Seeding Gap?

You are here: Home > Forum > General > General questions, comments, and issues > Seeding Gap?

Page 1 of 1

Seeding Gap? 29/11/2019 at 00:53 #121986
VInce
Avatar
420 posts
Hi all,

On the Peterborough sim I've been trying to seed a failed train with dead loco attached on the Flyash Line and then have a loco run onto it to assist it forward.

If I create the train and seed it in the normal way and in the location list I just show Peterborough Flyash line with the activities which is a join and a next train. I set the seeding gap at P73 GPL to be 40m to allow plenty of room for the assisting engine but it does not go in clear of P73 GPL no matter what figure is in the seeding gap box.

If however I seed the train in the normal way and insert just a join activity at Peterborough Flyash Line and in the location list I show an actual train movement starting at Peterborough Flyash Line with locations New England North, Helpston etc.... then the assisting engine will go in clear

Just for the record, after numerous test timetables I can say that this has nothing to do with the length of the train.

I need to do the former method in order not to generate a length warning, but the only thing that works is the second example, which means the length of the train is incorrect.

Seems odd to me but is this working as designed and does the fact that the seed is at a GPL have anything to do with it?

Vince

Last edited: 29/11/2019 at 00:56 by VInce
Reason: None given

Log in to reply
Seeding Gap? 29/11/2019 at 01:51 #121987
postal
Avatar
3897 posts
Online
VInce in post 121986 said:
Hi all,

On the Peterborough sim I've been trying to seed a failed train with dead loco attached on the Flyash Line and then have a loco run onto it to assist it forward.

If I create the train and seed it in the normal way and in the location list I just show Peterborough Flyash line with the activities which is a join and a next train. I set the seeding gap at P73 GPL to be 40m to allow plenty of room for the assisting engine but it does not go in clear of P73 GPL no matter what figure is in the seeding gap box.

If however I seed the train in the normal way and insert just a join activity at Peterborough Flyash Line and in the location list I show an actual train movement starting at Peterborough Flyash Line with locations New England North, Helpston etc.... then the assisting engine will go in clear

Just for the record, after numerous test timetables I can say that this has nothing to do with the length of the train.

I need to do the former method in order not to generate a length warning, but the only thing that works is the second example, which means the length of the train is incorrect.

Seems odd to me but is this working as designed and does the fact that the seed is at a GPL have anything to do with it?

Vince
Not sure I am following what you say, but how do you mean that the length of the train is incorrect? Suppose you have a train set up with a train type showing a length of 250m. You then add a joining loco (say 20m.) and send the train forward. The train type without the joined portion is still 250m. in the TT but the core code keeps book on the lengths and the F2 display will show the length as 250m. plus join or 270m. in total and will apply that length when clearing track circuits, drawing clear of points etc.

Or have I totally misunderstood what you are driving at?

"No question is too stupid, there are just some stupid answers" - Dr. Michael Reece, 12/08/1927 - 03/06/2019. Electrical engineer and inventor
Log in to reply
Seeding Gap? 29/11/2019 at 02:59 #121988
VInce
Avatar
420 posts
Yes, I'm sorry its not very clear.

Attached are two test timetables.

Test Timetable 1

The assisting loco gets is clear of GPL P73

There are two train types, a failed Class 45 and and assisting loco a Class 47.

When they have joined, the core code correctly shows the length as 40m for the forward train but the description of the train is incorrect in that 0Z00, now consists of a Class 47 and a Class 45 but is only shown as a Class 45 (failed)

To rectify the descriptions we need to introduce a third train type which is a Class 45 + Class 47 and a third train (0Z02).

This is Test Timetable 2.

The assisting loco arrives, and runs onto the failed loco - it doesn't get in clear of GPL P73 although its position is still the same as it was in Test Timetable 1. When the join is complete, if ARS is being used, the combined train (0Z02) sets off wrong line towards New England because is there's no signal to stop it.

Hope that's a bit clearer.

Vince

Post has attachments. Log in to view them.
Last edited: 29/11/2019 at 03:17 by VInce
Reason: None given

Log in to reply
Seeding Gap? 29/11/2019 at 04:14 #121989
MarkC
Avatar
888 posts
I think I understand what you are trying to do, however in this case seeding gap will not work as the train on seeding in the sim will always approach a signal or bufferstop.

What you would need to do is adjust the stopping position on the flyash you would need to set stopping position to farend with an adjustment of 40 going from what you stated in your first post, that should enable it to stop 40m from the signal and in this case it may be best to seed the train at a signal before the flyash line.

It's not my fault! I was left unsupervised.
Log in to reply
Seeding Gap? 29/11/2019 at 04:51 #121990
VInce
Avatar
420 posts
MarkC in post 121989 said:
I think I understand what you are trying to do, however in this case seeding gap will not work as the train on seeding in the sim will always approach a signal or bufferstop.

What you would need to do is adjust the stopping position on the flyash you would need to set stopping position to farend with an adjustment of 40 going from what you stated in your first post, that should enable it to stop 40m from the signal and in this case it may be best to seed the train at a signal before the flyash line.
Thanks Mark, but that doesn't explain why it gets in clear in one instance and not in the other.

Vince

Log in to reply
Seeding Gap? 01/12/2019 at 23:06 #122027
VInce
Avatar
420 posts
Hi all,

I've given it a few days for an answer so can anyone explain, in the two test timetables I've attached to another message in this thread why 0Z01 gets in clear in example 1 but doesn't in example 2 when the position of the 0Z00 doesn't change.

Vince

Last edited: 01/12/2019 at 23:14 by VInce
Reason: None given

Log in to reply
Seeding Gap? 02/12/2019 at 00:56 #122028
postal
Avatar
3897 posts
Online
VInce in post 122027 said:
Hi all,

I've given it a few days for an answer so can anyone explain, in the two test timetables I've attached to another message in this thread why 0Z01 gets in clear in example 1 but doesn't in example 2 when the position of the 0Z00 doesn't change.

Vince
Perhaps someone with Tester access to the sim can confirm whether the position of 0Z00 is actually the same between the 2 TTs (or perhaps Clive would be willing to grant you tester access so you can check for yourself).

"No question is too stupid, there are just some stupid answers" - Dr. Michael Reece, 12/08/1927 - 03/06/2019. Electrical engineer and inventor
Log in to reply
Seeding Gap? 02/12/2019 at 01:40 #122029
JamesN
Avatar
1129 posts
Online
In example 1:

0Z00 seeds at 295/335 on the Flyash line in motion, then proceeds to stop at 310/335

In example 2:

0Z00 seeds at 295/335 on the Flyash line in motion, then proceeds to stop at 334/335

The joining logic is more complicated than I fully understand; but if I recall correctly, if the sim knows it needs X amount of room to join, it will leave that amount of room available. For it to know how much room it needs; it needs to know which way the timetable (not train, they are separate things in SimSig) is going. In example 1, the join activity is mid-timetable. The train knows that after joining it needs to continue in the down direction - and it can determine it needs to leave room at the northern end of the train as the timetable for 0Z01 arrives in the down direction. In example 2 it only has one location in its timetable, it can't work out which way its meant to be going next. Seeding it at signal forces the train to head in a down direction, but the timetable is in an indeterminate no-mans state. It still knows 0Z01 arrives down, but can't work out what's happening next from the info it has.

Trains will always first try to move when seeded, so the seeding gap alone is a bit of a red herring, but necessary as I'll come to.

The solution I'd propose is not only to provide a seeding gap - so that the train appears before the stopping point (as much is necessary as otherwise the train might miss the stopping point), but I'd also suggest a stopping point adjustment against the location.

EG:-
0Z00 seed at S73
Seeding Gap 55m
Peterborough Flyash Line Stopping Position "Far End Exact" with Adjustment 40

You could reduce those values as you see fit, but I'd leave a dozen metres between seeding gap and adjustment to give the train a second or two to realise it is meant to be stopping and where it is meant to be stopping.

That way it won't matter which way round you do the timetable, the train will always stop in the right place.

HTH

EDIT to add: While a useful tidbit on timetabling and joining trains, in its raw form the above is probably a little too vague to go directly into the wiki. If 1) We can clarify the exact correct behaviour is as I described and 2) someone with a better command of the english language wants to reword it into friendly, documentation-speak then fire away

Last edited: 02/12/2019 at 01:50 by JamesN
Reason: None given

Log in to reply
The following users said thank you: postal, VInce, Steamer
Seeding Gap? 02/12/2019 at 18:34 #122041
VInce
Avatar
420 posts
Hi,

Thank you - I'm very grateful for this and I've learned some of the processes involved from your explanation.

I did try this on one of the numerous scenarios I tested but I guess I failed to get the numbers right.

But this is progress and I shall do some more experimentation based on this explanation.

Thank you.

Vince

Last edited: 02/12/2019 at 18:46 by VInce
Reason: None given

Log in to reply