. For example, they can monitor system health and generate alarms when a particular input is outside a specified range.

Figure 1: a) An analog voltmeter and b) a digital voltmeter

There are various implementations available for digital panel meters. Each of these implementations offers specific advantages and some disadvantages, as discussed below.

ASIC and ASSP based solutions
Application specific ICs (ASICs) and Application Specific Standard Products (ASSPs)for digital panel meters usually integrate the ADC and the display driver in a single device. Figure 2 below shows the basic block diagram of a digital panel meter implementation using an ASIC.

Figure 2: ASIC-based digital panel meter
In the above block diagram, the signal conditioning circuit varies based on the range and physical quantity being measured. Generally these ASICs support a couple of standard input voltage ranges; for example, from 0 to 199.9 mV or 0 to 1.999V.

For inputs other than those supported, external signal conditioning circuits have to be used to scale the input signal to within the range supported by the device. For example, if an input is 0-19.9mV, the signal conditioning circuit would be an amplifier with a gain of 10. For an AC input, a precision rectifier or an RMS converter would be used. And so on.

The advantage of an ASIC-based approach is ease of design. These designs can be made with discrete components and do not need any microcontroller or software programming. However, with these advantages come certain disadvantages as well:

* ASICs generally require capacitors with low dielectric loss that add significant cost to a design.

* External signal conditioning increases the number of components and cost.

* These devices support a fixed number of displays and counts; for example, a display might support 3 ½ digits with 1999 counts or 4 ½ digits with 19999 counts.

Displaying an input with a maximum range that is less than the full scale requires using only part of the input range, thus affecting the maximum resolution displayed.

* They require manual calibration using multi-turn potentiometers which increases cost and manufacturing time.

* Because different signal conditioning circuits are required for different inputs, separate boards have to be designed for different input ranges. This increases inventory cost.

* Implementing communication protocols like MODBUS require special ASICS which are expensive.

MCU-based approach
The flexibility of an MCU-based overcomes many of the disadvantages that are present in an ASIC-based solution. Figure 3 below shows the block diagram of an MCU-based digital panel meter implementation.

Clickon image to enlarge.
Figure 3: MCU (with on-chip ADC and PWMs) based digital panel meter

As with an ASIC-based approach, the analog front-end converts / scales the input signal to within the ADC’s range. The microcontroller measures the signal using the ADC (which is integrated in many microcontrollers), applies the desired calibration coefficients and scaling factor, and then outputs the result on the display.

As the MCU is used for post-processing the ADC results and drive the display, this gives more control to developers to apply any scale factor between the input and display. For example, the input to the ADC could be 0 to 2V, and the display could be from 0 to 750 RPM.

Another advantage of an MCU-based approach is the implementation of digital calibration – two-point usually – using a highly accurate signal source and a communication interface (UART, I2C, SPI etc).

A known input at around 10% of the full scale is applied to the meter and the value of the known input is transmitted to the MCU using the communication interface. Another known input at 90% of the full scale value is applied to the instrument, and the value is transmitted to the MCU.

The MCU measures the ADC output for these two known values and calculates the offset and gain errors of the system using the actual input vs. the corresponding ADC result. The MCU then stores these coefficients in non-volatile memory and uses them when making actual measurements.

The advantage of digital calibration is that multiple instruments can be calibrated at the same time. In addition, the whole process may be automated, thus reducing manufacturing time.

Communication protocols like MODBUS can be easily implemented using on-chip UART and external RS-485 transducers. Also, using an internal PWMs, external filter, and V-I converter, a 4 to 20 mA output may be implemented.

Summarizing, the advantages of an MCU-based approach include scaling of the input vs. display may be performed in the digital domain, thus simplifying the analog front end requirement; digital calibration reduces manufacturing time; and a communication protocol like MODBUS may be implemented.

Its disadvantages include that, however simplified it may be, an analog front-end is still required, thus increasing the size and cost. In addition, experience in microcontroller programming is required.

Mixed Signal System-on-Chip approach
Mixed signal devices have both analog and digital peripherals on-chip. Apart from offering all the advantages of an MCU-based approach, these devices can integrate the analog front-end into the same device to further reduce the size and cost of the system. Figure 4 below shows the block diagram of a digital panel meter using a mixed signal SoC processor.

Clickon image to enlarge.
Figure 4: PSoC 1-based digital panel meter solution
These devices have on-chip programmable gain amplifiers (PGA) that can be used for signal conditioning. Multiple input ranges can be supported by changing the gain of the amplifier in firmware.

These devices also have internal low pass and band pass filters that can improve signal conditioning, an internal accurate band-gap reference, and the ability to provide buffered Analog Ground (AGND) to bias bipolar and AC input signals. Different types of ADCs like incremental and Delta Sigma are available with configurable resolution.

Depending on the accuracy and speed requirements, the resolution may be selected. True RMS calculations may be performed using the on-chip MAC (Multiply Accumulate) registers which perform efficient hardware multiplication and accumulation.

Since SoC devices contain the signal conditioning circuit on chip, dynamic offset compensation called correlated double sampling may be performed. In this method, the input of the signal conditioning circuit is shorted to AGND and the output of the ADC is measured.

This ADC result represents the total offset error of the system, including the signal conditioning circuit and the ADC. This value is saved temporarily. The input signal is now connected to the signal conditioning circuit, and the ADC result is measured. The saved offset is subtracted to get the offset compensated value.

The advantage of this offset compensation over the offset compensation done in two-point calibration is that this method takes care of the offset drift due to temperature changes as well.

Also, as the offset calibration can be performed using correlated double sampling, the digital calibration can be performed using a single point compared to the two points required for an MC-based approach. This further reduces the calibration time.

SoC devices also have on-chip comparators that can implement Hysteresis comparators, window comparators, or Zero Crossing detectors. These comparators may be used in implementing frequency meters where the input signal is fed to a comparator and the digital output of the comparator is fed to the capture input of a timer to measure the period of the input signal.

Window comparators may be used to generate hardware alarm signals when the input signal is outside a defined range. Integrated on-chip UARTs can implement communication protocols like MODBUS, and on-chip PWMs and DACs can be used to implement 4 to 20mA outputs.

Some SoCs also integrate capacitive buttons and sliders. These can be used to implement touch-based keypads which do not have the disadvantage of wear and tear associated with mechanical buttons. Apart from increasing reliability, this also reduces the cost of the user interface. Summarizing, the advantages of a mixed signal SoC controller-based approach are:

1 – Lowest possible external component count due to integration of analog front end;
2 – Programmable Gain Amplifiers and filters on chip;
3 – True RMS calculation using hardware MAC;
4 – Frequency measurement using on-chip comparators and timers;
5 – Dynamic offset compensation using correlated double sampling;Single point digital calibration;
6 – 4 to 20mA output; support for MODBUS communication protocol;
7 – Capacitive touch interface capabilites; and
8 – Reduced inventory since the same hardware can support multiple input ranges.

Similar to the MCU-based approach, the disadvantage of an SoC-based approach is the need for knowledge in microcontroller system design and programming.

In price-sensitive markets where end-customers demand more features, batter accuracy, and flexible configurability, a mixed signal SoC device based approach is the best fit. SoCs enable multiple functions to be integrated onto a single chip and reduce system footprint size. Also, they help in reducing inventory as only a single device is needed to support different configurations and the fact that SoCs do not need external signal conditioning and peripherals.

Ganesh Raaja received his degree in Electronics and Communication Engineering from Motilal Nehru Government Polytechnic, Pondicherry. His expertise lies in developing with analog circuits, embedded systems, designing PCBs, and working with Assembly and C.

Sachin Gupta is working as Product Marketing Engineer with Cypress Semiconductor. He holds B.Tech in Electronics and Communication from Guru Gobind Singh Indraprastha University, Delhi. He has several years of experience in mixed signal applications development.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.