Jump to content
 

St Ruth

  • entries
    59
  • comments
    352
  • views
    85,284

Your Starter for Ten


D869

2,177 views

blog-0177123001339589171.jpgOur next planned outing should be TINGS near Leamington Spa in early September although the show web site is curiously shy about our presence and that of another 2FS layout of this parish which I'm told will be in attendance.

 

There are still a few more buildings needed and some of these may appear before September but I think it's time to make a start on another aspect (get it?) of the infrastructure - the signals.

 

St Ruth needs a lot of signals so there is zero chance that they will be finished any time soon but they certainly won't get finished until we start on them.

 

The first thing we need is a plan. This has been work in progress for a long time - the original plan was created before the control panel and then drawn using a PCB design tool. The printout became the template for the control panel. Our understanding of the way we will operate St Ruth has now moved on so there have been some changes such as adding full running signals to allow departures via the east crossover. I re-drew the control panel overlay using some free software called 'Dia'. The new plan is rather closer to the conventions used by the drawing office at Reading but it wasn't possible to get it exactly right because of limitations of the tool.

 

Here is a PDF of the new plan. It is oddly shaped in places because it needs to fit over the existing switches on the control panel. struthsigplan.pdf

 

We've tried to avoid inventing our own arrangements and have tried to find precedents from the prototype as much as possible. It's probably best described as a hybrid of the pre and post 1938 plans for Penzance with the addition of the branch line and goods loop. The signals for the east crossover are based on the postwar arrangement at Newquay which also gives us an excuse for a GWR backing signal. The trickiest part was the branch because there doesn't seem to be a perfect precedent for a single line junction joining a double track line within the limits of a terminus station. The closest I've found is the arrangement of the spur from Barnstaple North into Barnstaple Junction but it's not a perfect match and is only partly GWR.

 

I am sure that signalling lawyers could pick hols in the plan but we've done our best to stick to the rules and get it right. Looking at the plan again I am thinking of some further minor changes - arm 62 might become a disc because it only covers shunt moves unless the branch points fail again and 65 should probably be lower than 15. The three doll bracket might also gain a slotted distant arm for the next box to the east.

 

So far we have one signal which is mostly finished. This is the starting signal from the arrivals platform (number 68) and is one of the very few single arm signals on the plan. It has the extra advantages of lacking a rule 55 diamond and having its balance weights hidden below platform level. The intention is to operate all of the signals using servos as described in MRJ 201. This signal has been fitted with a servo on a test rig and seems to work OK.

 

The signal also has a working lamp. This consists of a tiny white LED to which some enamelled wire is soldered. This is contained inside a short length of heat shrink sleeving in which two holes are drilled for the lens and backlight. A short length of 1mm styrene rod is pushed into the top and the whole thing painted black. It is pretty close to the shape of a GW signal lamp but perhaps a little 'chunky'. The power feed for the lamp comes via a resistor under the baseboard and then up the ladder as

described in the 2mm handbook. The lamp man's platform is some thin double sided PCB with one side used for structural solder joints and the other side used for the lamp power feed.

 

There is still some more 'development' work to do. We haven't yet finalised our servo controller approach. I used a 555 timer circuit (see photo) to get the thing working initially and we tried out a MERG SERVO4 unit yesterday. There will be some more head scratching about powering the LEDs on multi-arm signals because the easy thing to do would be to wire them in parallel but I'm told that having LEDs in parallel sharing a single resistor is a bad idea. I'm sure that using the first few signals at a real exhibition will also reveal some more problems for us to think about. The jury is still out on the colour for the post. I tried metalcote 'steel' but this was too dark so the signal was repainted in light grey which looks less wrong. I now have some metalcote 'aluminium' which I will try on the next signal.

 

blogentry-9623-0-37061500-1339589246_thumb.jpg

 

Here are some pics of the signal on its test rig and temporarily fitted to the platform at St Ruth. The clear signal was a bit of a cheat because the photo was taken before the servo was fitted. The servo is now in place, albeit with no controller or wiring yet.

 

blogentry-9623-0-09995000-1339589299_thumb.jpg

 

blogentry-9623-0-51356400-1339589269_thumb.jpg

 

Now I'm working on the second signal. This one is rather more complex and provides lots of new challenges for me to think about, but more about that another day.

  • Like 10

17 Comments


Recommended Comments

  • RMweb Gold

...I thought I had my work cut out having to make two junction semaphores one day...but having seen your plan... :O

 

That looks terrific Andy...and the lamp really effective.

 

Did you use a mixture of mse etches and scratchbuild items overall?

Link to comment

...I thought I had my work cut out having to make two junction semaphores one day...but having seen your plan... :O

That looks terrific Andy...and the lamp really effective.

Did you use a mixture of mse etches and scratchbuild items overall?

Thanks Pete. I'm not finding it too daunting - just taking it one signal at a time I find it quite satisfying. Each one individually is not too much work compared to say a coach kit so there is a real feeling of progress. I'm sure you'll manage your signals just fine although I'm guessing that you might not have a nice easy one to practice on?

 

The MERG bounce is the bees knees imho :sungum:

 

Tweaking and possibly modifying the MERG is definitely something we plan to play with but nothing to report as yet other than plugging in the vanilla SERVO4 and proving that it works.

Link to comment
  • RMweb Premium

Get the modified PIC I think they are about a £1 and play it really is quiet impressive even in 2mm :no:

 

I'm really impressed how far you've ALL got since Expo 2010 :happy_mini:

Link to comment

Get the modified PIC I think they are about a £1 and play it really is quiet impressive even in 2mm :no:

 

I'm really impressed how far you've ALL got since Expo 2010 :happy_mini:

 

Thanks Nick. I think we might already have the bouncy code but that's not my department... which makes a change because software is my day job. I believe that we also have the technology to reprogram the PIC but we weren't brave enough to try reprogramming our one and only controller yesterday - we had only just got it working. We definitely have lots of new things to try out over the next few weeks.

Link to comment

Did you use a mixture of mse etches and scratchbuild items overall?

 

Oops - didn't answer that one, did I? Yes, we're using MSE parts wherever they suit. For the signal in the photo that means just the arm the ladder. If we could use more off the shelf bits then we would do - not having to make finials would be quite nice but I don't know of anything else that is suitable.

  • Like 1
Link to comment
  • RMweb Gold

Nice work. Its amazing how much difference just one signal on the platform makes. Regarding your comment about the branch joining within station limits. I suspect the problem is that the GWR would not have done it quite like that. It is more likely to be a post steam revision done that way. So perhaps it doesn't matter if you adopt something not quite GWR.

Don

Link to comment

Nice work. Its amazing how much difference just one signal on the platform makes. Regarding your comment about the branch joining within station limits. I suspect the problem is that the GWR would not have done it quite like that. It is more likely to be a post steam revision done that way. So perhaps it doesn't matter if you adopt something not quite GWR.

Don

 

Thanks Don. To be honest the more real signalling plans I look at the more I realise that there were lots of very odd arrangements out there in the real world. I suspect that if we had invented the arrangement of carriage sidings and signals part way along the departure platform then nobody would have believed it (including us). Actually it is the genuine pre-1938 arrangement which was forced on the GWR by circumstances such as a restricted site and the previous broad gauge arrangement. The way that the goods yard sidings join a platform road is also pretty odd, but again is taken straight from the real thing. I'm pretty sure that this particular platform face would not have been used for loaded passenger trains though.

 

I have had my reservations about the branch junction because it forces some 'wrong line' running for a short distance on the double track but in the end I decided that within the limits of a terminus station there are plenty of examples where this is inevitable. Finding the Barnstaple Junction arrangement (which isn't even a terminus) also helped.

Link to comment

At an risk of this being interpreted as putting my hand up to do this...

 

Are the shunting disks going to stay fixed, or will they be moving as well?

Link to comment
  • RMweb Gold

Are the shunting disks going to stay fixed, or will they be moving as well?

 

...as they say...if you don't like the answer...then don't ask the question ;)

  • Like 2
Link to comment
  • RMweb Gold

There is, of course, the view expressed by some that GWR/WR lower quadrant signals don't bounce :no:

 

Nick

 

I don't remember a promounced bounce but there definately was a soild sort of clunk and things shook a little - not exactly the sort of thing you could model in 2mm unless someones comes up with a sound chip to work on DCC?

Don

Link to comment

There is, of course, the view expressed by some that GWR/WR lower quadrant signals don't bounce :no:

 

I don't remember a promounced bounce but there definately was a soild sort of clunk and things shook a little - not exactly the sort of thing you could model in 2mm unless someones comes up with a sound chip to work on DCC?

 

I suggest that we consult the real thing for the answer on this question.

 

 

A couple of nice photos of a ground disc on that thread though.

 

At an risk of this being interpreted as putting my hand up to do this... Are the shunting disks going to stay fixed, or will they be moving as well?

 

I'll let you have a disc etching and a servo once you have made good your escape from Cornwall.

  • Like 2
Link to comment

I suggest that we consult the real thing for the answer on this question...

Very interesting, but I'm left with the feeling that the first two are well worn and listening to the audio on the third, it is clearly not in the best of health. Overall, I'm left with the impression that whilst they may bounce the selection of evidence here is a little partial :no:

 

Nick

Link to comment

Very interesting, but I'm left with the feeling that the first two are well worn and listening to the audio on the third, it is clearly not in the best of health. Overall, I'm left with the impression that whilst they may bounce the selection of evidence here is a little partial :no:

 

Oh well, I didn't select anything - these are just the first 3 videos that I found that show a WR signal returning to danger. Admittedly they are all fairly recent so it is possible that there was a golden age of non-bouncy WR semaphores in the past.

 

My gut feel is that there are some heavy lumps of metal involved. If you put the signal back to danger quickly enough then all of that kinetic energy has got to go somewhere when it reaches the end of its travel. I don't know exactly what stops the motion, but I'm guessing that it has a cushioning effect built in to avoid breakages, so bounce seems like a distinct possibility.

Link to comment

Fair enough! I've just been looking at this mid-1940s film and there is a seriously bouncy signal at about 10:40 and 11:15. It appears to bounce more when pulled off than when returned. Maybe the trick will be to have a variety of bounce patterns across your collection of signals?

 

Nick

Link to comment

Fair enough! I've just been looking at this mid-1940s film and there is a seriously bouncy signal at about 10:40 and 11:15. It appears to bounce more when pulled off than when returned. Maybe the trick will be to have a variety of bounce patterns across your collection of signals?

 

Yes, definitely. I think it is highly likely that some signals were more prone to bouncing than others. I also think that some of my model signals will likely be less able to stand up to the repetitive strain of simulated bouncing than others. I foresee some of the more complex signals having the servo controller speed set quite low to keep the stresses to a minimum.

  • Like 1
Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×
×
  • Create New...