Australian Motorways

[jackal]

The duck-n-dive element seems to be lifted from the A690 here https://goo.gl/maps/rDbvpKLd9K2FqyQp9

Yes, similar in principle. Exactly the same distance between the roundabouts, and use of C/D roads. But those roundabouts are more UK standard, and are bigger, with more lanes. And despite similar proximity to housing, etc, the A690 vibe is minor rural HQDC rather than the built-up urban style of Armadale Road.

The UK road that this one most reminds me of is the A50 entering Stoke, westbound, between those vertical brick retaing walls for a mile or so. Not the same, but still that urban, bricky, comfortably-confined feeling.

I dislike this design, which basically functions as one GSJ, with capacity to match, but which involves a lot of infrastructure and expense so looks like it achieves more. Much better is the version where the link roads join directly to the mainline, effectively turning it into one and a half GSJs. Here's an example on the A24: https://www.google.com.au/maps/@51.0654 ... a=!3m1!1e3 Unfortunately this is deprecated as part of the current DMRBs war on multiple entries and exits.

The Perth and A50 examples do at least have the excuse of not having space to do much more.

[Bryn666]

The duck-n-dive element seems to be lifted from the A690 here https://goo.gl/maps/rDbvpKLd9K2FqyQp9

Yes, similar in principle. Exactly the same distance between the roundabouts, and use of C/D roads. But those roundabouts are more UK standard, and are bigger, with more lanes. And despite similar proximity to housing, etc, the A690 vibe is minor rural HQDC rather than the built-up urban style of Armadale Road.

The UK road that this one most reminds me of is the A50 entering Stoke, westbound, between those vertical brick retaing walls for a mile or so. Not the same, but still that urban, bricky, comfortably-confined feeling.

I dislike this design, which basically functions as one GSJ, with capacity to match, but which involves a lot of infrastructure and expense so looks like it achieves more. Much better is the version where the link roads join directly to the mainline, effectively turning it into one and a half GSJs. Here's an example on the A24: https://www.google.com.au/maps/@51.0654 ... a=!3m1!1e3 Unfortunately this is deprecated as part of the current DMRBs war on multiple entries and exits.

The Perth and A50 examples do at least have the excuse of not having space to do much more.

Apparently multiple exits are confusing to drivers, who presumably are now dumbed down to the point they can't read signs. How we've coped with the likes of M8 J15 or M60 J15/14 for five decades remains a mystery.

[Peter Freeman]

I dislike this design, which basically functions as one GSJ, with capacity to match, but which involves a lot of infrastructure and expense so looks like it achieves more. Much better is the version where the link roads join directly to the mainline, effectively turning it into one and a half GSJs. Here's an example on the A24: https://www.google.com.au/maps/@51.0654 ... a=!3m1!1e3 Unfortunately this is deprecated as part of the current DMRBs war on multiple entries and exits.

The Perth and A50 examples do at least have the excuse of not having space to do much more.

Actually, I also dislike the design, though for a slightly different reasons -
1. It's not really required and therefore is not good value, for this particular road, which is not fully grade separated and is definitely not a motorway or expressway.
2. My usual reason: the grade-separated roundabout is a dumb design for most well-trafficked routes, in my opinion. If it gets too busy, it will need signalisation, and then you're better off with a diamond. Making things worse, most GSR's have the roundabout on top, with two curved bridges, making subsequent conversion to anything else quite difficult.

I do, illogically, like this Perth instance though, simply for its unusualness and willingness to try something different. I acknowledge that it's not new in UK terms.

Regarding your A24 roundabout pair: yes, the intermediate re-join ramps, facing one way but not both, do increase capacity, as not all leaving/joining traffic has to negotiate both RA's. However, while there is some similarity to Perth's 'duck-and-dive', there are also significant differences. First, the separation between those roundabouts is much greater: about 900m. Second, like Bryn's A690 comparison, it's on a larger scale, one of the roundabouts having a very large diameter. And they're not on local suburban roads: A690 and A24 are both HQDC's.

Regarding my A50 comment, I was only remarking the fabulous brickwork-walled trenches about 4km east of Sideway. I don't know whether they're to everyone's aesthetic taste, but they're just so apt for the potteries towns. Love 'em! I do see, now I look, that the road thereabouts also has some roundabout pairs with similarities to the 'duck-and-dive'.

[jackal]

Come to think of it, this "roundabout pair" design is basically a dumbbell GSJ flipped 90 degrees and elongated - which does rather bring out the low capacity compared to the cost.

[Peter Freeman]

Come to think of it, this "roundabout pair" design is basically a dumbbell GSJ flipped 90 degrees and elongated - which does rather bring out the low capacity compared to the cost.

I can't quite see this analogy. Please explain.

[crazyknightsfan]

Come to think of it, this "roundabout pair" design is basically a dumbbell GSJ flipped 90 degrees and elongated - which does rather bring out the low capacity compared to the cost.

I can't quite see this analogy. Please explain.

Picture a dumbbell, but the freeway goes left-right across the full length of the dumbbell instead of up-down between the two ends.

[Peter Freeman]

Come to think of it, this "roundabout pair" design is basically a dumbbell GSJ flipped 90 degrees and elongated - which does rather bring out the low capacity compared to the cost.

I can't quite see this analogy. Please explain.

Picture a dumbbell, but the freeway goes left-right across the full length of the dumbbell instead of up-down between the two ends.

OK, I've pictured that. I've even drawn it out, with some variants. None of them makes any sense as an interchange. A major difference between a dumbbell and the 'duck-n-dive' is that one has four arms, the other has six. One can hardly be just a warp of the other.

I could accept that warp theory if each end of the duck-n-dive only had one minor road (in addition to the C/Ds and ramps) leading off them. That's four arms overall. But then you'd have four structures built across the freeway, instead of the dumbbell's two (or even one). And you couldn't simply tear-drop them away, as in a dumbbell, because the enforced detour (to the other end of the pair) is too far away.

Jackal, this is a bit like the warping of a freeway stack interchange to achieve a stagger - as we discussed some time ago. But this one doesn't work ... !?

[Pgd]

Hi, lurker here, bored at the end of a long working week and with PowerPoint sitting open in front of me

Yes there is the complication of an extra two minor road connections, and given that dumbbell roundabouts tend to be small, the "warped" junction in my attached image probably wouldn't function nearly as well as the "duck and dive". But still, you can see the relationship.

(I too love thinking about morphology of junction layouts!)

[Peter Freeman]

Pgd, that is an interesting warp, and a nice little diagram showing the intermediate stage. However, it breaks certain rules that I'd self-impose for genuine warping: links are moved and re-connected, and bridges eliminated.

So yes, there is a relationship, just, but it's not much more than the fact that both extremes of the warp (dumbbell and duck-n-dive) have two roundabouts .

Now, staggered crossroads, with two T's, morphing into a 4-level stack, is more fascinating - but not for now ... !

[Peter Freeman]

An influx of British engineers within Main Roads WA (HE equivalent) over the past decade means that there's starting to be a lot more of a British influence, the most obvious example being Main Roads' current obsession with roundabouts over traffic signals.

A discovery in Perth using Google Maps has just ruined my day. There's a hamburger at the intersection of Morley and Alexander.

https://www.google.co.uk/maps/place/Tre ... 15.8243698

[crazyknightsfan]

An influx of British engineers within Main Roads WA (HE equivalent) over the past decade means that there's starting to be a lot more of a British influence, the most obvious example being Main Roads' current obsession with roundabouts over traffic signals.

A discovery in Perth using Google Maps has just ruined my day. There's a hamburger at the intersection of Morley and Alexander.

https://www.google.co.uk/maps/place/Tre ... 15.8243698

This intersection actually works quite well, keeping the local road connections that would have otherwise been cut off with normal four-way signals. The phasing also operates as two-phase only so more efficient than the normal four-way intersection with 4 phases.

The odd thing is that the original intersection here was a very elongated oval-about orientated NW to SE (i.e. The Strand was the centreline of the roundabout) but when rebuilt to a hamburger in 1985 they didn't re-use any of the old intersection at all, instead orientating it north-south.

Not an intersection type I would select in a greenfields environment as it's a massive waste of land but here it seems to work.

[Peter Freeman]

Melbourne spent over 200 days in lockdown during 2020, with traffic variation ranging from -60% (hardly anyone going anywhere, work from home, online school) to +5% (semi-normal, avoidance of public transport). Here are the 2020 figures anyway. I'll aim to revise them after a normal-ish 2022.

Well, no new figures, as 2022 hasn't been normal after all. In most AU cities, traffic is still down about 10% from pre-covid levels. This is surmised to be due, mainly, to a persistence of work-from-home. And it is despite a minor lingering avoidance of public transport. Is the UK similarly affected?

Edit 17-02-2024: Official figures for 2023 show, for all locations in my table, AADT either same as 2020 or slightly reduced! I am surprised - the roads seem and look busier to me. The reason must be change of habits and/or lack of economic growth, since covid lockdowns. I'll check again in a couple of years. I've revised my table anyway, in my original 13-10-2021 post.

[traffic-light-man]

Melbourne spent over 200 days in lockdown during 2020, with traffic variation ranging from -60% (hardly anyone going anywhere, work from home, online school) to +5% (semi-normal, avoidance of public transport). Here are the 2020 figures anyway. I'll aim to revise them after a normal-ish 2022.

Well, no new figures, as 2022 hasn't been normal after all. In most AU cities, traffic is still down about 10% from pre-covid levels. This is surmised to be due, mainly, to a persistence of work-from-home. And it is despite a minor lingering avoidance of public transport. Is the UK similarly affected?

I believe that there has been a marked increase in traffic levels this side of the July-September school summer holidays at the 'traditional' AM and PM peak times, or at least in the areas I frequent. This is certainly the busiest I've witnessed it, and I think it's the first time I've witnessed definitive peaks this side of lockdown. It strikes me that perhaps more people have decided to go back to regular office working since the children went back to school for this academic year.

The PM school run also always tends to stand out as well, presumably because the AM is absorbed quite well with the general AM peak whereas the PM sits before the general peak, but this too seems to have got more severe of late.

[Peter Freeman]

The NorthConnex tunnel opened over the weekend, which fills in the missing link in Sydney's regional motorway network. It's now possible to drive from Melbourne to Newcastle on motorways, and in a couple of years, when the Coffs Harbour bypass is built, all the way up to north of Brisbane apart from the one junction at Newcastle/Hexham.

The complete fix for Hexham should be underway by early next year -
https://caportal.com.au/tfnsw/m1rt

(It's interesting to see that the small village Heatherbrae will be bypassed by the new motorway. The road through there currently has a large roundabout with provision for a flyover to be added, creating a UK-style GSR. Clearly the intention was that the Pacific Highway would remain through town. Fortunately, that future-proofing will not be used now. Also very British-looking are the transverse white lines across the southbound approach to that roundabout, to promote speed-reduction.
https://www.google.com.au/maps/@-32.781 ... a=!3m1!1e3)

[Peter Freeman]

Bump!

Come to think of it, this "roundabout pair" design is basically a dumbbell GSJ flipped 90 degrees and elongated - which does rather bring out the low capacity compared to the cost.

As an aside to rejecting the relationship here with a dumbbell, I also question the general contention that a dumbbell has "low capacity compared to the cost"? I don't think so. Capacity low? Yes, low-ish. Cost? definitely low. I think it's one of the most cost-effective designs for a low-medium-volume GSJ.

In topology terms, with 2 arms plus motorway ramps, it's almost identical to a classic UK 2-bridge roundabout. If it's of teardrop/dog bone shape, then it's exactly identical. It's as if a circular or oval roundabout has been squeezed across its diameter, to bring opposite traffic directions into contact where they pass over or under the motorway. Assuming equal carriageway widths to its circular analogue, there's no reason why its capacity will be lower. It will operate at lower vehicle speeds, but lower speed doesn't mean that vehicles pass a point at a lower rate, as any roadway can carry a higher load at reduced speeds.

The dumbbell is seen as a poor relation to the classic 2-bridge circular-ish roundabout. The belief is obvious in many Sabre posts, and probably is the public view, but it is the ubiquity of 2-bridge rbt's in the UK that drives this entrenched feeling.

Dumbbells have these advantages -
1. Safer (owing to lower speed, enforced by smaller roundabouts).
2. Lower cost (thanks to straight bridges instead of curved).
3. Better suited to supporting additional arms.

Advantage number 2 (straight bridges) also facilitates future upgrade to a different, high-volume GSJ type. Such upgrades are often performed in Australia.

Number 3 can be important for capacity. It means that local, minor-road traffic whose destination is on the same side of the motorway as its source, need not cross the bridge(s). That unloads the opposite-side roundabout, leaving more space for other users, and thereby increases the total interchange capacity. It's analogous to (and topologically identical to) roundabout short-cuts such as the short-lived one at M1J24 for A50 sb, but it doesn't require signalisation and it doesn't freak out unfamiliar drivers.

It's time to recognise the dumbbell. If roundabouts must be inflicted on GSJ's, then this type should be considered first.

[Chris5156]

Dumbbells have these advantages -
1. Safer (owing to lower speed, enforced by smaller roundabouts).
2. Lower cost (thanks to straight bridges instead of curved).
3. Better suited to supporting additional arms.

I broadly agree with your post, but I wanted to reply to this bit because your item 2 has come up a few times in other posts you've made too. In the UK, it is decidedly rare for a two-bridge roundabout interchange to have curved bridges. The roundabout is almost always oval shaped so that the bridges are straight. IME curved bridges on junctions of this type are more common in European versions of the junction.

[jackal]

Bump!

Come to think of it, this "roundabout pair" design is basically a dumbbell GSJ flipped 90 degrees and elongated - which does rather bring out the low capacity compared to the cost.

As an aside to rejecting the relationship here with a dumbbell, I also question the general contention that a dumbbell has "low capacity compared to the cost"? I don't think so. Capacity low? Yes, low-ish. Cost? definitely low. I think it's one of the most cost-effective designs for a low-medium-volume GSJ.

In topology terms, with 2 arms plus motorway ramps, it's almost identical to a classic UK 2-bridge roundabout. If it's of teardrop/dog bone shape, then it's exactly identical. It's as if a circular or oval roundabout has been squeezed across its diameter, to bring opposite traffic directions into contact where they pass over or under the motorway. Assuming equal carriageway widths to its circular analogue, there's no reason why its capacity will be lower. It will operate at lower vehicle speeds, but lower speed doesn't mean that vehicles pass a point at a lower rate, as any roadway can carry a higher load at reduced speeds.

The dumbbell is seen as a poor relation to the classic 2-bridge circular-ish roundabout. The belief is obvious in many Sabre posts, and probably is the public view, but it is the ubiquity of 2-bridge rbt's in the UK that drives this entrenched feeling.

Dumbbells have these advantages -
1. Safer (owing to lower speed, enforced by smaller roundabouts).
2. Lower cost (thanks to straight bridges instead of curved).
3. Better suited to supporting additional arms.

Advantage number 2 (straight bridges) also facilitates future upgrade to a different, high-volume GSJ type. Such upgrades are often performed in Australia.

Number 3 can be important for capacity. It means that local, minor-road traffic whose destination is on the same side of the motorway as its source, need not cross the bridge(s). That unloads the opposite-side roundabout, leaving more space for other users, and thereby increases the total interchange capacity. It's analogous to (and topologically identical to) roundabout short-cuts such as the short-lived one at M1J24 for A50 sb, but it doesn't require signalisation and it doesn't freak out unfamiliar drivers.

It's time to recognise the dumbbell. If roundabouts must be inflicted on GSJ's, then this type should be considered first.

I like dumbbells for some of the reasons you mention. But now imagine that you rotate the dumbbell 90 degrees, while the mainline stays where it was. Each roundabout now has to be built over the mainline, so you need two bridges per roundabout and four bridges in total.

This is what the 'roundabout pair' design achieves - i.e., the same decent capacity as a dumbbell but at vastly inflated cost as you have four times as many bridges. That is the design I was criticising, not the dumbbell.

[Peter Freeman]

Oh, I see. I read your post as critiquing the particular Armadale Road project as low value (with which I agree), but then making a separate critique of dumbbells generally. My apologies! And Viva la Dumbbell.

[Peter Freeman]

Dumbbells have these advantages -
1. Safer (owing to lower speed, enforced by smaller roundabouts).
2. Lower cost (thanks to straight bridges instead of curved).
3. Better suited to supporting additional arms.

I broadly agree with your post, but I wanted to reply to this bit because your item 2 has come up a few times in other posts you've made too. In the UK, it is decidedly rare for a two-bridge roundabout interchange to have curved bridges. The roundabout is almost always oval shaped so that the bridges are straight. IME curved bridges on junctions of this type are more common in European versions of the junction.

You are right, and I stand corrected: UK curved GSR bridges are rare indeed. I just scanned right along M1 and M4, and the only ones I found are at M1 Lofthouse, where there are four of course (and their curvature is so slight that they may as well have been straight). As you point out, I've harboured this false assumption for a long time. I suppose I should edit my praise of dumbbells!

[Bryn666]

Dumbbells have these advantages -
1. Safer (owing to lower speed, enforced by smaller roundabouts).
2. Lower cost (thanks to straight bridges instead of curved).
3. Better suited to supporting additional arms.

I broadly agree with your post, but I wanted to reply to this bit because your item 2 has come up a few times in other posts you've made too. In the UK, it is decidedly rare for a two-bridge roundabout interchange to have curved bridges. The roundabout is almost always oval shaped so that the bridges are straight. IME curved bridges on junctions of this type are more common in European versions of the junction.

You are right, and I stand corrected: UK curved GSR bridges are rare indeed. I just scanned right along M1 and M4, and the only ones I found are at M1 Lofthouse, where there are four of course (and their curvature is so slight that they may as well have been straight). As you point out, I've harboured this false assumption for a long time. I suppose I should edit my praise of dumbbells!

The use of straight alignments on two bridge roundabouts is why most of them have high speed problems and end up signalised. The curved bridges you lament would actually be safer.