|
smackfu posted:OK, can you explain the I-95 North to Rt 8 ramp in Bridgeport? It's a one lane exit that goes into a tight curve and comes out as two lanes on the overpass. It's like they decided, this bridge is wide enough to fit two lanes so we better have two lanes even if it makes no sense. I agree, it would be a bit more practical to have the second lane added at the on-ramp from 95 south. The pavement widens about 1/3 of the way through the loop, but loops are only supposed to carry one lane, since cars going around a tight curve at that speed have trouble staying in their lane. I'm sure the volumes are pretty high, but they're coming from one lane to begin with, so...
|
#
?
Aug 28, 2010 02:54
|
|
|
|
| # ? Jun 8, 2024 11:05 |
|
|
I was doodling a bit to see if I could optimise the design for the new Badhoevedorp Interchange near Amsterdam but apparantly it couldn't work out. Something about needing multiple lanes on the connectors withing a reasonable ROW, even though it will lead to lovely merging situations. Here's the official design and my potential solution.
|
|
#
?
Aug 28, 2010 12:32
|
|
|
Koesj posted:I was doodling a bit to see if I could optimise the design for the new Badhoevedorp Interchange near Amsterdam but apparantly it couldn't work out. Something about needing multiple lanes on the connectors withing a reasonable ROW, even though it will lead to lovely merging situations. Here's the official design and my potential solution. Wow, I have to say, I really love the overall design of the bypass and the interchanges. Plenty of provisions for future growth, but without providing excessive capacity in the present. Having the opportunity to design something like this would keep me happy for years. I'll never get that chance here, because the DOT doesn't have the money or right-of-way to do anything like this. So, for your design: I should mention first that the weaving area you identified to the left of your picture 1 is likely to be minimal, because almost nobody will want to change lanes there, unless they took the wrong exit. Heck, I'd be ok with making that two separate ramps, as you did. The other weaving area, as you noted, will be tougher, as is shoving two lanes of traffic around a loop ramp. In your design, though, you keep the loop ramp, and at one lane. One lane of cars going 70 kph doesn't handle nearly as much volume as when they're going 110. The loop is really there to provide adequate weaving distance on the mainline. Since you got rid of the weaving area, why not get rid of the loop entirely? Do a flyover, up and around the mainline and the other ramps. Keep the capacity great with a single lane, and then have them merge in with the rest as in your design, except I'd keep two lanes on the ramp and then merge them into the mainline over the course of a kilometer or two. Heck, you've got that next exit down the line, may as well make an auxiliary lane. Simple.
|
|
#
?
Aug 28, 2010 16:32
|
|
|
I went out to get myself a bagel earlier and used my cell phone to grab a pic of the bike path I had never noticed before. I really oughta pay more attention to my surroundings. 
|
|
#
?
Aug 28, 2010 20:42
|
|
|
BklynBruzer posted:I went out to get myself a bagel earlier and used my cell phone to grab a pic of the bike path I had never noticed before. I really oughta pay more attention to my surroundings. It's amazing how easy it is to miss things when you're so used to the status quo. A stop sign was installed near my childhood home about 5 years ago, and I run through it every single time because I lived there for 20 years with no sign. It's almost invisible to me.
|
|
#
?
Aug 29, 2010 05:04
|
|
|
Cichlidae posted:I should mention first that the weaving area you identified to the left of your picture 1 is likely to be minimal, because almost nobody will want to change lanes there, unless they took the wrong exit. Heck, I'd be ok with making that two separate ramps, as you did. It's needed to provide downstram access to the airport though, it seems weaving will actually be at its worst along the north/south local connectors from Amsterdam towards Schiphol since so many people work or travel from there. quote:The other weaving area, as you noted, will be tougher, as is shoving two lanes of traffic around a loop ramp. In your design, though, you keep the loop ramp, and at one lane. One lane of cars going 70 kph doesn't handle nearly as much volume as when they're going 110. I guess so, the loop starts out with only one lane in the original design though so I gathered it wouldn't be much of a problem, same goes for the north/west ramp. quote:The loop is really there to provide adequate weaving distance on the mainline. Since you got rid of the weaving area, why not get rid of the loop entirely? Do a flyover, up and around the mainline and the other ramps. Keep the capacity great with a single lane, and then have them merge in with the rest as in your design, except I'd keep two lanes on the ramp and then merge them into the mainline over the course of a kilometer or two. Heck, you've got that next exit down the line, may as well make an auxiliary lane. Simple. The designers had to go back to the drawing boards since a fly-over was considered to be too costly and alternate routing via local roads for the Schiphol/Haarlem-connection was deemed unsatisfactory (no poo poo): http://www.rijkswaterstaat.nl/images/Presentatie%20Informatieavond%20december%202009_tcm174-277621.pdf (pages 5 and 6).
|
|
#
?
Aug 29, 2010 13:12
|
|
|
Koesj posted:It's needed to provide downstram access to the airport though, it seems weaving will actually be at its worst along the north/south local connectors from Amsterdam towards Schiphol since so many people work or travel from there. Ah, I didn't see the airport exit. It wouldn't be too hard to just make a second exit for the airport, which would eliminate that weave, but it's likely not worth the extra cost. I suppose they came to the same conclusion about a flyover. Personally, I'd rather do it right the first time than deal with an accident problem later on, but unfortunately, engineering is constrained by budgets.
|
|
#
?
Aug 29, 2010 15:31
|
|
|
I've only read the first page of this giant thread and then the last few pages, so I apologize if this has been asked already, but: Why in the hell is the USA so anti-roundabout? I live in the northwest Chicago suburbs with have a whole mess of hellish 6-way intersections that pretty much always have epic backups approaching them, and I can't help but think that a roundabout would be a FAR more efficient and cheaper approach. ESPECIALLY when traffic lightens up at night. I can't tell you how infuriated I get sitting at some of these intersections in the middle of the night waiting all of 90 AGONIZING SECONDS for the lights to switch (and many times I'll just say hell with it and run them). Another question: How do places determine if a left-turn lane is signaled-only, in that the drivers may turn left only with a green arrow, or if when corresponding lanes both have green, the drivers may turn at their discretion?
|
|
#
?
Aug 29, 2010 22:41
|
|
|
Tsaven Nava posted:I've only read the first page of this giant thread and then the last few pages, so I apologize if this has been asked already, but: That's changing, slowly, but we have a century of hate for circular intersections to undo. The short story is that the traffic circles built in the 20th century in the US are inconsistent, designed generally for high speeds and, often, for traffic in the circle to yield to entering traffic. These are horrible characteristics for a traffic circle, because they lead to high accident rates and gridlock. That sort of thing has poisoned the average American's mind. The roundabouts being built now are very different from the old style. They were popularized by France in the 1970s, and they feature much slower speeds (leading to less severe accidents and making them pedestrian-friendly) and higher capacity than a stop-controlled intersection. The last 10 years have seen these modern roundabouts built all across the country, and most traffic engineers have warmed up to them. The public at large is slower to accept new things, but in cities where modern roundabouts have been built, the reaction is overwhelmingly positive. So, give it another 10 years, and I'm sure you'll see a lot more. quote:Another question: How do places determine if a left-turn lane is signaled-only, in that the drivers may turn left only with a green arrow, or if when corresponding lanes both have green, the drivers may turn at their discretion? There's no standard, but I posted a flowchart many pages (~20?) back that gives a general idea of when protected-only (green arrow only) or protected-permissive (both green arrow and green ball) are used. The basic criteria depend on both the turning and oncoming volumes, the number of lanes the cars have to cross, and the accident history at the intersection. Edit: Here it is: 
Cichlidae fucked around with this message at 23:13 on Aug 29, 2010 |
|
#
?
Aug 29, 2010 22:59
|
|
|
Oh hey, that reminds me - Cichlidae, can you talk a bit about places that do/don't do right turn on red? It's always confused me.
|
|
#
?
Aug 29, 2010 23:04
|
|
|
BklynBruzer posted:Oh hey, that reminds me - Cichlidae, can you talk a bit about places that do/don't do right turn on red? It's always confused me. It's a confusing subject. Every engineer seems to have a different policy, and so you'll find that whether or not an NTOR is posted depends on the engineer who was in charge when it was designed. Here are the things that will almost always result in NTOR: - Bad sight distance. If you can't see if someone's coming, you shouldn't enter the intersection. - An accident problem. A few people get run over by right-turners, and we'll put up that sign. - Strange geometry. If it's an offset intersection, like two T-intersections controlled by the same signal, "turning" right could involve significantly disrupting through traffic on another leg. Here are some things that sometimes cause NTOR: - The signal is in a school zone. This used to be ConnDOT policy; any signal near a school would automatically be NTOR. - Intersections at an acute angle. Complicates sight distance as well as the turning maneuver itself. - Heavy pedestrian volumes. Right turn on red poses a pretty significant hazard to pedestrians who think the "WALK" signal means they don't have to look both ways. And some things that rarely cause NTOR: - To discourage cut-through traffic. Making the cycle length longer and posting NTOR can discourage people from cutting through neighborhoods. - Heavy volumes. If there are almost no gaps in traffic, it would be hard to turn anyway.
|
|
#
?
Aug 29, 2010 23:35
|
|
|
All 50 states allow RTOR, but aren't there a number of US/Canadian cities where it's illegal? I travel to a lot of countries and always worry about either getting pulled over for an illegal turn, or rear-ended for not turning if I was supposed to...
|
|
#
?
Aug 29, 2010 23:41
|
|
|
grover posted:All 50 states allow RTOR, but aren't there a number of US/Canadian cities where it's illegal? Until a few years ago, in China, it was legal to go straight through a red signal at a T-intersection, if the side street was on your left. It's topologically equivalent to RTOR, but it caused a lot of accidents and was thus summarily forbidden.
|
|
#
?
Aug 29, 2010 23:49
|
|
|
We call this Kamikaze Curve locally, but going through this thread, it's downright tame comparatively: It's an older picture, but the basic structure is still there, I-81 splitting to go either into the City of Binghamton, or head out toward the local shopping mall. The mall road links in with Route 17, coming off the bridges, which is supposed to become I-86 at some point. That route is the inside highway, and the bottom bridge. The outside highway, and top bridge is traffic going in the opposite direction. And the whole thing is starting to sweep around the hill, as you can see at the bottom corner. Not shown is the fun that awaits just over the bridge, where if you are in the wrong lane getting off 81, you have to pull a quick dodge through 2 lanes of merging traffic on the bridge to catch the exit for Binghamton.
|
|
#
?
Aug 30, 2010 00:38
|
|
|
Kwik posted:We call this Kamikaze Curve locally, but going through this thread, it's downright tame comparatively: That's a Directional Y (also called Wye, because engineers can't spell) interchange, and they were very common before engineers realized that left exits/entrances were bad. We've got dozens up here in the northeast, and it'll be decades before we finally get rid of them all.
|
|
#
?
Aug 30, 2010 00:49
|
|
|
Cichlidae posted:So, give it another 10 years, and I'm sure you'll see a lot more. Well that's nice to know. On another topic, do you have any statistics or evidence for how much traffic congestion can be reduced by motorcycles lane-splitting? Or generally, any sort of information in support of lane-splitting so that I have even more evidence that it should be legal everywhere, and not just in California? If only to assist me when I start ranting all frothy-mouthed about it at people who don't care much.
|
|
#
?
Aug 30, 2010 01:23
|
|
|
Tsaven Nava posted:Well that's nice to know. I couldn't find anything about it, positive or negative, on the FHWA website. I'm sure someone's done studies. For a capacity increase, it'd be trivial to calculate if you knew the percentage of motorcycles in a vehicle stream. Just assume a speed, and suddenly you have an X vph increase in capacity during congestion.
|
|
#
?
Aug 30, 2010 21:41
|
|
|
Cichlidae posted:I agree, it would be a bit more practical to have the second lane added at the on-ramp from 95 south. The pavement widens about 1/3 of the way through the loop, but loops are only supposed to carry one lane, since cars going around a tight curve at that speed have trouble staying in their lane. I'm sure the volumes are pretty high, but they're coming from one lane to begin with, so... Is there any acceptable way to mark a one to two lane width change that doesn't just wait until it's two lanes wide?
|
|
#
?
Aug 31, 2010 00:24
|
|
|
smackfu posted:Yeah. It's kind of weird because some people know it splits into two lanes and try to hug the sides so they end up in the lane they want at the end of the curve, while other people just go right down the middle and then end up splitting the lanes at the end. Mix the two and it's a bit chaotic. My favorite approach is to make the shoulder lines lead into one lane, and then taper out to two in a short distance (50-100 feet). People who aren't paying attention will be shunted into one lane, those who know the area will take the second.
|
|
#
?
Aug 31, 2010 01:33
|
|
|
And now, the long-awaited answer to the Bridge Door Mystery! So, why does this bridge on Route 11 have doors in its abutments?   Answer: they're not abutments! They're piers. Since the bridges are so long (over 200 feet each), a single unbroken span would have some pretty massive stresses. To help compensate, the bridge is actually cantilevered over each "pier," with the section of the span beyond the pier helping to balance the main span. The pier is walled in, which is unusual, making it look more like an abutment. And the doors are for maintenance access (and hobo storage).
|
|
#
?
Aug 31, 2010 21:12
|
|
|
Hey, thanks for looking into that! I'll have to check it out more closely next time I go by.
|
|
#
?
Aug 31, 2010 22:43
|
|
|
OK, now do these My money is on hobo storage.
|
|
#
?
Aug 31, 2010 23:34
|
|
|
GWBBQ posted:OK, now do these Oh God, not another one! Looks like pretty normal maintenance access to me. Of course, a door is kind of pointless if it's open all the time.
|
|
#
?
Sep 1, 2010 00:03
|
|
|
Cichlidae, Back East do you guys ever use cameras at intersections as opposed to the in ground car detectors? It is something that we have at here in California and it can be irritating at times. Why you ask: fog. Living close to the ocean there are many mornings where fog comes pretty far inland. There is one of these cameras on my way to work that will flip out when it is foggy and will start going through cycles even when there are no cars on the approaching intersection. Of course, this at an intersection that is rarely used (Old Bolsa Chica Rd here: http://maps.google.com/maps?f=q&sou...5,0.009989&z=17) and the light will sit red on Bolsa Chica for a good two minutes until the light cycles and the southbound traffic backs up over the freeway. Very annoying!
|
|
#
?
Sep 3, 2010 16:18
|
|
|
Wiggly posted:Cichlidae, We do use video detectors here. In fact, several towns use them exclusively over loop detectors. Technology continuously improves for video detection, but it's still rather primitive. When I worked in the TMC, we had an experimental camera set up under a big interchange in downtown Providence by the DHS to watch for terrorists. We also tried setting up video detection on the freeways to search for stopped cars. The success rate was around 20%. Most of the false positives were caused by rain or snow, especially by headlight reflections in puddles. I can't say I've seen any caused by fog, but that's just a result of the algorithm; this one looked for a block of pixels that slowed down and stopped within a certain area. Video detection cameras at intersections are typically purpose-built, and have a much simpler job. Of course, as you saw, they are rather unreliable in bad weather conditions. Radar detectors and microwave detectors seem to be somewhat more reliable and easier to configure, but they still have their drawbacks. For one, they have to be mounted very high so that one truck won't block visibility to the other lanes. Weather doesn't affect them at all, so perhaps in your situation, that would be the best solution.
|
|
#
?
Sep 3, 2010 17:22
|
|
|
Any cameras at intersections in CT are for traffic monitoring and not red light enforcement, right?
|
|
#
?
Sep 3, 2010 18:26
|
|
|
GWBBQ posted:Any cameras at intersections in CT are for traffic monitoring and not red light enforcement, right? At state intersections, sure. I don't think it was ever voted into law to allow them. The issue was in the news last year; I read a transcript of the general assembly hearing, and there were good arguments on both sides of the debate. The general argument seems to be safety vs. privacy.
|
|
#
?
Sep 3, 2010 22:06
|
|
|
Red light cams do not make things safer; in fact, accident rates increase due to people trying to beat the yellow (and due to shortened yellow times by greedy municipalities) and/or increased rear end accident rates due to people slamming brakes to not get tickets. The only argument for those cameras is revenue, and it's a slimy one. It's the same reason I'm opposed to speed cameras.
|
|
#
?
Sep 4, 2010 04:56
|
|
|
Wiggly posted:Cichlidae, Why not use something like radar range detectors? Point it at where the cars would stop and set it to detect when the range shortens so it knows a car is there?
|
|
#
?
Sep 4, 2010 17:11
|
|
|
I make it a point to always drift through redlights sideways when I run them. It confuses the cameras, makes them think I'm part of the other traffic that just got the green light! Right? 
|
|
#
?
Sep 4, 2010 18:02
|
|
|
grover posted:I make it a point to always drift through redlights sideways when I run them. It confuses the cameras, makes them think I'm part of the other traffic that just got the green light! If I saw that at a red light, I wouldn't even be mad.
|
|
#
?
Sep 4, 2010 18:38
|
|
|
kefkafloyd posted:Red light cams do not make things safer; in fact, accident rates increase due to people trying to beat the yellow (and due to shortened yellow times by greedy municipalities) and/or increased rear end accident rates due to people slamming brakes to not get tickets. Do you have some scientific evidence for this? Even if rear-ends increase, the number of angle accidents could be reduced, and those are much more severe. I'm interested to see if some research has been done, so I'll have something to fire back at my boss the next time he goes on his "let's put speed and red-light cameras on every road" tirades. grover posted:I make it a point to always drift through redlights sideways when I run them. It confuses the cameras, makes them think I'm part of the other traffic that just got the green light! At the very least, you'd give the guy who reviews the photos something to hang on his cubicle wall!
|
|
#
?
Sep 4, 2010 20:54
|
|
|
Where did this come from? On Sepulveda Boulevard in LA going under I-105, there's this orange "No U Turn" sign. It must have been custom-made because it's clearly non-standard, but why would they do that? Was it a construction sign left up for some reason? (There are a couple of correct, standard signs next to the far signal head in addition to this one.)
|
|
#
?
Sep 4, 2010 20:56
|
|
|
Cichlidae posted:Do you have some scientific evidence for this? Even if rear-ends increase, the number of angle accidents could be reduced, and those are much more severe. I'm interested to see if some research has been done, so I'll have something to fire back at my boss the next time he goes on his "let's put speed and red-light cameras on every road" tirades. If we do start getting speed cameras here, I sincerely hope there will be groups of angry drivers destroying them like there are in the UK.
|
|
#
?
Sep 5, 2010 00:15
|
|
|
The recommendation from numerous studies is that red light cameras cause more accidents and more severe accidents than they reduce, and that the actual real solution in problem intersections is to extend the yellow period and pause between giving the other lanes the green light. http://blog.motorists.org/red-light-cameras-increase-accidents-5-studies-that-prove-it/ Ontario Study posted:“Exhibit 2 indicates the red light running treatments have: VA Study posted:“After cameras were installed, rear-end crashes increased for the entire six-jurisdiction study area… After controlling for time and traffic volume at each intersection, rear-end crash rates increased by an average of 27% for the entire study area.” NC Study posted:“Using a large data set, including 26 months before the introduction of RLCs, we analyze reported accidents occurring near 303 intersections over a 57-month period, for a total of 17,271 observations. Employing maximum likelihood estimation of Poisson regression models, we find that:
|
|
#
?
Sep 5, 2010 00:53
|
|
|
Choadmaster posted:Where did this come from? On Sepulveda Boulevard in LA going under I-105, there's this orange "No U Turn" sign. It must have been custom-made because it's clearly non-standard, but why would they do that? Was it a construction sign left up for some reason? (There are a couple of correct, standard signs next to the far signal head in addition to this one.) Construction signs get forgotten all the time. They usually belong to the Contractor, so if he forgets to take them down at the end of the project, they'll be there until they fall apart or someone complains. Even the state signs don't always get taken down. I see orphaned Road Work Ahead, Fines Doubled signs all over the place, as well as their End Road Work counterparts. grover posted:The recommendation was that red light cameras cause more accidents and more severe accidents than they reduce, and that the actual real solution in problem intersections is to extend the yellow period and pause between giving the other lanes the green light. Thanks, that's some good evidence, and I'll whip it out the next time someone tells me that I should be putting up red light cameras. Extending yellow periods, though, isn't the proper solution. British research showed that using a uniform yellow interval of 3 seconds is safest, since most drivers will either try to stop if they think they have three seconds before red, or run the light if they don't, regardless of how long the yellow is. Extending the all-red clearance does improve safety, though, since it'll give those light-runners time to get across the intersection before opposing movements can go.
|
|
#
?
Sep 5, 2010 01:07
|
|
|
http://www.bbc.co.uk/news/world-us-canada-11203656 posted:Barack Obama announces $50bn infrastructure plan Link This must have put a smile on your face, Cichlidae? Maybe some of those unfunded projects have a chance after all; or at least maybe there can be slightly fewer condemnable bridges out there just waiting for a chance to collapse.
|
|
#
?
Sep 7, 2010 00:18
|
|
|
mike_F_ posted:Link If it manages to get past Congress, that'll be a small step in the right direction. ASCE puts our current need at $2.2 trillion, and that's only going to get more expensive over the next few years. $50 billion isn't even enough to maintain the current network; not even close. But hey, it could mean I don't get laid off next Summer. That's something to look forward to.
|
|
#
?
Sep 7, 2010 00:34
|
|
|
mike_F_ posted:Link The biggest issue with these types of stimulus plans is that they are patch jobs. They don't drive demand for anything but direct labour jobs, and fail to create the need for higher education and management jobs that help expand the middle class. This isn't a dig against Cichlidae or his coworkers, but this type of spending doesn't create need for any more of them, so it doesn't drive growth in higher education and better prospects. This type of infrastructure spending should be ongoing, not this hodge podge of 'fix the worst of it' poo poo that is coming out.
|
|
#
?
Sep 7, 2010 17:59
|
|
|
|
| # ? Jun 8, 2024 11:05 |
|
|
EoRaptor posted:The biggest issue with these types of stimulus plans is that they are patch jobs. They don't drive demand for anything but direct labour jobs, and fail to create the need for higher education and management jobs that help expand the middle class. This isn't a dig against Cichlidae or his coworkers, but this type of spending doesn't create need for any more of them, so it doesn't drive growth in higher education and better prospects. It would've been great if, sixty years ago, someone had set aside a trust fund for road maintenance and renewal. That way, we could spend a steady amount each year keeping things in good shape, save money overall, and allow the economy to grow. Road construction by itself doesn't do much for the middle class, but without the fast and easy transport of goods, other industries stagnate. Kevin Lynch compared the growth of Boston and New York in the 19th centuries. They were on even economic footing until Boston's poor planning effectively isolated its port and inner city. New York continued to grow, while Boston became comparatively weaker. Of course, Lynch also mentions that stagnant - or even shrinking - cities can still be healthy, but no president would shoot for stagnation as a goal.
|
|
#
?
Sep 7, 2010 21:57
|
|