By Dick Caro
The ISA SP-100 standards committee met for three days following ISA’s Wireless Summit held in Vancouver, BC. Several major goals for the standard to be named ISA100 have been met, with the prospect of the first SP-100 committee ballot scheduled to be following this year’s ISA Conference and Expo to be held in Houston, Texas, October 2-4. The committee is planning to hold a demonstration of ISA100 at this Expo to stress the readiness of the networking technology, and the cooperation among the potential suppliers, most of which are participating in the demo.
The committee has now agreed upon all of the major sections of the Principles of Operation, an overview of the protocol describing how each layer will accomplish its goals. A more detailed version of this document is already completed and is called the Functional Description, which will serve as the framework for the ISA100 standard itself. Each of the layers now appears to be completely defined using as much standard technology as possible.
At the lowest layer, the PHY/MAC, is based on the use of IEEE 802.15.4:2006 chips that are also used for both WirelessHART and ZigBee. This layer was chosen because these chips are designed for use in very low power, battery-operated networks, and have the speed and range required for industrial wireless networks. ISA100 only uses the 2.4 GHz ISM (Industrial, Scientific, and Medical) band supported by these chips, which are available at low cost from several silicon suppliers. The MAC layer is the portion of the Data Link Layer (DLL) that is necessary to support these chips.
The Upper portion of the Data Link Layer (UDLL) immediately above the PHY/MAC has been originally designed by this committee for the channel hopping that will be needed to provide frequency diversity necessary to overcome the multipath reflections expected in processing areas often referred to as “canyons of steel.” The UDLL also has used the 16 channels that were defined by IEEE 802.15.4 so that ISA100 will avoid interference with WiFi or IEEE 802.11, which will often be used in manufacturing areas for both supervisory and other data networks.
One of the most far-reaching decisions reached at the Vancouver meeting was the selection of an Internet draft standard, 6LoWPAN, as the Network layer protocol. This forward-thinking technology is based on Internet Protocol Version 6 (IPv6), but in a reduced form suitable for low power networks. The significance of this choice is that ISA100 will be able to communicate with common IP-based protocols such as UDP, SMTP, and FTP. For example, FOUNDATIONTM Fieldbus HSE is based on the use of UDP.
6LoWPAN at the network layer allows the use of a simple well-defined protocol, RDP at the Transport Layer. RDP (Reliable Data Protocol) is also an Internet standard (RFC 1151) last updated in 1990. RDP is simple subset of TCP, the most common Transport layer protocol. Simplicity in this case means not implementing those features of TCP that are only useful on large high speed networks such as the Internet. Using existing but simple internet protocols at the network and transport layers is very important to ISA100 since it will enable many future applications such as location of a web server in wireless field devices as wireless and battery technologies advance. The committee evaluated both 6LoWPAN and RDP to be sure that they did not introduce excess overhead to ISA100 messages. ISA100 does not implement either the Session or Presentation layers of the ISO seven-layer data communications model.
The Application layer remains as an object-oriented data transfer mechanism using EDDL to make it compatible with FOUNDATIONTM Fieldbus, Profibus, HART, and OPC. Each of these consortiums support EDDL and are expected to create interfaces to the ISA100 Application layer. So far, only the Fieldbus Foundation has made a formal announcement that they will support ISA100. Additionally, a full data encapsulation protocol has been defined to tunnel messages between wireless field devices and their host systems that do not use EDDL. Tunneling will be used to allow Modbus, DeviceNet, EtherNet/IP, and other such common wired industrial network protocols access to the wireless network.
Network security is being addressed for both the process of forming the network (joining) as well as for integrity of the data being transferred. Both operations will be encrypted to assure that only authorized and properly configured units are admitted to the field network, and that data exchanges will be protected from interception. Details of the security protocols are still being worked out, but will use well-accepted practices defined in the ISA99 draft standard.
The issue of WirelessHART continues to be debated at these meetings. Users have consistently indicated their dissatisfaction with the fact that WirelessHART and ISA100 will be different and not interoperable. Previously, the users were assured that each network, both operating at 2.4 GHz, could be assigned a set of the 16 available channels to avoid interference, if that ever becomes a problem. While this suggestion is technically workable, dissatisfaction remains. Users continue to press for greater cooperation from both organizations, and meetings between SP100 and the HART Communications Foundation continue. The SP-100/HART collaboration committee released this statement: “An active dialog is underway between ISA100 and the HART Communications Foundation to define how collaboration between the two organizations can best be accomplished to satisfy the acknowledged needs of industry for clear and understandable wireless standards.”
As part of the ISA100 process of forming its mesh network, called discovery, a unique format for ISA100 devices has been defined to differentiate from WirelessHART, ZigBee, or any other message using IEEE 802.15.4 chips. At least this will allow manufacturers that wish to build devices recognizing multiple protocols (such as ISA100 and WirelessHART) to be able to use their ISA100 stack for ISA100 messages. Cooperation with the HART Communications Foundation continues in order to find a better way integrate ISA100 with WirelessHART. The technical problem is the protocols used by ISA100 at the layers above the PHY/MAC must be robust enough to be used for many industrial applications intended for ISA100, whereas WirelessHART needs only to provide relatively limited HART data. Several suppliers have already indicated their intentions to build adapters to interface wired HART field devices with ISA100 networks (referred to as “HART over wireless.”) Most of them have also indicated their intent to build WirelessHART devices and adapters for wired HART transmitters and valve positioners. The users working on the ISA100 standard are pressing for a less confusing situation than to install and configure two different networks in their plants.
The schedule for the ISA100 standard remains as previously announced:
Principles of Operation draft | Q2, 2007 |
Principles of Operation release | Q3, 2007 |
Preliminary draft ISA100 Standard | Q4, 2007 |
Draft ISA100 Standard letter ballot | Q4, 2007 |
ISA100 Standard release | Q1, 2008 |
Test Plan | Q4, 2007 |
Test Procedure | Q1, 2008 |
To interpret this schedule: the Principles of Operation describes how each part of the standard will accomplish its work. The standard adds actual protocol specifications to the Principles of Operation. The letter ballot is the formal voting process used by ISA and all standards committees. This ballot will be conducted by ISA simultaneous with the ANSI (American National Standards Institute) Public Comment and Review process. This means that a very broad audience will be able to examine the draft standard in order to provide comments, catch errors, and recommend changes. The standards voting process requires all negative ballots to be accompanied by appropriate technical comments. Editorial comments with changes are also accepted, but are not sufficient to reject the standard. The committee must process every comment and make the suggested changes, if appropriate, and request any negative votes to be changed to positive if those changes are made. It is only reasonable that technical comments submitted without a suggested change in the standard will be rejected. The committee may also reject any comment with reasons for the rejection communicated back the source of the comment. [Authors note: WirelessHART is a private specification that does not need this formality for approval by its membership, but this lack of public review precludes it from being an ANSI standard.]
Meanwhile, the ISA Automation Standards Compliance Institute (ASCI) is facilitating the creation of the ISA100 Wireless Compliance Institute (WCI) for the purposes of administering the conformance testing of products to the ISA100 Wireless Standards. This organization will be responsible for completion of the scheduled tasks in the ISA100 Standards Committee work plan called the “Test Plan” and the “Test Procedure”.
ISA has already begun the formation of WCI, which is administered via ASCI, an independent non-profit membership-driven organization. ISA will retain ownership of the ISA100 standard while the WCI will be responsible for the conformance testing of devices expected to bear the “ISA100 Wireless” compliance logo. This organization is expected to be somewhat similar to the Fieldbus Foundation in its responsibilities and governorship. It will be self-sustaining through membership dues and testing fees. The ISA is currently accepting initial donations from the automation controls industry as a demonstration of support for WCI. These donations are tax deductible and will be used to accelerate the completion of the ISA100 Wireless Standards.
At the same time, ISA through ASCI is facilitating the business development and organization building tasks to form WCI. A separate solicitation for membership will be offered to the automation controls community to fund this organization, and to secure a seat on the initial board of directors.
Richard H. Caro is a certified Automation Professional, CEO of CMC Associates. You can reach him at [email protected]