Train spotting at Finsbury Square

[Mallard60022]

12 hours ago, Baby Deltic said:

It does in kettles

Not if you use special equipment and substances. The latter is not used externally so that you don't notice if anything is happening or don't care.

Ar$£

[great northern]

On 08/01/2020 at 12:58, CUTLER2579 said:

 

2.368421052631578 Seconds approximately.

 

 

 

 

 

2.

Oh, come on man, why can't you be accurate.

[manna]

G'day Folks

 

Wonder what that figure is in Fractions ...................

 

manna

[Dr Gerbil-Fritters]

2 hours ago, manna said:

G'day Folks

 

Wonder what that figure is in Fractions ...................

 

manna

 

In fractions?  A smidge under 3 seconds

[Clive Mortimore]

31 minutes ago, Dr Gerbil-Fritters said:

 

In fractions?  A smidge under 3 seconds

I ain't that energetic.

[PhilJ W]

But that figure is only one dimensional and you are talking of the turntable that is 3 dimensional. If you cubed the figure you would come to a more reasonable 24/25 seconds.

[Baby Deltic]

1 hour ago, PhilJ W said:

But that figure is only one dimensional and you are talking of the turntable that is 3 dimensional. If you cubed the figure you would come to a more reasonable 24/25 seconds.

Now that would be interesting; A one dimensional layout..

[St Enodoc]

4 minutes ago, Baby Deltic said:

Now that would be interesting; A one dimensional layout..

A bit pointless really...

[CUTLER2579]

1 hour ago, St Enodoc said:

A bit pointless really...

 

That would be boring having NO points.

[PhilJ W]

3 hours ago, Baby Deltic said:

Now that would be interesting; A one dimensional layout..

What in the USA is called V-scale. A virtual layout created inside a computer.

[Baby Deltic]

6 hours ago, PhilJ W said:

What in the USA is called V-scale. A virtual layout created inside a computer.

So it’s not full of figures which are reptilians disguised a humans then?

[Dungrange]

On 07/01/2020 at 22:58, 31A said:

I'm inclined to think maybe it's one of those things that doesn't 'scale' very easily (like speed, or time).

 

There is no such thing as 'scale time': those who try to tell you otherwise are talking out of their ar$e.  Time doesn't scale in the same way as the first three dimensions - it technically doesn't scale at all.

 

Let's say that you're interested in a one mile stretch of line where the trains travel at 60 mph.  The prototype trains would traverse that section of line in 60 seconds.  If you build a perfect scale model of that section of line (ie one without compromises), then the model trains should take 60 seconds to traverse the model irrespective of what scale the model has been constructed to.  Time scales at 1:1 even although your model has been scaled at, say, 1:76 for the first three dimensions.  Since speed is distance divided by time, speed scales in exactly the same way as length, width and height on a perfect scale model.  That is a scale model traversing a scale mile of track in 60 seconds is travelling at a scale 60 miles per hour.  That's all fairly straightforward.

 

Angles don't scale either, so since the number of revolutions that a turntable would make doesn't scale and neither does time, the number of revolutions per hour should be the same in the model as the prototype.  Therefore, if it takes two to three minutes to turn a locomotive in reality, then it should take two to three minutes to turn a locomotive on a perfect scale model.  Easy.

 

The problem is that very few of us have the space to build a perfect scale model.  We need to make compromises and one of the most common is linear compression.  We don't have the space to represent our favoured mile of track, so we squeeze the key features into maybe half of the distance.  This is the source of the perceived 'scaling' issues with speed and time, as we are now presented with a choice.  If we run our models at what would be scale speed on a perfect scale model, then we will traverse the stretch of line that we have modelled, in just 30 seconds (ie in half of the real time).  Alternatively, we could drive the train so that it still takes 60 seconds to travel our representation of our favoured mile of track, but that would be travelling at half of scale speed.

 

I could argue that neither one of these choices is inherently more accurate than the other, but I think that the majority of us prefer to drive the train at something closer to true scale speed and pretend that the 30 seconds that it took us to drive the train from one end of the layout to the other was actually 60 seconds (as per the prototype).  We haven't scaled time as such, but have pretended that the clock was running faster to overcome the effects of linear compression in our less than perfect model.  This is really where the idea of operating with fast clocks come from.  If we have to compress the stretch of line that we are interested in by a factor of two, then we want our model time to pass twice as quickly so that we can run our trains at scale speed.  If we had less space and had to compress the stretch of line that we are interested in by a factor of three, then we want our model time to pass three times as quickly.  This isn't scaling time, but pretending that it is passing faster in the model than it is in reality and the extent to which we perceive that is related to the amount of linear compression that we were forced to adopt.

 

Where does this lead us to with regards your turntable?  Well, although revolutions per hour technically don't scale, if we are imagining a faster clock to offset the effects of linear compression on the mainline, then we need to assume that same faster clock in relation to other aspects of the model as well, such as turning a locomotive on the turntable.  Therefore, if you have been forced to compress the length of your running lines such that it would be appropriate to assume that time in the model is passing three times quicker than reality, then that means that the speed of your turntable needs to be increased by a factor of three from reality.   That is, if it takes three minutes to turn a locomotive in reality, because you are perceiving time to run three times faster, it should take just one minute in your model.

 

The problem is that the amount of linear compression that you have adopted is not easy to define and will no doubt be more in some locations than in others and this means that it's not possible to accurately derive a formula to calculate how long it should take to rotate.  Furthermore, every individual looking at your layout will perceive slightly different levels of linear compression and therefore have slightly different perceptions of what they think 'scale time' is: it would be more accurate to call it something like compressed time.  Ultimately all I can say is that the people who say "if it looks right, it is right", are the ones who are right.  If you are happy with how fast it rotates, then that is all that matters - there is no formula to say you have got it wrong.  I think the only guidance that anyone can give, is that it's probably better to operate a little faster than reality, so if it were to take two minutes to turn a locomotive on your model, I think most people would perceive that as being too slow.

[St Enodoc]

14 hours ago, CUTLER2579 said:

 

That would be boring having NO points.

Absolutely - quite one-dimensional in fact!

[St Enodoc]

3 hours ago, Dungrange said:

 

There is no such thing as 'scale time': those who try to tell you otherwise are talking out of their ar$e.  Time doesn't scale in the same way as the first three dimensions - it technically doesn't scale at all.

 

Let's say that you're interested in a one mile stretch of line where the trains travel at 60 mph.  The prototype trains would traverse that section of line in 60 seconds.  If you build a perfect scale model of that section of line (ie one without compromises), then the model trains should take 60 seconds to traverse the model irrespective of what scale the model has been constructed to.  Time scales at 1:1 even although your model has been scaled at, say, 1:76 for the first three dimensions.  Since speed is distance divided by time, speed scales in exactly the same way as length, width and height on a perfect scale model.  That is a scale model traversing a scale mile of track in 60 seconds is travelling at a scale 60 miles per hour.  That's all fairly straightforward.

 

Angles don't scale either, so since the number of revolutions that a turntable would make doesn't scale and neither does time, the number of revolutions per hour should be the same in the model as the prototype.  Therefore, if it takes two to three minutes to turn a locomotive in reality, then it should take two to three minutes to turn a locomotive on a perfect scale model.  Easy.

 

The problem is that very few of us have the space to build a perfect scale model.  We need to make compromises and one of the most common is linear compression.  We don't have the space to represent our favoured mile of track, so we squeeze the key features into maybe half of the distance.  This is the source of the perceived 'scaling' issues with speed and time, as we are now presented with a choice.  If we run our models at what would be scale speed on a perfect scale model, then we will traverse the stretch of line that we have modelled, in just 30 seconds (ie in half of the real time).  Alternatively, we could drive the train so that it still takes 60 seconds to travel our representation of our favoured mile of track, but that would be travelling at half of scale speed.

 

I could argue that neither one of these choices is inherently more accurate than the other, but I think that the majority of us prefer to drive the train at something closer to true scale speed and pretend that the 30 seconds that it took us to drive the train from one end of the layout to the other was actually 60 seconds (as per the prototype).  We haven't scaled time as such, but have pretended that the clock was running faster to overcome the effects of linear compression in our less than perfect model.  This is really where the idea of operating with fast clocks come from.  If we have to compress the stretch of line that we are interested in by a factor of two, then we want our model time to pass twice as quickly so that we can run our trains at scale speed.  If we had less space and had to compress the stretch of line that we are interested in by a factor of three, then we want our model time to pass three times as quickly.  This isn't scaling time, but pretending that it is passing faster in the model than it is in reality and the extent to which we perceive that is related to the amount of linear compression that we were forced to adopt.

 

Where does this lead us to with regards your turntable?  Well, although revolutions per hour technically don't scale, if we are imagining a faster clock to offset the effects of linear compression on the mainline, then we need to assume that same faster clock in relation to other aspects of the model as well, such as turning a locomotive on the turntable.  Therefore, if you have been forced to compress the length of your running lines such that it would be appropriate to assume that time in the model is passing three times quicker than reality, then that means that the speed of your turntable needs to be increased by a factor of three from reality.   That is, if it takes three minutes to turn a locomotive in reality, because you are perceiving time to run three times faster, it should take just one minute in your model.

 

The problem is that the amount of linear compression that you have adopted is not easy to define and will no doubt be more in some locations than in others and this means that it's not possible to accurately derive a formula to calculate how long it should take to rotate.  Furthermore, every individual looking at your layout will perceive slightly different levels of linear compression and therefore have slightly different perceptions of what they think 'scale time' is: it would be more accurate to call it something like compressed time.  Ultimately all I can say is that the people who say "if it looks right, it is right", are the ones who are right.  If you are happy with how fast it rotates, then that is all that matters - there is no formula to say you have got it wrong.  I think the only guidance that anyone can give, is that it's probably better to operate a little faster than reality, so if it were to take two minutes to turn a locomotive on your model, I think most people would perceive that as being too slow.

An excellent summary of the challenges of scaling. Of course, you can scale time and you can scale size but it's not usual to scale both at the same time - unless you are modelling, for example, river flows and you want to be able to see in a few hours what might take years in real life; or using a model in a film to simulate a train falling off a bridge where smaller size and shorter time are manipulated to give the illusion on the screen of larger size and a longer time.

 

Generally it makes more sense for us in the world of model railways to use scale size and real time as that fits our perception of reality more closely.

[31A]

Crumbs some good stuff here; not all of which I can pretend I understand, but thanks to all for the contributions!

 

Having watched the video clips linked above and others, I think quite possibly the turntable does spin a bit too quickly, and will investigate using a lower voltage.  I expect there will come a point where the voltage is too low for the motor to get the turntable started, but there's certainly room for experiment.

 

The turntable electrics need some tidying up anyway; I intend to put a DPDT centre off switch in the circuit to act as a reversing / emergency stop switch, and also an interlock with the track feed section switch so that the table can't be operated whilst a loco is moving on it, and vice versa.

[Baby Deltic]

All this talk about turntable drives and speeds has reminded me of something I used to come across regularly as a factory engineer back in the day - Geneva drives. What better way to get perfect indexing between tracks every time as long as the angles are equal?

Edited January 11, 2020 by Baby Deltic

[St Enodoc]

Just now, Baby Deltic said:

All this talk about turntable drives and speeds has reminded me of something I used to come across regularly as a factory engineer back in the day - Geneva drives.what better way to get perfect indexing between tracks every time as long as the angles are equal?

I think the original Tri-ang turntable used a Geneva - or similar perhaps - drive.

[31A]

4 minutes ago, Baby Deltic said:

All this talk about turntable drives and speeds has reminded me of something I used to come across regularly as a factory engineer back in the day - Geneva drives.what better way to get perfect indexing between tracks every time as long as the angles are equal?

 

1 minute ago, St Enodoc said:

I think the original Tri-ang turntable used a Geneva - or similar perhaps - drive.

 

Derrrr.....  Wot's one of them, then?

[Baby Deltic]

1 minute ago, St Enodoc said:

I think the original Tri-ang turntable used a Geneva - or similar perhaps - drive.

The only disadvantage I suppose is that they stop momentarily on each index because of their nature, plus of course they can move quite fast between indexes so you would need a slow crank motor. The best thing is probably still a stepper motor and encoder.

[Baby Deltic]

5 minutes ago, 31A said:

 

 

Derrrr.....  Wot's one of them, then?

Google is your friend. Loads of spiel on there, but I couldn’t get my iPad to copy and paste any links.

[Baby Deltic]

In hindsight Google isn’t anybody’s friend because it tracks everything. DuckDuckGo or (a VPN if you must use Google) is better.

[31A]

Just now, Baby Deltic said:

Google is your friend. Loads of spiel on there, but I couldn’t get my iPad to copy and paste any links.

 

Thank you; of course I should have thought of that before typing!  The Wikepedia entry has a good explanation, and it does look as if it would work quite well.  A very ingenious mechanism.

 

https://en.wikipedia.org/wiki/Geneva_drive

 

However I'm not sure whether it would cope where there the inlet tracks are close together as there would need to be a short 'swing' followed by a long 'swing' - could a locking segment be designed to allow this?  Also, I don't think I could make one from Meccano!!  

[Baby Deltic]

4 minutes ago, 31A said:

 

Thank you; of course I should have thought of that before typing!  The Wikepedia entry has a good explanation, and it does look as if it would work quite well.  A very ingenious mechanism.

 

https://en.wikipedia.org/wiki/Geneva_drive

 

However I'm not sure whether it would cope where there the inlet tracks are close together as there would need to be a short 'swing' followed by a long 'swing' - could a locking segment be designed to allow this?  Also, I don't think I could make one from Meccano!!  

Ideally you need about 8 equispaced tracks for it to work well and pay dividends. Of course with CAD packages and 3D printers being readily available at reasonable prices, the world is your oyster if you want to design,  simulate and create one to your own requirements.

Edited January 11, 2020 by Baby Deltic

[31A]

The timetable sequence / loco rostering spreadsheet can throw up some strange combinations at times, and this is what a nocturnal spotter would have seen if he poked his nose over the wall at City Road loco at the end of today's session.  If he'd been to the pub, he might think he'd had one too many!  Perhaps it is the "Claud" that was rumoured to have been set aside for preservation?

 

 

At the end of the sequence, the loco yard will contain one loco from of each of the Express Passenger, GN Main Line, GE Main Line and Outer Suburban links, and two from the Inner Suburban link, one of which will remain in the yard "Spare" the following day.  Apart from fuelling, it isn't a place where much maintenance would be carried out so the "Spare" loco is just there to make the place look busier, get in the way and save me having to find somewhere to plonk it in the fiddle yard.  In this case the "Spare" loco is 69531 on the right, which has been put on the coal wagon road for now but will be in the way when more coal arrives around midday tomorrow.

 

 

 

[lezz01]

Personally I have always thought that the best way to operate a turntable is to just wire it to a controller and drive it like an engine. If you want to make it more of a challenge you wire in two motors one of them free running with a big flywheel and just get used to which way and how much juice you give it. Now days of course you just wire it to a decoder with only one motor and set the inertia to your desired level and learn how to drive it. Of course the fun comes when you let someone else drive it........but then I always did a a mischievous streak or is that sadistic....

Regards Lez.