Micrologix 1100 o 1400?

[asteroide 0]

Hello Friends

I need to install 4 Slave PLC and 1 Master.

The slave PLC will send some bits and integers to the master PLC.

The Master PLC will receive this data and will show in an HMI, and maybe also i will use a message display to show some messages.

Is ML1100 enough to do that through Ethernet IP messaging??

According with the specs:

Micrologix 1100:  EtherNet/IP messaging only

Micrologix 1400: EtherNet/IP messaging, DNP3 over IP and Modbus TCP/IP

 

What do you recommend?

 

Thanks in advance.

 

 

 

 

 

 

 

 

 

 

 

 

 

[TimWilborne 108]

The 1100s can do messaging in between each other.  Given the preference I usually choose the 1400 mainly because of its additional expansion module capabilities but if your I/O count isn't an issue then the 1100 should do just fine.

[asteroide 0]

Thanks for your fast response.

The input/output is not a problem, I think i will go with ML1100. Maybe a ML1400 for the Master to approach the DB9 port.

Another question:

Could be the 5 PLC´s in different IP range?

For example:

Master:     192.168.1.10

Slave1:      192.168.10.5

Slave2:      192.168.20.5

Slave3:      192.168.30.5

Slave4:      192.168.40.5

Is it possible?

Or should be the 5 PLC´s in the same IP range?

Master:     192.168.1.10

Slave1:      192.168.1.15

Slave2:      192.168.1.25

Slave3:      192.168.1.35

Slave4:      192.168.1.45

 

Thanks

 

 

 

 

 

 

 

 

 

 

 

 

[TimWilborne 108]

They would need to be in the same IP range to communicate with one another unless you used a router to coordinate packets in between them.

[Michael Lloyd 85]

If there is even the slightest chance that you'll want to be able to remote access the PLC's it's a good idea to use an IP range that's available. If the system is behind a router the IT people can exclude a range of IP's and give you the Subnet Mask and router IP. We have so many PLC's on the cloud that I get a full block of IP addresses from the IT people and Subnet them accordingly. Even if you don't it's not a bad idea to subnet an IP range and it's easy to do

As an example, for a Class C .224 subnet on .202

The first station would have:

Router: IP: 192.168.202.1 Subnet Mask: 255.255.255.224 <--- this means that it's a Class C network with 30 hosts and 32 addresses. This will give me 8 stations of IP addresses on the 224 Class C subnet

The IT guys want us to leave them the first 10 addresses so the next block of addresses are assigned:

DHCP IP: 192.168.202.11 thru 15 (same Subnet Mask) I do that so that when I plug in locally my laptop is given an address that works with the PLC I want to connect to

The next block is subdivided, by text only, ie I can put anything I want to on the address but to have consistency I put specific pieces of hardware in specific places. In reality all of the addresses are available to use however I or another programmer chooses. 

Printers, Peripherals, Third Party devices: 192.168.202.16 thru 20 (same Subnet Mask)

PLC's, Flow Computers, etc: 192.168.202.16 thru 20 (same Subnet Mask): 192.168.202.21 thru 24 (same Subnet Mask)

HMI's: 192.168.202.25 thru 27 (same Subnet Mask): 192.168.202.21 thru 24 (same Subnet Mask)

Whatever else comes along: 192.168.202.28 thru 30 (same Subnet Mask): 192.168.202.21 thru 24 (same Subnet Mask)

 The second station would be the same arrangement but just add 32 to the last octet. Which simply means the router is on 192.168.202.33, DHCP starts at .43, etc.

The third... at 32 to the second ... and so on

You can subnet all the way down to only 2 hosts but the smallest range that we use is 14 Hosts, 16 addresses. Use a .240 subnet to do that. The range is smaller so:

DHCP: 192.168.202.4 thru 7 

Everything else: 192.168.202.8 thru 14

Increment by 16 for the next block. Thus will give you twice as many subnets as a .224 subnet but 1/2 of the addresses. 

I built a spreadsheet to subnet my IP's for me but it's not very hard to do 

 https://www.aelius.com/njh/subnet_sheet.html

 

[gleblanc 1]

1 hour ago, Michael Lloyd said:

1 hour ago, Michael Lloyd said:

d... at 32 to the second ... and so on

You can subnet all the way down to only 2 hosts but the smallest range that we use is 14 Hosts, 16 addresses. Use a .240 subnet to do that. The range is smaller so:

DHCP: 192.168.202.4 thru 7 

Everything else: 192.168.202.8 thru 14

Increment by 16 for the next block. Thus will give you twice as many subnets as a .224 subnet but 1/2 of the addresses. 

I built a spreadsheet to subnet my IP's for me but it's not very hard to do 

I'd like to apologize in advance for the formatting being screwed up.  The quality of the forum software here has really tanked of late.  I spent 15 minutes trying to figure out how to clear it out, or get a basic text only editor, or something, before I gave up and figured you guys could figure it out.

 

1 hour ago, Michael Lloyd said:

If there is even the slightest chance that you'll want to be able to remote access the PLC's it's a good idea to use an IP range that's available. If the system is behind a router the IT people can exclude a range of IP's and give you the Subnet Mask and router IP. We have so many PLC's on the cloud that I get a full block of IP addresses from the IT people and Subnet them accordingly. Even if you don't it's not a bad idea to subnet an IP range and it's easy to do

As an example, for a Class C .224 subnet on .202

The first station would have:

Router: IP: 192.168.202.1 Subnet Mask: 255.255.255.224 <--- this means that it's a Class C network with 30 hosts and 32 addresses. This will give me 8 stations of IP addresses on the 224 Class C subnet

The term Class A,  (well, also B, C, D, and E) are obsolete, and have been since 1993-ish.  https://en.wikipedia.org/wiki/Classless_Inter-Domain_Routing

The original definition of a class C was a subnet with a 24-bit subnet mask, with addresses from 192.0.0.0 to 223.255.255.255.  What you're discussing is classless inter-domain routing (CIDR).  It's far superior to the old class system, and does what you're talking about.  If you drop the "Class C" terminology, you'll avoid confusing old fogies like me.

1 hour ago, Michael Lloyd said:

The IT guys want us to leave them the first 10 addresses so the next block of addresses are assigned:

DHCP IP: 192.168.202.11 thru 15 (same Subnet Mask) I do that so that when I plug in locally my laptop is given an address that works with the PLC I want to connect to

Do you issue DHCP from your border router?  What devices are you having good luck with for routers?

1 hour ago, Michael Lloyd said:

The next block is subdivided, by text only, ie I can put anything I want to on the address but to have consistency I put specific pieces of hardware in specific places. In reality all of the addresses are available to use however I or another programmer chooses. 

Printers, Peripherals, Third Party devices: 192.168.202.16 thru 20 (same Subnet Mask)

PLC's, Flow Computers, etc: 192.168.202.16 thru 20 (same Subnet Mask): 192.168.202.21 thru 24 (same Subnet Mask)

HMI's: 192.168.202.25 thru 27 (same Subnet Mask): 192.168.202.21 thru 24 (same Subnet Mask)

Whatever else comes along: 192.168.202.28 thru 30 (same Subnet Mask): 192.168.202.21 thru 24 (same Subnet Mask)

 The second station would be the same arrangement but just add 32 to the last octet. Which simply means the router is on 192.168.202.33, DHCP starts at .43, etc.

I don't like to have printers and workstations on the same collision domain (VLAN, these days) as my PLCs and HMIs.  I suppose there's no reason that they can't be on the same subnet, but I'm not sure what advantages that gets you, except that the allen-bradley software isn't great about crossing subnets.

[Michael Lloyd 85]

Just now, gleblanc said:

 

I could have also left out the "class" of the subnet. The web is FULL of class references. But, good point

Quote

Do you issue DHCP from your border router?  What devices are you having good luck with for routers?

We have a tunnel through cell providers (keeping it a little generic here...) and initially use a cell modem configured to serve IP addresses, etc. You have to do some port forwarding to allow for access to the hardware. We usually have a flow computer that spits out a 7AM report to the local printer and a centralized printer on the network otherwise it wouldn't be on the list. Geographically speaking if you draw a line from McAllen (TX) to San Antonio to Victoria to Corpus Christi and back to McAllen it'll form a box around most of the hardware we have scattered about. It's important to have connectivity at the time of startup due to leak detection requirements and the fact that we don't have a lot of manpower which makes drive time to the site to troubleshoot important. At some point the IT guys bring in wireless or hard line network connectivity (with a Cisco router) and the cell modem is relegated to back up duty. The point of subnetting ahead of time is to make the transition to the permanent network connectivity easy on our side. Easy as in we don't have to do anything.

Quote

I don't like to have printers and workstations on the same collision domain (VLAN, these days) as my PLCs and HMIs.  I suppose there's no reason that they can't be on the same subnet, but I'm not sure what advantages that gets you, except that the allen-bradley software isn't great about crossing subnets.

Local printing in the station. Other than that there's no point. Our HMI's are a client. The server(s) are in the main office or at a remote backup site.

I routinely send "messages" from one subnet to another with CLX PLC's. If there is a cell modem at the site I have to initiate the "conversation" from the end with the cell modem. I never figured out why. I suspect it has something to do with the cell modem / tunnel. I gave up on using produced / consumed tags though.