scale time and clocks

[worzel]

Apologies if this has been asked before but i can't find anything on this topic on the site

 

I am running my prototype based layout (oo) to the actual working timetable currently on a sequence basis and want to run it against the pressure of real ( compressed) time

 

Could anyone point me in the direction of any articles on how to speed a clock up to either triple (20 mins to the hour) or double( 30mins to the hour) I also want to be able to switch the clock off part way through the operation session to restart the next day at the same point

 

I saw something from america on the web but it was in the region of 355 which i thiught was a bit stiff

 

If anyone has tried this iwould really appreciate any advice

 

ps i know hornbys elite has this function but i'm not dcc yet and am probably going dynamis anyway for the walk round ability

 

thanks in advance

[Stuart-AU]

Hi..

 

you could build a kit...

http://electronics.mdodd.com/fastclock.html

http://railwaykits.net/FClkP.htm

 

or use your PC

http://www.stanstrains.com/Software.htm

http://tt-module.de/index.php?seite=fast-clock-for-model-railroads&sprache=en

http://www.dotric.com.au/trains/FastClock.htm

 

I think that will do, a simple GOOGLE search will get more

 

Stuart-AU

[andyrush]

We have just started using the Apple Fast Clock app for iPhones and iPods on a G scale garden railway, set at x4 when everything is going all right or adjusted to x2 when it isn't (or stopped altogether at tea time!). The timetable for the line has been devised with this in mind and first impressions are good - further experience next year may reveal flaws.... See http://itunes.apple.com/gb/app/fast-clock/id361958630?mt=8

 

Andy

[Gloucester Road]

Good

Apologies if this has been asked before but i can't find anything on this topic on the site

 

I am running my prototype based layout (oo) to the actual working timetable currently on a sequence basis and want to run it against the pressure of real ( compressed) time

 

Could anyone point me in the direction of any articles on how to speed a clock up to either triple (20 mins to the hour) or double( 30mins to the hour) I also want to be able to switch the clock off part way through the operation session to restart the next day at the same point

 

I saw something from america on the web but it was in the region of 355 which i thiught was a bit stiff

 

If anyone has tried this iwould really appreciate any advice

 

ps i know hornbys elite has this function but i'm not dcc yet and am probably going dynamis anyway for the walk round ability

 

thanks in advance

 

Good question. It is a very popular way to run model trains here in the states. I have tried making a fast clock and gave up for one of the online version.

[Guest jim s-w]

Time doesn't scale.

 

if you have a scale 10 mile branch line and you run a train along it at an scale average of 20mph it will take half an hour just like the real thing.

 

The time it takes a model loco to run round a model train, if using scale speeds, should be exactly the same as the real thing.

 

Its obvious when you think about it.

 

Cheers

 

Jim

[RJS1977]

I think where the speeded-up clock comes in is that most people don't have room for a 10 mile branch line - most people would be lucky to manage half a mile!

 

So by speeding the clock up 20 times, a model train that takes a minute and a half from end to end has taken a "scale" half hour. It's just not the same scale as most of the rest of the layout!!

[jjnewitt]

A model train that has taken a minute and a half to traverse a set distance has taken a minute and a half, not 'half an hour'. How fast it is going depends on how far it's travelled in that time. As Jim says you can't scale time, it makes no sense at all. How can you scale something that is essentially an abstract concept? You can't go out, find a peice of time, measure it and then devide it by whatever scale your using. If you make a clock go twice as fast as it would normally all you end up with is trains going twice as fast as they should.

[Coombe Barton]

Isn't what we're trying to do, if we run at scale speeds, is to reduce the "time where nothing happens"?

 

For instance if a train comes into a branch terminus every two hours, runs round and departs twenty minute later, we need the arrival, run round and departure to be "to scale" but the majority of the twenty minutes and hour and 40 minutes to be reduced. We want to see the action, but not the inaction.

[zarniwhoop]

Isn't what we're trying to do, if we run at scale speeds, is to reduce the "time where nothing happens"?

 

For instance if a train comes into a branch terminus every two hours, runs round and departs twenty minute later, we need the arrival, run round and departure to be "to scale" but the majority of the twenty minutes and hour and 40 minutes to be reduced. We want to see the action, but not the inaction.

 

Very true, but my impression of "scale" clocks is that they continually run at a fast time (but perhaps not 76x or 87x real time). So, what is needed is a clock that runs at both real time and a faster time - and then you have to signal to it to speed up and slow down. Seems perfectly feasible if you are sufficiently into computerised control, but the control won't be trivial : either a "timekeeper" changes it and the operators then either do something or relax, or you need detectors to tell when a train is in public view (ignoring, for the moment, the branchline train which stays at its terminus overnight .

 

Conceptually, I find the idea appealing - I'd like to run a small doubletrack layout, automated, with say 4 trains in each direction per hour, and complete an hour in maybe 10 minutes, but the technicalities are *interesting*. And then you find, if it is on public display, that you really need to shrink the hour to 9 minutes, because the spectators are bored, or perhaps you can increase it to 12 minutes and still get a "realistic" feeling.

 

My general impression is that the system is used in America where they have big layouts, and run a full 24-hour timetable, but that they probably only use the clock to signal when an operation should start - unless they want to run the trains at silly fast speeds. No doubt someone will give a more detailed explanation of how it works if I'm wrong

 

ĸen

[Edwin_m]

Presumably on those big layouts the clock tells them when a particular move such as a loco runround should start, but during the move no attention is paid to the clock. The runround itself takes about as long in real time as it would on the prototype, but longer as measured by the clock. However moves between one station and the next now take a "realistic" amount of time as measured by the clock.

 

I believe Railroad and Co software can be configured so its fast clock affects the acceleration/deceleration rate of trains, which seems to me to be totally ridiculous.

[Danemouth]

NCE Powercab also has scale time available - can be adjusted IIRC,

 

Not used it myself!

 

Dave

[clecklewyke]

what is needed is a clock that runs at both real time and a faster time - and then you have to signal to it to speed up and slow down.

 

Peter Denny did something like this with his "Automatic Crispin", which was a mechanical computer which replaced his fiddle yard operator (his son, Crispin) when he left home. It had a home-made clock and, using stored details of the timetable, switched appropriate points and sent and responded to block bell codes at the appropriate times. Details can be found in his books on Buckingham Great Central and articles in Railway Modeller and MRJ. The clock would stop if trains were running too late.

 

I made a simplified version of this which I called "Digital Crispin". AC/DC - gettit? This was a programme written in Basic, originally on a ZX Spectrum, then for Micorsoft DOS (remember those?) It used the computer's clock, which could be set to any speed and speeded up by tapping on any key, but slowed down when trains were actually running. A description is in MRJ 61.

 

I would love to resurrect it, if only I could run Basic on my Macbook. Automatic Crispin still exists, and its restoration is on Tony Gee's to-do list.

 

But! You don't always need to slow down clocks when running a timetable. When we operate Mike Norris' Preston we usually have a three hour session (any more and our brains would fry...) Starting the timetable at midnight we are doing well if we get to 2 o'clock in the morning, there are so many trains during the night!

 

Ian

[beast66606]

Time doesn't scale.

 

if you have a scale 10 mile branch line and you run a train along it at an scale average of 20mph it will take half an hour just like the real thing.

 

The time it takes a model loco to run round a model train, if using scale speeds, should be exactly the same as the real thing.

 

 

Good points Jim,

--

 

If you want to scale time your best bet (imho) is to scale the gaps, rather than the moves, so if your timetable had a departure at say 11:00 and another at 11:10, scale the gap to be 1 minute but then run at normal speed when actually performing the move, maybe rather than scale time, you need the ability to jump forward a number of minutes on a mouse click to cut out/down the gaps in movements.

[28XX]

Surely the best thing to do during 'periods if inaction' is to get on with some modelling. No layout is ever truly finished as we all know. Keep an eye on the clock like a good signalman would and run the next train when its due. Keep an electric clock running 'layout time' and take the battery out when the session ends, put it back in when the next session starts.

 

If yours is a country branch and the 'box is closed 22:00 to 06:00, when the late shift bobby goes home, simply advance the clock by hand to the start of the next day.

 

This approach mirrors real life on the railway when, during such lulls in the service, the staff got on with paperwork, gardening or giving hair cuts.

[andyrush]

For those of us with big layouts where the various 'stations' (and I use the term loosely to include fiddle yards and hidden 'handover tracks'), it is possible to devise a timetable to overcome the inconsistencies mentioned above. No attempt is being made to 'scale' time, which is clearly impossible, but rather to use the speeded up clock to co-ordinate the activities of widely separated operators in a way that keeps the 'flow' of operations working in a realistic way.

 

On our Swiss fantasy, I wanted to replicate the 'rhythm' of the 'railway day', with overnight freight workings etc, building up to early morning light engine and empty stock workings, followed by a daytime cycle of passenger services interspersed by mainly local goods trains. The whole thing then reverses towards the end of the day. A key feature is that at the end of each operating session (2½ hours weekdays, 5 hours on Saturdays) the position is 'frozen' and recorded so that we can start from the same place in the timetable at our next session.

 

I recognise that this style of operation is not possible (or desirable!) for everybody and it is relatively easy to adjust the timetable to suit when a freelance layout is the subject. As I said in my earlier post we are only in the early stages, but initial results have made the concept worth pursuing.

 

Andy

[Guest jim s-w]

I made a simplified version of this which I called "Digital Crispin". AC/DC - gettit? This was a programme written in Basic, originally on a ZX Spectrum, then for Micorsoft DOS (remember those?) It used the computer's clock, which could be set to any speed and speeded up by tapping on any key, but slowed down when trains were actually running. A description is in MRJ 61.

 

I would love to resurrect it, if only I could run Basic on my Macbook. Automatic Crispin still exists, and its restoration is on Tony Gee's to-do list.

 

Hi Ian

 

You can - see Chipmonk basic - http://www.nicholson.com/rhn/basic/

 

HTH

 

Jim

[shortliner]

Don't re-invent the wheel http://www.wesleysteiner.com/mr/mrfc/mrfc.html Change the time compression as indicated, and the clicki in the top yellowish bar

[Dungrange]

Time doesn't scale.

 

The time it takes a model loco to run round a model train, if using scale speeds, should be exactly the same as the real thing.

Change the time compression as indicated

 

The question of 'scaling' time is often asked and of course it is not possible to 'scale' time. However, it is possible to 'compress' and 'skip' time in an operating sequence.

 

If the track layout is genuinely reproduced to scale, then as Jim has highlighted, the time it takes a model loco to run round a model train, if using scale speeds, should be exactly the same as the real thing. Time therefore obviously scales at 1:1 irrespective of the scale of the models. However, few of us have the space to replicate our chosen station or railway without some selective linear compression. Platform lengths, turnout lengths, run round loops, signal sections, etc may all be slightly shorter than simply scaling down the prototype. Therefore, if our runround loop is only half of the length of the prototype and we drive a model locomotive at a scale speed, it would be possible to complete the runround move in less time than the prototype. To overcome this compression of linear distance on the model and therefore time taken to complete a movement at scale speed, it may be desirable to adjust the rate of change of time during this movement to achieve the desired effect. However, this is compressing time in the same ratio as we have compressed linear distances on the layout over and above the scaling down of the prototype dimensions, rather than 'scaling' time per se.

 

If the length of the runround loop on the layout was compressed such that it was only half of the length of the prototype, then making the clock run twice as fast as real time only makes sense whilst the locomotive is actually in motion over the compressed distance. The time taken to change the points, signals etc is probably close to real time, so the overall time taken to complete the movement will be more than half of the time taken to run round on the prototype. Running a clock at twice or three times real time is therefore probably too fast for many typical layouts (particularly those with shunting moves). Such time compression is perhaps more realistic when considering large multi-station layouts (as I understand are popular in the USA) or garden railways, where there is likely to be significant linear compression of distances between stations.

 

The other point highlighted is of course 'skipping' time during the 'boring' bits when nothing much happens. Again this can be achieved by making a clock run faster, but is neither scaling nor compressing time and is wholly dependent on personal preference and how busy the prototype operations are. That is, there is no specific formula for deriving how much faster than real time a clock could be set during these periods.

 

Regards

 

David

[Kenton]

Jim is absolutely correct - time does not scale. Always been a source of long arguments on RMWeb and in the modelling press.

 

But it is correct only in the absolute. Most layouts - in likelihood nearly all layouts are neither absolute or even prototypical. We readily compress distance in addition to scaling it. The correct term is compressing time. A choice some of us choose to make to effectively speed up our interpretation of the scene. The value you use will effectively alter the true speed of the trains and can range from a small amount to the unrealistic speeds seen on children's train sets.

 

All of this is relative. We can choose to make all of our trains run at unrealistic speeds - that industrial shunter moving through the scene at the same effective speed as an express - simply compressing time for the shunter relative to the express loco. A similar mismatch is frequently seen on layouts between goods and passenger traffic. We can also choose to change the compression of time with respect to white time - those intervals in the proceedings where nothing is happening on the layout. That period the public hates at an expo when nothing in reality happens. We accept it as much as we accept seeing a tail chaser re-emerge from the opposite side of the scene just moments after having seen it leave the other side. A prototypical impossibility.

 

So in conclusion time does not scale but we choose to compress time at varying degrees to suit our own operational characteristics. That compression is a variable divergence from the prototypical.

 

Wouldn't it be dull if we didn't, 4 or 5 trains a day on some of those branch lines?

[Guest jim s-w]

If you dive to roughly scale speeds the interesting bits will take roughly the same amount of time. (running round a half length train will be the same as if the real thing is half length - you dont need to drive slower to make it look better) If you omit the dull bits then you really just follow a sequence the timings of which are dictated by the trains. The whole need for a clock is completely removed.

 

Cheers

 

Jim

[The Stationmaster]

For those of us with big layouts where the various 'stations' (and I use the term loosely to include fiddle yards and hidden 'handover tracks'), it is possible to devise a timetable to overcome the inconsistencies mentioned above. No attempt is being made to 'scale' time, which is clearly impossible, but rather to use the speeded up clock to co-ordinate the activities of widely separated operators in a way that keeps the 'flow' of operations working in a realistic way.

On our Swiss fantasy, I wanted to replicate the 'rhythm' of the 'railway day', with overnight freight workings etc, building up to early morning light engine and empty stock workings, followed by a daytime cycle of passenger services interspersed by mainly local goods trains. The whole thing then reverses towards the end of the day. A key feature is that at the end of each operating session (2½ hours weekdays, 5 hours on Saturdays) the position is 'frozen' and recorded so that we can start from the same place in the timetable at our next session.

I recognise that this style of operation is not possible (or desirable!) for everybody and it is relatively easy to adjust the timetable to suit when a freelance layout is the subject. As I said in my earlier post we are only in the early stages, but initial results have made the concept worth pursuing.

Andy

 

Very interesting Andy and while I realise the Ruschban is not like lots of (most?) model railways you still presumably have to take account of the really time consuming tasks such as a spot of shunting or running-round as well as the more straightforward point-to-point running. So how do you account for those activities or do you just accept that they will take 'a longer time' on your speeded-up clock in order to keep your train times in a sensible relationship to these activities?

 

Coming away from that another answer to the initial question is why bother with time at all? Because of the difficulty impossibility of carrying out some tasks in a compressed time (good choice of words there I think Kenton) effectively what happens is that you construct your timetable around a sequence of events, i.e. the passenger train can only depart after it has been platformed and has got an engine etc on the front, once it has departed something else can happen (or depending on protypicality of operating can't happen until ...). Thus one thing follows another irrespective of what the clock says and if you want to put a time against it you can do it however you like - even if it is just turning over a card to show the next time and event(s) which take place at that time.

[andyrush]

Very interesting Andy and while I realise the Ruschban is not like lots of (most?) model railways you still presumably have to take account of the really time consuming tasks such as a spot of shunting or running-round as well as the more straightforward point-to-point running. So how do you account for those activities or do you just accept that they will take 'a longer time' on your speeded-up clock in order to keep your train times in a sensible relationship to these activities?

 

Coming away from that another answer to the initial question is why bother with time at all? Because of the difficulty impossibility of carrying out some tasks in a compressed time (good choice of words there I think Kenton) effectively what happens is that you construct your timetable around a sequence of events, i.e. the passenger train can only depart after it has been platformed and has got an engine etc on the front, once it has departed something else can happen (or depending on protypicality of operating can't happen until ...). Thus one thing follows another irrespective of what the clock says and if you want to put a time against it you can do it however you like - even if it is just turning over a card to show the next time and event(s) which take place at that time.

 

Hello Mike

 

In answer to your first point, the answer is yes, allowances have been made in the timetable for activities which take longer than others. For example, using the pilot from the goods loop to take a van off the rear of a train at the principal station and deposit it at the loading dock, then returning the pilot to the headshunt takes about 3 minutes real time (assuming no other movements hinder the process), whereas the train the van has come off may have only taken 3 actual minutes to come from its previous station. However, this won't be very obvious to the staff at the main station as the train has come from an intermediate hidden handover section and so long as it arrives and departs (to another hidden section) at timetable times it doesn't really matter.

 

In answer to the second question, why bother, I would answer that the Ruschbahn used to be run to a sequence, but the difficulty of co-ordinating people scattered around the garden and the constant 'where are we in the sequence' questions led me to try out speeded up clocks, principally as a means of synchronising things. We could have adopted a CTC dispatcher type system I suppose, but we have a problem finding enough people with railway operating aptitude as it is, without going for something as technical as that....

 

You are always welcome to come and see how it is working in practice!

 

Regards

 

Andy

[clecklewyke]

Hi Ian

 

You can - see Chipmonk basic - http://www.nicholson.com/rhn/basic/

 

HTH

 

Jim

Thanks, Jim. Looks like I've a lot of revision to do. I haven't used Basic for nearly twenty years. I cannot believe it's been that long!

 

Ian

[The Stationmaster]

 

In answer to the second question, why bother, I would answer that the Ruschbahn used to be run to a sequence, but the difficulty of co-ordinating people scattered around the garden and the constant 'where are we in the sequence' questions led me to try out speeded up clocks, principally as a means of synchronising things. We could have adopted a CTC dispatcher type system I suppose, but we have a problem finding enough people with railway operating aptitude as it is, without going for something as technical as that....

 

You are always welcome to come and see how it is working in practice!

Regards

 

Andy

 

Thanks Andy - it's definitely on my 'to do' list but alas it's not the only thing on it - really must get the garden here sorted before coming to look at yours (or rather what's in it )

[Clive Mortimore]

If you dive to roughly scale speeds the interesting bits will take roughly the same amount of time. (running round a half length train will be the same as if the real thing is half length - you dont need to drive slower to make it look better) If you omit the dull bits then you really just follow a sequence the timings of which are dictated by the trains. The whole need for a clock is completely removed.

 

Cheers

 

Jim

 

Hi Jim

 

I agree that keeping the sequence of movements to real time and omitting the non-train movements does compress the time. Trouble is the "dull bits" on the real railway are when the railwaymen are loading/unloading the wagons, selling the tickets, changing the signal lamps, feeding Hercules the horse before his next delivery round etc. etc. Most of this we cannot replicate but must be mindful that this is what happens so that we have a "reason" for our stations on our layouts.

 

I personally like to see gaps in the train movements on layouts that do not represent Clapham Junction in the rush hour. It gives me more time to view the other modelling.