Signalling Littlemore

[D869]

It does from the point of view of electrically interlocking the locomotive movement depending on the mechanically worked lever frame position (that is not locked - at the moment). At the moment the wiring does not allow power to the track unless the route is correctly set. This is achieved by double isolating with the switch operating. Unfortunately this method needs a tweak to include the signals and I'm not sure how to do it. 

 

Maybe a second feed to some pieces of rail each with a diode will allow directional running depending on the signal being off? In my head this is already starting to get very complicated the more I think about the practicality of it.

 

Tricky... you could ignore the question of direction and power the track if either signal is clear, but I doubt whether you have signals covering every shunt move (which would be protected by having both running signals returned to danger)

 

I've tried diodes in fiddle yard feeds and would avoid them now. It probably depends on your controller, but I've seen some odd effects with certain locos and controllers that use pulse width modulation. Can't say what would happen on other types of controller but it's 100% clear to me that you can guarantee circa 0.7V drop on the diode sections as well as completely blocking current in the reverse direction. Strangely, my experience is that some locos speed up on reaching the diode section... not 100% sure why though.

 

Regards, Andy

[richbrummitt]

Thinking more about this diodes will mess with a feedback controller and I have reached the same conclusion.

[richbrummitt]

I'm going to calculate the locking table and consider mechanically interlocking the levers. With the plan not fully signalled there is no incentive for an operator to put the locks in and electrical locking would force the issue I'm sure the locks would have only worked for the switches normal and are certainly only required when running over in the facing direction. I could require the locks be in for travelling in either direction as a compromise to avoid the FPL levers being 'optional'. It might get a bit annoying when shunting to have to engage the lock to run on past the switch with it normal but I think it might be the most acceptable compromise for my requirement to force the operator to use all the levers?

[richbrummitt]

Would it be permissable to have one lever operate all the four switches on the right hand board? Another lever could operate both locks. My thinking being that two levers would be spare to use for the other signals that there should be.

[Grovenor]

Well, I guess using one lever for all 4 points would be useable but not prototypical, if you are going to go that far away from real why not just take the simple route of omitting the fpl levers that do nothing on the model anyway, then you would have enough for the signals. Or splash out on another 5 lever frame so you can do it properly.

Regards

Keith

[The Stationmaster]

Assume Prince's Locks for the FPLs and that's 3 levers saved  (the GWR did use Prince's Locks for combined point and FPL levers although a bit later than Littlemore opening date and they didn't last all that long I believe - although the WR installed one in 1965 having somewhat miscalculated when renewing a ground frame and with no more room to add a lever they had to use a combined FPL and point lever).

[richbrummitt]

Well, I guess using one lever for all 4 points would be useable but not prototypical

I won't do it then. Thank you.

[richbrummitt]

After thinking for even longer I'm back around to this:

 

 

The quality of the image is really starting to suffer now where I've chopped the numbers about so much. Sorry. 

 

Having the signals at the east end didn't make sense to me because there would be shunting movements that would have to pass them. I know that the two signal posts that I will model did exist.

 

I will make the FPLs release the home signal in the direction of travel and allow the possibility of FPLs in and the signals in each direction off. I believe the locking table should look like this:

 

No.  Locks  Releases

1      3,7,8

2                      4

3         2           1

4                                   

5

6                     

7         6           1

8                     

9         8          10

10    2,6,9

 

If I've got that right then I can have a go at the dog chart next.

[The Stationmaster]

I'm assuming you are locking it as a ground frame and not as a signal box frame but even so I suspect not all the locking would have been altered when it was down-rated, so the following should also still be there -

 

2 locks 6

 

4 - locks 5 BW (both ways)

 

6 locks  2 & 8

 

8 locks 6

 

and as far as 1 is concerned it should read :-

 

1 locks 2, 6, 8

 

and as far as 10 is concerned it should read (to correct what looks like a typo) :-

 

10 locks 2, 6, 8

 

(When the frame was a signal box 1 would have locked 8 BW but that should have been altered on reduction to ground frame status)

[richbrummitt]

I'm assuming you are locking it as a ground frame and not as a signal box frame

Er, pass. I think you covered both options in your reply though. It was classified as a ground frame by the period modelled so you perhaps assume correctly.

even so I suspect not all the locking would have been altered when it was down-rated, so the following should also still be there -

 

2 locks 6

 

4 - locks 5 BW (both ways)

 

6 locks  2 & 8

 

8 locks 6

 

and as far as 1 is concerned it should read :-

 

1 locks 2, 6, 8

 

and as far as 10 is concerned it should read (to correct what looks like a typo) :-

 

10 locks 2, 6, 8

 

I knew I didn't have it all but it's not an easy thing to get your head around trying to understand the 'rules' and apply them to part of a slightly obscure arrangement.

 

I'm not clear why 2 locks 6, 6 locks 2&8 and 8 locks 6 would have all been required except to force a man to keep a tidy frame? I can kind of see why 6 locks 2 because the move would be from the siding at the north west (bottom right) corner and presumably out of the station limits to the east (left) but I don't understand the others. Please would you elaborate...

 

A brief bit of thinking and scribbling suggests all the locking fits easily in one slot until I try to add these.

 

(When the frame was a signal box 1 would have locked 8 BW but that should have been altered on reduction to ground frame status)

I understand this one.

[D869]

I could use a table of where this discussion has reached - it's hard to keep track even after just a few posts... but not sure whether it's easy to do tables on RMWeb.

 

I'd suggest adding 1 locks 10 and vice versa (except when the box is closed, which you've said that you don't want to do)... but you'll definitely need another channel in your locking tray for that one.

 

The practice of not allowing conflicting sets of points to be reversed is one that I've seen before. 2 and 6 definitely fit that category - you couldn't sensibly be making (or hand signalling) moves across 6 if 2 was reversed. 6 and 8 are not actually conflicting, so I'm less sure on that one but there may be precedents - possibly a principle of 'only set up one move at a time' for such a small location?

 

You should show the reciprocal of each lock since it will happen anyway with mechanical locking. For instance 6 locks 10... since you already have 10 locks 6. The same will be true with your FPLs since you are have them locking the point normal (locking either way would not have a reciprocal).

 

FWIW, I usually concentrate on the 'released by' column rather than 'releases' because you have all of the release info (e.g. for a signal) in one place that way. The prototype charts have both columns though.

 

Some WR locations would indicate that 1 locks 9 both ways but if the responses to my queries elsewhere on this are correct then it is for protection of the FPL machinery and has no particular relevance on a model other than to stop you moving a lever when you probably wouldn't move it anyway.

 

Naturally this has operational significance - for instance at a first glance at the track plan you can depart from the loop, but the starter won't clear if 8 is reversed, so you need to get your train fully onto the main line and put 8 back before you can depart... better get your operators trained.

 

Regards, Andy

[richbrummitt]

We're here

 

No.  Locks  Releases

1    2,6,8BW

2         6            4

3         2           1

4       5BW

5

6        2,8           

7         6           1

8         6      

9        8          10

10    2,6,8

 

I think.

 

What you wrote makes sense. 1 doesn't lock 10 and vice versa because I want to be able to have the locks in and the signals off.

 

Thank you for reminding me that I need to work out the released by table to make the proper plates for the levers. I'm still not sure i really understand what I am doing, although I do seem to be getting there.

[The Stationmaster]

Er, pass. I think you covered both options in your reply though. It was classified as a ground frame by the period modelled so you perhaps assume correctly.

I knew I didn't have it all but it's not an easy thing to get your head around trying to understand the 'rules' and apply them to part of a slightly obscure arrangement.

 

I'm not clear why 2 locks 6, 6 locks 2&8 and 8 locks 6 would have all been required except to force a man to keep a tidy frame? I can kind of see why 6 locks 2 because the move would be from the siding at the north west (bottom right) corner and presumably out of the station limits to the east (left) but I don't understand the others. Please would you elaborate...

 

A brief bit of thinking and scribbling suggests all the locking fits easily in one slot until I try to add these.

  I understand this one.

I purposely omitted 1 locks 10 (and the converse) as that will impose complications you would not wish to, and need not, address.

 

As far as the points are concerned the normal convention is conflicting points lock each other and while I realise that 6 reverse and 8 reverse do not create conflicting routes I'm fairly sure that they would normally have locked each other in a situation like this as much to avoid potential run throughs when shunting as anything else.

 

I have never come across the business of a signal locking an FPL in rear of it on Western locking and it was certainly not one which the last real expert on the subject ever mentioned to me when he was checking my 12":1ft scale locking chart for frame I was planning.  The critical thing (and the standard locking rule) is that signal locks both ways  (usually both ways) a trailing point in rear of it and the reason for that is to prevent premature movement of the point before the train has passed clear.  There is no purpose whatsoever in a signal locking an FPL in rear of it as that would in any case be locked by the presence of the train - which is far more reliable than using a signal lever to do it.

 

Here's a pic showing the latter style of Western lever lead - if you want the style of teh brass, earlier, version I shall have to take one of one of my paperweights!

 

[D869]

The Stationmaster, on 08 Mar 2013 - 00:21, said:

I have never come across the business of a signal locking an FPL in rear of it on Western locking and it was certainly not one which the last real expert on the subject ever mentioned to me when he was checking my 12":1ft scale locking chart for frame I was planning. The critical thing (and the standard locking rule) is that signal locks both ways (usually both ways) a trailing point in rear of it and the reason for that is to prevent premature movement of the point before the train has passed clear. There is no purpose whatsoever in a signal locking an FPL in rear of it as that would in any case be locked by the presence of the train - which is far more reliable than using a signal lever to do it.

I came across this arrangement in the charts from the SRS for Penzance and Newquay (for FPLs on trailing points both in advance and in rear of the signal) and was puzzled by it so I asked on the signalbox forum. Here's the discussion thread...

 

http://www.signalbox.org/forum/viewtopic.php?f=2&t=5462

 

Anyway, it's not something that Rich needs to worry about.

[D869]

We're here

I re-did this in Excel but can't post a CSV here, so here's an image. I added the other columns and also the reciprocal locks in square brackets so you can see what I added.

 

Somewhere the lock between 1&8 has changed from locks normal to locks either way. Was this delibarate?

Thank you for reminding me that I need to work out the released by table to make the proper plates for the levers. I'm still not sure i really understand what I am doing, although I do seem to be getting there.

Yes. It would be good to put some words on the diagram (up and down would be a start) and then somebody needs to name each lever... which has its own conventions which can be a discussion topic on their own.

 

Also, the 'released by' column may not be quite the same as the numbers on the lever leads. The lever leads would be in the pull sequence whereas the locking chart is in numerical order.

 

Regards, Andy

[The Stationmaster]

I came across this arrangement in the charts from the SRS for Penzance and Newquay (for FPLs on trailing points both in advance and in rear of the signal) and was puzzled by it so I asked on the signalbox forum. Here's the discussion thread...

 

http://www.signalbox.org/forum/viewtopic.php?f=2&t=5462

 

Anyway, it's not something that Rich needs to worry about.

Alas the thread doesn't come up, not even once I have logged in.  I am aware that the arrangement could be applied where trains are likely to 'stand out' in advance of a signal  (although it would only happen where they were likely to stand between the signal and the lock bar I believe) or to lock in rear of a platform starting signal at a terminus (which is in fact what I did on a locking chart anyway)  but those were more specialised circumstances than the looking at a through station.

[D869]

Alas the thread doesn't come up, not even once I have logged in.

Oh dear. Maybe the URLs aren't portable from one user to another. The thread is in the historical discussion forum, started on 24th Jan and is titled 'Penzance/Newquay: why lock an FPL both ways?'

[richbrummitt]

For some reason the quote button is non functional on the iPad just now.

 

Andy,

The change to 1 locks 8 BW was deliberate. This makes more sense because the operator (me) is more likely to remember that is how things work It might be a little hybrid between box and ground frame but I think it will suit me.

 

If my thinking is correct only the information in the first two columns appears on the levers - the lever number and the released by numbers (in sequence order)?

 

Up and Down are the same through line btw. The line at the bottom right is a siding that runs alongside the main in the down direction for about 500 yards to a sand pit where it splits into three. This closed and opened variously over the years but from photos it seems to have been used as additional storage space between the switch and the bridge.

 

Mike,

I must have a picture of a brass type somewhere in my library. There is probably one in Vaughan. Your paperweight does sound like a nice piece to have though.

[The Stationmaster]

The main problem with the lever leads will be naming for the point levers I think as the two on the same side will need to be differentiated from each other in some descriptive way, No.6 is miore straightforward.

 

Beyond that it's all very straightforward although in one or two cases there will be an or  situation on some leads, e.g the lead for No.4 will be 2 OR 2 5  (set out vertically of course)

[D869]

The change to 1 locks 8 BW was deliberate. This makes more sense because the operator (me) is more likely to remember that is how things work It might be a little hybrid between box and ground frame but I think it will suit me.

So now you can depart directly from the loop.

If my thinking is correct only the information in the first two columns appears on the levers - the lever number and the released by numbers (in sequence order)?

On Mike's picture of the real thing it has the lever number (large font), a horizontal line, the lever name (small font), another line and then the other levers that need to be pulled beforehand (small font)... I doubt whether you'll fit all of that on and still be able to read it though.

Up and Down are the same through line btw. The line at the bottom right is a siding that runs alongside the main in the down direction for about 500 yards to a sand pit where it splits into three. This closed and opened variously over the years but from photos it seems to have been used as additional storage space between the switch and the bridge.

Normally there would be a couple of arrows with labels indicating up and down, at either end would be a label to tell you what the next box is. Individual lines may or may not be labeled - most often something very generic like 'down siding' or 'up loop' but occasionally something more specific and local.

[richbrummitt]

After spending some time unsuccessfully working out the tappets and slides I've sent an enquiry to modratec about them supplying just the locking frame without levers, which has been met with a positive response. 

 

My next signalling task is to construct the signals but I am unsure of the detail of one post. The starter (1) must have been mounted on the platform but the photograph is not clear enough to make out the position of the balance weight and how the wire would have been routed to the base of the post. The clarity is further reduced by leaves obscuring the post around half way up. I've found some other examples of signals mounted on the platform along the line at Wheatley and Thame. At Wheatley the photograph shows the down platform had an opening that the post passes through, which is filled in with some cover plates around the base of the post. I presume the signal wire passed into the platform face at ground level and the balance weight was beneath the platform and access was possible by removing the cover plates. At Thame at the end of the up platform the photograph shows the balance weight half way up the post but the surface of the ramp is not visible due to the angle of the photograph to ascertain whether the signal wire would have been routed up the surface of the ramp. A later picture shows the same signal but the wooden post has been replaced by a tubular type and the balance weight must be below a cover in the platform because it is no longer visible. I cannot see that there is an opening in the platform face at either location to allow the signal operating wire into the space below the platform surface. If I can't see this then I probably don't need to model it but why would the balance weight be half way up a post rather than at the base?

[Miss Prism]

Don't know, and maybe it depended what kind of bits were available at the time of post installation. (It does seem a bit odd, from a maintenance point of view, to put a crank below the platform surface level.) It is possible that in some cases the signal position was established before platforms were upgraded and extended, so the new platform was built 'around' the existing signal. Cover plates were usually wooden planks. I've never seen a pulley wire go up a ramp (or indeed on top of any platform surface), so I think there will be some kind of opening in the platform face somewhere - probably too small to spot from a picture.

[buffalo]

The lower picture on this page shows an example with a cover plate and the lever etc. are hidden below the platform.

 

Nick

[richbrummitt]

The covers are bigger than those in the photo on that site. I'll scan and crop the photos I refer to in my previous post so you can see.

 

As far as I know the platforms at these locations were never extended but the balance arm on the Up inner home at Thame definitely moved to below the platform surface when the post was changed from wood to tubular steel.

[The Stationmaster]

The covers are bigger than those in the photo on that site. I'll scan and crop the photos I refer to in my previous post so you can see.

 

As far as I know the platforms at these locations were never extended but the balance arm on the Up inner home at Thame definitely moved to below the platform surface when the post was changed from wood to tubular steel.

As far as I can tell from what I saw over the years the usual practice with tubular steel signals was to put the balance weight etc below platform level with a timber decking around the post  so access could be gained.  There were sometimes exceptions to this but I think they were fairly infrequent and they normally occurred for particular reasons - e.g unavoidable mechanism such as that for route indicators but on a short signal post so it couldn't be mounted above head height - by the 1960s such signals tended to be 'fenced' to keep folk away from them.  

 

I can recall one timber post signal where the balance weight was above platform level but also above head height and in its case the signal wire came up through the platform in a length of small diameter pipe - darned if I can remember where it was tho'.