Solving Analog Integration Conundrum

These days, there are several options to solve the integration problems with analog sensors such as measurement or temperature sensors. This blog explains the several options for analog integration and the “expected” benefits.

Before we describe the options, let’s get a few things cleared up.  First, most controllers out there today do not understand analog at all: whenever a controller needs to record an analog value, an analog-to-digital converter is required.  On the other end of the equation is the actual sensor measuring the physical property, such as distance, temperature, pressure, inclination, etc.  This sensor, a transducer, converts the physical property into an analog signal.  These days with the advanced technologies and with the cost of microprocessors going down, it is hard to find a pure analog device.  This is because the piezo-electronics inside the sensor measures the true analog signal, but it is converted to a digital signal so that the microprocessor can synthesize it and convert it back to an analog signal.  You can read more about this in a previous blog of mine “How Do I Make My Analog Sensor Less Complex?

Now let’s review the options available:

  1. The classical approach: an analog to digital converter card is installed inside the control cabinet next to the controller or a PLC. This card offers 2, 4, or even 8 channels of conversion from analog to digital so that the controller can process this information. The analog data can be a current measurement such as 0-20mA or 4-20mA, voltage measurement such as 0-10V, +5- -5V etc., or a temperature measurement such as PT100, PT1000, Type J, Type K and so on.  Prior to networks or IO-Link, this was the only option available, so people did not realize the down-side of this implementation.  The three major downsides are as follows:
  • Long sensor cable runs are required from the sensor all the way to the cabinet, and this required careful termination to ensure proper grounding and shielding.
  • There are no diagnostics available with this approach: it is always a brute-force method to determine whether the cable or the actual transducer/sensor has the issue. This causes longer down-times to troubleshoot problems and leads to a higher cost to maintain the architecture.
  • Every time a sensor needs to be replaced, the right tools have to be found (programming tools or a teaching sequence manual) to calibrate the new sensor before replacement. Again, this just added to the cost of downtime.
  1. The network approach — As networks or fieldbuses gained popularity, the network-based analog modules emerged. The long cable runs became short double-ended pre-wired connectors, significantly reducing the wiring cost. But this solution added the cost of network node and an additional power drop.  This approach did not solve the diagnostic problems (b) or the replacement problem above (c ). The cost of the network analog module was comparable to the analog card, so there was effectively no savings for end users in that area.  As the number of power drops increase, in most cases, the power supply becomes bigger or more power supplies are required for the application.
  2. The IO-Link sensor approach is a great approach to completely eliminate the analog hassle altogether. As I mentioned earlier, since the sensor already has a microprocessor that converts the signal to digital form for synthesis and signal stabilization, why not use that same digital data over a smarter communication to completely get rid of analog? In a nutshell, the data coming out of the sensor is no longer an analog value; instead it is a digital value of the actual result. So, now the controller can directly get the data in engineering units such as psi, bar, Celsius, Fahrenheit, meters, millimeters, and so on. NO MORE SCALING of data in the controller is necessary, there are no more worries of resolution, and best of all enhanced diagnostics are available with the sensor now. So, the sensor can alert the controller through IO-Link event data if it requires maintenance or if it is going out of commission soon.  With this approach, the analog conversion card is replaced by the IO-Link gateway module which comes in 4-channels or 8 channels.

Just to recap about the IO-Link sensor:

  1. IO-Link eliminated the analog cable hassle
  2. IO-Link eliminated the resolution and scaling issue
  3. IO-Link added enhanced diagnostics so that the end users can perform predictive maintenance instead of preventative maintenance.
  4. The IO-Link gateway modules offers configuration and parameter server functionality that allows storing the sensor configuration data either at the IO-Link master port or in the controller so that when it is a time to replace the sensor, all that is required is finding the sensor with the same part number and plugging it in the same port — and the job is done! No more calibration required. Of course, don’t forget to turn on this functionality on the IO-Link master port.

Well, this raises two questions:

  1. Where do I find IO-Link capable sensors? The answer is simple: the IO-Link consortium (www.IO-Link.com) has over 120 member companies that develop IO-Link devices. It is very likely that you will find the sensor in the IO-Link version. Want to use your existing sensor?  Balluff offers some innovative solutions that will allow you bring your analog sensor over to IO-Link.
  2. What is a cost adder for this approach? Well, IO-Link does a lot more than just eliminate your analog hassle. To find out more please visit my earlier blog “Is IO-Link only for Simplifying Sensor Integration?

Balluff offers a broad portfolio of IO-Link that includes sensors, RFID, SmartLights, Valve connectors, I/O hubs, and the gateway modules for all the popular fieldbuses and networks. Learn more at www.balluff.com

This entry was posted in All posts, Industrial Networking, IO-Link and tagged , , , , , . Bookmark the permalink.

Leave a Reply