Theory of General Minories

[simon b]

2 hours ago, Compound2632 said:

What's missing there is the possibility of simultaneous arrivals and departures involving anything other than the upper platform line and one other. If the lower lines are given an independent connection into the departure line, clear of the double crossover, there's a lot more flexibility. 

 

It would but there are only 3 platforms available on that layout, those are 1,4, and 5.  As per the prototype the short platforms 2 and 3 are only used at night for parcel traffic, access to these is blocked by the other platforms during the day. I thought this could be an interesting feature to add to operations.

 

On CJF's plan, I could never work out how to shunt the 2 goods sidings on the kickback from platform 3. In the steam era the pilot could be used, but there wouldn't be a pilot in the era of EMU operation so this is my alternative arrangement.

 

It is getting away from the minories concept a little though.

Edited November 29, 2020 by simon b

[KeithMacdonald]

On 20/11/2020 at 22:05, simon b said:

I drew up a plan inspired by Clive Mortimer's Sheffield exchange mk1. It puts the approach pointwork on a curve to fit in the space I had available, and by using the Tillig single slip avoids the sharp angle of the Peco item. The only thing I wasn't sure of is the loco siding being a facing point on the up line, but trying to put it anywhere else involves takes too much space. Platforms are 5ft long which will take 4 mk1 suburban coaches and a class 31 at either end.

 

 

 

This is the tillig slip in question.

 

Thanks for that, it reminded me I've seen the Tillig Slip discussed here before (a Tillig -v- Peco topic).

 

 

I've tried to achieve something similar with Peco Streamline.

 

The track in red is a piece of flextrack, with a radius of 117cm.

 

Does it look better with a couple of short straights joined to the single slip?

 

That gives the flextrack a radius of 153cm, to smooth it out a bit more.

 

 

 

[Satan's Goldfish]

2 minutes ago, KeithMacdonald said:

 

Thanks for that, it reminded me I've seen the Tillig Slip discussed here before (a Tillig -v- Peco topic).

 

 

I've tried to achieve something similar with Peco Streamline.

 

The track in red is a piece of flextrack, with a radius of 117cm.

 

Does it look better with a couple of short straights joined to the single slip?

 

That gives the flextrack a radius of 153cm, to smooth it out a bit more.

 

 

 

 

I think many moons ago (when i had access to a machine with xtrack on) i tried this formation but using Peco curved points on the approach line and Peco slip. The problem with using them isn't the radius of the approach curve along side the slip, but the radius on the slip itself (60cm ish i think) and the radius of the inside leg of curved points (75cm ish?) 

 

i deliberately gave away my only Peco single slip to avoid the temptation...

[simon b]

That was the issue I had. I could get rid of the reverse curves in the point work, but the sharp angle of the peco slips just ruined it for me. I was hoping their bullhead version would be better, but sadly not.

[KeithMacdonald]

34 minutes ago, Satan's Goldfish said:

The problem with using them isn't the radius of the approach curve along side the slip, but the radius on the slip itself (60cm ish i think) and the radius of the inside leg of curved points (75cm ish?) 

 

i deliberately gave away my only Peco single slip to avoid the temptation...

 

Curses! Foiled again! (Muttley, do something!)

 

Out with the curved point then, back to the drawing board...

[KeithMacdonald]

Does anyone know what the radius is for the SL-89 long straight points and the SL-98 Y points?

 

Edited November 30, 2020 by KeithMacdonald
Edit: I nearly forgot to say, the flextrack (in red) is now a piece 52cm long on a 246cm radius. The short straights are Peco ST203

[KingEdwardII]

1) SL-89 large radius straight points = 1524mm / 60 inches

 

2) SL-98 large radius Y points = 1828mm / 72 inches

 

I've attached a file which is my gospel for Peco point geometry (not created by me as the page headers indicate...!)

 

Mike.

Peco_turnout_dimensions.pdf

[Harlequin]

22 minutes ago, KingEdwardII said:

1) SL-89 large radius straight points = 1524mm / 60 inches

 

2) SL-98 large radius Y points = 1828mm / 72 inches

 

I've attached a file which is my gospel for Peco point geometry (not created by me as the page headers indicate...!)

 

Mike.

Peco_turnout_dimensions.pdf 56.49 kB · 2 downloads

 

The "Nominal radii" stated by Peco and reproduced in that PDF don't match up with the real world unfortunately.

 

If you do the maths you can show that the largest possible radius for a straight turnout using the Peco geometry (12° turn at 1 inch offset) is 45.76in (1162.34mm).

 

So more like a nominal 48in - and that makes some sense in the context of the range:

Small = 24in (nominal) = 2ft

Medium = 36in (nominal)  = 3ft

Large  = 48in (nominal) = 4ft

Curved and Y = 60in (nominal) = 5ft

 

Why Peco keep on claiming Large radius turnouts are 60in radius is unfathomable!

 

[Chimer]

5 minutes ago, Harlequin said:

Why Peco keep on claiming Large radius turnouts are 60in radius is unfathomable!

 

 

Especially as seven and a half of them (at least as plotted in XTrackCad) complete a 90 degree turn in 50 inches ..... until today, I'd always just believed the catalogue .....

[Pacific231G]

18 hours ago, KeithMacdonald said:

Does anyone know what the radius is for the SL-89 long straight points and the SL-98 Y points?

 

Nominally five foot for the long straight and six foot for the Y . Those are the radii (in mm) given by Peco but the actual radius varies along the legth of the points so that may be the average not the minimum. But see later posts for more realistic figures.

Beware though, because of its curved divergence track (to get to the same 12 deg final divergence) a  straight crossover made with a pair of long straight points, though a great improvement in terms of throwover between vehicle ends,  won't give as much advantage over one made with medium points as you should expect. Real crossovers are normally straight between the two frogs (common crossings)  The same is true of Peco's medium Y but, because the angle on each leg is only 6 deg and the radius is greater, it seems to be less of a problem.

How well a particular throat works is somewhat dependent on the geometry of the stock you're using (mainly the relationship between overal length and the pivot points of the bogies) so it's worth testing the arrangement with your actual coaches (you can probably do this by printing out Peco's helpfully provided templates though I use some gash code 100 points not fit for actual service but good for this)  

I'm currently testing a 'Minories' derived throat with the  outer crossover formed  from a a single long point forming a crossover with a Y replacing the first of the two back to back points, the other points in the actual throat are the same though the entrance to the loco spur (a fourth parcels and railcar departure bay platform in my plan) also seems to work best with a Y. That seems a good compromise with the actual H0 stock I'm using but it might not work so well with say BR Mk 1s in OO. The 'problem' route between the inbound and the upper platform which in the classic Minories is the one unseparated reverse curve still has the greatest throwover but nothing like as much.

Edited November 30, 2020 by Pacific231G
reflectino of new information

[KeithMacdonald]

18 minutes ago, Chimer said:

 

Especially as seven and a half of them (at least as plotted in XTrackCad) complete a 90 degree turn in 50 inches ..... until today, I'd always just believed the catalogue .....

 

I get almost exactly the same in AnyRail (7 and a half on 90 degrees, 30 in a complete 360 circle). With what looks like a 50 inch outside radius, 48 inch inside radius, 12 degree turn per point.

[Pacific231G]

21 minutes ago, KeithMacdonald said:

 

I get almost exactly the same in AnyRail (7 and a half on 90 degrees, 30 in a complete 360 circle). With what looks like a 50 inch outside radius, 48 inch inside radius, 12 degree turn per point.

I've just tried this in Anyrail for all four radii , short, medium, long and Y . The results are interesting

and I get (to about the nearest inch

2'9" short, 3'4" medium, 4'1" long, 6'10" medium Y 

I'd expect the overall replacement radius to be greater than the nominal because there are  straight rails beyond the frog in small and medium and between the toe end and the start of the actual switch rails on all of them but the nominal five foot radius long points are definitely anomalous. The average radius can't be less than the minimum or the nomjnal radius so I agree that the long point ought to be quoted as a four foot radius.

 

This is all four on a twelve inch grid

The other tihng you can do with Anyrail is to cut a point from a plan having glued the tracks it connects with and replace it with a smoothed piece of flex track. It then analyses the minimum radius and that's always going to be a bit larger than the actual minimum radius in the diverging road of the turnout simply because there's always at least a smidgeon of straight track in a straight turnout. 

Turnout radii as used by us have always been a bit nomnal though so it's how a vehicle behaves going through it tha tmatters.

[Harlequin]

There are a number of other mistakes in that PDF regarding the OO turnout radii. I suggest chucking it in the bin!

 

[Pacific231G]

On 30/11/2020 at 07:20, Harlequin said:

There are a number of other mistakes in that PDF regarding the OO turnout radii. I suggest chucking it in the bin!

 

Mistakes in transposing the data suppled by Peco or mistakes in that data?

If the former then the Queensland  pdf is still a useful summary.

 

Given that some of Peco's data for nominal radii is, shall we say, questionable (though the geometry of a Peco nominal three foot radius medium turnout seems to be fairly consistent with other nominal three foot radius turnouts including SMP) 

If Peco's divergence angle is 12.0  degrees (rather than to the nearest degree) then the crossing is more like 1:4.7 than 1:5. If I've got my geometry right the divergence angle of a no 5 crossing is about 11.5 degrees.

Given that the curve through a turnout is effectively a transition rather than a single radus curve I'm not sure how useful the nominal radius is anyway. It's how vehicles behave and the relationship between them when passing over pointwork that we're really interested in.

 

Edited December 1, 2020 by Pacific231G

[Joseph_Pestell]

5 minutes ago, Pacific231G said:

Mistakes in transposing the data suppled by Peco or mistakes in that data?

If the former then te Queensland  pdf is still a useful summary.

 

Given that some of Peco's data for nominal radii is, shall we say, questionable (though the geometry of a Peco nominal three foot radius medium turnout seems to be fairly consistent with other nominal three foot radius turnouts including SMP) 

If Peco's divergence angle is 12.0  degrees (rather than to the nearest degree) then the crossing is more like 1:4.7 than 1:5. If I've got my geometry right the divergence angle of a no 5 crossing is about 11.5 degrees.

Given that the curve through a turnout is effectively a transition rather than a single radus curve I'm not sure how useful the nominal radius is anyway. It's how vehicles behave and the relationship between them when passing over pointwork that we're really interested in.

 

 

We can't really blame anyone for believing that data provided by Peco about their own products would be accurate.

 

Quite why Peco feel that it is a good thing to advertise a turnout as 5' radius when it isn't is a bit of a mystery.

[Pacific231G]

14 hours ago, KeithMacdonald said:

Does anyone know what the radius is for the SL-89 long straight points and the SL-98 Y points?

 

Hi Keith

Looking at this I'm seeing both crossovers with reverse curves. Though the inner one with a Y will be  smoother, a train travelling from the inbound line to the centre platform will encounter a double reverse curve.

I think you should be able to take advantage of the overall curved throat to eliminate reverse curves entirely. It depends on whether, in terms of appearance, you want a gentler overall curve through the throat or to reduce the throwover between coaches as trains negotiate the crossovers.  

[Chimer]

Using Streamline curved points, you can indeed eliminate reverse curves entirely.  The effective radius of the inner track here is 38" - but one section of it is 32", and of course the inner track of the points is around 26".

 

 

The layout itself has already been discussed in another thread in this section (Variations on a couple of themes by lots of you).  Won't let me add the URL today, for some reason ....

 

 

Edited November 30, 2020 by Chimer

[KingEdwardII]

Yes - neatly done to eliminate the reverse curves. The irony is that with the available Streamline points, the overall effect is to tighten the overall curvature of the tracks - but perhaps that is appropriate for a layout designed for such a small space.

 

Mike.

[Pacific231G]

19 hours ago, KeithMacdonald said:

Does anyone know what the radius is for the SL-89 long straight points and the SL-98 Y points?

 

Hi Keith

I had a play around with this in Anyrail and came up with this . It's on  25cm grid.

 

I've not tried to match your curves beyond the throat each side exactly but that could be done quite easily and the throat is the same length . This  is made up entirely of left and right medium length turnouts. Though you have to look quite carefully to convince your eyes, there are no unseparated reverse curves and the only reverse curve of any type is the one into the lower platform. That has a radius of 447 cm (14'8")  so won't cause any problems. 

The equivalent of your piece of track marked in red now has a minimum radius of 103 cm (40") so still a significantly shallower curve than the turnouts albeit tighter than your original. With no reverse curves there'll be no buffer locking and the passengers might even be able to use the corridor connections without falling to their doom. .

 

[Zomboid]

19 hours ago, Pacific231G said:

I've just tried this in Anyrail for all four radii , short, medium, long and Y . The results are interesting

and I get (to about the nearest inch

2'9" short, 3'4" medium, 4'1" long, 6'10" medium Y 

I'd expect the overall replacement radius to be greater than the nominal because there are  straight rails beyond the frog in small and medium and between the toe end and the start of the actual switch rails on all of them but the nominal five foot radius long points are definitely anomalous. The average radius can't be less than the minimum or the nomjnal radius so I agree that the long point ought to be quoted as a four foot radius.

 

This is all four on a twelve inch grid

The other tihng you can do with Anyrail is to cut a point from a plan having glued the tracks it connects with and replace it with a smoothed piece of flex track. It then analyses the minimum radius and that's always going to be a bit larger than the actual minimum radius in the diverging road of the turnout simply because there's always at least a smidgeon of straight track in a straight turnout. 

Turnout radii as used by us have always been a bit nomnal though so it's how a vehicle behaves going through it tha tmatters.

For the academic interest, here's peco's code 83 range.

the #5s look to be about 3'6", the #6s about 4'6", and the #8s form a curve of noticably over 8' radius.

[Gilbert]

7 hours ago, Chimer said:

Using Streamline curved points, you can indeed eliminate reverse curves entirely.  The effective radius of the inner track here is 38" - but one section of it is 32", and of course the inner track of the points is around 26".

 

 

The layout itself has already been discussed in another thread in this section (Variations on a couple of themes by lots of you).  Won't let me add the URL today, for some reason ....

 

 

Looks like a homage to Borchester Market....

[Keith Addenbrooke]

On 29/11/2020 at 20:57, KeithMacdonald said:

Does anyone know what the radius is for the SL-89 long straight points and the SL-98 Y points?

 

Further to the earlier responses (which I agree with), I use one of two approaches when creating parallel sidings at standard 2” Streamline spacing.  One is to put in a point (as if creating a crossover) and then swap it for a Streamline curve - this gives the slightly different radius, as @Pacific231G explained above.

 

The other method I use is to insert a 12 degree section of Flextrack curved at 48” radius - this appears to give the second 1” offset I need.  I can’t remember enough geometry to check the maths, but it seems to work:

 

 

It doesn’t matter whether the point is short, medium, long or curved, as they are all set to 12 degrees.

[Flying Pig]

39 minutes ago, Keith Addenbrooke said:

Further to the earlier responses (which I agree with), I use one of two approaches when creating parallel sidings at standard 2” Streamline spacing.

 

I'm a bit puzzled here.  If you are describing how you add parallel track in Anyrail, there is an 'add parallel track' function on the right click menu when you have a length of plain track selected.  You need to enter the track spacing and choose the relative position of the new track.

[Keith Addenbrooke]

17 minutes ago, Flying Pig said:

 

I'm a bit puzzled here.  If you are describing how you add parallel track in Anyrail, there is an 'add parallel track' function on the right click menu when you have a length of plain track selected.  You need to enter the track spacing and choose the relative position of the new track.

Good point.  My explanation was a bit brief, but that is how I hit upon this (I was using Anyrail, as you indicate):

 

I’d drawn a straight main line with a point for a siding, which I wanted to be parallel to the main line.  I then added the parallel track at 2” spacing as you describe.  Using the Curve Flex function in Anyrail I knew I then needed a 12 degree curve to connect the point to the parallel track, and found that setting it at 48” radius just seemed to work.

 

It is an approximate swap for a long point - that’s not a surprise: @Harlequin explained the anomaly in the Peco descriptions for me a while back, which was very helpful.

 

Where I now find this quick and useful is if I am using up spare straight Setrack in my sketch for the mainline or siding, as I don’t think you can add a parallel line in the same way (I could be wrong on that: I’m not an expert).

 

Incidentally, if I use the alternative approach of putting in a second medium point then swapping it for a piece of smooth curved Flextrack, I think the radius comes out as something like 31.35” from memory - quite a bit tighter.

Edited December 1, 2020 by Keith Addenbrooke

[Pacific231G]

1 hour ago, Keith Addenbrooke said:

Good point.  My explanation was a bit brief, but that is how I hit upon this (I was using Anyrail, as you indicate):

 

I’d drawn a straight main line with a point for a siding, which I wanted to be parallel to the main line.  I then added the parallel track at 2” spacing as you describe.  Using the Curve Flex function in Anyrail I knew I then needed a 12 degree curve to connect the point to the parallel track, and found that setting it at 48” radius just seemed to work.

 

It is an approximate swap for a long point - that’s not a surprise: @Harlequin explained the anomaly in the Peco descriptions for me a while back, which was very helpful.

 

Where I now find this quick and useful is if I am using up spare straight Setrack in my sketch for the mainline or siding, as I don’t think you can add a parallel line in the same way (I could be wrong on that: I’m not an expert).

 

Incidentally, if I use the alternative approach of putting in a second medium point then swapping it for a piece of smooth curved Flextrack, I think the radius comes out as something like 31.35” from memory - quite a bit tighter.

I don't think the trial version of AnyRail includes the curved flextrack function so  I used to use the method of adding a point then replacing it with a piece of flexible track. The method I now more often use is to add a parallel track (for blind sidings or platform roads without releasing crossovers you don't have to use the standard BRMSB OO/Peco  50mm spacing ) then glue it, cut it at a suitable  distance from the end of the point, insert a short piece of flex and then use the smooth flex function. That way you can, within limits, make the radius whatever you like. The catch with using a point and replacing it is that, if you're coming off the diverging route of another set of points, you may get an excessively sharp reverse curve especially if using medium radius- the very problem that Minories was designed to avoid.  This method is also very useful for creating island platforms with a constant width and parallel bays as well as goods yard roads separated for loading/unloading.