Upcoming Games

(UTC times)


Full list
Add a game

Upcoming Events

No events to display

Upper Fern Tree Gully to Belgrave - Melbourne, Victoria

You are here: Home > Forum > Wishlist > Simulation wish list > Upper Fern Tree Gully to Belgrave - Melbourne, Victoria

Page 1 of 1

Upper Fern Tree Gully to Belgrave - Melbourne, Victoria 01/11/2021 at 23:45 #142145
Anothersignalman
Avatar
96 posts
Per attached, I think this would make a nice, relatively simple sim to operate, which could reasonably be played at accelerated time for a quick session. Peak trains are every 10min or so each way. The track shown is about five miles long, and has about sixty trains each way per day including empties and shunts to/from Upper Fern Tree Gully sidings. In 1974 the line had about fifty trains each way per day, including two goods per week and two parcels coaches per day.

1995 is a bit of a guess, not really sure when the Upper Fern Tree Gully yard was reduced. I'd also like the chance to experiment with the existing timetable on a modified track layout with Belgrave No.1 road slewed over to Siding A, and points 51 moved further out - my hypothesis is that doing so would improve reliability significantly, if the train normally stabled there could be moved elsewhere.

I've used BR style signals for this but the real area uses speed signalling. It might also be worth renaming some of the sidings for clarity, and/or adding track circuits and additional signals to/from the sidings to make it more interesting to operate?






Post has attachments. Log in to view them.
Log in to reply
Upper Fern Tree Gully to Belgrave - Melbourne, Victoria 02/11/2021 at 01:35 #142146
Anothersignalman
Avatar
96 posts
Example train graphs based on working time tables: weekdays 1974, weekends 1975 - note big PDFs.

The map would include stations 4 to 8 only.





Post has attachments. Log in to view them.
Last edited: 02/11/2021 at 01:36 by Anothersignalman
Reason: None given

Log in to reply
Upper Fern Tree Gully to Belgrave - Melbourne, Victoria 02/11/2021 at 02:49 #142147
flabberdacks
Avatar
636 posts
That's a neat mock-up, well done. That whole line from Camberwell down would make a good sim. The Australian developers have several current and planned projects incorporating the greater Sydney area at the moment, however.

Programming Victorian speed signalling into SimSig will be no mean feat. It would be a tremendous amount of work for such a small area. It may be considered at some point in the future, once all the Sydney projects are completed. Never say never.

Last edited: 02/11/2021 at 02:50 by flabberdacks
Reason: None given

Log in to reply
Upper Fern Tree Gully to Belgrave - Melbourne, Victoria 02/11/2021 at 05:39 #142149
Anothersignalman
Avatar
96 posts
Thanks flabberdacks.

This could evolve into a larger sim at a later date, but I figured it'd be a good one to experiment with. It doesn't really need speed signalling, either, because the track layout is quite simple. Ignoring shunting signals for a bit, every signal is either Normal only, Medium only, or Medium-plus-Normal-Warning, which effectively devolves to four-aspect. The only possible exception is 36, which can show normal aspects only to the down platform and medium to either platform, but this is approach-cleared anyway and a post-1995 modification. For every other signal, standard 3- or 4-aspect signals combined with curve speed limits give enough detail.

Log in to reply
Upper Fern Tree Gully to Belgrave - Melbourne, Victoria 02/11/2021 at 09:44 #142152
Anothersignalman
Avatar
96 posts
Random thought: Does Simsig's timetable program system allow for one train to have various chances of including wagons for different destinations, i.e. the down goods could have a 25% chance of needing to drop wagon/s in the Council siding, independent of 50% chance of dropping off at the goods shed, etc? That sort of thing would need to be communicated to the signaller before the train bells on at Ferntree Gully, but not much earlier than that. Basically, adding an Inglenook function.
Log in to reply
Upper Fern Tree Gully to Belgrave - Melbourne, Victoria 02/11/2021 at 09:48 #142153
flabberdacks
Avatar
636 posts
Nah, if it's done at all, it must have the correct aspect sequences. As a signalling simulator, it wouldn't be worth the price of purchase if it didn't... you know... simulate the signalling.

One of SimSig's greatest strengths is the detail on display, and we have worked long and hard to get the two NSW systems pretty spot-on.

edit: There is a function in timetabling to decide between, and give the player, one out of several different versions of the 'same' train, that is possible. Very complex and requires a lot of testing but it's definitely doable.

Last edited: 02/11/2021 at 09:50 by flabberdacks
Reason: None given

Log in to reply
Upper Fern Tree Gully to Belgrave - Melbourne, Victoria 02/11/2021 at 10:07 #142155
Anothersignalman
Avatar
96 posts
Ordinarily I'd agree with you, but how much difference does it make here given that:
a) the signaller at Upper Fern Tree Gully can't see the specific aspects of any of the signals, except a handful of them by sticking their head out the window (and they're all normal speed or low speed only), and
b) the short block sections and tight curve radii mean that the train physics would be more or less the same whether the trains are running to 3/4-aspect signalling or speed signalling?

You might want a mode which displays different aspects on the panel controls - I probably wouldn't use it but I understand the appeal - but the train operations (application of acceleration/braking) probably don't need to change all that much?

Log in to reply
Upper Fern Tree Gully to Belgrave - Melbourne, Victoria 02/11/2021 at 20:36 #142161
flabberdacks
Avatar
636 posts
This mode does exist, 'Panel Signals' in the F3 options menu, supported on all sims. This reflects what signallers see versus what the signals are doing in the field. I believe only IECC-type workstations actually display the signal aspect to the signaller. I always have Panel Signals set to 'on', so I only see green or red.

You ask how much difference it makes, I'm happy to answer. All sims are programmed with the full aspect sequence available in the real life system, and the sim knows the 'correct' order of aspects (to simulate Adverse Change of Aspect phone calls). Trains moving along a sim respond to the aspect in front of them like a driver would, including moving at a restricted speed through the block if the most recent aspect they saw requires it.

The Victorian 'clear normal -> reduce -> medium -> stop' four-aspect sequence is not the same as the British 'clear - advance caution - caution - stop' sequence. The absolute requirement to travel at 40kph through the block on a medium aspect in ABS territory, or 40kph until clear of the points on a clear medium in ATC, does not apply to the UK scheme (to name a couple of examples - there are more). This means that if the Victorian aspect sequence isn't programmed, trains will not behave correctly. They'll move too quickly through sections (among other things), and the experience of signalling the Upper Gully area will be poorly represented as a result.

Don't get me wrong, I think it would be good to have a sim of that line, and your mockups are really good - I've never seen an old layout of the area. I'd just hate to see a sim rushed. It is worth taking the time to do it right, and perhaps one day it will be.

Log in to reply
The following user said thank you: Anothersignalman
Upper Fern Tree Gully to Belgrave - Melbourne, Victoria 02/11/2021 at 20:52 #142162
TUT
Avatar
532 posts
flabberdacks in post 142161 said:
This mode does exist, 'Panel Signals' in the F3 options menu, supported on all sims. This reflects what signallers see versus what the signals are doing in the field. I believe only IECC-type workstations actually display the signal aspect to the signaller. I always have Panel Signals set to 'on', so I only see green or red.
Ah well you'd be surprised...

https://www.wbsframe.mste.co.uk/public/West_London_Junction.html

Log in to reply
The following users said thank you: Anothersignalman, flabberdacks
Upper Fern Tree Gully to Belgrave - Melbourne, Victoria 02/11/2021 at 23:29 #142163
Anothersignalman
Avatar
96 posts
flabberdacks in post 142161 said:
The Victorian 'clear normal -> reduce -> medium -> stop' four-aspect sequence is not the same as the British 'clear - advance caution - caution - stop' sequence. The absolute requirement to travel at 40kph through the block on a medium aspect in ABS territory, or 40kph until clear of the points on a clear medium in ATC, does not apply to the UK scheme (to name a couple of examples - there are more). This means that if the Victorian aspect sequence isn't programmed, trains will not behave correctly. They'll move too quickly through sections (among other things), and the experience of signalling the Upper Gully area will be poorly represented as a result.

Thanks for that. I was hoping the relatively short block sections for the Upper Gully panel would address the simulation aspect, i.e. depart Ferntree Gully 2 on clear medium, would the simulated train achieve >40 km/h before clearing the points anyway? Also, does Simsig have a function for introducing curve speed limits in the middle of a random block section, perhaps based on the TSR designs? i.e. you could hardcode all medium aspect signals to introduce a TSR at 25mph for the relevant length.

I'm also assuming that altering the bitmap to show speed signalling aspects instead of 3/4-aspect signals is a separate issue to the underlying coding and train operations?

Log in to reply
Upper Fern Tree Gully to Belgrave - Melbourne, Victoria 03/11/2021 at 01:51 #142166
GeoffM
Avatar
6376 posts
Anothersignalman in post 142163 said:
flabberdacks in post 142161 said:
The Victorian 'clear normal -> reduce -> medium -> stop' four-aspect sequence is not the same as the British 'clear - advance caution - caution - stop' sequence. The absolute requirement to travel at 40kph through the block on a medium aspect in ABS territory, or 40kph until clear of the points on a clear medium in ATC, does not apply to the UK scheme (to name a couple of examples - there are more). This means that if the Victorian aspect sequence isn't programmed, trains will not behave correctly. They'll move too quickly through sections (among other things), and the experience of signalling the Upper Gully area will be poorly represented as a result.

Thanks for that. I was hoping the relatively short block sections for the Upper Gully panel would address the simulation aspect, i.e. depart Ferntree Gully 2 on clear medium, would the simulated train achieve >40 km/h before clearing the points anyway? Also, does Simsig have a function for introducing curve speed limits in the middle of a random block section, perhaps based on the TSR designs? i.e. you could hardcode all medium aspect signals to introduce a TSR at 25mph for the relevant length.

I'm also assuming that altering the bitmap to show speed signalling aspects instead of 3/4-aspect signals is a separate issue to the underlying coding and train operations?
We can add speed limits anywhere in a track section, from 1m to its entire length, from/to any position.

For speed signalling we do have things like "after passing aspect X, reduce speed to Y immediately, and approach next signal at X speed" kind of thing. I believe it covers most US aspect rules, and likely the Australian ones too (Flabberdacks and others were instrumental in designing this).

SimSig Boss
Log in to reply
The following users said thank you: flabberdacks, Anothersignalman
Upper Fern Tree Gully to Belgrave - Melbourne, Victoria 03/11/2021 at 18:10 #142169
Anothersignalman
Avatar
96 posts
Anothersignalman in post 142149 said:
Thanks flabberdacks.

This could evolve into a larger sim at a later date, but I figured it'd be a good one to experiment with. It doesn't really need speed signalling, either, because the track layout is quite simple. Ignoring shunting signals for a bit, every signal is either Normal only, Medium only, or Medium-plus-Normal-Warning, which effectively devolves to four-aspect. The only possible exception is 36, which can show normal aspects only to the down platform and medium to either platform, but this is approach-cleared anyway and a post-1995 modification. For every other signal, standard 3- or 4-aspect signals combined with curve speed limits give enough detail.

Correction - signal 10, not 36. It gives normal aspects for following moves to the down platform only, else medium aspect.

Log in to reply