Skip to content

Connecting the hardware

This chapter covers getting a LITES MHU116 or MHU117 fire panel talking to Protequ. The tree can come from a live import once the panel is connected, or from a configuration-program XML export. Fields and addresses are on the configuration page. What the panel can and cannot do is in Notes.

How Protequ reaches it

Over an IP network, directly. There is no converter. Protequ is the TCP client. The panel side is Modul periferií 2 (DPE-2) — the Ethernet module on the MHU, listening as a TCP server. The protocol is the native LITES link, not Modbus.

The TCP port is not factory-fixed. Read it from the LITES configuration program (Ethernet / DPE-2). A site may use 502; that still is the native port, not Modbus TCP. Do not assume 502.

DPE-2 accepts one TCP client. The LITES demo program, the configuration program, a leftover Protequ session, or a probe will hold the socket. Protequ then cannot open it.

The module closes a silent socket after 5 seconds. Protequ keeps the session alive by itself. A drop after a few seconds of quiet is the module idle rule, not a cabling fault.

There is no login and no password on this link. After connect the traffic is not encrypted; the site network must be trusted.

Before you start

A LITES technician needs the configuration program and the project that is loaded into the panel.

  • Protequ on Windows or Linux with the LITES adapter selected (see In Protequ).
  • The Protequ host can reach the DPE-2 address (same VLAN, or a permitted path to the native TCP port).
  • The display master number of the panel Protequ should own (1…16). On a network with a tablo (MHS 817) or a second MHU, that number is the connected master, not the tablo.
  • Nothing else is already connected to that DPE-2 port.

Fill in before you arrive:

DPE-2 IP:                       ........................
Native TCP port:                ........................  (from the config program)
Master on the display:          1 … 16
Config-program project:         ........................
XML export (optional):          ........................  (root Konfigurace)

Two things that do not change:

What Value
Protocol Native LITES over TCP. Do not point Protequ at Modbus.
Client One TCP session on DPE-2. Close the demo / config-program link first.

On the equipment

A LITES technician does this in the configuration program. Without these steps Protequ will not connect, even when the IP address answers ping.

  1. DPE-2 on the LAN. The Ethernet module has an IP the Protequ host can reach. Note the native TCP port from the Ethernet / DPE-2 screen — not a Modbus port, even when the number happens to be 502.

  2. Load the project. The running panel must match the project you will import. Saving the file on disk is not enough.

  3. One client. Disconnect the configuration program and any LITES demo software from that port before Protequ starts. A second client will not share the socket.

  4. Optional XML. Export the configuration (Konfigurace, UTF-16). The file is for Add device when the panel is offline. When the panel is already connected, Import configuration reads the tree live and the XML is not required.

In Protequ

Install the adapter first, then add the panel. The panel steps are the same on Windows and Linux.

Windows. Run the Protequ installer and select LITES (fire) in the adapter catalog. The adapter runs as a service. You do not install Docker or a separate .NET runtime.

Linux. On Ubuntu 24.04 or 22.04 (x86_64) download pq-installer-online-v…-linux-x86_64.tar.gz from repo.protequ.com, unpack it and run sudo ./run.sh. The installer installs Docker Engine if it is missing. In the catalog select LITES (fire) — that pulls pqlitesmhu and starts it as a container. Adding the adapter later means running the installer again.

Confirm the port, not only ping:

Windows:         Test-NetConnection <DPE2_IP> -Port <PORT>
Linux / macOS:   nc -z -v -w3 <DPE2_IP> <PORT>

TcpTestSucceeded : True means the port answers. Timeout or refused is VLAN, firewall, a wrong IP, DPE-2 not in the project, or the native port was never written down.

If the panel card is Stopped after a host restart, restart only the adapter in the UI (Extra adapter), not the whole server.

Devices → Add device. Adapter LITES, type MHU Fire Panel, name it. Enter IP, the native port, and the display master (1…16). Wait until the card is no longer Stopped. Then Import configuration on that panel so groups, detectors, inputs and outputs appear.

An XML export is only for the offline path: Load configuration file on Add device, root element Konfigurace. It does not contain the IP or port — those stay as typed on the card.

You set on the equipment You enter in Protequ
DPE-2 IP IP address
Native TCP port from the config program Port (required, no default)
Master number on the display Address 1…16
Site time zone Time zone on the panel card (clock write + daily alarm counter)

Site extension — new devices

When groups or detectors are added on site, do not Add device again. That creates a second panel and a second session. DPE-2 will not accept both.

Use Import configuration on the panel that is already in the tree (live), or Devices → Import devices → Update existing with a new Konfigurace XML. IP, port and master stay as typed on the card.

Identity after import:

Element On the panel / in XML In Protequ (address)
Group, detector, input, output SysAdr the same SysAdr number
Display Adresa (01-01L-001…) label only not the Protequ address
Another master (tablo, second MHU) master 2…16 on the display a network-master child; its points sit under the panel in that master's address block

The wrong identity silently drives the neighbouring element. Always use SysAdr, never the printed loop label.

Inputs vs outputs. An output (relay or sounder) has Activate / Deactivate. An input is a contact on the terminals — Protequ cannot close it, only see it, bypass it and put it in test.

Confirm it works

  • The panel is connected. Firmware appears on the card (lab MHU 116 reports 20.2).
  • Import configuration fills the tree. Detectors and I/O show the same SysAdr as the configuration program.
  • Raise a harmless condition (a known fault, a test input). The state appears on the correct element, not as an unresolved device.
  • A reset on the panel or from Protequ clears a standing fire.
  • Unplug Ethernet: the card goes offline; after reconnect the states come back without adding the device again.

More on what was verified in the laboratory, and what not to treat as a failure: Notes.

If it does not connect

What you see What to do
The native port does not answer VLAN, firewall, wrong IP, or DPE-2 not loaded in the project
Port answers, Protequ stays Stopped Another client holds DPE-2 (demo software, config program, a second Protequ panel). Disconnect it
Connects, then drops after a few seconds Expected if the session goes silent — Protequ must stay running; do not probe the same port at the same time
Link is up, tree is empty Run Import configuration, or load a Konfigurace XML
Unresolved device The element is not in the tree. Import again (this also reads the tablo's points); restart the adapter if a command then reports "not found"
Clocks off by a whole hour Time zone on the panel card ≠ the site time zone
Reset / mute does nothing during T1 Current builds send those commands anyway; if the panel refuses, wait for the local acknowledgement window
Day / night button missing Expected — Protequ observes day/night from the panel, it does not set it
Test or Activate reports success but the element does not change Do not treat as a site fault yet — see Notes