Wireless technology offers many advantages over traditional wired systems, helping in the creation of a new generation of products for remote monitoring applications. Brett James, design engineer at Mantracourt Electronics, explains some of the issues to be considered when designing low power telemetry for instrumentation
Advances in wireless technology have enabled instrumentation manufacturers to expand their range, creating a new generation of products for remote monitoring. Whether it is measuring torque on a rotating system or measuring loads in a hazardous environment such as a construction site, wireless technology has many advantages over a more traditional wired system.
Designing low power systems is always a difficult compromise between the required or desired battery life and the frequency of readings or the reactiveness of the device. With some applications demanding small battery volumes, the choice of battery also becomes critical.
The basic power-saving model is to get your device into a very low current state in between times when it has to be awake to take readings or transmit data. From waking up it is imperative that each operation is conducted as fast as possible, using the minimum amount of current, which may mean powering different parts of the circuitry during the cycle.
In some telemetry modules that wake periodically to take a reading from a strain gauge, for example, the sequence of events is critical. We would first wake the MCU and power up the analogue circuitry. While the minimum number of readings are averaged from the A/D we would start to power up the radio module.
As soon as the readings have been taken, the analogue circuitry is powered down and the data transmitted. The radio is then powered down and the entire device re-enters a low current sleep mode until the next wake-up. The wake, take readings and transmit sequence can take as little as 20mS, so even with transmission rates as high as 25Hz we can still save some power between transmissions. Once the transmission rate drops to one per second or below we can start to see very long battery life from reasonably small batteries.
The choice of battery can be determined by availability. In the Mantracourt
T24 telemetry range we designed around a 3V supply so that a pair of D, AA or AAA batteries can be used to power the modules. Many customers feel happier that batteries can be replaced in an emergency by sourcing from the local garage.
It is very important that the devices can operate down to low supply voltages. The lower they can operate, the more capacity of the battery can be utilised. For example, a pair of D cell batteries may be specified by the manufacturer as having a capacity of 10Ah, but this is usually down to a voltage such as 0.9V. The circuitry usually cannot operate that low so given that it can only operate down to 1.1V this may mean that we can only use 8Ah capacity of the battery. Battery self-discharge also determines the choice because sometimes the batteries will discharge themselves long before the attached electronics.
Improvements in alkaline battery technology means that shelf life is very good – so these batteries can now be used instead of non-rechargeable lithium batteries when very long life (five years, for example) is required.
For rechargeable batteries, lithium polymer has proved to be a very reliable and efficient battery chemistry. The 3.7V output is ideal for telemetry modules and the associated charging and regulating modules means that batteries can be recharged in situ.
An area that must be sacrificed for longer battery life is the use of LEDs to inform users of operating modes and errors. A better approach is the use of pulsed LEDs and sequenced flashes. By utilising the techniques mentioned above it is possible to provide transmissions every 30 seconds from a load cell sensor powered from a small lithium button cell, for months.
The low power requirements of radio modules mean energy-harvesting solutions such as solar cells can also be used. This means the products can be used without having to worry about changing batteries. This is an area Mantracourt are currently researching for development.
Antenna and radio frequency choice
The antenna choice will mostly be determined by the application. As discussed above, wireless instrumentation tends to operate in harsh environments, so a chip antenna mounted on the PCB provides a solution that is both low cost and robust.
A similar alternative is a PCB antenna with omni-directional characteristics. Of course, an external antenna can give additional range but may not be practical in all circumstances. At Mantracourt we chose 2.4GHz for our latest radio modules after previous experience of 915MHz and 868MHz equipment. With different geographical regions requiring different frequencies, we previously not only had to manufacture and stock two variants of all products but also had to make decisions for customers as to which type to purchase, as not all retained the product themselves.
Coupled with the limitation of duty cycles for 868MHz band, we decided to investigate the 2.4GHz licence exempt band as a solution which, in consultation with our customers, proved to be a good solution. Our distribution partners had less stock to carry with only one version, and an easier decision for customers who may have not known exactly where the devices would operate in the world.
We did have concerns for adopting the 2.4Ghz frequency band including reduced range and the crowding of this band with other radio equipment such as BlueTooth and Wi-Fi, not to mention microwave ovens. However, the radio techniques employed by our module meant that we had no issues operating with any of the other equipment. The reduction in range compared to 868MHz and 915MHz was apparent, but the line of sight range was more than acceptable and indoor range could be boosted with repeaters.
We state a range of 100m in an open field site for our chip antenna modules and this has been tested to 200m in a true open field site. About 20m indoors through one wall can also be achieved. Using the larger antennas as in our boxed modules results in a specified open field range of 200m where we were able to achieve 400m in a true open field site. Indoors range is about 40m through a single wall and less if passing between two walls.
Transmitting the data
From the original design concept of the radio module it was decided that a general purpose module would be more useful than one that specialised in a particular area – so we designed our own protocol. We needed the ability to transmit up to 1000 packets per second and run for up to five years on a pair of AA batteries (at usable delivery rates).
We also wanted to avoid the complexity of requiring and configuring coordinators. For speed, flexibility, low power and complexity, we decided against mesh networking.
This has proved to be a good decision as a very high percentage of applications can be solved using standard modules along with smart repeaters in the few cases that require them.
This approach has proved to be simple to configure and working systems can be added to at later dates with minimal fuss and usually no change to existing modules.
A beneficial technology
Whether you are considering developing, or simply buying-in wireless instrumentation, hopefully this article has provided an insight into the issues surrounding wireless instrumentation and a level of expectation as to what is possible at this time.
This technology is helping engineers worldwide to monitor loads, torques and strains in difficult to access locations and on moving systems.
T: 01395 232020