M3 Junction 9 Improvements

[BOH]

Wasn't there originally a plan for the M3 to follow what is now the A3 London - Portsmouth route with a split just after Guildford that went on to Winchester and Southampton loosely following the line of the A31? Am talking 50-60+ years ago though, possibly when the original motorway systems were being planned.

[si404]

The original purpose of the M3 was to relieve the A30, and of course a lot of traffic on the A30 was from/to the A303. Those facts determined its approximate route as far as its original terminus at J8.

Which is fair enough. But it could've remained as motorway, multiplexing with the A303 for a few miles, before taking a more southerly course.

You mean westerly?

The big barrier to the west-of-Winchester route is that phasing it would be hard - you would have to do all the A34-M27 section in one go. By the time it came to it, a plug in to the A33 Winchester was a cheap fix, and then an online upgrade of the A33 Chandlers Ford bypass to deal with commuter traffic, followed by the Winchester bypass itself.

In fact, if it had done that, and met with the M271, the last couple of miles of M27 westbound wouldn't have been necessary (or just been A31(M) instead.

The M27 was about bypassing the South Coast Trunk Road, which met the A31 at Cadnam. The J1-3 bit of M27 is very important for Totton.

I can't help thinking that would've made better sense?? Than shoving all Southampton traffic down past Winchester, mixed in with the local commuter traffic,

the commuter traffic is Southampton traffic!

Worth pointing out that most other cities have no dedicated road for docks or through traffic, segregating it from commuter traffic. The main reasons why the east of Winchester route is not ideal is that J11-13 hasn't been widened, and the whole Twyford Down debacle.

and arriving at Bassett

Chilworth

so everyone has to either drive through the city, or the long way round to the docks.

It's not really a long way around, especially for freight traffic - most of which doesn't take The Avenue - no least as it is slower.

[wrinkly]

Wasn't there originally a plan for the M3 to follow what is now the A3 London - Portsmouth route with a split just after Guildford that went on to Winchester and Southampton loosely following the line of the A31? Am talking 50-60+ years ago though, possibly when the original motorway systems were being planned.

This extract from Hansard in 1962 shows that it was already known at that date that the M3 would go to Basingstoke:

http://hansard.millbanksystems.com/comm ... 205_HOC_37

There's also this from Jan 1960 referring to a proposed motorway from London to Kempshott without mentioning any M number:

http://hansard.millbanksystems.com/comm ... 127_HOC_59

As for any possible earlier plans, I've looked at Figs 2-4 in Motorways by Sir James Drake et al. These are very crude and schematic; I don't know whether the proposals they summarise were more detailed than the maps suggest.

Fig. 2, on the Institution of Highway Engineers 1936 plan, seems to show a motorway in a straight line from a junction with a London orbital to a junction with a south coast motorway about a third of the way from Salisbury to Southampton.

Fig. 3, on the County Surveyors' Society National Plan for Motorways of 1938, shows a motorway as a straight line from a point somewhere east of Reading, where it turns off an M4-like motorway, to Southampton, with a spur turning off southwards to what is presumably Portsmouth.

Fig. 4, on the Ministry of Transport Road Programme, 1946, has continuous lines (presumably meaning motorways) and dashed lines (presumably meaning all-purpose roads). There are no continuous lines south of (what became) the M4, except for the last bit of (what is roughly) the M5 to Bristol. There is a dashed line from London to Salisbury, Exeter and Plymouth, with another dashed line turning off it to Southampton.

I suppose it's possible that the plan that you describe existed for a time, say in the 1950s, but I've found nothing to suggest that it did.

[jackal]

Consultation on the proposed option is now open. Maybe they could've done something about the weaving up to the A33 junction but it's basically good

Brochure: https://highwaysengland.citizenspace.co ... 0FINAL.PDF
Consultation page: https://highwaysengland.citizenspace.co ... rovements/

[Bryn666]

That's remarkably similar to something a few of us drew a long time ago.

Do you think the junctions department has finally realised you can do things without using signalised junctions?

[RichardA35]

Consultation on the proposed option is now open. Maybe they could've done something about the weaving up to the A33 junction but it's basically good

Brochure: https://highwaysengland.citizenspace.co ... 0FINAL.PDF
Consultation page: https://highwaysengland.citizenspace.co ... rovements/

Indeed as I see it local traffic (from Spitfire Link or Tescos etc) heading up the A33 will now have to weave across fast freeflowing traffic rather than being positioned in the right hand lane from a standing start at the roundabout currently.
Reversed slip roads - could be fun!

[jackal]

If I'm understanding that correctly, you basically have the same movement replicated (i.e. with slips on both sides), so there is no reason for anyone to weave across. I don't think this is presently used in the UK but there are similar layouts in California and South Africa. It's quite a high end solution, usually used for merging freeways, so maybe overkill here.

Do you think the junctions department has finally realised you can do things without using signalised junctions?

Never!

[Herned]

Seems to be very sensible apart from the A33 nbound. The weaving distance looks very short, perhaps an off-slip to the B3047 would work better?

In fact RichardA35 has an even better solution, there would be no need for the nbound merge at all, and no need for extra bridges... wonder why that hasn't been considered?

[jackal]

I still think the ideal solution is to put in an additional northbound carriageway (in blue, because it would be under motorway regs). The current northbound A34 (in red) just becomes the northbound slip road to the A33 and A34. The merge with the A34 is then north of the A33 diverge so there is no weaving, and there is no TOTSO for the A33 either.

[Bryn666]

I still think the ideal solution is to put in an additional northbound carriageway (in blue, because it would be under motorway regs). The current northbound A34 (in red) just becomes the northbound slip road to the A33 and A34. The merge with the A34 is then north of the A33 diverge so there is no weaving, and there is no TOTSO for the A33 either.

M3 A34 - Copy.PNG

My doodle was this: https://drive.google.com/open?id=1FapNT ... sp=sharing

That said, the A33 gets a bit of a raw-deal as you will have to cross a roundabout and use an S2 link road to join the existing J9. The idea was to keep the A34 separate as much as possible.

[jackal]

I took a look at the Technical Appraisal Report from the consultation page. There are five options (with drawings):

Option 11 - basically similar to my drawing above and Bryn's (minus the B3047 bypass), with 120kph freeflow in both directions and the weaving and A33 TOTSO removed.
Option 14 - the consulted option, and largely a lower cost version of option 11, with three bridges rather than six. There is still freeflow in both directions, but less generous southbound geometry (100kph with three relaxations), while northbound has weaving and a TOTSO. Some detail on the TOTSO: 'North of the existing River Itchen Bridge the existing A34/A33 diverge will be widened to allow two lanes to run continuously on the A34 with a non-recommended offside type A taper diverge to the A33. TD22 states that an offside diverge is not recommended due to safety reasons, however, this is unavoidable in this case without significant construction work similar to Option 11. As a potential alternative the existing A34 northbound carriageway could be widened to three lanes in the vicinity of the tie in of the new A34 northbound link with the existing to provide a length of three lane carriageway which allows two lanes to proceed for A34 northbound and the offside lane to proceed to A33'. From the drawing in the brochure I don't think they're widening to three lanes.
Option 16A - freeflow southbound only. Essentially half of option 14.
Option 16B - freeflow northbound only. Essentially the other half of option 14 - but with no replacement for the northbound onslip at J9! ('Traffic from Winchester wishing to join the M3 northbound will be required to do so via Junction 11 or the A33/A30 and Junction 7'.)
Option 18 - hamburger for A34>M3sb plus changes to A33/A34 diverge arrangements: 'For Option 18 the benefits are all driven by improvements at the A34/A33 diverge and the proposed hamburger junction arrangement at M3 Junction 9 is over capacity and actually results in small disbenefits to users'.

That said, the A33 gets a bit of a raw-deal as you will have to cross a roundabout and use an S2 link road to join the existing J9. The idea was to keep the A34 separate as much as possible.

The TAR reached the same conclusion: 'the benefits of Option 11 would be expected to be greater than Option 14 but Option 14 benefits are increased as a result of the A34/A33 diverge arrangement performing with fewer wider impacts than that in Option 11. More specifically the diverge arrangement in Option 11 is less beneficial for trips from the M3 to A33 than Option 14 as a result of needing to negotiate the proposed dumbbell arrangement at Junction 9'.

Overall the consultants clearly know what they're doing, and in spite of apparent drawbacks, the proposed option appears to be well justified.

[Truvelo]

More specifically the diverge arrangement in Option 11 is less beneficial for trips from the M3 to A33 than Option 14 as a result of needing to negotiate the proposed dumbbell arrangement at Junction 9'.

That can be overcome by adding a short link between the new northbound carriageway and the existing weaving section. As M3 to A34 traffic is removed from this section there won't be any hassle with weaving.

[Paul7755]

North of the existing River Itchen Bridge the existing A34/A33 diverge will be widened to allow two lanes to run continuously on the A34 with a non-recommended offside type A taper diverge to the A33.

Why was that never provided in the first place, or in the period since? It is plain as day that the capacity is throttled by the present 50/50 lane division, and as people merge into the left lane for half a mile the stop start ripples back to the Winnal roundabout. ISTM it could be fixed over a weekend by a lane marking crew...

Paul

[jackal]

More specifically the diverge arrangement in Option 11 is less beneficial for trips from the M3 to A33 than Option 14 as a result of needing to negotiate the proposed dumbbell arrangement at Junction 9'.

That can be overcome by adding a short link between the new northbound carriageway and the existing weaving section. As M3 to A34 traffic is removed from this section there won't be any hassle with weaving.

Yes, that would be a good improvement - specifically you'd link into the northbound A33 just after it diverges. But it would be yet another cost for option 11 (including an additional bridge over the river). Option 11 is already far beyond the scheme budget of £50m-£100m:

Option 11 14
Cost (£m, 2014) 186.8 134.1
BCR 1.31 1.88

In fact, even option 14 exceeds the budget, and could be considered a great outcome IMO.

[Richardf]

That's remarkably similar to something a few of us drew a long time ago.

Do you think the junctions department has finally realised you can do things without using signalised junctions?

Yes I thought it looked familiar. Had similar ideas myself in the past!

[mikehindsonevans]

North of the existing River Itchen Bridge the existing A34/A33 diverge will be widened to allow two lanes to run continuously on the A34 with a non-recommended offside type A taper diverge to the A33.

Why was that never provided in the first place, or in the period since? It is plain as day that the capacity is throttled by the present 50/50 lane division, and as people merge into the left lane for half a mile the stop start ripples back to the Winnal roundabout. ISTM it could be fixed over a weekend by a lane marking crew...

Paul

Every time we leave M3j9 to head northbound in lane one leading to the A34, we play "how long will the numpty in the (insert brand name) car/van/*COACH* leave it before starting to indicate left and annoy everyone by pulling in?". Two lanes northbound, continuous from the roundabout to the A34 proper, make sense. (Indeed, when the layby immediately north of the roundabout (yes, it was a REALLY safe place to stop to collect a hitchhiker!) was closed some years ago, we had high (and as it turned out, wrong) hopes that a new far-left A34 lane would emerge following the roadworks. Our hopes were cruelly dashed!).

Frankly, anything would be an improvement on the overstressed mess that is M3J9 today. I suspect that, northbound on the M3 heading past J10, a two-mile warning that the left-most lane is the lane for the A34 might reduce *SOME* last-minute weaving from lane 2 (which is the first M3 London-bound lane after M3J9). Not a lot, given the Audi kamikaze brigade, but some motorists might move left slightly earlier, before the left-hand bend (and overgrown hedge) which blocks the view of the one-mile ADS sign.

[RichardA35]

North of the existing River Itchen Bridge the existing A34/A33 diverge will be widened to allow two lanes to run continuously on the A34 with a non-recommended offside type A taper diverge to the A33.

Why was that never provided in the first place, or in the period since? It is plain as day that the capacity is throttled by the present 50/50 lane division, and as people merge into the left lane for half a mile the stop start ripples back to the Winnal roundabout. ISTM it could be fixed over a weekend by a lane marking crew...

Paul

Every time we leave M3j9 to head northbound in lane one leading to the A34, we play "how long will the numpty in the (insert brand name) car/van/*COACH* leave it before starting to indicate left and annoy everyone by pulling in?". Two lanes northbound, continuous from the roundabout to the A34 proper, make sense. (Indeed, when the layby immediately north of the roundabout (yes, it was a REALLY safe place to stop to collect a hitchhiker!) was closed some years ago, we had high (and as it turned out, wrong) hopes that a new far-left A34 lane would emerge following the roadworks. Our hopes were cruelly dashed!).

Frankly, anything would be an improvement on the overstressed mess that is M3J9 today. I suspect that, northbound on the M3 heading past J10, a two-mile warning that the left-most lane is the lane for the A34 might reduce *SOME* last-minute weaving from lane 2 (which is the first M3 London-bound lane after M3J9). Not a lot, given the Audi kamikaze brigade, but some motorists might move left slightly earlier, before the left-hand bend (and overgrown hedge) which blocks the view of the one-mile ADS sign.

The 1 mile ADS northbound has been missing for at least 20 years so I suspect it is deliberate.

[Johnathan404]

When I had a quick look at the leaflet this morning I was taken in by how elegant the solution was. Now I've thought about it, I'm disappointed the easy goal has been missed.

Firstly, it's good to see the A33/A34 split will be re-marked with two lanes for the A34. It's OK if you're the first one through the lights at J9, but if you're at the back of the queue then you will come to a standstill on the A34 due to the volume of vehicles piling in to one lane. That's dangerous and should have been changed years ago, and can be done within the existing tarmac. It wasn't changed because right-hand exits go against standards, but it's good to see a blind eye will finally be turned.

In light of the comments above I thought about it a little more, and then wondered: why is the exit there anyway? The traffic flows are very low and it is all local traffic heading to one of the villages not served by the M3 or trying to sneak in to northern Winchester without having to use the one way system. There is no need for it to be part of the free-flow link to the M3, even if it makes for a great drive.

It feels like the split was outside the true scope of the investigation so has been left substantially as it is. This is a shame because the aspiration must surely be to make the A34 an "expressway" thingy and therefore the opportunity should be taken to remove unnecessary and sub-standard exits.

Meanwhile, we are introducing another exit to the congested M3. That's not the end of the world, but again, I wonder if it was really necessary. Was a study done into who uses the south-facing sliproads at M3 J9, and did it ask whether they could just use J10 instead?

On another note, it's good to see confirmation that the goal is to make the M3 D4, and that the scheme will be compatible with it. Just by lengthening the 2+2 split it should reduce the extent of domino-braking on the M3 considerably.

It's also good to see a hamburger was rejected

[jackal]

Looking at the video they've now uploaded the weaving on the A34 does look a bit scary. Northbound, the tapers for the merge and diverge almost overlap! And there's even a bit of an issue southbound because they show a long ghost island for the merge, bringing it close to the diverge.

The A33 isn't busy enough to justify a two lane merge so the ghost island can easily be removed. Northbound is trickier but it would help if they just made the nose for the A33 diverge less ridiculously long.

[Richardf]

I have to say this plan looks really good, I just hope it gets done like thus abd not watered down or cancelled.

While the A34/A33 split is a problem, especially NB, am not sure what substantially could be done about it, not without a lot if extra money being spent, which given the traffic levels isnt justified. Not even sure how you could cut out the A33 movement, not without sending A33 traffic into Winchester and out again to reach the M3 and other routes.

One option would be to leave the A34 NB as a sliproad and build a new NB carriageway as jackal suggested. Then you can either leave the rest of the junction alone, with the A33 leaving and joining as now or make the old A34 and A33 NB two way with a roundabout where they currently split. The SB A33 slips could then be removed and the whole thing is simplified.