• Our new ticketing site is now live! Using either this or the original site (both powered by TrainSplit) helps support the running of the forum with every ticket purchase! Find out more and ask any questions/give us feedback in this thread!

Why are AWS faults not being fixed? Anyone on here from network rail pway?

Status
Not open for further replies.

DriverScot

New Member
Joined
10 Feb 2026
Messages
3
Location
Scotland
Edit:

Thanks for clarifying it’s S&T not Pway that deals with these faults.

We have many AWS magnet faults both code 8 and 2 in our area that have been reported multiple times but network rail keep claiming no knowledge that they exist. Does anyone know what the hold up could be apart from the obvious trying not to spend money? Some faults are approaching 10 years old which is crazy!
 
Last edited:
Sponsor Post - registered members do not see these adverts; click here to register, or click here to log in
R

RailUK Forums

MadMac

Established Member
Joined
13 Jun 2008
Messages
1,504
Location
Moorpark, CA
We have many AWS magnet faults both code 8 and 2 in our area that have been reported multiple times but network rail keep claiming no knowledge that they exist. Does anyone know what the hold up could be apart from the obvious trying not to spend money? Some faults are approaching 10 years old which is crazy!
AWS is generally an S&T matter. Code 2 is “Horn instead of bell” and Code 8 is “Horn where there shouldn’t be anything”. Those should certainly be getting fixed.
 

Skeletor

Member
Joined
15 May 2024
Messages
22
Location
Glasgow
It’s not operating as intended but is failing safe.

There’s areas where there’s been AWS faults for years as you say. Maintenance don’t care and must not see it as a matter of priority as long as it fails safe.
 

swr444

Member
Joined
26 Aug 2008
Messages
1,018
Location
London
There are constant problems with AWS faults with the new signalling on the Windsor lines since it all went live a few years ago. There's been an AWS fault at Ascot P1 for over a year now; I have no idea why it still hasn't been fixed.
 

irish_rail

On Moderation
Joined
30 Oct 2013
Messages
4,530
Location
Plymouth
There's 4 of these faults (at least) between Taunton and Reading, and all have been faulty at least 6 months. In the past they'd be repaired pretty quickly, but about a year or two ago they just stopped fixing them. Can only assume budget cuts is the issue?
 

Falcon1200

Established Member
Joined
14 Jun 2021
Messages
6,608
Location
Neilston, East Renfrewshire
network rail keep claiming no knowledge that they exist.

This surprises me - Any such fault reported by a Driver to a Signaller should be passed to Control who would input this to the Fault Management System, unless it had been reported previously, in which case a note might be added? And local S&T management should certainly be chasing up long-outstanding faults.
 

Tomnick

Established Member
Joined
10 Jun 2005
Messages
5,933
We have many AWS magnet faults both code 8 and 2 in our area that have been reported multiple times but network rail keep claiming no knowledge that they exist. Does anyone know what the hold up could be apart from the obvious trying not to spend money? Some faults are approaching 10 years old which is crazy!
In some cases (albeit not this one, clearly), drivers not reporting such faults doesn't help, or in some cases being put under pressure to stop reporting them. I know of one where the S&T attended on numerous occasions in response to repeated reports; tested it and booked it back in order each time, but missed the actual fault that had been present in the interlocking somewhere since commissioning. The false-feeding that they'd done to the electromagnet to test it had bypassed the actual fault each time. It was only by a handful of drivers continuing to report it that eventually they came to realise that the fault still remained.

Another Code 2 fault was fixed remarkably quickly when the signallers became fed up of receiving reports from drivers, and took to keeping the distant signal in question at caution instead!
 

Gaelan

Member
Joined
3 Apr 2023
Messages
1,012
Location
Edinburgh
Maintenance don’t care and must not see it as a matter of priority as long as it fails safe.
Surely this is nonsense? If a given magnet has been incorrectly triggering a horn every time a train passes for years, drivers will almost certainly learn to ignore it and it might as well not be fitted.
 
Joined
15 Apr 2020
Messages
409
Location
Wakefield
Budget cuts and access issues mainly from my experience. There’s often just a 1 or 2 person fault team in my patch nowadays, and a right-side AWS fault is pretty low priority comparatively. They also require track access to fix, and usually signals clearing for testing etc so involve finding gaps in traffic etc.

All in it can lead to them being left for a long time.
 

172007

Member
Joined
2 Jan 2021
Messages
1,171
Location
West Mids
Budget cuts and access issues mainly from my experience. There’s often just a 1 or 2 person fault team in my patch nowadays, and a right-side AWS fault is pretty low priority comparatively. They also require track access to fix, and usually signals clearing for testing etc so involve finding gaps in traffic etc.

All in it can lead to them being left for a long time.
The very real danger is no one reports right side failures and leave it to the infrasturucure moniting trains to do it.
 

Tomnick

Established Member
Joined
10 Jun 2005
Messages
5,933
Surely this is nonsense? If a given magnet has been incorrectly triggering a horn every time a train passes for years, drivers will almost certainly learn to ignore it and it might as well not be fitted.
Absolutely. Fails "safe" is perhaps misleading in this context – it's not a wrong side failure, but it does introduce additional risks. The one that I referred to above, which was outstanding for many, many months, was for a distant signal which was followed by a short 30mph PSR and then a left-hand curve finally leading to the poorly-sighted stop signal. Most drivers would routinely check the AWS visual indication upon clearing the PSR, but naturally became used to disregarding it as the fault meant that it was always lit. I don't think anyone actually had a SPAD as a result, but a number of drivers came close to it.
 

Harpo

Established Member
Joined
21 Aug 2024
Messages
3,950
Location
Newport
Sounds like a job for ASLEF to start rattling cages. Routinely cancelling (and potentially ignoring) any AWS horn sets a safety trap based on assumptions, especially if any loss of situational awareness occurs.
 

DriverScot

New Member
Joined
10 Feb 2026
Messages
3
Location
Scotland
Sounds like a job for ASLEF to start rattling cages. Routinely cancelling (and potentially ignoring) any AWS horn sets a safety trap based on assumptions, especially if any loss of situational awareness occurs.
This is the exact point I’m making. The fault in isolation is not dangerous as it’s failing safe but may of these faults are located in busy areas with complex signalling. Also thanks to those pointing out it’s not pway that deals with this as I’m not familiar with the Network rail teams. The are also located in areas fitted with TPWS+ which is why I find the habitual cancellation of these magnets concerning.

== Doublepost prevention - post automatically merged: ==

AWS is generally an S&T matter. Code 2 is “Horn instead of bell” and Code 8 is “Horn where there shouldn’t be anything”. Those should certainly be getting fixed.
Thanks for clarifying this
It’s not operating as intended but is failing safe.

There’s areas where there’s been AWS faults for years as you say. Maintenance don’t care and must not see it as a matter of priority as long as it fails safe.
the problem being that whilst it is failing safe it is creating distraction risks and completely defeating the point of the AWS audible safety system, as drivers habitually cancel without thinking about why.

== Doublepost prevention - post automatically merged: ==

This surprises me - Any such fault reported by a Driver to a Signaller should be passed to Control who would input this to the Fault Management System, unless it had been reported previously, in which case a note might be added? And local S&T management should certainly be chasing up long-outstanding faults.
At first I thought fair enough perhaps other drivers are not reporting as they are fed up of having their time wasted with faults not being fixed. But I have personally reported via a RT3185 multiple times now and the latest was they claimed no knowledge which I know to be untrue.

== Doublepost prevention - post automatically merged: ==

In some cases (albeit not this one, clearly), drivers not reporting such faults doesn't help, or in some cases being put under pressure to stop reporting them. I know of one where the S&T attended on numerous occasions in response to repeated reports; tested it and booked it back in order each time, but missed the actual fault that had been present in the interlocking somewhere since commissioning. The false-feeding that they'd done to the electromagnet to test it had bypassed the actual fault each time. It was only by a handful of drivers continuing to report it that eventually they came to realise that the fault still remained.

Another Code 2 fault was fixed remarkably quickly when the signallers became fed up of receiving reports from drivers, and took to keeping the distant signal in question at caution instead!
This is the most frustrating aspect that instead of a proper fix it’s patch up to stop any more complaints. I also suspect in our area that because anytime an AWS magnet fault is fixed another fault closely follows in very close proximity as if the magnets are purely being switched instead of replaced. I have no technical knowledge to confirm this, it’s only my opinion looking in.

== Doublepost prevention - post automatically merged: ==

Absolutely. Fails "safe" is perhaps misleading in this context – it's not a wrong side failure, but it does introduce additional risks. The one that I referred to above, which was outstanding for many, many months, was for a distant signal which was followed by a short 30mph PSR and then a left-hand curve finally leading to the poorly-sighted stop signal. Most drivers would routinely check the AWS visual indication upon clearing the PSR, but naturally became used to disregarding it as the fault meant that it was always lit. I don't think anyone actually had a SPAD as a result, but a number of drivers came close to it.
Thanks for this as this is the angle I am looking at it too.
 
Last edited:

MadMac

Established Member
Joined
13 Jun 2008
Messages
1,504
Location
Moorpark, CA
Sounds like a job for ASLEF to start rattling cages. Routinely cancelling (and potentially ignoring) any AWS horn sets a safety trap based on assumptions, especially if any loss of situational awareness occurs.
As the old saying goes, “Familiarity breeds contempt”.
 

ChilliSauce

On Moderation
Joined
28 Oct 2025
Messages
328
Could be that either Network Rail don't have anyone left in the organisation who knows how to fix them or some manager (with a degree of course!) doesn't understand the significance of wrong announce.

Perhaps it needs the ORR to give NR an improvement notice ?
 

Annetts key

Established Member
Joined
13 Feb 2021
Messages
3,832
Location
West is best
Before I waffle on with some technical details, I need to explain that AWS can be installed in a number of configurations. For plain line on a unidirectional line, that's the simplest arrangement. Mounted in the four foot is a permanent magnet and immediately following it is an electro-magnet (as viewed in the direction of travel of the train). For bidirectional lines or reversible lines the arrangements depend on where the signals are and on a number of other requirements. If there is no opposite direction signal nearby, then the arrangement is normally a permanent magnet with a suppression coil (normally called a "suppressor") immediately followed by an electro-magnet (as viewed in the direction of travel of the train for the direction that the AWS applies for).

So, for code 2 faults, that's the train equipment being unable to detect the magnetic field from the AWS electro-magnet. It's the magnetic field from the permanent magnet with no further magnetic field being detected (from the electro-magnet) that causes the horn to sound in the cab.

There are lots of possible reasons why this may be the case.

Some of the most common are:
  • Problem with the train equipment (not applicable if multiple trains report the same signal or AWS),
  • Damaged cable to the electro-magnet in the four foot (fairly common fault),
  • Corroded wire or terminal in the terminal box for the electro-magnet (the terminal boxes sometimes fill up with water if the seal is not perfect, or it being deliberately disconnected because it's got water in it),
  • Defective electro-magnet (not a very common fault, unless affected by water, can last for twenty or more years),
  • The disconnection link left disconnected (either by mistake or intentional for some reason),
  • Defective main lineside cable,
  • Faulty power supply unit for the electro-magnet,
  • Blown fuse (sometimes caused by a defective diode),
  • Fault in the control system, such as a faulty relay contact.
The vast majority of these problems are not difficult to fix. some only require around one to two hours of work.

Code 8 faults are typically a failure of the suppression coil of a permanent AWS magnet. If the AWS applies to an opposite direction signal, then the suppression coil should cancel out the magnetic field from the permanent magnet so that nothing is received in the cab. There are some other circumstances where suppressor are used, but I won't go into the details here.

The list of problems is similar to those I listed for code 2 faults. However, unlike electro-magnets, some makes/designs of suppressors are not very reliable. Some lasting less than three years.

Some of the most common problems are:
  • Not technically a fault, but if an unsignalled move is made, because the signalling doesn’t know the intended direction, the suppressor may not be energised.
  • Damaged cable to the suppressor in the four foot (fairly common fault),
  • Corroded wire or terminal in the terminal box for the suppressor, (the terminal boxes sometimes fill up with water if the seal is not perfect, or it being deliberately disconnected because it's got water in it),
  • Defective suppressor,
  • The disconnection link left disconnected (either by mistake or intentional for some reason),
  • Defective main lineside cable,
  • Faulty power supply unit for the suppressor,
  • Blown fuse (sometimes caused by a defective diode),
  • Fault in the control system, such as a faulty relay contact.

Changing any of the magnets in the four foot is a bit of a ball ache, because they are heavy and awkward to handle. So the reduction of S&T teams from three people to just two does hinder the work.

But as others have said, the S&T being able to get access may delay fault investigation and therefore rectification. The testing required to fault find a code 8 fault may require a long length of the line to be clear of traffic for example. To test an electro-magnet, the signal must be cleared to a green. Both of these are difficult to do on a busy railway.

And alas, managers don't consider these faults to be urgent or important, as they don't bring trains to a halt or significantly delay trains. Other work therefore may take priority.

It's important that drivers keep reporting any defective equipment or any other defects. If you are concerned that something is not being fixed within a reasonable time, bring it to the attention of your manager. Your manager can make enquiries. All defects and faults should be reported by the signaller to fault control, where they will be logged in a computer system. The signaller is then given a fault number. Your company can ask Network Rail for details of a reported fault.

This is the exact point I’m making. The fault in isolation is not dangerous as it’s failing safe but may of these faults are located in busy areas with complex signalling. Also thanks to those pointing out it’s not pway that deals with this as I’m not familiar with the Network rail teams. The are also located in areas fitted with TPWS+ which is why I find the habitual cancellation of these magnets concerning.

== Doublepost prevention - post automatically merged: ==


Thanks for clarifying this

the problem being that whilst it is failing safe it is creating distraction risks and completely defeating the point of the AWS audible safety system, as drivers habitually cancel without thinking about why.

== Doublepost prevention - post automatically merged: ==


At first I thought fair enough perhaps other drivers are not reporting as they are fed up of having their time wasted with faults not being fixed. But I have personally reported via a RT3185 multiple times now and the latest was they claimed no knowledge which I know to be untrue.

== Doublepost prevention - post automatically merged: ==


This is the most frustrating aspect that instead of a proper fix it’s patch up to stop any more complaints. I also suspect in our area that because anytime an AWS magnet fault is fixed another fault closely follows in very close proximity as if the magnets are purely being switched instead of replaced. I have no technical knowledge to confirm this, it’s only my opinion looking in.

== Doublepost prevention - post automatically merged: ==


Thanks for this as this is the angle I am looking at it too.

Could be that either Network Rail don't have anyone left in the organisation who knows how to fix them or some manager (with a degree of course!) doesn't understand the significance of wrong announce.

Perhaps it needs the ORR to give NR an improvement notice ?

Edited to correct typos and the darn auto-correct / auto-suggest thing on my pad. Also some extra details added.
 
Last edited:
Joined
15 Apr 2020
Messages
409
Location
Wakefield
I can assure drivers that there is no one more annoyed than a signaller when every train reports a know fault, and it is signaller workload that then is affected, so the faults are constantly chased by operations.

This of course doesn’t specifically mean they get fixed but pushing from both sides is the best option.
 

Skeletor

Member
Joined
15 May 2024
Messages
22
Location
Glasgow
Before I waffle on with some technical details, I need to explain that AWS can be installed in a number of configurations. For plain line on a unidirectional line, that's the simplest arrangement. Mounted in the four foot is a permanent magnet and immediately following it is an electro-magnet. For bidirectional lines or reversible lines the arrangements depend on where the signals are and on a number of other requirements. If there is no opposite direction signal nearby, then the arrangement is normally a permanent magnet with a suppression coil (normally called a "suppressor") immediately followed by an electro-magnet.

So, for code 2 faults, that's the train equipment being unable to detect the magnetic field from the AWS electro-magnet. It's the magnetic field from the permanent magnet with no further magnetic field being detected (from the electro-magnet) that causes the horn to sound in the cab.

There are lots of possible reasons why this may be the case.

Some of the most common are:
  • Problem with the train equipment (not applicable if multiple trains report the same signal or AWS),
  • Damaged cable to the electro-magnet in the four foot (fairly common fault),
  • Corroded wire or terminal in the terminal box for the electro-magnet (the terminal boxes sometimes fill up with water if the deal is not perfect), or it being deliberately disconnected because it's got water in it),
  • Defective electro-magnet (not a very common fault, unless affected by water, can last for twenty or more years),
  • The disconnection link left disconnected (either by mistake or intentional for some reason),
  • Defective main lineside cable,
  • Blown fuse (sometimes caused by a defective diode),
  • Fault in the control system, such as a faulty relay contact.
The vast majority of these problems are not difficult to fix. some only require as on d one to two hours of work.

Code 8 faults are typically a failure of the suppression coil of a permanent AWS magnet. If the AWS applies to an opposite direction signal, then the suppression coil should cancel out the magnetic field from the permanent magnet so that nothing is received in the cab. There are some other circumstances where suppressor are used, but I won't go into the details here.

The list of problems is similar to those I listed for code 2 faults. However, unlike electro-magnets, some makes/designs of suppressors are not very reliable. Some lasting less than three years.

Changing any of the magnets in the four foot is a bit of a ball ache, because they are heavy and awkward to handle. So the reduction of S&T teams from three people to just two does hinder the work.

But as others have said, the S&T being able to get access may delay fault investigation and therefore rectification. The testing required to fault find a code 8 fault may require a long length of the line to be clear of traffic for example.

And alas, managers don't consider these faults to be urgent or important, as they don't bring trains to a halt or significantly delay trains. Other work therefore may take priority.

It's important that drivers keep reporting any defective equipment or any other defects. If you are concerned that something is not being fixed within a reasonable time, bring it to the attention of your manager. Your manager can make enquiries. All defects and faults should be reported by the signaller to fault control, where they will be logged in a computer system. The signaller is then given a fault number. Your company can ask Network Rail for details of a reported fault.
The reduction to two man teams may be having an impact because from my experience a suppressor should always be a three man lift. It can be done but it’s an awkward lift and there’s a risk of injury when you’ve only got two people.

I’ve opened a fair of the few termination boxes in a suppressor and its filled to the brim with water I’d say at least 50% of the time.
 

Belperpete

Established Member
Joined
17 Aug 2018
Messages
3,692
To add to what has already been said, there can also be interlocking problems that can affect the AWS indication, such as the interlocking switching off the electromagnet too soon, or switching the suppressor magnet on too late. These usually occur where the AWS magnets are close to track-circuit joints or axle counter boundaries. These type of problems are very difficult for trackside staff to diagnose, as when tested, the magnets will generate the correct magnetic field. And not every train may experience the problem, depending on how fast a particular train is travelling, and the type of train (and where it's AWS receiver is mounted).

Such problems are usually found a week or two after signalling alterations have been commissioned, when a succession of drivers all report the fault. However, they can also happen when linespeed is increased, a different type of stock starts using the line, or the type of AWS magnet is changed. I understand that some of the newer types of suppressor need to be switched on for longer before the field strength builds up sufficiently.

Unfortunately, as well as being difficult to diagnose, such problems can be difficult to rectify. It usually requires a change to the interlocking logic, which on today's railway is not something that can be achieved quickly, and not by the maintenance organisation. Alternatively, it may be possible to move the AWS magnet, but again this is not something that maintenance can do by themselves.
 

Krokodil

Established Member
Joined
23 Jan 2023
Messages
6,614
Location
Wales
One long-term code 8 fault up here was only rectified after one of our guards had a word with her Network Rail husband.
 

Surreytraveller

On Moderation
Joined
21 Oct 2009
Messages
3,905
Could be that either Network Rail don't have anyone left in the organisation who knows how to fix them or some manager (with a degree of course!) doesn't understand the significance of wrong announce.

Perhaps it needs the ORR to give NR an improvement notice ?
The ORR is full of people with degrees....
 

QueensCurve

Established Member
Joined
22 Dec 2014
Messages
2,203
Is AWS not redundant in the presence of TPWS?* Could the availability of TPWS be a reason to assign a lower priority to right side AWS faults?

* Moderators please assign to a new thread if considered appropriate.
 

Surreytraveller

On Moderation
Joined
21 Oct 2009
Messages
3,905
Is AWS not redundant in the presence of TPWS?* Could the availability of TPWS be a reason to assign a lower priority to right side AWS faults?

* Moderators please assign to a new thread if considered appropriate.
AWS is there to assist the driver. TPWS is there to sometimes mitigate the consequences when a driver makes a mistake.
 

skyhigh

Established Member
Joined
14 Sep 2014
Messages
6,901
Could be that either Network Rail don't have anyone left in the organisation who knows how to fix them
AWS is a very very very common system over the network. If there was nobody in NR who could fix it we'd be in huge trouble...
 

Annetts key

Established Member
Joined
13 Feb 2021
Messages
3,832
Location
West is best
Is AWS not redundant in the presence of TPWS?* Could the availability of TPWS be a reason to assign a lower priority to right side AWS faults?

* Moderators please assign to a new thread if considered appropriate.
TPWS is not a replacement for AWS. It's supplementary to it. They are different systems. The primary purpose of AWS is to try to reduce drivers missing distant signals (or for colour light signals, missing a single yellow or double yellow aspect).

The primary purpose of TPWS is to try and reduce the risk of collision (or the amount of damage if a collision does occur) if a train passes a fitted signal at red.

As others have said, TPWS is only fitted to a relatively small number of signals (normally those where there is a point of conflict ahead of the signal, such as points / junction or where other risk factors apply in the determination of the decision to fit it).

Attending and fixing AWS faults should be no different in priority to what it was in the past. The priority had not been changed when I led an S&T faulting and maintenance team back in 2023. As long as there were no train stopping or train delaying faults, we attended AWS faults as soon as we could. If the fault came in during my shift, we normally were on site within one hour. Depending on what we found, rectification could be within one hour, 24 hours or in some cases, the work would require planning at a weekend or night turn (due to a lack of access, because only then could a suitable line block or T3 worksite be obtained). If we were out of spare units, it sometimes took weeks, sometimes a couple of months though, that's because there was a lack of suppressors).
 

dk1

Veteran Member
Joined
2 Oct 2009
Messages
19,898
Location
East Anglia
Is AWS not redundant in the presence of TPWS?* Could the availability of TPWS be a reason to assign a lower priority to right side AWS faults?

* Moderators please assign to a new thread if considered appropriate.
Although both offer a safety aid they work in very different ways.

== Doublepost prevention - post automatically merged: ==

AWS is a very very very common system over the network. If there was nobody in NR who could fix it we'd be in huge trouble...
It’s often just the case that AWS is left inoperable for a period of time but always in the most restrictive (horn v bell) mode.
 
Last edited:
Status
Not open for further replies.

Top