[Sydney, Australia] - WestConnex and NorthConnex

Going on holiday? Just returned with pictures or news? Found an interesting website? Post everything international in here.

Moderator: Site Management Team

User avatar
jackal
Member
Posts: 8026
Joined: Tue Jan 08, 2008 23:33
Location: M6

Re: [Sydney, Australia] - WestConnex and NorthConnex

Post by jackal »

^ I can feel your enthusiasm for the interchange, but I'm afraid I don't find these arguments very convincing!

First off, take a look at how massive this interchange is compared to any at-grade junction. There is really not any kind of space-based argument for freeflow rather than at-grade.

A more cost and space efficient design, still with ample capacity, would keep the two carriageways and M8eb to airport connector as they are, but deletes all the other bridges. You then put in two north-facing slips with at-grade junctions on the east-west carriageway: the western one is just to access an M8eb onslip (on a similar horizontal but not vertical alignment to the deleted top level ramp), the eastern one is a crossroads connecting an offslip from M8wb as well as the two local road spurs.

This maintains all the movements of the constructed interchange as well as freeflow between the airport and both directions of M8. But it has one rather than five big freeflow ramps, and a fraction of the cost and land take.

PS - I think the junction as built is only 4 level? I'd also say it's a bit stack-turbiney rather than purely stack-derived, as the westbound to northbound ramp crosses the eastbound to southbound ramp twice. In a true stack, opposite turns like these do not cross each other.
User avatar
Bryn666
Elected Committee Member
Posts: 36790
Joined: Thu Jun 20, 2002 20:54
Contact:

Re: [Sydney, Australia] - WestConnex and NorthConnex

Post by Bryn666 »

https://maps.app.goo.gl/wqohUviWZs7ymCYD7

Australia has tackled the "blank signal" problem. Another big failure red cross against UK dynamic motorways.
Bryn
Terminally cynical, unimpressed, and nearly Middle Age already.
She said life was like a motorway; dull, grey, and long.

Blog - https://showmeasign.online/
BlueSky - https://bsky.app/profile/showmeasignbryn.bsky.social
YouTube - https://www.youtube.com/@BrynBuck
User avatar
Chris5156
Deputy Treasurer
Posts: 17350
Joined: Thu Sep 06, 2001 21:50
Location: Hampshire
Contact:

Re: [Sydney, Australia] - WestConnex and NorthConnex

Post by Chris5156 »

Bryn666 wrote: Thu Sep 12, 2024 12:03https://maps.app.goo.gl/wqohUviWZs7ymCYD7

Australia has tackled the "blank signal" problem. Another big failure red cross against UK dynamic motorways.
They make it look so easy! :laugh:
Peter Freeman
Member
Posts: 1618
Joined: Thu Aug 30, 2012 07:52
Location: Exits 9 & 10, M1 East, Melbourne, Australia

Re: [Sydney, Australia] - WestConnex and NorthConnex

Post by Peter Freeman »

Chris5156 wrote: Thu Sep 12, 2024 21:42
Bryn666 wrote: Thu Sep 12, 2024 12:03https://maps.app.goo.gl/wqohUviWZs7ymCYD7

Australia has tackled the "blank signal" problem. Another big failure red cross against UK dynamic motorways.
They make it look so easy! :laugh:
Most of these signs are at or near to tunnels. All our tunnels, and some inner-city motorways, are 80-limited.

We also fulfill another one of your wishes Bryn: lane-use management signals (LUMS) are only ever blank under fault conditions. They normally show the speed limit, this also implying that the lane is open. The surrounding red circle flashes if the limit is below the normal value. If not showing a speed limit, they show either a red X, or a white diagonal-down arrow (either way) to instruct 'change lane' (similar to UK). There is also, not often seen, a slowly flashing orange circle, meaning ' for goodness sake change lanes as instructed, as I'm about to go red X ! '.

A blank (ie. failed) LUMS means 'continue with the last-indicated limit' (usually 100 or 80). I don't have proof of this (it's presumably documented somewhere), but it's certainly de-facto practise.
Peter Freeman
Member
Posts: 1618
Joined: Thu Aug 30, 2012 07:52
Location: Exits 9 & 10, M1 East, Melbourne, Australia

Re: [Sydney, Australia] - WestConnex and NorthConnex

Post by Peter Freeman »

jackal wrote: Wed Sep 11, 2024 16:04 ^ I can feel your enthusiasm for the interchange,
I don't let enthusiasm displace objectivity, and I'm sure you know that I'm not obsessed with grade separation if at-grade is adequate and more cost-appropriate.
... but I'm afraid I don't find these arguments very convincing!
First off, take a look at how massive this interchange is compared to any at-grade junction.
Any average freeflow interchange is much larger than an average at-grade. This stack, by itself, is compact: 270m square*. Compare -
600m M25 J7
590m M25 J15
410m M5 J15
400m AU M4/M7
270m St Peters

The space east of the stack's N-S carriageways, before you hit existing roads and buildings, is a rectangle 350m E-W x 100m N-S.
There is really not any kind of space-based argument for freeflow rather than at-grade.

A more cost and space efficient design, still with ample capacity, would keep the two carriageways and M8eb to airport connector as they are, but deletes all the other bridges. You then put in two north-facing slips with at-grade junctions on the east-west carriageway: the western one is just to access an M8eb onslip (on a similar horizontal but not vertical alignment to the deleted top level ramp), the eastern one is a crossroads connecting an offslip from M8wb as well as the two local road spurs. This maintains all the movements of the constructed interchange as well as freeflow between the airport and both directions of M8. But it has one rather than five big freeflow ramps,
I find that description hard to follow. I have a couple of visions of what I think you mean. They include 3 levels, but more ramps than you mention. A better description or a drawing is required. Worth the trouble? No.
... and a fraction of the cost and land take.
The degraded/polluted land here was probably almost free, but any encroachment outside of the garbage dump would have been extremely expensive (Sydney prices ...). If the hole hadn't been used to bury the stack, it would have had to be almost filled to make it into useful land.

The issue with using the 350x100 space is that a new at-grade intersection would be close to the existing intersection of Euston Road and Campbell Road, considering the vehicle storage required and the resultant wide carriageways. Cramming in an intersection for Gardeners Road too would exacerbate that situation. I too can imagine possible layouts that would fit, and work, but they would be poorly-performing.
PS - I think the junction as built is only 4 level?
Yes. I've always previously referred to it as 4-level, 5-arm. It seems that the number 5 got lodged in my brain for a while there.
I'd also say it's a bit stack-turbiney rather than purely stack-derived, as the westbound to northbound ramp crosses the eastbound to southbound ramp twice. In a true stack, opposite turns like these do not cross each other.
"Crosses eb-sb twice" - that depends on which eb-sb you mean: eb-seb, or eb-swb. In any case, yes it is indeed impure, as it must be: (a) because it has 5 arms, and (b) because two of those arms are not actual M8, but are two pairs of ramps servicing the underground M8 that runs right past here.

Interestingly, a normal 4-level stack contains 12 pathways through it: L, R and straight at each arm: 4x3=12. This one, because certain turns are not required or not provided, also has 12 pathways: 3+3+2+2+2.

My continuing conclusion: not overbuilt, performs well, and still will even in many years' time when the underground M8, projected for D5 at that point and currently being extended in both directions, approaches 200K AADT. My concern is that it is possibly under-built at its eb connection to Euston Road signals. In future that piece may need widening and/or ramp-metering and/or a signalised merge.

* stack measurements are between opposing left turns, 2 perpendicular, averaged.
User avatar
jackal
Member
Posts: 8026
Joined: Tue Jan 08, 2008 23:33
Location: M6

Re: [Sydney, Australia] - WestConnex and NorthConnex

Post by jackal »

Peter Freeman wrote: Mon Sep 16, 2024 07:16
jackal wrote: Wed Sep 11, 2024 16:04 I'd also say it's a bit stack-turbiney rather than purely stack-derived, as the westbound to northbound ramp crosses the eastbound to southbound ramp twice. In a true stack, opposite turns like these do not cross each other.
"Crosses eb-sb twice" - that depends on which eb-sb you mean: eb-seb, or eb-swb. In any case, yes it is indeed impure, as it must be: (a) because it has 5 arms, and (b) because two of those arms are not actual M8, but are two pairs of ramps servicing the underground M8 that runs right past here.
The fact that one of the eastbound to southbound ramps crosses the westbound to northbound ramp is precisely what makes it stackturbine-like.

If none crossed - e.g., because westbound to northbound was moved northeast, avoiding both eastbound to southbound ramps in plan view - it would be stack-like.

If both crossed - e.g., westbound to northbound was moved southwest - it would be turbine-like, and have nothing in common with a stack.

This is premised on right turn ramps being classifiable based on their alignment and interaction with other right turn ramps even outside of the classic four-arm, full access context. So for instance, regarding four-arm limited access, Catthorpe has stack-like ramps (because opposite turns don't cross), whereas M60 J5 has turbine-like ramps (because opposite turns cross).

The basic principles are, to me, just as applicable at a 5-arm interchange like St Peters.
Last edited by jackal on Mon Sep 16, 2024 12:17, edited 1 time in total.
Peter Freeman
Member
Posts: 1618
Joined: Thu Aug 30, 2012 07:52
Location: Exits 9 & 10, M1 East, Melbourne, Australia

Re: [Sydney, Australia] - WestConnex and NorthConnex

Post by Peter Freeman »

^ Oh, yes, I see exactly what you mean now :oops:.

So, about 55% stack, 45% turbine? I'm still 'enthusisastic' !
User avatar
jackal
Member
Posts: 8026
Joined: Tue Jan 08, 2008 23:33
Location: M6

Re: [Sydney, Australia] - WestConnex and NorthConnex

Post by jackal »

We're not going to agree about the practicality at this site, but I do agree that it's nicely compact for what it does while maintaining decent radii.

Out of curiosity I counted the number of pathways through the LTC/A13 junction, which is also 5 arm (if you exclude the links to Orsett Cock). I think there are 13 freeflow pathways: four along A13 and LTC mainlines, four turns between A13E and LTC, and five involving the A1089. There are, additionally, numerous pathways via at-grade turns at Orsett Cock.

I was basically wondering if a more elegant LTC/A13 design could be based on the St Peters Interchange. The arms line up quite nicely if you rotate St Peters 45 degrees clockwise (so the M4-M8 arm serves as A13E). It doesn't work as the movements aren't right and there's too much going on at A13/LTC. But here they are anyway:

Image

Image

Also, for what it's worth, the blue stuff is what I was saying I would keep as freeflow in my cheaper redesign. The remaining movements would still be there but via at-grade junctions.
Last edited by jackal on Mon Sep 16, 2024 13:56, edited 1 time in total.
User avatar
Bryn666
Elected Committee Member
Posts: 36790
Joined: Thu Jun 20, 2002 20:54
Contact:

Re: [Sydney, Australia] - WestConnex and NorthConnex

Post by Bryn666 »

Peter Freeman wrote: Fri Sep 13, 2024 00:34
Chris5156 wrote: Thu Sep 12, 2024 21:42
Bryn666 wrote: Thu Sep 12, 2024 12:03https://maps.app.goo.gl/wqohUviWZs7ymCYD7

Australia has tackled the "blank signal" problem. Another big failure red cross against UK dynamic motorways.
They make it look so easy! :laugh:
Most of these signs are at or near to tunnels. All our tunnels, and some inner-city motorways, are 80-limited.

We also fulfill another one of your wishes Bryn: lane-use management signals (LUMS) are only ever blank under fault conditions. They normally show the speed limit, this also implying that the lane is open. The surrounding red circle flashes if the limit is below the normal value. If not showing a speed limit, they show either a red X, or a white diagonal-down arrow (either way) to instruct 'change lane' (similar to UK). There is also, not often seen, a slowly flashing orange circle, meaning ' for goodness sake change lanes as instructed, as I'm about to go red X ! '.

A blank (ie. failed) LUMS means 'continue with the last-indicated limit' (usually 100 or 80). I don't have proof of this (it's presumably documented somewhere), but it's certainly de-facto practise.
See how amazingly simple it could be? Nope, not for National Highways. Equally they'd never be able to design the interchange we're discussing because you can't do 70 mph around the slip roads so it would clearly have to be a roundabout. Those are the two options: warp 9 or crawling. There's just going to be no design or operational competence left here whatsoever.
Bryn
Terminally cynical, unimpressed, and nearly Middle Age already.
She said life was like a motorway; dull, grey, and long.

Blog - https://showmeasign.online/
BlueSky - https://bsky.app/profile/showmeasignbryn.bsky.social
YouTube - https://www.youtube.com/@BrynBuck
Post Reply