Skip to content
English - United Kingdom
Contact us
  • There are no suggestions because the search field is empty.

Why do several devices on my addressable loop report random communication faults?

Random faults on several addresses on one loop usually point to a loop-level fault, not a single device. Disable the affected zones(if necessary) and notify the alarm receiving centre before isolating anything.

Direct answer

Systematic isolation and testing will distinguish between a faulty device and a loop-level fault on an addressable loop. Removing a device or splitting the loop takes automatic detection out of service, so before you do either, agree the work with the responsible person for the premises, disable the affected zones or loop at the panel, and notify any alarm receiving centre. Confirm the panel configuration (device list) matches the physical devices and upload the configuration file if required, test the suspect device in a mini-loop connected directly to the panel, re-learn devices after isolation, and use a divide-and-conquer approach on long loops to narrow the physical location of communication problems. Check terminations and joint work carefully once the problematic segment is identified.

Applies to
  • Products / Models: Hochiki ESP analogue addressable loop devices (ALN-EN, ACD-EN, ATJ-EN, ACC-EN sensors; CHQ- series modules; HCP- series manual call points) on a Hochiki ESP-protocol addressable loop. Confirm your panel range with Hochiki Europe Technical Support if you are unsure it uses the ESP protocol.
  • Firmware / Version: All versions
  • Region: UK / Europe
  • Standards: BS 5839-1 (system integrity; maintenance and modification); BS EN 54-2 (control and indicating equipment); BS EN 54-17 (short-circuit isolators)
Before you isolate anything

Removing devices or splitting a loop takes automatic detection out of service for that section of the system. BS 5839-1 recommends that building occupants, and any alarm receiving centre to which the system reports, are notified before testing or maintenance that might generate a fire alarm signal, and lists testing or maintenance without appropriate disablement or warning as a recognised cause of unwanted fire alarm signals. Before starting:

  • Agree the work and its duration with the responsible person for the premises.
  • Notify the alarm receiving centre and disable the transmission path if fire signals are monitored remotely.
  • Disable the affected zones or loop at the panel rather than leaving them in an unknown state, and note which areas are unprotected while you work.
  • Arrange interim measures (for example a fire watch) if a large area will be without detection.
  • Reinstate detection and remote monitoring as soon as the work is complete, verify the reinstatement, and record the work and any faults in the system logbook.
Symptoms
  • Random communication errors are reported against several different addresses on the same loop, rather than consistently against one address.
  • One or more devices are missing when the panel runs auto-learn, even though they are physically present and wired correctly.
  • A fault appears to clear after a panel reset but returns.
Likely causes

Check these in priority order:

  1. Loose, over-tightened or incorrectly terminated connections at call points and junctions, or excessive cable length.
  2. A short-circuit isolator that has operated and is isolating devices beyond it from the panel.
  3. A single device transmitting corrupted data, affecting communication for other addresses sharing the loop.
  4. The panel configuration (device list) not matching the physical devices on the loop.
  5. Loop length or device count approaching the protocol's address capacity.
  6. An Earth fault is present (most likely water ingress)
Resolution / Steps

Confirm the panel configuration matches the physical devices

  • Verify that the panel configuration (device list) matches the physical devices on the loop, and inspect visible cabling and terminations for obvious damage or loose connections before starting more invasive isolation procedures.
  • If you re-upload the panel configuration, treat it as a modification to the system rather than a diagnostic step. BS 5839-1 recommends that the user agrees the change in writing, that at least one device on every circuit is tested after the control equipment has been modified, that further random testing is carried out where software has changed, and that a modification certificate is issued on completion.

Test a suspect device in a mini-loop

Follow the notification and disablement steps above before you begin.

  1. Remove the suspect device from the main loop.
  2. Connect the device and its base directly to the panel (create a small "mini loop").
  3. Put the panel into learn/auto-learn mode for that device.
  4. Observe whether the device appears and communicates correctly on the panel.
  5. If the device works correctly in the mini-loop, the issue is likely in the main loop cabling or another device; if it fails, replace or repair the device.

Re-learn a device on the main loop

Follow the notification and disablement steps above before you begin.

  1. Take the device out of the loop and put the panel into auto-learn.
  2. Reintroduce the device at one end of the loop and perform auto-learn.
  3. If required, try relocating the device within a different area or at the other end of the loop and repeat auto-learn to confirm behaviour.

Narrow a fault on long loops by divide-and-conquer

Follow the notification and disablement steps above before you begin. Confirm the address capacity of your panel first (1 to 127 as standard, up to 254 with extended addressing), because it sets how many halving steps you need.

  1. Remove one end of the loop from the panel so the loop is effectively split (works for radial or advanced loop configurations).
  2. If communication issues persist on one half, the fault lies within that half.
  3. Continue halving the problematic section until the faulty segment is localised. A Hochiki ESP loop uses addresses 1 to 127 as standard, and up to 254 where the panel supports the extended addressing feature. Check which applies on your panel first: on a loop of 127 addresses, halving works down from roughly 60 devices, then 30, then 15; on a loop of 254 it is one step longer.
  4. Inspect devices and cabling within the identified segment particularly for water ingress. Measure for an earth fault by using standard practices for resistance readings.

Check wiring and call points in the identified segment

  • Check that terminals within call points and junctions are correctly terminated and not excessively loose or over-tightened, and that cable lengths are within specification.
  • Under BS 5839-1, a single open or short circuit fault should not disable detection over more than 2,000 m² or more than one floor plus five devices above and below — so a fault that takes out more than that indicates a wiring or isolator problem, not simply a failed device.
When to contact technical support

Contact Hochiki Europe Technical Support if the fault persists after isolation and re-learn testing, or if you are unsure how to safely divide a loop for divide-and-conquer testing on your system.

Safety note

This article is general guidance for Hochiki Europe products. Design, installation, commissioning, maintenance and modification of fire detection and alarm systems should be carried out by, or under the responsibility of, a competent person in accordance with the applicable standards (for example BS 5839-1 in the UK). Statutory requirements imposed by the relevant enforcing authority also apply. Always refer to the current product manual and data sheet for the exact model before carrying out any work.

References

Last reviewed: 2026-10-21 — Reviewed by: Hochiki Europe Technical Support