An end-user view on wireless instrumentation
What do end users want from wireless, and a wireless standard? ExxonMobil's instrumentation team leader weighs in on the ISA100.11a standard.
Those who have been following developments of the ISA100 standard should know the name Patrick Schweitzer. He is co-chair for ISA100 and the user member of the ISA100 Wireless Compliance Institute board. If that isn't enough, his day job is instrumentation team leader for ExxonMobil. As someone who will have to work with the outcome of the standard development efforts, he keeps his company objectives in mind. We sat down for a talk at the ARC Advisory Group summit in February.
CE: ISA100.11a is complete. Are you happy with it, and what is the next step?
"In the sense that we knew it would work, 11a has worked. Now it's a matter of going to the next level and finding out what are the other things we can do with it. It's nice to have a network. It's nice to say,‘I'm going to save money on wires.' That gets you the infrastructure. The next thing you want to put in is the innovation part. The base standard will get us into the applications so we can start thinking about the next thing we can do with it. We haven't even touched it yet. Sure, I can put a pressure sensor in wirelessly and get information we didn't before. We've done it. But what else can we do? What else can we get? Can I go over to this area and get some energy numbers that I didn't have before? Can I enhance my margins because I'm looking at something else? You put the infrastructureout, and then you start building the rest of it."
CE: Can one wireless device standard or network do everything?
"You need a single logical network at the device level that everything is built upon. You don't want multiple networks. They're just too hard to maintain and way too expensive to install and deploy. End users know that. We went through it in the wired world with all the field networks. I think we counted 18 wired networks at one point. It's very expensive.
"You want your network to support communication, but you want something you can innovate off of. It's hard for me to say what that's going to be yet, because I don't know. I think the application profiles, those optional behaviors that we've written into the standard, are going to get us to the next level. We'll start with an installed base, and now where do you go from here? Is there something innovative out there that's going to allow me to connect with another device's network? ZigBee's one. They want to be part of the ISA100 process. Let's say I have an 11a network, and ZigBee's got a new application that I want to get to. Is there a way of using that optional behavior to allow them to be part of my network without changing the whole network? That's what I'm hoping this is all going towards. I have a single logical network but with the greatest diversity of applications that I can get. I think we're headed to that kind of flexibility.
"Let's say we discover that new device that truly innovative, but uses a different protocol. We know we're going to make a billion dollars using this sensor, but now it's disjoint from everything else. What do we do? Set up a new network? I have to build a method into my original standard where I can go to the maker and say,‘I understand you've built this thing, but we need you to insert this little application.' So now I can bring that into my network and I don't have to change anything."
CE: How far ahead are you thinking?
"I'm an oil and gas guy, and we take a long view on everything. If we're going to set up a network today, I'm looking for some stability so I don't have to reconsider this every five years. This looks like a great idea today, but what's it going to look like in another 10 or 20 years? Let's go back to the analog control days. Who would have thought back then that today we'd have plant services that have everything integrated in one system? But the 4-20 mA standard that was developed back then has spanned 30 years and more. Wireless needs to be as durable as that standard."
CE: How have you been incorporating end-user requirements?
"I think one of the things that has been misunderstood is the idea of meshing. For a while, everybody was saying that everything needed to mesh. But if you start talking to some users and seeing some presentations, you begin to realize that there are limitations to mesh networks. It's not the answer to all. So when the committee asked the users,‘What do you want?' the answer was, ‘We want to be able to engineer our system.' There will be times when mesh will get us what we need and it will be very robust, but there will be times when we may need to go point to point, or set boundaries around our network. Some of that flexibility allowed us to engineer the unique solutions we needed for our sites."
CE: So, when you wear your ExxonMobil hat, you want all the flexibility you can get, but within a standard.
"You want to keep it as open as possible, which is the reason we have consistently supported the ISA100 standard. It was a place where we could propose our stuff and be vetted in the open market. When we started working on wireless, WirelessHART didn't exist. We went to the open consensus-based standard, and that was ISA100."
CE: What about WirelessHART? Where do you see it?
" WirelessHART stuff works, there's no doubt about it. We wanted them to come to the table. We were working on ISA100 and we knew it was not going away. So now we've come to the point where we now have two competing standards-or at least two, I hear there are others-that are addressing the same industrial space. So, now what do we do? ExxonMobil is still supporting ISA100, but really what we'd like is a single standard that everybody can deal with. We're pushing for convergence which we think will be beneficial to users as well as suppliers. I'm starting to hear the rest of the user community going the same way. Having to deal with two standards is very confusing. Users don't know which one is right. If producers have to support both, what's that going to do? Eventually the additional costs will filter down into the price of the product and some of the gains we hoped to see from wireless will go out the window.
CE: So there is a solution?
"The end goal of a standard is to provide a basis for everybody to build off of so they can achieve the goals they've set for themselves. I think the solution is out there. The users have to get together and tell the suppliers that if they can come to the table on this one, it will benefit everybody. I don't know how we're going to do that. I'm trying to lead the bandwagon on this, and I'm not going to give up. I know the solution has to be there, I just have to keep pushing to find it."
Check out Control Engineering's wireless resource page .
Read recent posts on wireless instrumentation at our Pillar to Post blog.
-Peter Welander, process industries editor, firstname.lastname@example.org
Control Engineering Process Instrumentation & Sensors Monthly eNewsletter
Register here to select your choice of free eNewsletters .