TCP/IP is a way of sending and receiving data reliably over a network.  TCP/IP functions are built into LabView, and give the instrumentation and control engineer a way of transmitting information across the network reliably.  Here are the some general comments on using TCP/IP.

• TCP/IP gives you a way to use measurement data at locations remote from where the measurements are performed.
• You might take measurements on water quality in a stream - or in many streams - and transmit data over a wireless network to a central point for analysis.  In many cases the volume of data might exceed what a floppy disk can store, and the limitations on data storage are considerably easier to work with when data is transmitted easily.
• In a production facility, you might want to do control from a central location, and "keep tabs" on a number of controlled processes from one place.  That would be especially true if the processes interact - as they would if they were different parts of a production line.

        If you want to use TCP/IP you need to do the following in LabView.

• Establish a TCP/IP Listen.  That's on the Functions - Communication - TCP palette.  Here is the icon for the TCP/IP Listen.
• 
• The TCP/IP Listen listens over the network for another computer trying to connect over the port specified in the integer input at the left side of the icon.  That port number must be a number not used by other parts of the system, and we have used 2055, arbitrarily.
• When the TCP/IP Listen hears another computer trying to connect over the correct port, it establishes a Connection ID, and an Error Signal, and those outputs are used in later TCP/IP vi blocks.
• You cannot use another TCIP/IP Listen with the same port number once the connection has been established.
• After a connection is established and the Connection ID is available, you can sent and recieve data over the network, and the connection is a reliable connection between the two computers involved.  You use either of the following vis.
• TCP/IP Write - which writes data to the other computer over the network.  Here is the icon.
• 
• The TCP/IP Write uses the Connection ID, which it also passes to the next vi, and the Error signal which it also passes to the next vi.
• The input is a string.  It's much harder to use numerical information directly, so you will have to convert your data to a string format.
• There are other inputs and outputs which we have ignored, but you can specify a timeout, and you can get information on the number of bytes written over the network with this vi.
• TCP/IP Read - which reads bytes from the remote computer.  Here is the icon.
• 
• There are the usual inputs, the Connection ID and the Error Signal.
• The integer input (blue line at left) is the number of bytes that you want to read.
• The output is the string composed of the bytes that are read.

E1 -  Using TCP/IP Write.  This is the diagram of a sub-vi we have developed that will write a string to a remote computer. Click here for a more detailed look at this vi.

Note the following steps that are needed to send a single string.

• Assuming that you have a string input and an open connection, you must first compute the length of the string.
• The length of the string is a number, and that number must be converted to a string.
• That String-Length string is sent using a TCP/IP Write.
• Then, the string itself is sent.
• The Connection ID and Error Signal are passed to further TCP/IP blocks, and eventually the connection should be closed.

Note the following steps that are needed to receive a single string.

• Assuming that you have an open connection, you assume that the length of the first string received will be 2.  The first string you receive has the data for the string length of the string you want.  For example, the string could be 17 bytes long, and you need to know that value - 17.  You use a TCP/IP Read to get the string length.
• The length of the string is a string, and that number must be converted to a number, and that number is used in the second TCP/IP Read.
• If you want to perform numerical calculations on the data you received you will need to convert that data to a number.  As it stands, the data out is in string format.