IMDTouchKnowledge BaseShop main store

How-to

Configuring Dual LAN HMI Ports for Two PLC Networks

HMI and touchscreens Published 7 min read

Short answer

Yes. A dual-LAN HMI terminal with two independently configurable interfaces reads a PLC on each network when LAN1 and LAN2 carry non-overlapping subnets and sit on physically separate segments. A remote client on the LAN1 side reaches the terminal itself over VNC, cMTViewer, OPC UA or Modbus TCP, but not the devices behind LAN2.

On this page
  1. Can one HMI terminal read PLCs on two separate Ethernet networks?
  2. How do I confirm the panel has two independent LAN interfaces?
  3. What settings keep LAN1 and LAN2 on separate networks?
  4. Which network should face LAN1?
  5. How do I commission the first PLC connection?
  6. How do I add the second PLC without disturbing the first?
  7. What can a remote client on LAN1 actually reach?
  8. When is one network on LAN1 alone the better choice?
  9. Questions asked

Yes — a dual-LAN operator terminal can read one PLC on each of two networks, provided its two Ethernet interfaces are independently configurable. The recipe: set LAN1 and LAN2 on non-overlapping subnets, wire each interface to its own segment, configure one PLC connection per network in the project, and prove each link with a live value before enabling remote access. The split also fixes the remote boundary: a client on the LAN1 side reaches the terminal itself — VNC, cMTViewer, OPC UA or Modbus TCP — but not the devices behind LAN2. The steps below follow a Weintek terminal with two independent LAN interfaces; the addressing rules apply to any dual-interface panel.

Can one HMI terminal read PLCs on two separate Ethernet networks? #

Yes, when the terminal exposes two independently configurable LAN interfaces: it then acts as an endpoint on both subnets, polling each PLC through its own port without forwarding packets between the two networks.

The panel does not need to pass traffic between the segments to serve both; each driver connection binds to the interface whose subnet contains its PLC. That same refusal to route is what turns the two-port layout into a security boundary in the remote-access arrangement described later.

Two connectors alone are not proof of the capability. Some dual-port devices place both jacks behind one internal switch that shares a single IP; such a unit cannot separate networks at all. Over a shared serial link instead of two Ethernet networks, the problem becomes sharing one port, covered in connecting one HMI panel to two PLCs over serial.

How do I confirm the panel has two independent LAN interfaces? #

Check the network settings on the terminal itself: each interface must expose its own IP-address and subnet-mask fields, because on Weintek HMIs each physical port takes exactly one IP address with no alternate-address option.

Look for separate LAN1 and LAN2 configuration pages in the terminal's system settings, and confirm the arrangement against the model's documentation. Record the current addresses of both interfaces and of every PLC before changing anything, so you can undo cleanly.

Decision: if the settings show one shared address behind two jacks, stop — the arrangement in this article is not available on that unit. The alternatives are one common subnet, or a routed network with a router between subnets.

What settings keep LAN1 and LAN2 on separate networks? #

Two things: non-overlapping subnets, decided by the subnet mask rather than by how the address looks, and physically separate segments, because one shared unsegmented switch merges the networks however you address them.

The mask determines which destinations the terminal treats as directly connected. Two addresses that look different can still describe overlapping networks when one mask is wide.

Worked example, illustrative addresses. Network A carries PLC 1 at 192.168.10.61; network B carries PLC 2 at 192.168.20.61.

  • LAN1 = 192.168.10.50, mask 255.255.255.0. Masking octet by octet: 192 with 255 gives 192, 168 with 255 gives 168, 10 with 255 gives 10, 50 with 0 gives 0. Connected network 192.168.10.0/24, spanning 192.168.10.0 through 192.168.10.255, with 254 usable addresses.
  • LAN2 = 192.168.20.50, mask 255.255.255.0 gives 192.168.20.0/24, spanning 192.168.20.0 through 192.168.20.255.
  • Overlap check: the two ranges share no address, so the assignment passes.

The trap: give LAN2 the mask 255.255.0.0 and its network becomes 192.168.0.0, spanning 192.168.0.0 through 192.168.255.255 — 65,536 addresses that include all of LAN1's range. Overlapping directly connected subnets make interface selection ambiguous, and changing the gateway does not fix that; correct the mask instead.

Setting LAN1 LAN2
Terminal address 192.168.10.50 192.168.20.50
Subnet mask 255.255.255.0 255.255.255.0
Connected network 192.168.10.0/24 192.168.20.0/24
PLC on the segment 192.168.10.61 192.168.20.61
Wiring first network only second network only

Apply each address through the terminal's own interface settings, then read both back before wiring the second segment.

Which network should face LAN1? #

Put the Internet-connected or company network on LAN1 and the machine PLC network on LAN2; that orientation is what lets a remote session reach the machine side while the company network stays off the remote path.

LAN1 then attaches to a network with Internet access, for which an LTE router is one documented example. Passing the panel's Modbus TCP service through that cellular path has its own failure modes, covered in HMI Modbus TCP works locally but not remotely over cellular.

LAN2 carries the PLCs. This split is what guarantees that a session arriving through EasyAccess 2.0 can work on the machine network and cannot spill onto the company network. If you also need data from a PLC that lives on the company network, the panel still polls it through LAN1 — acquisition on both sides and isolation between sides are two different properties, and this arrangement gives you both.

If neither network carries remote access and both are internal plant networks, assign either network to either interface, keep the subnets disjoint, record the choice and move on.

How do I commission the first PLC connection? #

Wire one segment, configure that PLC's driver with its real address, and prove the link with a live read before you add the second network, so any later fault has exactly one new variable.

  1. Connect the first PLC's network to its assigned interface and confirm the link indication at the terminal and at the switch or PLC port.
  2. In the HMI project, add the PLC device, select the driver that matches it, and enter the PLC's actual IP address plus any protocol settings its own configuration requires.
  3. Place one known readable value and the connection's communication status on a diagnostic screen.
  4. Run the project, check the value against the PLC's engineering tool, then make the PLC value change and watch the display follow it.
  5. Treat the link indication as proof of the physical connection only; the successful read is what proves addressing, driver and data location together.

Test with reads. If you must exercise a write, pick a value that cannot start motion or energize an output.

How do I add the second PLC without disturbing the first? #

Leave the first connection running, wire the second segment to the remaining interface, create a second device connection with its own driver and address, and verify that both values update at the same time.

  1. Attach the second PLC's network to the remaining interface and confirm its link indication.
  2. Add the second PLC as a separate device connection with a distinct name, its own driver and its real address.
  3. Show one known value from each PLC on the same diagnostic screen, each labeled with its device connection.
  4. Exercise both values and confirm each against its own PLC engineering tool while both connections stay active.

If the second connection fails, recheck its address, mask, cable placement, driver and requested data address before touching the working first connection. If bringing up the second interface disrupts the first, recheck subnet overlap and confirm the two segments share no switch.

What can a remote client on LAN1 actually reach? #

The terminal, and only the terminal: VNC or cMTViewer for the screen, OPC UA or Modbus TCP for its data services — devices connected behind LAN2 stay unreachable from the LAN1 side.

Remote need Result in this arrangement
Panel polls both PLCs Works, each over its own interface's subnet
View the panel screen VNC or cMTViewer, reachable from the LAN1 side
Read panel data as a client OPC UA or Modbus TCP, reachable from the LAN1 side
Reach a LAN2 PLC directly Not possible from the LAN1 side
Reach one chosen LAN2 device cMT Ethernet Pass-Through, with its destination IP configured

An EasyAccess 2.0 session therefore lands on the panel and, through pass-through, on the machine network — never on the company network. Ethernet Pass-Through opens a path to the one destination address you configure, so verify model support and re-check that destination after any renumbering. For panels that also run CODESYS control in the same chassis, the same split applies, as covered in configuring CODESYS LAN1 vs LAN2 on cMT HMI panels.

If the remote session should watch the machine but not operate it, restrict what the session can change on the terminal; for Windows-based panels the equivalent discipline is making remote HMI access read-only over RDP.

When is one network on LAN1 alone the better choice? #

Use LAN1 only, with every device in one subnet, when company-side clients must reach the PLCs directly or the panel, PLCs and company network should form a single connection — that is the documented single-network alternative.

The trade is simplicity against the boundary. One flat network needs one addressing plan and one switch design, but a remote session that reaches the panel then sits on the same network as everything else, and the separation built in this article is gone.

Decision: choose two networks when remote access must stay off the company side; choose one network on LAN1 when company-side clients must poll the PLCs themselves. Nothing in between is worth commissioning.

Questions asked

Can one physical LAN port carry a second IP address for the other subnet?

No. Weintek allows exactly one IP address per physical port, with no alternate-address option, so one port belongs to one subnet. To reach a second subnet through a single port you need a router with routes to both networks, and that router becomes the boundary instead of the panel.

Does Ethernet Pass-Through expose the whole machine network to a remote client?

No. Pass-through is configured toward one specified destination IP address, so the remote path ends at the device you name rather than the whole segment. Verify your model supports the feature, and re-check the configured destination any time the machine network is renumbered.

What happens if both networks are later connected to the same switch?

Do not do it. One shared unsegmented switch merges both segments onto one broadcast domain, and the separation that keeps remote access off the company network stops existing. If the two networks must share a cable path, use a managed switch and keep each network in its own VLAN, with no route between them.

Can the two PLCs exchange data with each other through the panel?

No. In this arrangement the panel is an endpoint, not a bridge: it reads and writes each PLC through its own interface but passes no packets between the networks. If the two PLCs must exchange data directly, give them their own routed path or a PLC-to-PLC link designed for that purpose.

Keep reading

  • How-to

    Programming a Fatek PLC Through a Weintek HMI Ethernet Link

    Yes — the FBs-20 can be programmed from a PC through the panel with only the console's one Ethernet cable. Bridge the panel's RS232 line over Ethernet with the panel's pass-through feature or an Ethernet-to-serial converter, present it to the PC as a virtual COM port, then open that port in WinProladder with the PLC's own baud, parity and station number and a raised timeout, because the panel's polling shares the line.

    HMI and touchscreens

  • Troubleshooting

    Why Two PanelViews Slow Down an SLC 5/04 DH+ Network

    The slowdown is display and alarm polling from the two terminals, not a fault in the SLC 5/04 — the sparse-screen test already proves it. Group the polled data into contiguous blocks, move alarming to the alarm-word method, and if the trunk is still saturated, move one PanelView to the controller DF1 port with pass-through routing, after confirming the firmware supports it.

    HMI and touchscreens

  • Troubleshooting

    Diagnosing intermittent HMI to PLC Ethernet comms hangs

    Stop changing the topology. Put one packet capture where it can see the PLC-to-HMI unicast exchange — a mirror port or a tap, not a spare switch port — run it through at least three hangs with device logs on the same clock, set every panel update cycle to 1000 ms, and confirm the three devices hold unique IP addresses. Whether panel requests stop or PLC responses disappear decides which side you repair.

    HMI and touchscreens