Flashing yellows question

[PudseyBearHST]

Hello,

I’ve noticed when I’m being held for another train at a particular location is that when a train goes past the single flashing yellow, it then changes to green for a few seconds before turning to red. I can’t say I’ve ever seen or heard this before. Should this not be happening or is it because it is an older system or is this completely normal?

Thanks

 

[hexagon789]

Hello,

I’ve noticed when I’m being held for another train at a particular location is that when a train goes past the single flashing yellow, it then changes to green for a few seconds before turning to red. I can’t say I’ve ever seen or heard this before. Should this not be happening or is it because it is an older system or is this completely normal?

Thanks

It's just the aspect sequence where the aspect changes by more than one 'step' and the various relays open and close in turn until the appropriate aspect is obtained.

It varies signal to signal how noticeable it is sometimes.

 

[Railsigns]

I’ve noticed when I’m being held for another train at a particular location is that when a train goes past the single flashing yellow, it then changes to green for a few seconds before turning to red. I can’t say I’ve ever seen or heard this before. Should this not be happening or is it because it is an older system or is this completely normal?

That shouldn't normally happen, and it would be interesting to see the control tables for the signals involved.

I wonder if at this location the track circuit occupied condition that allows the junction signal's aspect to step up from single yellow is the same condition that replaces the aspect of the previous signal to red, and the momentary green aspect on the latter (which a driver would never see) is just down to timing.

 

[PudseyBearHST]

It's just the aspect sequence where the aspect changes by more than one 'step' and the various relays open and close in turn until the appropriate aspect is obtained.

It varies signal to signal how noticeable it is sometimes.

I know when signal steps up that you may see the entire sequence. But in this scenario, when the section becomes occupied, it changes from flashing yellow to green then red when it should go straight to red.

That shouldn't normally happen, and it would be interesting to see the control tables for the signals involved.

I wonder if at this location the track circuit occupied condition that allows the junction signal's aspect to step up from single yellow is the same condition that replaces the aspect of the previous signal to red, and the momentary green aspect on the latter (which a driver would never see) is just down to timing.

Ahh I see what you’re saying, that’s interesting to know! Thanks!

== Doublepost prevention - post automatically merged: 26 Oct 2023 ==

Does anyone else know of any oddities or anomalies with regard to flashing yellows around the network?

 

[Surreytraveller]

It could be that the train is approaching an approach controlled signal, and has just occupied the track circuit which causes the approach controlled signal to step up, which causes the signal you're seeing to step up to green.
Then a moment later the train occupies the track circuit that the signal is protecting, causing it to show danger

 

[PudseyBearHST]

It could be that the train is approaching an approach controlled signal, and has just occupied the track circuit which causes the approach controlled signal to step up, which causes the signal you're seeing to step up to green.
Then a moment later the train occupies the track circuit that the signal is protecting, causing it to show danger

That also makes sense, thanks. Interesting stuff

 

[Annetts key]

I’ve noticed when I’m being held for another train at a particular location is that when a train goes past the single flashing yellow, it then changes to green for a few seconds before turning to red. I can’t say I’ve ever seen or heard this before. Should this not be happening or is it because it is an older system or is this completely normal?

Is this signal an automatic signal?
The overlap for an automatic signal is typically 200yds to 300yds beyond the signal. Hence the signal will continue to show a proceed aspect with a train beyond it until said train occupies the track circuit or axle counter section beyond the overlap. This includes stepping up to the next aspect.

It’s perfectly safe, as the signal in rear will still be held at red because the train is still on the overlap track circuit (or in the axle counter section) until the whole train has made it beyond the overlap.

 

[PudseyBearHST]

Is this signal an automatic signal?
The overlap for an automatic signal is typically 200yds to 300yds beyond the signal. Hence the signal will continue to show a proceed aspect with a train beyond it until said train occupies the track circuit or axle counter section beyond the overlap. This includes stepping up to the next aspect.

It’s perfectly safe, as the signal in rear will still be held at red because the train is still on the overlap track circuit (or in the axle counter section) until the whole train has made it beyond the overlap.

Actually yes, when you put it that way, it makes perfect sense. It would be no different for a signal stepping from single yellow to double yellow, etc… just weird to see flashing single yellow to green as I would have thought the flashing yellow sequence would be maintained. Thanks for the reply.

 

[dk1]

Such a shame flashing yellows couldn’t appear at more locations so easily. Great signalling idea.

 

[Annetts key]

Actually yes, when you put it that way, it makes perfect sense. It would be no different for a signal stepping from single yellow to double yellow, etc…

Well, I mention it because we once had someone riding on a HST (to London Paddington) somewhere near the back of it, who reported that the signal changed up to a green after the front of the train had passed the signal. The signal in question was an automatic. About one and a half hours of my life taken up checking this out… Yes, the signal was working as designed and was working correctly.

 

[PudseyBearHST]

Well, I mention it because we once had someone riding on a HST (to London Paddington) somewhere near the back of it, who reported that the signal changed up to a green after the front of the train had passed the signal. The signal in question was an automatic. About one and a half hours of my life taken up checking this out… Yes, the signal was working as designed and was working correctly.

Lol

Wouldn’t this be the case for semi-automatic and controlled signals (tcb/axle counters) too as the signaller just sets the route and the colour of the signal just depends on how many sections are free/have been set in front of it? The particular signal I was referring to is a controlled signal on an NX panel.

 

[Annetts key]

Ahh, controlled signals generally operate slightly differently. They have a separate overlap track circuit (or axle counter) that starts between the signal post, or beyond the signal post by a couple of meters, and which goes to the overlap point (normally 200 to 300 yards beyond the signal).

As soon as the overlap track circuit (axle counter) detects a train, the signal should go to red. So the opportunity for the signal to change to a less restrictive aspect as a train passes it is very short. Depending on the design of the interlocking, the control that determines if a signal can show proceed or show red can take up to nearly one second to operate (this is by design and is required for another interlocking function).

But, there is an exception to the above. In areas where a train may routinely (*) stop with the front of the train just beyond the signal at a station, or where shunting or propelling moves may take place, again normally in the area of a station (*), “last wheel replacement” may be specified in the design.

* as anticipated to be requirements at the time of the design.

With “last wheel replacement”, the occupation of the overlap track circuit (axle counter) is ignored until either the next track circuit (axle counter) is occupied, or the previous track circuit (axle counter), often called the berth track circuit (axle counter) goes clear.

With semi-automatic signals, it depends on the actual design. Some are more like automatic signals that can be put to red (not to be confused with automatic signals with either a signal post emergency replacement switch or where the signaller has an emergency replacement switch). Others are more like controlled signals, but which normally work automatically. Not to be confused with controlled signals that have an automatic working control.


In the above diagram, for an automatic signal section, the berth and and overlap (as shown for example, for block A123) are all a single track circuit (or axle counter) section as far as the interlocking is concerned. There is also a single indication for the signaller.
But for a controlled signal, the berth and overlap track circuits (or axle counter sections) are separate as far as the interlocking is concerned. They are also indicated separately to the signaller.

 

[PudseyBearHST]

Ahh, controlled signals generally operate slightly differently. They have a separate overlap track circuit (or axle counter) that starts between the signal post, or beyond the signal post by a couple of meters, and which goes to the overlap point (normally 200 to 300 yards beyond the signal).

As soon as the overlap track circuit (axle counter) detects a train, the signal should go to red. So the opportunity for the signal to change to a less restrictive aspect as a train passes it is very short. Depending on the design of the interlocking, the control that determines if a signal can show proceed or show red can take up to nearly one second to operate (this is by design and is required for another interlocking function).

But, there is an exception to the above. In areas where a train may routinely (*) stop with the front of the train just beyond the signal at a station, or where shunting or propelling moves may take place, again normally in the area of a station (*), “last wheel replacement” may be specified in the design.

* as anticipated to be requirements at the time of the design.

With “last wheel replacement”, the occupation of the overlap track circuit (axle counter) is ignored until either the next track circuit (axle counter) is occupied, or the previous track circuit (axle counter), often called the berth track circuit (axle counter) goes clear.

With semi-automatic signals, it depends on the actual design. Some are more like automatic signals that can be put to red (not to be confused with automatic signals with either a signal post emergency replacement switch or where the signaller has an emergency replacement switch). Others are more like controlled signals, but which normally work automatically. Not to be confused with controlled signals that have an automatic working control.

View attachment 145436
In the above diagram, for an automatic signal section, the berth and and overlap (as shown for example, for block A123) are all a single track circuit (or axle counter) section as far as the interlocking is concerned. There is also a single indication for the signaller.
But for a controlled signal, the berth and overlap track circuits (or axle counter sections) are separate as far as the interlocking is concerned. They are also indicated separately to the signaller.

Interesting, thanks.

 

[Mag_seven]

A gentle reminder that posts in this thread should be confined to discussing "Flashing Yellows". Thanks.

 

[dk1]

I’m hoping that when Haughley Junction gets redoubled that flashing yellows will be installed on the down main to assist the frequent freight services taking the Bury line. It’s a currently a curved uphill climb where the signal protecting the junction is only visible around 1/4 mile beforehand. It would greatly assist with efficiency and flow.

 

[Signal Head]

One reason could be that the same track circuit, when occupied, both releases the junction signal and replaces the signal in rear to Red (although you say the stepped up aspect is displayed for a few seconds, I've seen similar behaviour in the past but the change was much quicker).

The observed aspects *may* mean that the controls are arranged in such a way that they are susceptible to a wrong side failure whereby the junction signal 'sees' the approach track circuit as occupied, so it clears, whilst the rear signal 'sees' the same track circuit as clear (at the same time), so remains Off (the track circuit is actually clear, the fault causing the junction signal to clear is in the repeating circuits to or within the interlocking).


The technical term for this failure scenario is "Raynes Park Control", it is supposed to be checked and tested for these days to ensure it cannot happen, but there will have been many instances in older installations which predated the realisation of the problem. I've found a few in the past and was able to have them corrected by amending the circuit design. It's generally associated with relay-based systems, though I suspect it's possible to have the same in SSI under certain circumstances.


This fault isn't limited to FY sequences, I know of one location where the same 'step up and revert' behaviour could be seen regularly, and that was a simple 'Approach Release from Red'. After some signalling renewals took place the behaviour changed such that the rear signal went straight back to Red as the train passed it.

It would be interesting to know where this one is, and whether it's relay or SSI system, then it could be looked into.

 

[PudseyBearHST]

One reason could be that the same track circuit, when occupied, both releases the junction signal and replaces the signal in rear to Red (although you say the stepped up aspect is displayed for a few seconds, I've seen similar behaviour in the past but the change was much quicker).

The observed aspects *may* mean that the controls are arranged in such a way that they are susceptible to a wrong side failure whereby the junction signal 'sees' the approach track circuit as occupied, so it clears, whilst the rear signal 'sees' the same track circuit as clear (at the same time), so remains Off (the track circuit is actually clear, the fault causing the junction signal to clear is in the repeating circuits to or within the interlocking).


The technical term for this failure scenario is "Raynes Park Control", it is supposed to be checked and tested for these days to ensure it cannot happen, but there will have been many instances in older installations which predated the realisation of the problem. I've found a few in the past and was able to have them corrected by amending the circuit design. It's generally associated with relay-based systems, though I suspect it's possible to have the same in SSI under certain circumstances.

Thanks for the reply. That’s my thoughts too. My assumption would have been that the flashing yellow sequence would be maintained because the junction signal has the junction indicator lit and route set for a diverging path so it seems wrong that the previous signal is allowed to show anything less restrictive than one flashing yellow, albeit even briefly, when the route is set for a diverging route.
A signal stepping up is fine when it‘s for the straight route which several users posted about and thought it somehow answered my question but IMO it shouldn’t happen when it’s for a diverging route whether it’s for flashing yellows or approach release as you say. Thanks again for the reply and glad to hear that it mirrors my initial thoughts so it is technically a fault and not something that should be happening ideally.

 

[Signal Head]

Thanks for the reply. That’s my thoughts too. My assumption would have been that the flashing yellow sequence would be maintained because the junction signal has the junction indicator lit and route set for a diverging path so it seems wrong that the previous signal is allowed to show anything less restrictive than one flashing yellow, albeit even briefly, when the route is set for a diverging route.
A signal stepping up is fine when it‘s for the straight route which several users posted about and thought it somehow answered my question but IMO it shouldn’t happen when it’s for a diverging route whether it’s for flashing yellows or approach release as you say. Thanks again for the reply and glad to hear that it mirrors my initial thoughts so it is technically a fault and not something that should be happening ideally.

Under normal circumstances it will never step up with the train on approach, as the junction signal doesn't step up until the train has passed the FY, so there's little harm in the step-up itself as the driver will never see it.

If the Raynes Park fault is present though, there is the possibility that a right side failure (eg a blown fuse on the TC repeating circuit) will 'trick' the junction signal into releasing early, before the train is actually approaching. The problems come when that circuit isn't proved in the rear signal, so each signal is seeing the track circuit as being in a different state, each of which allows both signals to clear together, which you don't want for a diverging route.

Are you able to share the location (by PM if preferred), I might be able to have a look, just out of interest?

 

[PudseyBearHST]

Under normal circumstances it will never step up with the train on approach, as the junction signal doesn't step up until the train has passed the FY, so there's little harm in the step-up itself as the driver will never see it.

For sure which is why I have never reported it to the signaller. Just thought it was odd.

Also did you mean wrong side failure or is it right side?

== Doublepost prevention - post automatically merged: 28 Oct 2023 ==

Is this signal an automatic signal?
The overlap for an automatic signal is typically 200yds to 300yds beyond the signal. Hence the signal will continue to show a proceed aspect with a train beyond it until said train occupies the track circuit or axle counter section beyond the overlap. This includes stepping up to the next aspect.

It’s perfectly safe, as the signal in rear will still be held at red because the train is still on the overlap track circuit (or in the axle counter section) until the whole train has made it beyond the overlap.

Looking back at this again, I don’t think I got my point across correctly. I completely get why a signal steps up and have no problem with that when it’s for a straight route and I am not challenging this at all. It happens all the time and I have a decent understanding of how track circuits work. But it’s the diverging route aspect which seems odd to me.

Thanks for the reply. That’s my thoughts too. My assumption would have been that the flashing yellow sequence would be maintained because the junction signal has the junction indicator lit and route set for a diverging path so it seems wrong that the previous signal is allowed to show anything less restrictive than one flashing yellow, albeit even briefly, when the route is set for a diverging route.
A signal stepping up is fine when it‘s for the straight route which several users posted about and thought it somehow answered my question but IMO it shouldn’t happen when it’s for a diverging route whether it’s for flashing yellows or approach release as you say. Thanks again for the reply and glad to hear that it mirrors my initial thoughts so it is technically a fault and not something that should be happening ideally.

 

[Signal Head]

For sure which is why I have never reported it to the signaller. Just thought it was odd.

Also did you mean wrong side failure or is it right side?

== Doublepost prevention - post automatically merged: 28 Oct 2023 ==

It's both, ie if the Raynes Park fault is present in the wiring, a right side failure of the repeat relay circuit sending the track circuit state to the interlocking containing the junction signal causes it to see the track circuit as 'Occupied', so it 'thinks' the train has arrived and clears the junction signal. However, if the signal in rear gets its information from a different repeat relay circuit (of the same track circuit), which hasn't failed, then it 'thinks' the line ahead is clear (which it is), so that signal clears (or stays clear).

Now the Approach Release has been defeated as you can get Green aspects leading up to a junction signal which has cleared prematurely, with no advance warning for the driver that a diverging route is set up ahead. That's a wrong side failure.

If the track circuit itself is occupied, either by a train or because of a fault, then both signals see it as occupied and while the junction signal can still clear, the signal in rear will be held at Red, so there is no problem.

Any signalling introduced or altered within the last (say) 30 years, if not more, ought to have had this scenario taken into account in the design and testing so the failure shouldn't be present.

 

[PudseyBearHST]

It's both, ie if the Raynes Park fault is present in the wiring, a right side failure of the repeat relay circuit sending the track circuit state to the interlocking containing the junction signal causes it to see the track circuit as 'Occupied', so it 'thinks' the train has arrived and clears the junction signal. However, if the signal in rear gets its information from a different repeat relay circuit (of the same track circuit), which hasn't failed, then it 'thinks' the line ahead is clear (which it is), so that signal clears (or stays clear).

Now the Approach Release has been defeated as you can get Green aspects leading up to a junction signal which has cleared prematurely, with no advance warning for the driver that a diverging route is set up ahead. That's a wrong side failure.

If the track circuit itself is occupied, either by a train or because of a fault, then both signals see it as occupied and while the junction signal can still clear, the signal in rear will be held at Red, so there is no problem.

Any signalling introduced or altered within the last (say) 30 years, if not more, ought to have had this scenario taken into account in the design and testing so the failure shouldn't be present.

Fascinating, thanks