Galil Motion Control DMC-42 0 Series, Руководство пользователя

Страница: 43 / 218

Ethernet Configuration
Communication Protocols
The Ethernet is a local area network through which information is transferred in units known as packets.
Communication protocols are necessary to dictate how these packets are sent and received. The DMC-42x0
supports two industry standard protocols, TCP/IP and UDP/IP. The controller will automatically respond in the
format in which it is contacted.
TCP/IP is a "connection" protocol. The master, or client, connects to the slave, or server, through a series of packet
handshakes in order to begin communicating. Each packet sent is acknowledged when received. If no
acknowledgment is received, the information is assumed lost and is resent.
Unlike TCP/IP, UDP/IP does not require a "connection". If information is lost, the controller does not return a colon
or question mark. Because UDP does not provide for lost information, the sender must re-send the packet.
It is recommended that the motion control network containing the controller and any other related devices be
placed on a “closed” network. If this recommendation is followed, UDP/IP communication to the controller may be
utilized instead of a TCP connection. With UDP there is less overhead, resulting in higher throughput. Also, there is
no need to reconnect to the controller with a UDP connection. Because handshaking is built into the Galil
communication protocol through the use of colon or question mark responses to commands sent to the controller,
the TCP handshaking is not required.
Packets must be limited to 512 data bytes (including UDP/TCP IP Header) or less. Larger packets could cause the
controller to lose communication.
NOTE: In order not to lose information in transit, the user must wait for the controller's response before sending
the next packet.
Addressing
There are three levels of addresses that define Ethernet devices. The first is the MAC or hardware address. This is
a unique and permanent 6 byte number. No other device will have the same MAC address. The DMC-42x0 MAC
address is set by the factory and the last two bytes of the address are the serial number of the board. To find the
Ethernet MAC address for a DMC-42x0 unit, use the TH command. A sample is shown here with a unit that has a
serial number of 3:
Sample MAC Ethernet Address: 00-50-4C-20-04-AF
The second level of addressing is the IP address. This is a 32-bit (or 4 byte) number that usually looks like this:
192.168.15.1. The IP address is constrained by each local network and must be assigned locally. Assigning an IP
address to the DMC-42x0 controller can be done in a number of ways.
The first method for setting the IP address is using a DHCP server. The DH command controls whether the DMC-
42x0 controller will get an IP address from the DHCP server. If the unit is set to DH1 (default) and there is a DHCP
server on the network, the controller will be dynamically assigned an IP address from the server. Setting the board
to DH0 will prevent the controller from being assigned an IP address from the server.
The second method to assign an IP address is to use the BOOT-P utility via the Ethernet connection. The BOOT-P
functionality is only enabled when DH is set to 0. Either a BOOT-P server on the internal network or the Galil
software may be used. When opening the Galil Software, it will respond with a list of all DMC-42x0’s and other
controllers on the network that do not currently have IP addresses. The user must select the board and the
software will assign the specified IP address to it. This address will be burned into the controller (BN) internally to
save the IP address to the non-volatile memory.
Chapter 4 Software Tools and Communication ▫ 39 DMC-42x0 User Manual