Mike Geis,director of business development at AMETEK Automation & Process Technologies Sensor Technology Division, which used to be known as Patriot Sensors back in the day, writes:
"I think you have a point and believe senor manufacturers are ready to develop new products. However, consider the following:
Have you ever notice how many different level companies there are...including Walchem and Garnet? Why do you think this is the case?
--Have you ever noticed how many different types of tanks there are at processing plants? How many standardized vessels are there?
--What is the accuracy, repeatability, linearity and resolution specifications for industrial and process control instrumentation? What environmental and power requirements have to be addressed?
--Why do you think wireless instruments have had a difficult time penetrating the market? How well do they meet the end user's needs?
I doubt if many of your readers know very much about magnetostrictive level sensors, and the application you noted is for a commodity product not a configured sensor. The end-user needs to make demands, before many manufacturers will respond. We need to hear the voice of the customer. "Build it and they will come" is usually a dream, and not a reality."
Well, Mike, one of the points I was trying to make is that manufacturers have resisted the idea that many applications in industrial process automation can easily be handled by "commodity" products, and configured sensors are overkill. This, of course, comes from the need, as a Honeywell staffer rather vigorously at the top of his lungs pointed out to me, to maintain margins.
Over 15 years ago, I helped to design a level transmitter that could autoconfigure for any size or shape of tank, and which had all the equations for any conceivable tank shape built in, and could run them in real time, without the use of strapping tables. That same transmitter could be built today for a hard cost of less than $20, and sold at a 75% Gross Margin, for use with any sensor.
If by a configured product, you mean choosing one from column A, and one from column B, anybody who can order at a Chinese restaurant can configure a sensor. Omega Engineering has proven that there is a huge market for pick, pack and ship sensors in the process industries.
Manufacturers who start designing for the 80/20 rule are going to flourish, while manufacturers who insist that mystic passes are necessary for most level applications are not going to grow. As Dave Johnson, president of Yokogawa Corporation of America told me on Tuesday, you have to provide the services and infrastructure, and then the customers will buy. Dave is a very smart guy.
As far as wireless is concerned, please see my article in the upcoming September CONTROL, which will be mailing tomorrow. If you are a subscriber, you should get it early next week. If you aren't a subscriber, you can read the article on the web at www.ControlGlobal.com shortly. (I don't think it is posted yet...). The bottom line is that wireless sensors and especially wireless networking protocols aren't ready for prime time. They are coming, though, and you vendors will need to be ready with the kind of sensors that "ubiquitous data" will require.
Walt
