Manual interlocking?

[Overmod]

I propose that we at least think about adopting some conventions for naming things, as poor LO is now getting visibly confused by the terminology.

A tower with mechanical 'armstrong levers' actuating pipes and bellcranks is NOT yet an 'interlocking'.  There has to be some method of ensuring that routes, and not just individual switch positions, can be set. 

On any armstrong plant built before 1929, this had to be done with 'mechanical interlocking', which is the thing Saxby and Farmer pioneered and built, and Buchanan et al. did a different way a few years later.  These are the metal bars that move, analogous in a sense to wards in a lock, to ensure that multiple levers move together and others can't be moved.

By 1891, the armstrong levers no longer needed to move pipes on rollers and the like; they could actuate relatively simple valves that direct compressed air to cylinders or air motors at the switches or other devices to be moved.  Only a few years later, the same could be done with relatively simple electrical switches to individual solenoids or electric motors.  However in all these cases the actual interlocking is still implemented mechanically, via levers and bars, even though it would now be technically possible to use much simpler control means.

Only in 1929 was the first plant built that replaced levers for route selection with the equivalent of pushbuttons, with the 'interlocking' action now done with relay logic.  This incidentally allowed remote operation of tower plants.  This facilitated the relatively early introduction of CTC, where the 'interlocking' action was now extended to route integrity selected for train movement, rather than just setting tracks a certain way.  Note that the use of the word 'manual' here means what Jeff was saying -- that the routes are still set deterministically by humans working controls, the difference being only that the lockout of improper routes is now done electrically rather than mechanically.

It is theoretically possible to servo-equip armstrong-style plants to make them 'automatic' but to my knowledge this was not done; relay logic would have been simpler and far less maintenance-intensive. and permitted other features.

There are two basic styles of 'automatic' (vs. manual) operation once you have electric logic controlling remote electric switches.  These again have to do with the way the 'interlocking' functionality for safety is provided.  One approach is to control the inputs to relay logic (or its functional equivalent) that acts to select 'only one route' and resolve or flag any errors or hangups in the overall switch-lining.  The other is to handle the interlocking 'in software' and use error-detection and correction routines to catch misalignments; the individual commands to the individual switches are not interconnected electrically at all.  This gives considerably more flexibility (and tolerance to failure of individual devices) but requires very careful programming to be safe.

We need terminology that reflects all these differences as unambiguously in discussion as possible. 

[MMLDelete]

BaltACD

In my experience, on my carrier(s) Manual Interlocking has always referred to 'armstrong' type plants with pipelines from the levers in the tower operating the switches and crossovers.

This is the type of plant I was asking about. Something built around 1925 or so, I guess.

Are there any towers still in use which currently control a plant, and even though they don't use the armstrong levers, the levers are still there to see? And in my fantasy, there is a super-friendly operator there who loves to show off the place to railfans. (As you can tell, I don't want MUCH).

Is there any railroad museum or tourist road that still operates a fully-manual interlocking? (I am guessing not.)

Where are the best non-functioning restored interlocking towers, where one can visit and see all the armstrong levers? Seems like I've read about some, maintained by railfan groups. Are there any in GA/SC/NC/VA?

I appeciate all the input, guys.

[BaltACD]

In my experience, on my carrier(s) Manual Interlocking has always referred to 'armstrong' type plants with pipelines from the levers in the tower operating the switches and crossovers.

The CADS, as normally operated by CSX Train Dispatchers, is a manual system.  The system does have a 'automatic' feature built into it.  The track segment and trains can be manually defined to the system as being Automatic by the Train Dispatcher.  When various Dispatchers have used these features they have discovered too many curious decisions and results by the automatic system to catalog.  A number of Train Dispatchers incurred adverse discipline when using 'Automatic' and they did not supervise the operation closely enough to prevent unwarranted delay to trains; as a consequence 'Automatic' is rarely used.

Two other features of CADS is the ability to 'stack' signals, where the system can be told to line particular signal(s) a specified number of times.  Example - the VRE morning fleet of 6 trains will be operating North on #2 track by the Control Poit 'Fredericksburg' - Stack the signal at Fredericksburg to line North on #2 and the Dispatcher doesn't have worry about lining the signal for each individual train.

The second feature is known as Union Route - which permits the Dispatcher to set up for complex moves affecting multiple trains - trains crossing over, going into sidings, holding until another train makes a move past the affected train.  Once the Dispatcher has decided how the moves at the location will take place, he will input the Union Route instructions into CADS and then he can devote his attentions to the other issues taking place on his territory.

Knowing how the 'read' the model board and the various indications it displays, one can 'see' the logic the Dispatcher is using to run his railroad, and from a supervision stand point be able to question the moves he is making.

[BaltACD]

tree68

 

jeffhergert

You have to remember that 'manual' refers to the requirement that a person (control operator) operate the controls to line up a route of the interlocking plant.  Whether it be the old 'armstrong' levers, pistol grip electro pneumatic, or CTC type control panel.  Or now on a computer screen.  

Example of an automatic interlocking you can watch on-line:  Rochelle, IL.  First come, first serve.

Example of a manual interlocking you can watch on-line:  Deshler, OH.  You'll often see comments in the chat about how the dispatcher has "lined" a train through.  Back when you could see the ATCS on your own computer, you could see those routings when the dispatcher lined them up.

There area also hybrid Automatic railroad crossings at grade interlockings.  The logic for who gets the over the crossing first is automatic, however, the Train Dispatcher may withhold the request to cross by not lining his traffic over the crossing.  Once the Dispatcher lines 'his' traffic indicator the automatic logic of the crossing takes care of the rest.

The reason for this kind of arrangement is where there is main track switching or similar activities taking place closely beyond the crossing and for operational reasons the Dispatcher doesn't want to hold the 'next' train at the control point in advance of the railroad crossing at grade - which may be any number of miles in advance of the crossing.   A location where this kind of logic could potentially be implemented is a Rochelle on the Westbound signal on the track nearest the camera.  Just beyond the crossing (out of camera sight) is the East end of the Global 3 yard.  Such a arrangement would permit a engine to operate to or exit the East end of Global 3 while at Westbound is being held at the absolute signal protecting the crossing - once that engine has either cleared or begun Westward movement the absolute signal at the crossing would then be cleared and the automatic logic operating the crossing would then apply and the signal would display to the train.

[tree68]

jeffhergert

You have to remember that 'manual' refers to the requirement that a person (control operator) operate the controls to line up a route of the interlocking plant.  Whether it be the old 'armstrong' levers, pistol grip electro pneumatic, or CTC type control panel.  Or now on a computer screen. 

Example of an automatic interlocking you can watch on-line:  Rochelle, IL.  First come, first serve.

Example of a manual interlocking you can watch on-line:  Deshler, OH.  You'll often see comments in the chat about how the dispatcher has "lined" a train through.  Back when you could see the ATCS on your own computer, you could see those routings when the dispatcher lined them up.

[daveklepper]

And the amount of territory they cover.

[jeffhergert]

Lithonia Operator

 

 

jeffhergert

 

 

Lithonia Operator

Are there many (any?) manual-lever-controlled interlocking plants left in the US?

If so, where is the most complex one?

And is there an interlocking in greater Charleston SC? And what type is it?

Finally, where in GA/SC/NC/VA can I find a busy interlocking in a relatively safe rural or small-town environment?

 

 

 

 

There's a lot of manual interlockings left.  However, most are controlled by the train dispatchers.  For those on the scene, the interlocking doesn't look any different than a CTC installation.  

And those dispatcher controlled ones are all controlled electronically.  You would need a lot of piping and bellcranks to reach from the centralized dispatching offices to the interlockings out in the field.

Jeff 

 

 

 

 

I'm confused, Jeff. I don't understand how a remotely and electronically controlled interlocking can be considered "manual."

Maybe you're having some fun with me, I dunno. I don't get it.

But I'll agree that it would take a lot of pipe to reach from, say, Jacksonville to Waycross!!

 

You have to remember that 'manual' refers to the requirement that a person (control operator) operate the controls to line up a route of the interlocking plant.  Whether it be the old 'armstrong' levers, pistol grip electro pneumatic, or CTC type control panel.  Or now on a computer screen. 

An 'automatic' interlocking lines up a route automatically when a train hits the approach circuit. First train gets the light.  

I'm guessing that from the technical, hardware side of things, the bells and whistles for a manual interlocking aren't much different than a CTC control point in this day and age.  The biggest difference is probably the rules that govern the use of them.

Jeff

[ATLANTIC CENTRAL]

Not directly related to the OP's question, but maybe of some interest, in my model railroading efforts, I use a system to simulate CTC operations and control turnouts in interlockings that uses fail safe relay logic very similar to that once used by the prototype railroads.

It is intergrated into the signal system which is also all relay based using 24VDC ice cube control relays.

The dispatcher sets a route, the relays power all necessary turnouts to that route, return other turnouts to their prefered neutral/normal position, and provide all the required logic for the signals.

Then the dispatcher clears the route providing a clear signal if the route is both correct and clear.

And, making it even more like the OP's question/situation, all CTC panel controls are also duplicated on "tower panels" for each individual interlocking, allowing operation by "tower operators" with no main dispatcher on duty.

In another life, about 4 decades ago, I designed relay logic circuits for all sorts of industrial applications, then I learned how to program PLC's....... some of the very first ones.

Sheldon

 

 

[BaltACD]

daveklepper

thanks for a terrrific explanaition of the 737MAX

A different look at the 737 MAX issues

[daveklepper]

thanks for a terrrific explanaition of the 737MAX

[BaltACD]

Lithonia Operator

I trust metal bars more than software code. That could be wrong-headed. But my experience with computers hasn't made me think of them as infallible. Quite the contrary.

And think of Boeing's 737 MAX jets.

Computers may not be infallible, however, they are today's fact of life; especially on Class 1's Computer Aided Dispatching Systems.  There is a high level of protection built into the computer software to protect all concerned.  

If a signal is lined and then the Dispatcher figures that is the wrong move and he desires to line up a different, potentiall conflicting, route - it is not just a matter of taking down the lined signal and lining the new route.  Every signal once lined, has a 'Time Out' feature associated with it - the time is calculated based upon the 'expected' operation of a train that would be operating under the authority of that signal and the signals on the trains route preceding the location as they would be affected by the signal's operation.  The 'Time' varies, from none upto about 10 minutes, and in some cases even more.  While 'Time' is running the system will not let the Dispatcher change the route - this 'Time' function also existed all Interlocking plants as each of the appropriate levers were locked after their route was lined - to change a route, the operator could operate the lever to set the signal to STOP, however, he could not change any other lever that were part of the route until Time had been run and unlocked all the other levers in the route.

The following explains some things about the 737 MAX.

[BaltACD]

tree68

 

Lithonia Operator

But I'll agree that it would take a lot of pipe to reach from, say, Jacksonville to Waycross!! 

Or Deshler or Fostoria, Ohio...

That would be an interesting find as one wandered a ROW - pipe and bell cranks out in the middle of nowhere...

But, speaking of pipe and bellcranks, there was no tower there, or even what one would normally think of as an interlocking, but there are plenty of pipes and bellcranks laying in the bushes alongside the tracks in Big Moose, NY.  All I can think of is that said piping tied together switches, derails, etc to make life a tad easier (less walking) for the crews.  There was a wye there, too.

Most importantly, Big Moose is literally at the top of the hill (high point of the NYC), so if anything got away, it would get a good roll on fairly quickly (five miles of 1.1% in one direction).  Resetting derails would be crucial.

Don't know anything about Big Moose. 

With that being said, railroads 'back in the day' had numerous interlocking 'towers' in locations that today we view as the middle of nowhere.  Any number of such locations existed to facilitate the manipulation and 'servicing' of steam power (water most frequently) out on line of road.  Those locations that were abandoned, 50, 60 even 70 years ago would leave little evidence of their ever existing.  In most cases even the main line trackage at those locations has changed radically.  If Big Moose were the high point of it's territory - it is likely that in steam days that helpers may have been used to get trains over its summit; needless to say after getting trains over the summit, the additional power the provided the 'help' had to be dealt with (water, fuel, turning etc.)

[tree68]

Lithonia Operator

But I'll agree that it would take a lot of pipe to reach from, say, Jacksonville to Waycross!!

Or Deshler or Fostoria, Ohio...

That would be an interesting find as one wandered a ROW - pipe and bell cranks out in the middle of nowhere...

But, speaking of pipe and bellcranks, there was no tower there, or even what one would normally think of as an interlocking, but there are plenty of pipes and bellcranks laying in the bushes alongside the tracks in Big Moose, NY.  All I can think of is that said piping tied together switches, derails, etc to make life a tad easier (less walking) for the crews.  There was a wye there, too.

Most importantly, Big Moose is literally at the top of the hill (high point of the NYC), so if anything got away, it would get a good roll on fairly quickly (five miles of 1.1% in one direction).  Resetting derails would be crucial.

[MMLDelete]

I trust metal bars more than software code. That could be wrong-headed. But my experience with computers hasn't made me think of them as infallible. Quite the contrary.

And think of Boeing's 737 MAX jets.

[Overmod]

Lithonia Operator

In an electrically-controlled plant, is it the case that even though motors initiate the movement, that still happens in a central location? And the movement is still conveyed via pipes like in olden days? Or is there an array of electric switches, and no more pipes?

If I remember correctly, many manual interlockings had electric actuation, with the actual switches being pushed by what were previously the connection to the pipe rods and manual throws in the field.

What makes the mechanical interlocker 'special' is not just that it has levers to throw, but that all the complicated mechanism of locking bars that makes the plant interlocking is also mechanical.  It is still a bit awe-inspiring to me that Saxby and Farmer were manufacturing this stuff for very complicated switch layouts, and could design systems that could provide error-free continuance for given routes in a mechanically-assured foolproof manner, less than half a decade after the Anerican Civil War.

Electropneumatic plants were first used in 1891, which substituted compressed air both for the long mechanical pipe 'trackers' and the central mechanical power needed to move them.  Electric actuators (including geared motors at individual switches) came in a couple of years later.  But you'd still require some system of positive lockouts to prevent part of a route being left incomplete, or a given switch being thrown by mistake in a set route.  This can be provided electrically (through a system of relay logic rather than mechanical bars locking the levers), which was first done in practice in the late 1920s, and I'd assume that this, rather than the way the actual logic control and power is relayed to the switches, is what makes a plant 'electrical' instead of 'manual' interlocking.  (Modern solid-state microprocessor interlocking control does all the route selection and error correction in software -- hopefully complex, self-healing, and extremely conservative and redundantly provided -- one hopes using fairly unhackable/crackable protocols and isolated systems!)

[MMLDelete]

BaltACD

I could be mistaken - I don't think there are any truly manual interlockers left - Operated several of them in my day - when it came time to throw a switch or crossover - you had to summon your 'inner Hulk' and EXPLODE on that particular lever.  Crossover switch lock levers could also be difficult.

In those day a Signal Maintainer was assigned to each manual interlocking plant to maintain all the pipeline, switches, signals and the manual interlocking bed that all the levers and pipeline worked through and created the really manual form of locking that prevented conflicting moves from being lined up.

 

I can more than believe you regarding the oooomph required! Looking at photos and diagrams, I've wondered how it was even possible on the longer pipe runs. I assume that keeping everything well-greased was the key. But in really cold weather it must have been a back-breaker.

In an electrically-controlled plant, is it the case that even though motors initiate the movement, that still happens in a central location? And the movement is still conveyed via pipes like in olden days?

Or is there an array of electric switches, and no more pipes?

[MMLDelete]

jeffhergert

 

 

Lithonia Operator

Are there many (any?) manual-lever-controlled interlocking plants left in the US?

If so, where is the most complex one?

And is there an interlocking in greater Charleston SC? And what type is it?

Finally, where in GA/SC/NC/VA can I find a busy interlocking in a relatively safe rural or small-town environment?

 

 

 

 

There's a lot of manual interlockings left.  However, most are controlled by the train dispatchers.  For those on the scene, the interlocking doesn't look any different than a CTC installation.  

And those dispatcher controlled ones are all controlled electronically.  You would need a lot of piping and bellcranks to reach from the centralized dispatching offices to the interlockings out in the field.

Jeff 

 

I'm confused, Jeff. I don't understand how a remotely and electronically controlled interlocking can be considered "manual."

Maybe you're having some fun with me, I dunno. I don't get it.

But I'll agree that it would take a lot of pipe to reach from, say, Jacksonville to Waycross!!

[BaltACD]

I could be mistaken - I don't think there are any truly manual interlockers left - Operated several of them in my day - when it came time to throw a switch or crossover - you had to summon your 'inner Hulk' and EXPLODE on that particular lever.  Crossover switch lock levers could also be difficult.

In those day a Signal Maintainer was assigned to each manual interlocking plant to maintain all the pipeline, switches, signals and the manual interlocking bed that all the levers and pipeline worked through and created the really manual form of locking that prevented conflicting moves from being lined up.

[jeffhergert]

Lithonia Operator

Are there many (any?) manual-lever-controlled interlocking plants left in the US?

If so, where is the most complex one?

And is there an interlocking in greater Charleston SC? And what type is it?

Finally, where in GA/SC/NC/VA can I find a busy interlocking in a relatively safe rural or small-town environment?

 

There's a lot of manual interlockings left.  However, most are controlled by the train dispatchers.  For those on the scene, the interlocking doesn't look any different than a CTC installation.  

And those dispatcher controlled ones are all controlled electronically.  You would need a lot of piping and bellcranks to reach from the centralized dispatching offices to the interlockings out in the field.

Jeff 

[Lithonia Operator]

Are there many (any?) manual-lever-controlled interlocking plants left in the US?

If so, where is the most complex one?

And is there an interlocking in greater Charleston SC? And what type is it?

Finally, where in GA/SC/NC/VA can I find a busy interlocking in a relatively safe rural or small-town environment?