Hamburger/Throughabout Controversy

[Bryn666]

^ But - are signals installed without road-embedded detectors on each lane of each approach?

Properly designed junctions all have detector loops or microwave detection. Sites running fixed time are because there is a fault somewhere which either has been reported to the relevant UTMC or hasn't.

Very few, if any, junctions are built today to run fixed time and nothing else. It's another SABRE myth posted without any fact checking.

[RichardA35]

Then where there is a budget of say £100k for a signal installation, the hardware salesman will try to scoop every bit of that for the hardware side. The programming, which used to be bundled but is increasingly not, with a range of chargeable add-ons, then blows the budget, so the minimal programming is bought and it gets installed with the default settings.

Each of the signals installations I have been responsible for has been a complete budget for hardware, software and installation and never has it been pared back in such a fashion.
We don't deal with salesmen.
Tenders are invited on a clear specification.
To win the job, compliance is expected and achieved.
Price is negotiated.

^ But - are signals installed without road-embedded detectors on each lane of each approach?

Properly designed junctions all have detector loops or microwave detection. Sites running fixed time are because there is a fault somewhere which either has been reported to the relevant UTMC or hasn't.

Very few, if any, junctions are built today to run fixed time and nothing else. It's another SABRE myth posted without any fact checking.

Similarly each installation has been with a MOVA based controller with red light fault monitoring and the ability to fall back to other plans and also be modified via a comms link if needed. It suggests some Client HA's are cutting corners if some of the above claims are true.

[Bryn666]

Then where there is a budget of say £100k for a signal installation, the hardware salesman will try to scoop every bit of that for the hardware side. The programming, which used to be bundled but is increasingly not, with a range of chargeable add-ons, then blows the budget, so the minimal programming is bought and it gets installed with the default settings.

Each of the signals installations I have been responsible for has been a complete budget for hardware, software and installation and never has it been pared back in such a fashion.
We don't deal with salesmen.
Tenders are invited on a clear specification.
To win the job, compliance is expected and achieved.
Price is negotiated.

^ But - are signals installed without road-embedded detectors on each lane of each approach?

Properly designed junctions all have detector loops or microwave detection. Sites running fixed time are because there is a fault somewhere which either has been reported to the relevant UTMC or hasn't.

Very few, if any, junctions are built today to run fixed time and nothing else. It's another SABRE myth posted without any fact checking.

Similarly each installation has been with a MOVA based controller with red light fault monitoring and the ability to fall back to other plans and also be modified via a comms link if needed. It suggests some Client HA's are cutting corners if some of the above claims are true.

If they are, then the members of this website that suggest they are will be sitting on a massive investigative journalist scoop... why waste time telling us when those well known civil service loving Fleet Street publications would love to have hold of such dynamite?

[traffic-light-man]

SABRE Highways Consultancy Services, anyone?

[Ruperts Trooper]

^ But - are signals installed without road-embedded detectors on each lane of each approach?

Properly designed junctions all have detector loops or microwave detection. Sites running fixed time are because there is a fault somewhere which either has been reported to the relevant UTMC or hasn't.

Very few, if any, junctions are built today to run fixed time and nothing else. It's another SABRE myth posted without any fact checking.

If that's a SABRE myth then a lot of signalised roundabouts must be faulty and don't ever seem to get repaired - it's very noticeable in the early hours of the morning when there's little traffic about.

[Peter Freeman]

I'm saying it is better to have a roundabout that is congested for an hour or so, twice a day, than a signalised junction that is congested 24/7

Apart from the vast majority signalised junctions are not congested 24/7 because off-peak they have adaptive signal timings to minimise delays. The idea that there are endless queues at signals at 3am is laughable; what you really mean is you don't like having to wait a few seconds for a signal to change?

Roundabouts are never congested for "an hour or so", they either work or they don't.

Lets be realistic here. There are places where roundabouts work ok, and a signalised crossroads is unnecessary and would be an imposition. Then there are places where the traffic flow is so high that a roundabout simply can't cope (except in the dead of night), and a signalised crosssroads is clearly the answer.

However, there will be marginal cases where a certain roundabout is only congested for a short period each weekday, and the pragmatic action is to tolerate it. That's Micro's position, and a reasonable one.

Conversely, you may well approach a signalised crossroads, off-peak, and find that you need to briefly stop and wait, even with little conflicting traffic around. Or perhaps even with no other vehicle in sight. Signals must, even off-peak, have some minimum green-time (which is red-time for the opposing flow) programmed in, and minimum intergreen, etc. But does waiting for 15 seconds = "24x7 congestion"? Of course not.

Bryn's analysis is correct: some people just hate stopping.

[jackal]

Your daily reminder that the meta analysis of 54 Pinch Point schemes found that, overall, they increased journey times, predominantly due to signalisation. Reductions in the peaks were more than outweighed by increases offpeak. 88% of these schemes had journey time reductions as a target.

Traffic engineers that dismiss road users' concerns with signalisation out of hand would do well to remind themselves that their profession as a whole is actually pretty bad at designing signalised junctions even by their own metrics. A layperson can outperform sophisticated appraisal models where those models are systematically flawed.

That said, I expect that the poor performance of these particular schemes is predominantly due to signalisation of roundabouts. I agree with Bryn, Chris and others that conventional signalised junctions would perform better in most cases.

https://assets.publishing.service.gov.u ... _draft.pdf

[Micro The Maniac]

Thanks Jackal

Picking one at random

This (the M27/J5) scheme is no different and shows benefits in the PM peak and for movements from the Airport. However, all other periods and movements have overall adverse impacts.

Interesting that an official report disagrees so clearly with our resident (self-appointed) expert...

[Bryn666]

The big problem is trying to shoehorn in all movements and pedestrian/cycle facilities and a refusal to consider you can do things with subways and, for traffic, specific grade separation.

A local screw up was the "improvements" to Regent Road/Trinity Way in Manchester. This actually removed cycle provision by replacing it with shared footpaths whilst also adding numerous stages to try and separate traffic flows.

A simple one way flyover from Mancunian Way to Trinity Way would have allowed a simpler signal junction and cyclops layout for cyclists. Instead we have the worst of everything, and it still cost upwards of a million.

[DB617]

There definitely is some work to do on poorly timed (or indeed faulty) signalisations. From the research I've been doing to try and find some signage guidance for throughabouts, I stumbled upon a passage in LTN 1/09 which showed why part time lights for peaktime congestion have fallen out of favour, and it's quite justifiably pedestrian facilities. Without the lights being full time, it's difficult to fully and safely accommodate pedestrians. I don't think we have provision in the UK for signals that are part time but can also switch on off-time when a pedestrian is waiting to cross? That leads us to the need for 24/7 lights that should be programmed and constructed well enough to allow them to be essentially a non-factor during the times of day when congestion was never an issue. This is all within our grasp, as other users and engineers have said above, but it needs to be done properly by every authority and then properly maintained, which I can imagine is difficult at present given some cannot seem to even afford relining of roads which have turned into an expanse of blank tarmac (looking at you, South Glos).

[Bryn666]

Thanks Jackal

Picking one at random

This (the M27/J5) scheme is no different and shows benefits in the PM peak and for movements from the Airport. However, all other periods and movements have overall adverse impacts.

Interesting that an official report disagrees so clearly with our resident (self-appointed) expert...

No, it says some junctions don't work whereas you repeatedly (wrongly) claim all signal junctions don't work. What will be interesting is how these junctions perform safety wise and from an NMU perspective. I bet the results are far more positive there but if all you care about is "brum brum must drive fast" you won't be interested, naturally, because it doesn't fit with your worldview.

However as a more broad general point, since you're so knowledgeable about highways design with your constant debunkable posting why don't you come do it for a while? I'm sure we'll all have lots to learn from you. In fact, it's tedious bores like yourself which is why the general public is treated with contempt so readily by numerous consultancies and councils - why should we bother being reasonable with such drivel merchants who always know best despite their highways expertise extending no further than the fact they somehow passed a driving test?

The reports also ignore the fact that car journey times are not the priority in terms of the road user hierarchy, but a renowned highways expert like you would already know that. In fact you'll be able to tell the class who ranks first in said hierarchy without looking it up? Or maybe you can tell the class about WebTAG analysis, which also does not prioritise car journey times and looks at wider metrics for appraisal of highways schemes?

I'll wait. I mean I have so much to learn from you. Meanwhile, in the real world, I have actual useful schemes to design that factor in more than you and your solitary car that must never be held up at a single traffic signal.

[Bryn666]

There definitely is some work to do on poorly timed (or indeed faulty) signalisations. From the research I've been doing to try and find some signage guidance for throughabouts, I stumbled upon a passage in LTN 1/09 which showed why part time lights for peaktime congestion have fallen out of favour, and it's quite justifiably pedestrian facilities. Without the lights being full time, it's difficult to fully and safely accommodate pedestrians. I don't think we have provision in the UK for signals that are part time but can also switch on off-time when a pedestrian is waiting to cross? That leads us to the need for 24/7 lights that should be programmed and constructed well enough to allow them to be essentially a non-factor during the times of day when congestion was never an issue. This is all within our grasp, as other users and engineers have said above, but it needs to be done properly by every authority and then properly maintained, which I can imagine is difficult at present given some cannot seem to even afford relining of roads which have turned into an expanse of blank tarmac (looking at you, South Glos).

The answer is quite simple - grade separate the NMU movements. However this is expensive and that isn't allowed because people (including some in this very thread) complain about wasting money on any things that aren't "brum brum drive fast".

[Bryn666]

^ But - are signals installed without road-embedded detectors on each lane of each approach?

Properly designed junctions all have detector loops or microwave detection. Sites running fixed time are because there is a fault somewhere which either has been reported to the relevant UTMC or hasn't.

Very few, if any, junctions are built today to run fixed time and nothing else. It's another SABRE myth posted without any fact checking.

If that's a SABRE myth then a lot of signalised roundabouts must be faulty and don't ever seem to get repaired - it's very noticeable in the early hours of the morning when there's little traffic about.

That's a different story entirely - maintenance, or lack thereof, of equipment is nothing short of a scandal. I know several junctions that aren't operating as designed because of failure to maintain and they default to a fixed time state as a result. They weren't built as fixed time though, which is exactly mine and Richard's point.

The simple fact is some people on SABRE like to mouth off about how little they actually know about roads - hence why I say it's a SABRE myth. Fact checking is lost on these types because they think passing a test and driving on a road makes them qualified to design things. I use taps every day, it doesn't make me a plumber.

[WhiteBlueRed]

That's a case for more intelligently programmed signals, not a case against a signalised junction. There are plenty of junctions with flow control and vehicle actuation, but they're not universal unfortunately.

This subject came up a few months ago, in another thread. I was surprised then to read that the UK still uses traffic signals with fixed timings - that is, not vehicle-actuated. I thought that was a decades-old and now-disused technology. I can't imagine any modern installation providing efficient throughput without traffic-sensing loops.

I believe that all signal installations in Australia are fully programmable and vehicle-responsive. Any AU readers: am I wrong?

Here in Russia, most traffic lights still run on fixed timings. In a coordinated network, there may be multiple plans. However actuated control is becoming more common in the European part, locally known as "smart traffic lights".

[Bryn666]

That's a case for more intelligently programmed signals, not a case against a signalised junction. There are plenty of junctions with flow control and vehicle actuation, but they're not universal unfortunately.

This subject came up a few months ago, in another thread. I was surprised then to read that the UK still uses traffic signals with fixed timings - that is, not vehicle-actuated. I thought that was a decades-old and now-disused technology. I can't imagine any modern installation providing efficient throughput without traffic-sensing loops.

I believe that all signal installations in Australia are fully programmable and vehicle-responsive. Any AU readers: am I wrong?

Here in Russia, most traffic lights still run on fixed timings. In a coordinated network, there may be multiple plans. However actuated control is becoming more common in the European part, locally known as "smart traffic lights".

An emerging technology in the Netherlands is "flowtack" which is all about co-ordinating networks and optimising routes. It's quite technical so I won't pretend I can explain it but this probably represents a step up from technology deployed in the UK.

[WhiteBlueRed]

An emerging technology in the Netherlands is "flowtack" which is all about co-ordinating networks and optimising routes. It's quite technical so I won't pretend I can explain it but this probably represents a step up from technology deployed in the UK.

Sounds similar to SCOOT.

[JackieRoads]

Thoughfully, you have the eastern terminus of the M602 which is surprising because the motorway probably ploughes through it.

[Bryn666]

Thoughfully, you have the eastern terminus of the M602 which is surprising because the motorway probably ploughes through it.

It doesn't work very well, I use it regularly. However the presence of the A57 rules out a flat crossroads which would have fewer signal stages to deal with. The congestion is usually caused by the 3 into 2 merge beyond the signals rather than the signals themselves. Saturation flow theory has a lot to answer for.

[jackal]

Regents Road generally is just a big planning failure. You could argue that it's deliberate, to deter commuting, but when you've been trying deterrence for 40 years with an artificial bottleneck between two decent roads, and all you get is queues, maybe it's time to try something else.

Until now they had the option of putting in GSJs at either end and the A5066, closing/liloing everything else. But I see residential towers "with parking available" are going up right where the A5066 GSJ would be, which won't exactly help matters...

[M4Simon]

There definitely is some work to do on poorly timed (or indeed faulty) signalisations. From the research I've been doing to try and find some signage guidance for throughabouts, I stumbled upon a passage in LTN 1/09 which showed why part time lights for peaktime congestion have fallen out of favour, and it's quite justifiably pedestrian facilities. Without the lights being full time, it's difficult to fully and safely accommodate pedestrians. I don't think we have provision in the UK for signals that are part time but can also switch on off-time when a pedestrian is waiting to cross? That leads us to the need for 24/7 lights that should be programmed and constructed well enough to allow them to be essentially a non-factor during the times of day when congestion was never an issue. This is all within our grasp, as other users and engineers have said above, but it needs to be done properly by every authority and then properly maintained, which I can imagine is difficult at present given some cannot seem to even afford relining of roads which have turned into an expanse of blank tarmac (looking at you, South Glos).

The answer is quite simple - grade separate the NMU movements. However this is expensive and that isn't allowed because people (including some in this very thread) complain about wasting money on any things that aren't "brum brum drive fast".

Retrofitting grade separation for NMUs is a very difficult thing to get right because of the space required to build ramps, the severance ramps cause, the visual impact of the ramps, the personal safety issues caused by long structures with limited routes off in a threatening situation, and the sheer cost of building them.

I agree that if designed in from the outset, grade separation can work but it is not a good solution in many places. In the last few years, I've written more reports advising against footbridges, compared with promoting new pedestrian bridges and subways. And yes, I currently have a project for a new subway where we are making it spacious, and building it as a part of a wider public realm project.

Simon