Hardware

Wiring

The topology, or physical layout, of a LAN or Local Area Network depends on the type of wire used and the standards associated with that wire. In the early days of CBAS, we used coaxial cable or Thinnet to connect GWs to the DPU. Thinnet uses a bus or ring topology and a token passing protocol. Bus topology uses a straight line of wire to connect all computers and requires termination on the ends. Ring topology uses a ring of wire to connect the computers, thus termination is not required. The RS-485 protocol used in BASNET and OPTO implementations of CBAS systems also uses a Bus topology with a single twisted pair wire and token passing.

The future of building automation communications is Twisted Pair Ethernet, which employs a star topology by connecting each computer to a central Hub or Switch. Hubs are then connected in a star or daisy chained fashion. Star networks can also be connected using a bus or backbone.

Backbone Architecture:

10Base-T (10Mbps, Baseband, Twisted Pair) and 100Base-TX (100Mbps, Fast Ethernet) are wiring standards that use Unshielded Twisted Pair (UTP) wire, usually with a Plenum rated coating. Plenum rated means it is for use in walls and drop ceilings. There are several standards of UTP, but the one you need to know is CAT5, which handles speeds up to 100Mbps (Mega bits per second). The CAT5 standard specifies the size of wire, the number of twists per foot, and has a distance limitation of 100 meters or 328 feet per segment. RJ-45 connectors are used on the ends. For more information on how the wiring works, go to the website http://www.duxcw.com/digest/Howto/network/cable/cable4.htm.

For more information on wiring, see Section 4: Sample Wiring Layouts

Making a Cable

Twisted pair wire actually contains 4 pairs of wire but only two of the pairs are used. The standards designate those pairs as the Green pair and the Orange pair. The Blue and Brown pairs are not used but are still crimped into the RJ-45 connector. There are 2 standards for the order in which the wires go into the connector. 568A and 568B are actually opposites of each other. Here’s how they look:

A straight through cable has both ends made with the same standard and is used to go from a computer to a hub or a hub to a hub. A crossed-over cable has opposite standards on opposite ends and is used to go from a computer directly to another computer or controller without using a hub. A hub or switch actually reverses the polarity for you.

It doesn’t really matter which standard is used for a straight-through cable, but we recommend that you use 568B, so that all wall jacks purchased will be the correct standard.

To determine which wire is wire number 1, hold the RJ-45 with the open end to you and the clip facing down so that the copper side is facing up. When looking down at the copper side, wire 1 will be on the far left.

To make a cable:

1.Strip about 1 inch off the end of the wire.

2.Make sure there are no nicks in the coatings on the wires.

3.Cut the nylon string off completely.

4.Untwist the pairs and straighten the wires.

5.Place them in the order for the 568B standard with wire one to the left and no space between the wires.

6.Cut the wire off so that you have about ½” of exposed wire.

7.Holding the connector with the clip down, slide the wires into the back of the connector so that each wire goes into it’s own compartment.

8.Push the wires all the way in and make sure the outer coating is half way into the connector, or as far as it can go. This will make a much more reliable cable because only a small portion of wire is untwisted, and the outer coating is crimped into the connector, thus making it stable.

9.Check the order of the wires again before wasting a connector. The wire to the left should have a white background and they should alternate from there between striped and solid.

10.Crimp the connector.

11.Tug firmly on the connector to see if it is crimped properly.

12.If you are making a straight-through cable, make the other end the same way. If you are making a crossover cable, use the other standard (568A).

13.Test the cable.

Wall mount jacks are made in both standards, so be sure to use only 568B jacks. Actually, Leviton makes jacks that have color-coding for both standards on each jack. If you should happen to get a 568A jack, then the cable from the jack to the computer should be 568B. The wall side of the jack is color-coded.
Network Interface Card (NIC)

Computrols computers should come with a NIC already installed and configured. Actually, the newest Automation Servers have 2 NICs built into the motherboard, and the top NIC is configured for use. The send NIC is disabled, in order to avoid any conflicts.

If you should have to install a NIC yourself, make sure you ground yourself to the case to avoid damaging anything with static electricity. Make sure the card is firmly seated and held in place by a screw.

Many older Computrols computers are configured for CI1 boards, which could cause problems when installing NICs. CI1s use ISA slots and require that certain IRQs be reserved for those slots. IRQ stands for Interrupt Request. Each device in a computer is assigned an IRQ address, which allows it to interrupt the CPU or processor when it needs attention. At one time, production reserved all available IRQs for ISA slots in the computer’s setup.

To un-reserve an IRQ:

1.When the computer is booting, hit the delete key at the appropriate time to enter setup.

2.Go to the PCI/PNP section.

3.Scroll down to the IRQ listings. If they all say PCI/PNP, then you don’t need to do anything. Otherwise, you really only need to free IRQs 9 and 10 if they aren’t already.

4.Select IRQ 9 and hit enter.

5.Select PCI/PNP and hit enter.

6.Do the same for IRQ 10.

7.Hit ESC twice, and enter to save changes and exit

8.Now you have to install a driver for the NIC. If you are installing on a computer with Windows 2000 or XP, there’s a good chance it will have the drivers and Plug and Play will take care of it for you. When Windows starts up, it should find the NIC and it might ask you if you want to search for the drivers.

9.Insert the driver disk and type in the following path for the driver: A:\win98 for Windows 98, A:\Win2000 for Windows 2000. XP should not ask you for a driver, but if it does, the Win2000 driver should work. After installing the driver, you will have to reboot. In a later section, we will discuss how to test your NIC.

One other thing should be mentioned about NICs, and that is, the MAC address. The MAC address is also known as the hardware address and is a unique hexadecimal number hard coded to the NIC at the factory. These numbers are given out by the IEEE and are intended to be unique in the entire world. Computrols Internet ready controllers also have a MAC address.

Hubs and Switches

A hub is basically a multi-port repeater. When a packet of data arrives on a hub, it forwards it out to all ports. The problem with this is that traffic is multiplied. Another problem is that you are limited in the number of hubs that can be placed between 2 computers. On a 10Mbps network, you’re limited to 4 hops, 2 on a 100Mbps network. A hop is basically any hardware that the data goes through, like a hub or switch.

A switch is a more intelligent repeater, in that it forwards the packets to the intended PC port rather than all ports, thus reducing traffic. Also, the hop limitation is irrelevant with switches. Here’s a comparison.

Comparison of 10/100 Hub and 10/100 Switch:

1.Connections look similar, but the internal fabric is different.

•10/100 HUB Uses internal 10 <--> 100 bridge feature to interconnect both 10Mbps and 100Mbps Shared Bus.

•10/100 SWITCH Uses internal switched fabric to switch the data to the dedicated port.

2.Hop count limitation*:
•HUB Is Limited to 2 hop counts for 100Base-TX Fast Ethernet and 4 hop counts for 10Base-T Ethernet.

•SWITCH has no hop count limitations. Allows users to expand their network easily.

• From the IEEE802.3 and IEEE802.3u Ethernet standards, the HOP COUNT is defined as the numbers of HUBs that an Ethernet signal has to pass through to reach its final destination.

3.Total Bandwidth Comparison:

•HUB: Is normally 10Mbps for 10Base-T Ethernet and 100Mbps for 100Base-TX Fast Ethernet.

•SWITCH: The aggregated bandwidth depends on the port numbers of a switch, e.g. total bandwidth of an 8-port SWITCH is 800Mbps.

4.Full-Duplex Support:

•HUB: Is not popular for 10Base-T and 100Base-TX HUB.

•SWITCH: Most up-to-date SWITCHes support Full-Duplex transmission mode that will double the speed of each link, e.g. aggregated bandwidth of a 8-port SWITCH will become to 1.6Gbps, which means 200Mbps per link.

5.Cost vs. Performance Concern:

•HUB: Less expensive per port, however, performance depends on the network traffic. Suggested for use with SWITCH to get higher system performance.

•SWITCH: Affordable solutions no matter if the network configuration is simple or complicated. In addition, there’s no expansion limitation in the future.

Technicians won’t be making the choice to install a hub or switch. That should be specified in the job. There are only a few things about hubs/switches that you need to know. They may have an uplink port next to port 8 or whatever the highest port is. An uplink port is for the wire that goes to the next hub/switch down the line. It takes care of reversing the signal, like in a cross-over cable. If you plug into the uplink port, you cannot use the port next to it. Instead of having an uplink port, some hubs/switches have a switch next to one port that allows you to reverse the polarity. If the link light is lit on the uplink port, or any other port for that matter, you have a good connection to the next node. If the device at the next node is off or the wire is unplugged, then the link light will stay off. Most switches now sold have auto-sensing ports, which sense the polarity on the line and adjust to it, so that uplink ports are no longer necessary.

Controllers

A few things to know about the controllers:

•Presently, the only way to set the IP address, etc. is by using the handheld device.

•Controllers come with a default IP address of 192.168.1.199.

•There is no need to upgrade the handheld to communicate with the new controllers.

•It is possible to communicate to a RS-485 channel through a 10Mbps board. In other words, the new board can take the place of the CI1 card used in older CBAS systems.

•Controllers are hard coded with a unique MAC address at the Computrols production facility. This address cannot be changed, but can be viewed using a handheld.

To set the IP Address of a controller:

1.Plug in the handheld device to the female serial port on the terminal board of the controller. You will see the firmware version and date on the screen.

2.Press enter twice to see the Modes menu.

3.Press the number 3 for web, and press enter.

4.IP address is shown. Scroll down for other settings, like Subnet Mask and Default Gateway.

5.To change, press select. You will see a blank IP address.

6.Enter the numbers 192.168.1.X. When you enter a 3-digit number, the “dot” will be added for you. If you enter a 1 or 2-digit number, you will have to add the “dot” manually.

7.Hit enter when finished entering the address. You will be prompted to reboot for changes to take affect.

8.Turn the power switch off then back on.

Controller LEDs

There has been some confusion about what the LEDs on the 8X, 16X, 32X, and 64X really mean. Here’s a brief explanation:

Power Light: Sufficient power is supplied to the board when it is solid green.

Run Light: Blinking red light means normal operations. Solid red means the onboard computer is booting. No light/solid light means that the onboard computer is not running.

10Mbps Link Light: Solid orange means you have a good link to the next hub or switch on the network.

10Mbps Activity: Data is passing to and from the controller.

RS-232 LEDs are active when connected to the controller using HyperTerminal or the Commission program.

RS-232 Receive: Blinks yellow when receiving data on RS-232 port.

RS-232 Transmit: Blinks green when sending data on RS-232 port.

Port 1 and Port 2 LEDs signify activity on the Host and Secondary ports respectively. They act differently depending on the protocol in use. Following is a description of behavior when BASNet RS-485 is in use. The activity LEDs will be the most active.

Port 1 Receive: Blinks yellow when receiving a high level command from the DPU to a controller on the RS-485 Host channel.

Port 1 Transmit: Blinks green when sending data (token) on RS-485 Host channel.

Port 1 Activity: Blinks orange when receiving or passing tokens on RS-485 Host channel.

Port 2 Receive: Blinks yellow when receiving a high level command from the DPU to a controller on the RS-485 Host channel.

Port 2 Transmit: Blinks green when sending data (token) on RS-485 Secondary channel.

Port 2 Activity: Blinks orange when receiving or passing tokens on RS-485 Host channel.

*Generally, a solid RS-485 Activity light means that the polarity of the wires is reversed.
Software

TCP/IP Configuration

To communicate on a Local Area Network (LAN) using TCP/IP protocol you need:

•An IP Address

•A Subnet Mask

To access the Internet (outside of the LAN) you need:

•A Default Gateway

•A DNS Server Address

Each one of these TCP/IP requirements is described in the following sections. In addition, descriptions of Ping, Network Identification, and Workgroup are provided.

TCP/IP Address

Every computer on a TCP/IP network must have an IP address that uniquely identifies it and distinguishes it from the other computers on the Local Area Network (LAN). An IP address is a 32-bit number that consists of four numbers between 0 and 255 separated by periods. We are used to seeing the decimal version, which represents the binary form that the computer software understands. For example, the IP address 168.212.226.204 in binary form is 10101000.11010100.11100010.11001100. A portion of the address identifies the network that the computer is on and a portion identifies the node. The size of the network address depends on the class of the network address.

Since we will be dealing with mainly Class C addresses, here is an example of a Class C address. In a Class C address, the first 3 numbers identify the network and the last number identifies the node. So, in the number 192.168.1.200, 192.168.1 is the network or subnet, and the number 200 identifies the computer. There can be over 2 million Class C networks, but each one can only have 254 nodes. If we have a building that has more than 254 nodes requiring IP addresses, we will either have to create 2 Class C subnets or one Class B subnet.

In a class B network, the Subnet Mask is 255.255.0.0, so 192.168 identifies the network or subnet, 1.200 is the node address. So, you can have 65,025 addresses (255 X 255).

There are certain subnets of addresses that are reserved for use on internal networks. Internal networks are separated from the Internet by a router, and therefore, aren’t directly exposed to the Internet. Here at Computrols, we generally use the Class C addresses ranging from 192.168.1.1 to 192.168.1.254. There is also an address called the loop back address, which is used to test the operation of a NIC. This address is 127.0.0.1.

Computrols controllers come pre-programmed with an IP address of 192.168.1.199. Automation Servers generally come with an IP address of 192.168.1.2. Do not assign an IP Address of 192.168.1.0 or 192.168.1.255, as those are the Subnet and the Broadcast addresses respectively, and thus are not valid addresses.

Subnet Mask

The subnet mask is used to determine what subnet an IP address belongs to. A typical Subnet Mask for a Class C network is 255.255.255.0, which is represented in binary as 11111111.11111111.11111111.00000000. The 255 basically masks the corresponding portion of the IP address, making it the network address. A subnet can be further divided into subnets using a mask like 255.255.255.192, for instance. This kind of subnetting is beyond the scope of this course. If you have a controller that is exposed directly to the Internet, you will be given a Subnet Mask by the Internet Service Provider (ISP). This number will likely have a number other than 255 in it. Computrols controllers come pre-programmed with a Subnet Mask of 255.255.255.0.

In the case of a Subnet Mask of 255.255.255.0, there are only 254 possible nodes. If you have a job that requires more than 254 nodes, contact Computrols Tech Support for instructions.

Default Gateway

In order to get to the Internet from a LAN, you need to go through a router. The address of that router is the Default Gateway. To communicate with a server, web page, or another computer on the Internet, you might go through several different routers. The Default Gateway is the first router in that string of routers, and forwards your packets on to the next router. Computrols controllers come pre-programmed with a Default Gateway of 192.168.1.1.

DNS Server Address

When you type an Internet address such as http://www.computrols.com/, that address has to be translated into an IP address in order for your packets to reach their destination and receive a web page back. A Domain Name Service (DNS) server does translation of Internet addresses. Generally, the ISP provides DNS server addresses. Without a DNS server, you will never be able to surf the web.

Ping

You can use the ping utility to test a connection or a NIC. To test a NIC, click on Start, then Run, type “command” and hit enter. At the command prompt, type: Ping 127.0.0.1, and hit enter. If you get a reply, it means the NIC is installed properly and TCP/IP is set up properly.

To test a connection to another PC or controller, type ping, space, followed by the IP address of the PC or controller.

Network Identification

Each computer on a network must have a unique Computer Name, also known as NetBIOS Name. It is usually a name that describes the use, user, or location of the computer.

Workgroup

To share files in Windows’ My Network Places, the computers should be in the same workgroup or domain. Like the Computer Name, this is accomplished in the Identification section of Windows Network Configuration.

TCP/IP Computer Settings

To set up TCP/IP on a computer, follow the instructions for either Windows 98 or Windows 2000/XP below. The examples assume an IP address of 192.168.1.2, etc.

Windows 98

To set up a computer using Windows98:

1.From the Windows Desktop, right-click on Network Neighborhood, then click on Properties.

2.In the list of Network Components, double-click on TCP/IP.

3.On the IP Address tab, check the circle that says “Specify an IP Address” and type 192.168.1.2 for the IP Address and 255.255.255.0 for the Subnet Mask.

4.Set the Computer Name and Workgroup by going to the Identification Tab. Type in a descriptive name for the computer and make the workgroup name CBAS.

5.Click OK and you will be asked if you want to reboot. Close any open programs and click yes to reboot.

Windows 2000 or XP

To setup a computer using Windows 2000 or XP (with Classic View Start Menu and Control Panel):

1.From the Desktop, right-click on My Network Places, select Properties, and then click on Network and Dialup Connections.

2.Right-click on Local Area Connection icon and click Properties.

3.Double-click Internet Protocol (TCP/IP).

4.Check the circle next to “Use the following IP address” and type 192.168.1.2 for the IP address and 255.255.255.0 for the Subnet Mask.

5.Type 192.168.1.1 for the Default Gateway, click OK, and OK again.

6.Click on Advanced on the main menu.

7.Click on Network Identification and then the Properties Button. From here you can change the Computer Name and make the workgroup CBAS.

8.Click OK and you will be welcomed to the CBAS Workgroup. You will need to reboot.

CBAS Configuration

To configure CBAS:

Note: You must have version 1.5.11 (or later) of CBAS to setup TCP/IP channels and controllers.

1.If you aren’t using a pre-existing database, you must create a database.

2.Create a TCP/IP channel in addition to the one automatically created for workstations. The TCP/IP for Workstations channel is actually going to be the same channel because it uses the same NIC and IP address. It just provides a way to subdivide the workstations from the controllers.

To create the TCP/IP channel:

1.Go to Editor Mode, Hardware View.

2.Click on Add a Channel.

3.Give your channel a name that distinguishes it from the TCP/IP Channel for workstations.

4.Select TCP/IP for Controllers as the configuration for the channel. Click Add Channel Now! And the new channel will appear in the channel list.
To add a TCP/IP controller:

1.Click on the channel you created and click Add a BASNet Controller.

2.Put in a name that describes its location or use that will distinguish it from other controllers.

3.Choose a controller type from the next drop down list. Put in the IP Address of the controller and hit enter.

4.Select to Create a New Blank Database then click Add Controller Now!

If you have multiple controllers of the same type, that have identical point configurations, you can create a template by clicking on the controller in Hardware View, and click Save Database as Template. The Template will be stored in C:\CBAS2000\Templates (or C:\CBAS\Templates). When Creating a Database from a template, browse to that directory and select your template. You can also change all the point names to match the name of the new controller in the “Create Form Template” wizard.

To create a RS-485 over TCP/IP channel:

Note: These instructions apply to version 1.5.15+ of CBAS.

1.In Hardware View, click on your TCP/IP channel and then click on the controller that will act as the RS-485 interface.

2.Click on the Channels button.

3.In the next window, click on Add a Channel next to RS-485 Host or RS-485 Secondary. You will be prompted for the type of channel, which will be BASNet RS-485. A few other channel types are available here. Complete the wizard and the channel will now show up in Hardware View.

Since the advent of Internet controllers, there have been changes made to the way GWs connect to the DPU. The database folder no longer has to be shared in Windows in order to connect from a GW. You still have the choice of connecting through a network share but connecting without the share creates less traffic because the entire database is copied to the GW and only the changes are updated.

Setting Up a GW (Graphic Workstation)

In order to set up a GW, the Server and the GW must be able to communicate on a network basis. In other words, if you can see the GW in the Network Neighborhood window of the DPU, or you can ping the GW, it is possible. For more information on how to set this up, see the previous section.

When connecting as a GW, the CBAS version on the GW must be the same (or nearly the same) as the Server, or the Server will refuse the connection. Version 1.6.X will not work with 2.0.X, which will not work with 2.1.X, etc.

Before you can set up a GW, you will need the IP address of the Server and the GW, as well as the GW Name of both. To get this info:

•In Windows 98, from the windows desktop, click on Start, Run, type winipcfg, and enter.

•In Windows 2000/XP, click on Start, Run, type “command” or “cmd” and hit Enter. At the prompt, type ipconfig and hit Enter.

•Write down the IP addresses of both computers. Close the window.

First, you must set up the DPU to accept the GW:

1.In Editor Mode, click on Hardware View from the Main Menu.

2.Click the TCP/IP for Workstations channel, then Controllers.

3.On the last line, click Add A GW/Stringserver.

4.On the first line, enter a name for the GW. The default is “NEW GW”. Enter a name that describes the location or function of the GW computer.

5.For Configuration, choose GW. In the latest versions, the IP address of the GW is not needed. Click Add Controller Now!

6.Close the database and reopen in real mode. Now go to the GW computer.

7.In CBAS, click on System on the main menu, then Close Database.

8.Click on Database, Open Database, and Remote GW (see next figure).

9.Enter the IP address of the DPU, and the name of the GW as you entered it on the DPU. Click Connect and the database should open.

Note: In the workstation Channel on the Server, there is a Local GW listed. This is actually the Graphic Interface on the Server itself. It is added by default when a database is created, and cannot be removed or renamed.

Below is the GW Connection screen:

When you connect using the method above, the entire database is zipped and downloaded to the workstation. It resides in C:\CBAS\[GW Name] (or C:\CBAS2000\[GW Name]). If the database has changed since the last connection was made, the changes will be downloaded when you re-connect. You may have to close the connection manually to see newly added points and controllers.

There are a few things that can cause the GW Connection attempt to fail:

•TCP/IP Error: You will get this error message if TCP/IP is not configured correctly on both computers, or if you have used the wrong IP address for the Server (DPU).

•Workstation Name is not in the Database: Either you have not entered the workstation in the Workstation Channel on the server, or you spelled the name wrong on the TCP/IP Login screen.

•CBAS is not running in Real Mode on the Server.

•CBAS version is incorrect: Upgrade the GW to the same version that the Server is running. In some cases, it will ask you if you want to download the correct version from the DPU. If the Install.exe file is on the Server, then this will work.

•This GW is already in use: Another GW has logged in using the Workstation Name you specified. Try another Workstation Name, or add more GWs to the Server.

•Number of GW Connections has been exceeded: CBAS Professional comes with 2 Remote GW Connections by default. A license for more GWs can be purchased. Meanwhile, have someone log off in order to connect.

Notice the “Use File Sharing Connection” checkbox in the sample GW Connection screen. Advantages and disadvantages of File Sharing are outlined also.

If you want to connect using the file sharing method:

1.Check the box next to “Use File Sharing Connection.”

2.Click on the “Browse” button in the Database Path section. The window that opens will have selected the database you just closed.

3.Click on the minus sign next to the database then, scroll down to “Network Neighborhood” or “My Network Places.”

4.Click the plus sign next to “Network Neighborhood” and locate the DPU by its computer name.

5.Click the plus sign next to the DPU computer and select the shared database on the DPU.

6.Click OK and the network path will appear in the Database Path field.

7.Click “Connect” and the database will open.

Advanced Networking

Connecting Computrols controllers, DPUs, and GWs over a Wide Area Network (WAN) involves some advanced networking knowledge. Whenever making connections over a WAN, there are routers involved. This is the Default Gateway discussed earlier. A router basically protects the Local Area Network (LAN) computers from the Internet (hackers/viruses) and allows users on the LAN to access the Internet. In order to access or pass information to a computer or controller on the LAN side of a router, certain TCP/IP ports must be forwarded to the IP address of a computer or controller. Otherwise, the router will reject the request. The following illustration shows the ports needed. In a normal router, all ports are closed to access from the outside to the inside, unless communication is initiated from the inside through the same port.

Computrols Network Architecture

As of CBAS version 1.6.1, it is no longer necessary to forward TCP/IP port 8751 at the Router on the LAN where a Graphical Workstation exists. Now, the Graphical Workstation initiates communication to the DPU on port 8750 and the DPU communicates back on the same port. Version 1.6.1 is not backward compatible in respect to the GW connection.

Router Configuration Instructions

Computrols recommends the use of an inexpensive Cable/DSL router, in addition to the modem provided by the Internet Service Provider (ISP), when connecting CBAS systems to the Internet. This is recommended because it will provide a high level of security against viruses and hacker attacks. It does this by blocking packets intended for the internal network on all TCP/IP ports, except those specifically forwarded to the computers, programs, or controllers that will accept them.

In order for computers to work with multiple incoming and outgoing network communication packets, the TCP/IP protocol uses thousands of “ports”. Standard programs use specific ports, called “Well Known Ports”. Some examples of Well Known Ports are: port 80/html (Web pages); port 25/smtp (email); port 110/pop (email). CBAS uses some of the ports in the range from port 8750 to 8790. See CBAS Network Architecture for more specifics about CBAS ports.

The router, by default, is set up to allow all outgoing packets. However, most of the time, only port 80 is needed to access the Internet. NAT, or Network Address Translation, is used to disguise or hide the computer accessing the Internet. The router does this by changing the source IP address of the sender to the IP address of the router itself. Thus, the actual sending computer remains anonymous.

The following illustration shows the physical configuration of an Internet connection using a router:

In order to configure a router to allow access to the Internet, you must configure both the WAN interface and the LAN interface. The WAN is the Wide Area Network and the LAN is the Local Area Network.

Configure the WAN interface with the IP address, subnet mask, default gateway, and DNS server address that is provided by the Internet Service Provider. For more information on these terms, see Chapter 8, Section 2-TCP/IP Configuration.

The LAN interface usually comes pre-configured with an IP address and subnet mask. However, a computer will have to be configured with the correct IP address in order to access and configure both the LAN and WAN interfaces. Configuration is generally done through a web page based configuration tool on the router itself. Follow the manufacturers instructions to configure the router.

Once you can access the Internet from the computer on the LAN side of the router, you can forward the necessary ports through the router to the correct IP addresses.

Recommended routers:

•Linksys BEFSR11, w/1 LAN port

•Linksys BEFSR41, w/4 Port switch

•D-Link DI-604, w/4 port switch

Adding More Than 252 IP Controllers to a CBAS System

Let’s say you have a CBAS server with IP address 192.168.1.2 and subnet mask 255.255.255.0. This means you have a class C subnet and address range of 192.168.1.1 through 254. 1 is usually for the Default Gateway, which is the router that allows you access to the Internet. With the server on 2, you have 252 more addresses to work with. What happens if you need more addresses? Although most of us will never see this situation, there are 2 ways of dealing with it.

Change Your Subnet to a Class B Subnet.

If you change the Subnet Mask on the server to 255.255.0.0, you will then have a Class B subnet. Now you can work with IP addresses with different 3rd octet numbers, like 192.168.2.X, 192.168.3.X, etc. This gives you a possibility of over 64,000 addresses. The Subnet Mask on the controllers will also have to be changed to 255.255.0.0.

Add another IP Address/Subnet to Your Server.

Adding another Class C IP address to your server’s network card will allow it to talk to another subnet of 254 addresses, thus giving you over 500 possibilities. This IP address should be something like 192.168.2.2 and the controllers on that subnet would have addresses 192.168.2.3 through 254.

Go to Windows Network properties, Local Area Connection properties, and go to properties on TCP/IP. Click the Advanced button and add an IP address like 192.168.2.X and the subnet mask 255.255.255.0. Click ok, etc.

In CBAS Hardware View, add a TCP/IP Channel. Click on the newly created channel and click Program. Change the IP address to match the IP address you added to your Local Area Connection in the previous paragraph. You can now begin adding controllers to that subnet, using addresses 192.168.2.x.

Forwarding a Router to Multiple Controller Web Pages

Let’s say you have more than 1 controller in a building and you want to be able to access their web pages from the Internet. A visual representation is shown at the end of this article.

First of all, for this to work, the controller must be set up to offset ports. To do this you must access the controller with a handheld terminal.
1.Go to the main menu and hit 5 and Enter.

2.Hit the down arrow key twice and you will see “Select to Offset Ports.”

3.Hit the SEL button. You will see the following: 0:Off 1:On (OFF)

4.In parentheses is the current condition. To turn on Port Offset, select 1 and hit Enter.

5.The controller will reboot and then will be ready to accept http requests on port 80 and another port based on its IP address and the following formula:

10,000 + 80 + (Last Octet of IP Address) = Port Offset

“Last Octet of IP Address” means the number on the right. For example, in the IP address 192.168.1.200, 200 would be the last octet. 10,000 + 80 + 200 = 10,280

To access this controller’s web page from the LAN (inside the router), put this address in your browser: http://192.168.1.200:10280

Note: You must put in the http:// before the IP address or the controller will not know that this is a request for a web page.

To access the controller from the Internet (outside the router), you must first configure the router to forward that port to the internal IP address. This process will not be outlined here, because it is different on every router.

For this example, let’s assume that the WAN IP address of the router is 216.253.95.36. Once the port is forwarded to the right internal IP address, the web page can be accessed with the following address: http://216.253.95.36:10280

Since every controller on the LAN will have a different IP address, each one will have a different “Offset Port.”

For example:

LAN IP = 192.168.1.50, Offset Port = 10130

LAN IP = 192.168.1.99, Offset Port = 10179

LAN IP = 192.168.1.120, Offset Port = 10200

LAN IP = 192.168.1.155, Offset Port = 10235

LAN IP = 192.168.1.254, Offset Port = 10334

Note: This feature is not available on WorldSTAT products.

CBAS and The Internet

When Computrols introduced its line of Internet Controllers, it opened up a whole new way for people to communicate with controllers and automation systems. With that came a lot of confusion about what can and can’t be done in regards to remote access to controllers and whole CBAS systems. Let’s see if we can clear up some of the confusion by going through the different scenarios.

pcAnywhere Dial-up

In the beginning, there was pcAnywhere, the Symantec software that allows you to access and control another PC, either over a LAN (Local Area Network) or WAN (Wide Area Network or Internet). PcAnywhere works with dial-up using a standard telephone line, and is still used in many buildings that don’t have the budget to upgrade to some form of high-speed Internet. The same phone line can be used by CBAS to dial out and send alarms to pagers. Then, the engineer receives the page and can dial in to the system and make adjustments.

pcAnywhere High-speed

Many buildings have made the switch to high speed Internet, often because the engineers have Cable or DSL at home. It isn’t possible for them to use high speed Internet to get into a dial-up system. But, it is possible to dial up to the Internet and get access to a high-speed system. Anyway, pcAnywhere is a whole lot better when you have high-speed on both ends. Computrols has replaced the use of pcAnywhere with Log-Me-In. Go to www.logmein.com for more information.

Graphic Workstation (GW)

The same connectivity concepts apply to a CBAS GW over the Internet, and in most cases, works faster than pcAnywhere. The other benefit to GW is that it is less of a security risk if the CBAS DPU is on a LAN with other computers. When you gain access with pcAnywhere, you get access to everything on the computer, including access to other programs and other computers on the network. Many network administrators don’t allow pcAnywhere on their LANs because of this. With GW, you only gain access to the CBAS database, so there’s no security risk. So, engineers and technicians can be at home, at another building, or in a cyber café, and get access when they get a call, page, or email alarm.

Email Alarms

Another benefit of high-speed Internet is that you can program Email Alarms on any point that has an alarm programmed already. Emails can be sent to a dispatcher, people at home, or to a text-message-capable phone. Instructions for setting up email alarms were outlined in a previous section of this manual.

Controller Web Pages

Although this feature isn’t extensively used, you can access a web page on the Internet Controller itself. The web page can be set to require the user to log in to gain access. From there you have a choice of Text View or Hardware view, and command-able points can be commanded. However, the point program screen is not accessible, so schedules or logic cannot be changed.

CBAS-Web Server

Now that we’ve covered the basics, let’s get into some more unusual or complex situations. First, there’s the CBAS-Web Server. This product allows engineers and tenants to schedule overtime for their AHU by accessing a web page. A separate computer next to the DPU serves the web page. The web page is exposed to the Internet by forwarding a single TCP/IP port from the public (WAN) side of the router to the Overtime Web Server.

Stand-Alone Controllers

Stand-alone controller access is another possibility using the Internet. Let’s say you have a small building requiring only one 32-point controller, and the occupants of the building have a LAN with Internet access. That controller could be placed on the LAN by running a network wire from the controller to a nearby switch that is part of the network. Then, one port needs to be forwarded from the router to the controller. The technician would have the database for that controller in CBAS on his office computer or laptop. In that database, the IP address of the controller would be the IP address of the router, and the router would forward communications to the controller. Also, that database could contain a channel of RS RS-485 controllers like VAVs, or a Modbus RTU channel for some Veris Hawkeye power meters.

Multiple Stand-Alone

Let’s say you have a building management company that manages several small buildings that require one Internet controller with up to 64 hardware points and maybe a few more controllers or VAVs. Probably the best way to handle this situation would be to have the DPU at the management office and allow GW access from the outside by forwarding one port on the router. Business DSL or cable access with a static IP address would be preferable at each location, allowing the DPU to have a database containing a controller from each, connecting in the same way described in the previous paragraph. If more controllers are needed at a particular location, they could be placed on the RS RS-485 channel hosted by the one Internet controller.

People at the management office could monitor the building from there, but what happens when a technician is needed at the site? A permanent workstation could be placed at the remote building or, the technician brings a laptop, plugs his cable into the router or a switch on the LAN, and connects as a GW.

Multiple DPUs

On the other hand, you could have a DPU at each of the management company’s buildings and a workstation at the main office. However, this would not be a viable solution logistically or financially unless they have four buildings or less. A CBAS Commercial license would be required at each location, as opposed to one Professional license in the previous example. To connect from the main office, the user would have to connect as a GW separately to each building, one at a time. Unless the user was connected at the time, he would not see alarms as they occur.

WorldSTAT Enterprise

Let’s say a management company has 100 or more sites in many cities, maybe across multiple time zones. With a WorldSTAT or other Internet Controller in each location, and a DPU in the corporate headquarters, WorldSTAT Enterprise software would give them the features they need to achieve rapid deployment of new locations and regional management.

While these are just a few of the scenarios that are possible with Internet controllers and CBAS software, they cover the main concepts necessary to understand connectivity.

Example Diagram

Below is an example of one way to connect buildings over the Internet. It shows a management company that manages three buildings from its office using one CBAS Automation Server and one TCP/IP controller in each building. A laptop can be connected to the network in any building and log on as a workstation to the Server in the main office.

Sample Wiring Layouts

Sample Nine Story Building

Fully Loaded 25-Story Building

Nine Story Building With VAV Controllers