Signalized Merge

[traffic-light-man]

When I talk about it being exit blocked, I'm referring to nothing being able to leave the merge locally, not there being nothing to stop them at the next site downstream. If there's nothing leaving the merge, then there's likely to be little effective green time at the merge at all.

Assuming there's coordination between the merge and the entry to the gyratory, coordinating the signals at the merge with those at the gyratory almost certainly will result in one of the two approaches running into a red downstream - that's inevitable, but useful overall. It's a tactic probably most often seen at signalised roundabouts.

The Switch Island end of the M58 is a good example of the same arrangement. At busy periods, it would appear that the M58 approach usually runs into a red at the Island, whereas the A59 generally runs into a green. It's complicated slightly by the fact there's three stages at the Island itself, but it by-and-large works and despite it looking a bit chaotic, wait times in the reservoir are quite low.

[Owain]

I think I had it in my head that large stop line-merge point distances could actually be more efficient. Surely giving traffic more space to speed up before the merge point means you can tighten up distances between alternate platoons, whereas if the line is right up at the merge point, you have to hold the other traffic sooner to allow the newly started traffic time to limp past the merge as it accelerates. I don't know if UK signal rules permit you to let traffic into the junction before conflicting traffic has fully cleared the conflict point, which is essentially what you'd have to do here to extract full advantage. But even if you can't, there's probably a sweet spot where you can minimise buffer time between the tail end of the last phase's traffic passing the conflict point and the next green.

You're also thinking of this issue as "How do I maximise flow through a signalised merge?". That's not the point here (though it might be interesting in itself, elsewhere). The objective of a signalised merge is (usually) to avoid causing a problem** in the lead-up to the NEXT traffic obstruction (which, in this instance, is the signal to enter the gyratory). If there is no next traffic obstruction, then a normal (UNsignalised) merge is preferable.

It is true that a signalised merge can perform the function of balancing the queue lengths on the two inputs. And there may be, but not usually, situations where signalising a merge might increase overall flow. But these two possibilities are secondary issues or side benefits.

** So what is the 'problem' we're trying to fix? If the queue for the gyratory reaches back to this unsignalised merge point, traffic squeezes together in a take-it-or-leave it fashion. So, vehicles merging from Gelderd Road will most likely be in the left lane, and those from the A643 will likely be in the right lane. Before reaching the roundabout, those wishing or needing to swap lanes have 200m in a solid queue in which to do it. Much of the swapping will occur at the last minute, just as the roundabout light turns green. This disorganised swapping and jostling reduces flow onto the roundabout. That's the flow we're trying to maximise.

On the other hand, if the merge is signalised (so it's no longer really a merge!), drivers can lane-select at these lights, in a leisurely and safe way. The 200m queue up to the gyratory now flows in an orderly fashion, with less lane swapping, and enters the roundabout smoothly and densely.

Indeed - the problem there is caused by the proximity of the merge to the gyratory.

For info, what tended to happen in practice before the revamp began, is that traffic on the A58 merge would split during peak periods, even though there were no road markings to provide distinct lanes. This is what it looked like before.

With local knowledge, drivers would organise themselves during heavy traffic, depending on where they wanted to go from the gyratory:

- lane 1 to go into lane 1 on the mainline, which splits at the gyratory to provide lanes for the B6154 or A647 gyratory exits, or A58(M) to/through the city

- lane 2 to go straight to lane 2 (which becomes lane 3 at the gyratory) on the mainline, for A58(M) to/through the city

So there are conflicting movements, as you point out, but the traffic exercised a high level of discipline in organising itself to let everybody get into the right place. It was still gridlocked though, due to traffic on the mainline not moving, because it was held by red lights up the road at the gyratory.

Assuming there's coordination between the merge and the entry to the gyratory, coordinating the signals at the merge with those at the gyratory almost certainly will result in one of the two approaches running into a red downstream - that's inevitable, but useful overall. It's a tactic probably most often seen at signalised roundabouts.

I'm assuming that the signals will indeed operate in coordination with those on the gyratory. And in combination with the increased capacity currently being added to the gyratory, I expect the addition of signals - as well as lane markings on the A58 merge - will make quite a difference.

The only caveat around here is that there are no traffic signal cameras, and at least three vehicles will jump a red; this can block a junction, even where there's a box that's supposed to prevent that from happening. This can be particularly bad!

In fact I'm now wondering if the distance between the stop lines and the A643/A58 merge could be in anticipation of light-jumping? It's surely cheaper than installing enforcement cameras...

[Peter Freeman]

Switch Island signalised merge is extensively discussed upthread, starting in Keiji's post on 18-09-2023. That signalised merge works well. I have driven through there, though not at peak.

I didn't relise that those M58/A59 merge signals are coordinated with the next ones along. I would guess though, that the signalised merge would work ok there even without that coordination.

[Peter Freeman]

Indeed - the problem there is caused by the proximity of the merge to the gyratory.

For info, what tended to happen in practice before the revamp began, is that traffic on the A58 merge would split during peak periods, even though there were no road markings to provide distinct lanes. This is what it looked like before.

With local knowledge, drivers would organise themselves during heavy traffic, depending on where they wanted to go from the gyratory:

- lane 1 to go into lane 1 on the mainline, which splits at the gyratory to provide lanes for the B6154 or A647 gyratory exits, or A58(M) to/through the city

- lane 2 to go straight to lane 2 (which becomes lane 3 at the gyratory) on the mainline, for A58(M) to/through the city

So there are conflicting movements, as you point out, but the traffic exercised a high level of discipline in organising itself to let everybody get into the right place.

That's interesting. In some ways, the signalised merge, if it has a couple of nice lane-assignment gantry signs, will formalise, especially for non-locals less familiar with dealing with that particular battleground, what already happens.

It was still gridlocked though, due to traffic on the mainline not moving, because it was held by red lights up the road at the gyratory.

Well, yes, the real culprit in all of this! And my belief that the signalised merge is the correct thing to do, is because there might still be a long queue even after the roundabout works.

I'm assuming that the signals will indeed operate in coordination with those on the gyratory.

I wonder. I actually don't think it's necessary. We'll see, when it's all finished and I can watch G.Maps traffic layer!

The only caveat around here is that there are no traffic signal cameras, and at least three vehicles will jump a red; this can block a junction, even where there's a box that's supposed to prevent that from happening. This can be particularly bad!

In fact I'm now wondering if the distance between the stop lines and the A643/A58 merge could be in anticipation of light-jumping? It's surely cheaper than installing enforcement cameras...

Yes, that could be the reason for what I considered too far.

[Owain]

Regarding the ongoing works on the Armley gyratory, I would have been providing photos of this, but it's now a hostile environment for pedestrians with a lengthy diversion via the footbridge near the A58/A643 merge that we've been discussing.

The last time I walked via the gyratory itself, I was almost run over by an Audi that was in turn avoiding a MINI that seemed confused by all the orange cones and adjusted markings...

[Peter Freeman]

A well known one is just before the southbound Blackwall Tunnel.

Ah, yes -
https://www.google.com.au/maps/@51.5103 ... ?entry=ttu
And another one just north of there (northbound) -
https://www.google.com.au/maps/@51.5131 ... ?entry=ttu
They appear to be justified, on the usual grounds.

[Micro The Maniac]

Another "classic" merge - the A327 merging onto the A329 Reading IDR, before the A33 diverge.

https://maps.app.goo.gl/yZLHVZ4k9Gk8NdDg7

More surprisingly, there's one in the Southern Industrial Estate at Bracknell - which is a well known short-cut to avoid the Three Bridges Roundabout

https://maps.app.goo.gl/6KLizLfYH5smmeDz7

[Peter Freeman]

Another "classic" merge - the A327 merging onto the A329 Reading IDR, before the A33 diverge.
https://maps.app.goo.gl/yZLHVZ4k9Gk8NdDg7

Yep, classic usage there. The immediate area contains some very unusual skewed signalised intersections!

More surprisingly, there's one in the Southern Industrial Estate at Bracknell - which is a well known short-cut to avoid the Three Bridges Roundabout.
https://maps.app.goo.gl/6KLizLfYH5smmeDz7

A surprising location, but a classic reason: a trumpet leading shortly to a roundabout.

I'm actually quite surprised how many signalized merges have turned up.

[Micro The Maniac]

Yep, classic usage there. The immediate area contains some very unusual skewed signalised intersections!

The Reading IDR is a curious entity - fully grade-separated reasonable-quality D2 in places, and pretty meh! S2 in others... but never more than 40mph (which is reasonable) and with loads of traffic lights. While the IDR itself isolates the centre of Reading, there are plenty of safe routes to cross under or over it to the south and west, and myriad of light-controlled crossings to the north and east.

The real problem with it is that it is on the only viable north/south route from the M4 - plans to extend the A329(M)/A3290 around to the A4155 or even the A4074 have come to nought - the Chilterns AONB (now National Landscape) being a notable barrier to extension.

Apologies for the thread drift...

[M4Simon]

A well known one is just before the southbound Blackwall Tunnel.

Ah, yes -
https://www.google.com.au/maps/@51.5103 ... ?entry=ttu
And another one just north of there (northbound) -
https://www.google.com.au/maps/@51.5131 ... ?entry=ttu
They appear to be justified, on the usual grounds.

Both are required to manage the flow into the Blackwall Tunnel in spaces where there is both insufficient room for a safe merge layout, and where the flow into the tunnel is so far beyond free-flow limits that the flow frequently breaks down. Therefore, these junctions do not act as a traditional merge, and could be laid out as T-junctions. The signals balance demand from both the A12 and A13.

Simon

[Peter Freeman]

A well known one is just before the southbound Blackwall Tunnel.

Ah, yes -
https://www.google.com.au/maps/@51.5103 ... ?entry=ttu
And another one just north of there (northbound) -
https://www.google.com.au/maps/@51.5131 ... ?entry=ttu
They appear to be justified, on the usual grounds.

Both are required to manage the flow into the Blackwall Tunnel in spaces where there is both insufficient room for a safe merge layout, and where the flow into the tunnel is so far beyond free-flow limits that the flow frequently breaks down. Therefore, these junctions do not act as a traditional merge, and could be laid out as T-junctions. The signals balance demand from both the A12 and A13.

You're right that they don't act as traditional merges. They act as signalised merges (SM). Signalised merging means platoons taking turns.

They "could be laid out as T-junctions": I'm not sure what you mean. Do you mean the inputs from A13 could meet the A12 perpendicularly, insead of via a gore? Yes they could, and the conflict removal effect would be the same, but a 90 degree left turn is disruptive and inefficient, and you'd lose many metres of storage.

The northbound signalised merge is exactly a textbook case. A signalised junction (Zetland/Lochnagar Streets) lies 400m to the north. A queue filling up this space, without the SM intervening, would suffer lane-swapping and consequent inefficient flow. The SM allows entering vehicles from each source to first comfortably select a preferred lane, allowing smooth efficient flow up to and through the signals. Textbook.

The southbound one is different, in that there are no subsequent signals to cause queueing. However, entering a confined tunnel also causes slowdown and queuing. I don't know whether being in a non-preferred lane inside the tunnel (which could occur without the SM) is detrimental. Is lane-change prohibited in the tunnel?

"The signals balance demand from both the A12 and A13": Yes they can do that. For what reason? To prevent queue overflow onto A13? To prevent domination by A12? I suppose that could be another reason for the SM.

[Chris5156]

The southbound one is different, in that there are no subsequent signals to cause queueing. However, entering a confined tunnel also causes slowdown and queuing. I don't know whether being in a non-preferred lane inside the tunnel (which could occur without the SM) is detrimental. Is lane-change prohibited in the tunnel?

Yes, there are solid lines through the tunnel, and a fair distance beyond it where you can change lane, if you need to, before the next exit.

"The signals balance demand from both the A12 and A13": Yes they can do that. For what reason? To prevent queue overflow onto A13? To prevent domination by A12? I suppose that could be another reason for the SM.

Because the queues generated by the under-capacity tunnel often stack back onto both sliproads and far beyond. The lights ration out access to the tunnel; if it was a conventional merge (which it was originally, hence the physical layout) the A12 would dominate and vehicles trying to enter from the A13 would have even less opportunity to get in.

[A303Chris]

Yep, classic usage there. The immediate area contains some very unusual skewed signalised intersections!

The Reading IDR is a curious entity - fully grade-separated reasonable-quality D2 in places, and pretty meh! S2 in others... but never more than 40mph (which is reasonable) and with loads of traffic lights. While the IDR itself isolates the centre of Reading, there are plenty of safe routes to cross under or over it to the south and west, and myriad of light-controlled crossings to the north and east.

The real problem with it is that it is on the only viable north/south route from the M4 - plans to extend the A329(M)/A3290 around to the A4155 or even the A4074 have come to nought - the Chilterns AONB (now National Landscape) being a notable barrier to extension.

Apologies for the thread drift...

Don't I know it as I drive it every day. I created this post, I can not believe 10 years ago where I scanned in all the plans for proposals in Reading which were uncovered in an office move!!! The second post Third Thames bridge shows the proposals for either side of the town.

Reading and Wokingham are still looking at a third Thames bridge on the east, but given the bizarre county boundary which includes a small bit of Oxfordshire being surrounded by Berkshire on three sides to the north of the river and south of the Henley Road it constantly gets blocked by Oxfordshire as they say it will create rat running through South Oxfordshire.

[A303Chris]

Back to the original post, I am surprised this one on the A332 Windsor and Eton Relief Road has not been mentioned.

This was done in the early 90's, 30 years ago, and I worked on the project. I am sure this was the first one in the country and needed special dispensation from the DfT at the time.

[davidsouthwell]

Yep, classic usage there. The immediate area contains some very unusual skewed signalised intersections!

The Reading IDR is a curious entity - fully grade-separated reasonable-quality D2 in places, and pretty meh! S2 in others... but never more than 40mph (which is reasonable) and with loads of traffic lights. While the IDR itself isolates the centre of Reading, there are plenty of safe routes to cross under or over it to the south and west, and myriad of light-controlled crossings to the north and east.

The real problem with it is that it is on the only viable north/south route from the M4 - plans to extend the A329(M)/A3290 around to the A4155 or even the A4074 have come to nought - the Chilterns AONB (now National Landscape) being a notable barrier to extension.

Apologies for the thread drift...

Don't I know it as I drive it every day. I created this post, I can not believe 10 years ago where I scanned in all the plans for proposals in Reading which were uncovered in an office move!!! The second post Third Thames bridge shows the proposals for either side of the town.

Reading and Wokingham are still looking at a third Thames bridge on the east, but given the bizarre county boundary which includes a small bit of Oxfordshire being surrounded by Berkshire on three sides to the north of the river and south of the Henley Road it constantly gets blocked by Oxfordshire as they say it will create rat running through South Oxfordshire.

I've always thought the solution to this to get the problem of Reading traffic fixed would be to just jerrymander the boarder of Oxfordshire further north and give the northern side of the Thames, near Playhatch, to Reading. Its ridiculous that Oxfordshire has control over a tiny out cropping which has minimal houses, if any, yet they effectively create misery for all residents of Wokingham, Bracknell and Reading who just want to get north of the river without having to go through sleepy Sonning or the nightmare which is Cemetery Junction and central Reading. Its nuts, and I really think this is the nuclear option, as essentially Oxfordshire council need to be eliminated from this decades long farce.

[Peter Freeman]

Back to the original post, I am surprised this one on the A332 Windsor and Eton Relief Road has not been mentioned.

This was done in the early 90's, 30 years ago, and I worked on the project. I am sure this was the first one in the country and needed special dispensation from the DfT at the time.

Required because the river is just north, making a conventional merge unviable unless the bridge was widened nb.

I feel compelled to add (with the benefit of hindsight) that my solution would have been to NOT build a GSJ, thereby avoiding the merging difficulty. After all, southbound it leads almost directly onto a multi-lane roundabout which, if there was much more traffic than in reality, could force another signalised merge. More generally, A332 seems to fizzle out southwards, so it hardly needs near-motorway status at that point near the river.

Thanks for listing this example. And congratulations on successfully pioneering the dispensation.

[davidsouthwell]

Back to the original post, I am surprised this one on the A332 Windsor and Eton Relief Road has not been mentioned.

This was done in the early 90's, 30 years ago, and I worked on the project. I am sure this was the first one in the country and needed special dispensation from the DfT at the time.

Interesting, I've always wondered why this one doesn't just have a merge on the sliproad, and then a merge in turn on the main carriageway, which would then eliminate the need for the traffic lights. Traffic rarely seems that heavy here and there's ample stacking space on the flyover for the merge in turn on the main carriageway.

[Peter Freeman]

^ Compressing each input to single lane on their approaches is always the obvious other solution, avoiding signalisation. In most cases, the main road is of such status and traffic volume that that would be unacceptable. In this case, A332 might be borderline in its AADT.

[jackal]

They can't reduce the A332 to one lane. Its 60k AADT north of the merge. Maybe the issue was the short merge, which could be extended?

[Peter Freeman]

They can't reduce the A332 to one lane. Its 60k AADT north of the merge. Maybe the issue was the short merge, which could be extended?

Oh, that's more volume than I expected. You wouldn't try to constrict that to one lane !!!

The issue certainly was the short merge, which is one classic SM reason. The nb on-ramp is 200m long, the merge is 100m - total = 300m. Most motorway ramp+merge totals are >500m. This one couldn't be extended without widening the river bridge on its west side: $$$. So, in the circumstances, given the GSJ, in that era, the SM was the right solution.

(Though it must be said that many UK DC's do cope with merges that are this sub-standard).

Looking south from the GSJ, I would propose another SM on the sb on-ramp and sb mainline, for the other classic reason to signalise a merge: too close to the back of a multi-lane queue.

But let's forget the 1990's, and look at it from a modern, rational standpoint. Where is this junction situated? It's on the A332, purporting to be a high-capacity urban DC. Where does its traffic come from? To the north it comes from one signalised and one un-signalised roundabout, added together. To the south it comes from one 5-arm signalised roundabout. In other words, the GSJ is an 'orphan'. (Even just looking at it, on G.Maps plan view, it is so long N-S that it appears to be a mis-fit).

Yes, 60k AADT is higher than I expected, but in AU (for example - sorry!) we routinely have such flows on our urban at-grade arterials, without massive congestion. And I assume UK does too (A580?). An at-grade up-scale signalised crossroads would have handled, then and now, this junction, without having to worry about merging, without having to seek an exemption, and at radically lower cost.

I think this must be a relic (or an icon) of the happy UK days when money for grade separations was much easier to come by. (These words sound like 1960's/70's/80's ... but 1990's too?)