Signalling Confusion

[DelW]

Here's a photo of the co-actor at Thetford in days gone by

Thetford Signal Box 3

Thetford box and co-acting signal number 6 seen on a glorious day with just 9 days of their working lives left. The signal has two arms for sighting purposes. A McKenzie and Holland box of 33 levers built in 1883 working absolute block to Brandon and Harling Road.

www.flickr.com

Completely off-topic and pedantic, but that's a highly ungrammatical sign on that photo. I know what it's meant to say, but it's quite hard to work out what it's actually saying, it seems to say that passengers must only go along the footbridge as far as is necessary to cross the line, and no further. Then there's the "greengrocer's apostrophe" .....

 

[midland1]

It could also mean you walk over the tracks by the side of the footbridge not going over the bridge .

 

[John Webb]

Completely off-topic and pedantic, but that's a highly ungrammatical sign on that photo. I know what it's meant to say, but it's quite hard to work out what it's actually saying, it seems to say that passengers must only go along the footbridge as far as is necessary to cross the line, and no further. Then there's the "greengrocer's apostrophe" .....

It could also mean you walk over the tracks by the side of the footbridge not going over the bridge .

More recent photos of the station on the Geograph website - see Photos around Thetford station - show that the aberent apostrophe has been removed. Some of the photos also appear to show disused gates close to the bottom of the footbridge stairs, suggesting there was once a public footcrossing before the footbridge, which may explain the odd siting of the notice. Some pictures show the previous co-acting arms and some replacement LED signal.

 

[XAM2175]

Then there's the "greengrocer's apostrophe" .....

the aberent apostrophe has been removed...

Now if we could just start removing them from posts around here!

 

[Efini92]

The signal before a distant will only show a red, green or in some cases 2 yellows.

There are some anomalies on recently installed flashing yellows where momentarily you can get a yellow going onto a distant, but it will step up to a green before you reach it.

 

[Belperpete]

The signal before a distant will only show a red, green or in some cases 2 yellows.

There are some anomalies on recently installed flashing yellows where momentarily you can get a yellow going onto a distant, but it will step up to a green before you reach it.

Interesting. Modern solid state interlockings cause a signal to show yellow for about a second before stepping up to a higher aspect, just in case whatever is allowing that signal to clear is also causing the forward signal to be replaced. However, that shouldn't apply to a route that can only display a green aspect, even if the signal has a yellow aspect for other routes.

I am not clear about the situation you are describing. Are you saying that in certain situations the signal in rear can display flashing yellows for the route toward the distant? Or that the distant can display flashing yellow?

 

[MarkyT]

I don't claim to be a signalling expert but I'm a bit confused by how the term 'co-acting signal' is being used.
Where sighting was restricted by, for example, an overbridge a semaphore arm at drivers' eye level would be duplicated by another arm higher up on the same post to give sighting, at a distance, over the obstruction. Both arms were connected and therefore always displayed the same aspect, and this arrangement was called 'co-acting'. I realise that this arrangement is probably extinct being presumably replaced by banner repeaters in the modern world, but there was, for example, one on the Down Fast at Bletchley until power signalling.

There are some examples of colour light co-acting signals shown on this page, including the miniature 'close up' head at Penzance PZ70 which reminded me of the small low level road traffic light repeaters I'd seen in Paris back in the 1970s (don't know if they still have them)

Colour Light Signals

 

[MarkyT]

Interesting. Modern solid state interlockings cause a signal to show yellow for about a second before stepping up to a higher aspect, just in case whatever is allowing that signal to clear is also causing the forward signal to be replaced. However, that shouldn't apply to a route that can only display a green aspect, even if the signal has a yellow aspect for other routes.

I am not clear about the situation you are describing. Are you saying that in certain situations the signal in rear can display flashing yellows for the route toward the distant? Or that the distant can display flashing yellow?

Always a risk of it where there's a yellow where 'climbing aspects' relay circuitry provided though with filaments unlikely to heat up to visible illumination level before the feed was cut off by the DR picking. Might be a LED thing with a split second step up blink possible.

 

[edwin_m]

I think what @Belperpete is referring to is when, taking the example of a three-aspect area with original Solid State Interlocking, the second signal changes from red to yellow but the yellow lamp is defective. However because of the way the interlocking polls its trackside modules the loss of lamp proving won't be reported back for nearly a second, during which time the first signal will have cleared to green (other conditions permitting). It then remains at green for another period of nearly a second before dropping back to red as the loss of lamp proving makes its way through the logic. I think it could even be twice as long in some circumstances depending on how the ordering of the data affects timing issues. Hence the SSI software was changed so it always steps through yellow on the way to a less restrictive aspect.

I'm not an expert on relay circuits but I imagine something similar could happen here. However the relay circuit isn't polled so it reacts much faster and the incorrect green would be energized for a much shorter time.

 

[Efini92]

Interesting. Modern solid state interlockings cause a signal to show yellow for about a second before stepping up to a higher aspect, just in case whatever is allowing that signal to clear is also causing the forward signal to be replaced. However, that shouldn't apply to a route that can only display a green aspect, even if the signal has a yellow aspect for other routes.

I am not clear about the situation you are describing. Are you saying that in certain situations the signal in rear can display flashing yellows for the route toward the distant? Or that the distant can display flashing yellow?

Sorry it was badly worded. Basically after having flashing yellows the junction signal with the route indicator will display a yellow momentarily before stepping up to a green, even though the signal after the junction is a distant.

 

[Tomnick]

Interesting. Modern solid state interlockings cause a signal to show yellow for about a second before stepping up to a higher aspect, just in case whatever is allowing that signal to clear is also causing the forward signal to be replaced. However, that shouldn't apply to a route that can only display a green aspect, even if the signal has a yellow aspect for other routes.

I am not clear about the situation you are describing. Are you saying that in certain situations the signal in rear can display flashing yellows for the route toward the distant? Or that the distant can display flashing yellow?

We have a couple of examples where the junction signal is approach released from yellow with flashing yellows in rear, but where the first signal along the diverging route is a distant signal. You see (by design, presumably) the yellow before it steps up to green once you’re past a certain point, so it definitely happens! A colleague is insistent that, on one occasion, that the junction signal didn’t step up to green for some reason so he ended up passing the signal with a single yellow reading onto a distant signal.

 

[Efini92]

We have a couple of examples where the junction signal is approach released from yellow with flashing yellows in rear, but where the first signal along the diverging route is a distant signal. You see (by design, presumably) the yellow before it steps up to green once you’re past a certain point, so it definitely happens! A colleague is insistent that, on one occasion, that the junction signal didn’t step up to green for some reason so he ended up passing the signal with a single yellow reading onto a distant signal.

Was it in an area controlled by a ROC? I’ve noticed strange things happening in their areas. One is transitioning from a PSB with 4aspect to a ROC with 3 aspect. I’ve had a green at the 4 aspect and the 3 has been a yellow that stepped up on approach. It’s not a low speed diverging junction either, the linespeed actually increases.

 

[matchmaker]

Here's a photo of the co-actor at Thetford in days gone by

Thetford Signal Box 3

Thetford box and co-acting signal number 6 seen on a glorious day with just 9 days of their working lives left. The signal has two arms for sighting purposes. A McKenzie and Holland box of 33 levers built in 1883 working absolute block to Brandon and Harling Road.

www.flickr.com

Still one at Greenloaning.

 

[Tomnick]

Was it in an area controlled by a ROC? I’ve noticed strange things happening in their areas. One is transitioning from a PSB with 4aspect to a ROC with 3 aspect. I’ve had a green at the 4 aspect and the 3 has been a yellow that stepped up on approach. It’s not a low speed diverging junction either, the linespeed actually increases.

One’s under a ROC, one’s under a small workstation-based SCC, and I’m pretty sure there’s another example worked by a small (Absolute Block) box with a panel. I can give you at least one example of a green (on a four aspect signal) reading to a yellow - with, of course, adequate braking distance from there to the red.

 

[MarkyT]

We have a couple of examples where the junction signal is approach released from yellow with flashing yellows in rear, but where the first signal along the diverging route is a distant signal. You see (by design, presumably) the yellow before it steps up to green once you’re past a certain point, so it definitely happens! A colleague is insistent that, on one occasion, that the junction signal didn’t step up to green for some reason so he ended up passing the signal with a single yellow reading onto a distant signal.

That's the legitimate case I'd forgotten! The junction signal should remain at yellow until a point on approach at which the junction indicator is clearly readable. If you've already timed out on the berth after having approached the junction signal at red, then it should clear to green immediately where the next signal in advance on the route set is a distant, but I can definitely understand timing issues causing the yellow to be briefly stepped through in the process. Almost like a 'get ready to go' amber at a road traffic light! Perhaps most likely to occur at an interlocking boundary, where processing may take an extra major cycle or two?

 

[Tempest Malice]

Was it in an area controlled by a ROC?

Where control is from shouldn't change the on the ground signalling behaviour. only what interlocking is used should change that; Now of course there is a correlation that most areas controlled from a ROC have newer interlocking technologies, but there is no requirement for this, Several areas have migrated to a ROC with the actual signalling being completely unchanged.

A colleague is insistent that, on one occasion, that the junction signal didn’t step up to green for some reason so he ended up passing the signal with a single yellow reading onto a distant

There are failure modes which can cause this, though I believe that typically on a flashing aspect divergence (irrelevant of if the next signal is a distant), the controlling signal is typically to be held at yellow until the train "should" be down to the required speed (based on time spent occupying certain track circuits on the approach), so the most likely way in which the signal remains at yellow is actually if the train passing it has not slowed down enough yet, though there are definitely other ways for the equipment to fail to the same effect even with the trains slowing correctly, though given that most of the ways this could be seen involve wrong side failures of that last track circuit(s) they aren't all that common.

 

[4F89]

One’s under a ROC, one’s under a small workstation-based SCC, and I’m pretty sure there’s another example worked by a small (Absolute Block) box with a panel. I can give you at least one example of a green (on a four aspect signal) reading to a yellow - with, of course, adequate braking distance from there to the red.

I know a 4 aspect diverging signal that goes from red to green with no1 on approach release, to fixed distant, to stop board.

 

[edwin_m]

I know a 4 aspect diverging signal that goes from red to green with no1 on approach release, to fixed distant, to stop board.

Seems OK to me, as long as there is braking distance between the fixed distant and the stop board - in fact it would be wrong to display a yellow on the 4-aspect in that situation.

 

[4F89]

Indeed. More for an interesting variation thats all

 

[Merle Haggard]

Thank you to everyone who responded to my query about co-acting semaphore signals, all interesting.

When I was a youthful spotter it was easy to get a grasp of the principles of signalling. On a station or in a field close to a 'box one could hear (particularly in Summer, with the windows open) the bell codes and the clank of levers being pulled, and see the rodding or wire moving and what it worked. Travelling by train then, it was easy to have a forward view, and signals were generally easy to understand (apart from yellow arm ground signals, which puzzled me for a while).

In those days, I had noticed the method Edwin_m mentioned (speed of signal arm movement ) - sometimes accompanied by the signalman's own arm movements if there was a tight margin. Happy days.

 

[edwin_m]

Actually I'd say the principles of track circuit block and colour light signaling are far simpler than those of Absolute Block and semaphores. There are so many complexities and apparent oddities such as the principle of one distant applying to several stop signals (except when it doesn't) and all the extra controls added to the basic system in later years to protect against human errors.

So while TCB is less interesting it's also easier to learn. While textbooks such as the Kitchenside and Williams ones (again from distant memory on my part!) tend to go through several chapters on AB first, its rarity these days as well as its complexity means I'd probably recommend anyone wanting to learn about British signaling to ignore it until they fully understand TCB.

 

[Merle Haggard]

Actually I'd say the principles of track circuit block and colour light signaling are far simpler than those of Absolute Block and semaphores. There are so many complexities and apparent oddities such as the principle of one distant applying to several stop signals (except when it doesn't) and all the extra controls added to the basic system in later years to protect against human errors.

So while TCB is less interesting it's also easier to learn. While textbooks such as the Kitchenside and Williams ones (again from distant memory on my part!) tend to go through several chapters on AB first, its rarity these days as well as its complexity means I'd probably recommend anyone wanting to learn about British signalling to ignore it until they fully understand TCB.

Thank you for the observations.

One of my spotting stations (talking long ago...) had a 'box at each end of the station, and I never did work out which was the home and which the starter for each box - or indeed, whether all signals were actually present: reality is often not as straightforward as the theory!
One of me friends drives for SE Trains and he's mentioned that newer drivers (whose understanding of semaphores and block signalling is possibly limited to knowing which ones you have to stop at) are un-nerved by the apparent randomness of the location of semaphores - you go for miles without seeing one, then there's a cluster.
You may also recollect the Grand Central driver who encountered home over distant, showing stop and caution respectively, at night. What he saw was, of course, red over yellow; applying logic he assumed that only meant some sort of caution so carried on at pretty much full speed - I think a LC may have been involved (there was an RAIB report).
So I can see how the observation you make, that TCB is easier to understand, is very true.

 

[edwin_m]

Thank you for the observations.

One of my spotting stations (talking long ago...) had a 'box at each end of the station, and I never did work out which was the home and which the starter for each box - or indeed, whether all signals were actually present: reality is often not as straightforward as the theory!
One of me friends drives for SE Trains and he's mentioned that newer drivers (whose understanding of semaphores and block signalling is possibly limited to knowing which ones you have to stop at) are un-nerved by the apparent randomness of the location of semaphores - you go for miles without seeing one, then there's a cluster.
You may also recollect the Grand Central driver who encountered home over distant, showing stop and caution respectively, at night. What he saw was, of course, red over yellow; applying logic he assumed that only meant some sort of caution so carried on at pretty much full speed - I think a LC may have been involved (there was an RAIB report).
So I can see how the observation you make, that TCB is easier to understand, is very true.

Might even have had the home for one box being the same signal as the starter for the other … there were a few instances of stop signals being slotted when the boxes were really close.

I can understand the casual observer being confused by semaphores but it's worrying if drivers who sign the route are similarly confused. Here's the report you refer to:

RAIB Bulletin 03/2013: Signal NW36 passed at danger at Norton-on-Tees West, 16 January 2013 :: The Railways Archive

Archive of UK railways documents

www.railwaysarchive.co.uk

As train 1A60 approached signal NW36, the driver looked for the right-hand pair of signals (for the route to Norton-on-Tees South) and focused on the yellow light on the distant signal. Even though the driver recognised that signal NW36 was showing a red light (ie ‘stop’), the driver was thinking ahead to the next stop signal towards Norton-on-Tees South.

Some pretty sloppy working by the signalers in that report too.

 

[driver9000]

I always enjoyed working through AB semaphore areas and it brought a bit of character to the routes. We had short sections (section signal mounted with the distant for the box in advance), multiple home signals, level crossings, stacked shunting discs, stepped junction signals and Intermediate block signals all of which made things interesting. I work exclusively in TCB areas now so it's all a bit samey really. When I was a trainee driver the first thing they taught us in signalling was Absolute Block because everything else has evolved from it and is much easier to understand when you have your head around how AB works.