|
I'm surprised at how open you seem to be about talking about things, Cich. Most companies don't let their employees speak so openly about corporate goings on and processes and that kind of thing. Especially working for the state, I would think that they would want to keep many things under wraps, but perhaps the national highways aren't such a secretive subject.
|
#
?
Nov 14, 2010 22:23
|
|
|
|
| # ? May 12, 2024 03:23 |
|
|
Vino posted:I'm surprised at how open you seem to be about talking about things, Cich. Most companies don't let their employees speak so openly about corporate goings on and processes and that kind of thing. Especially working for the state, I would think that they would want to keep many things under wraps, but perhaps the national highways aren't such a secretive subject. It's not like we have any competition, right? Besides, so long as I don't mention any names specifically, it's not like I'm defaming someone. I'm sure to any of my coworkers, it would be immediately obvious who I am, but they have nothing to gain by ratting me out. Hey, if all else fails, you can consider the thread a work of fiction. As it is, even the risk of getting in trouble is worth it to help teach others about the subject and give some insight into the art and science of the job. Who knows? This thread might save a life or two (and will certainly save some of you a few hours of commute time), and that's much more important than getting yelled at by my manager.
|
|
#
?
Nov 15, 2010 00:01
|
|
|
Cichlidae posted:the FHWA has banned strobes on signs edit: saw another Darien cop running radar on an I95 ramp, an entrance this time. These guys have nothing to do whatsoever. GWBBQ fucked around with this message at 16:58 on Nov 15, 2010 |
|
#
?
Nov 15, 2010 05:35
|
|
|
GWBBQ posted:So would I have ground to stand on if I asked the town to remove those "Your Speed" radar signs with the xenon strobe that flashes if you go over the limit,or the car behind you goes over, or the car in front of you, the one going the other direction, a branch flails in front of it in the wind? Even if they turned them off at night so they didn't blind drivers I'd be fine with it. On top of that, the 25 limit during the day is fine because it's a school zone, but the road would be safe at 35 at night, probably 40 during the day, but Fairfield's speed limits are all at least 10 under what they should be. Well, here's what the MUTCD has to say about it: MUTCD posted:Section 4L.04 Speed Limit Sign Beacon Of course, I can already tell you what the town will say. They'll say that the sign isn't actually regulatory, nor is it a traffic control device, and therefore it doesn't have to comply with the MUTCD. "Your Speed" beacons on state roads are not put up by the state; they're placed by the town via encroachment permit. What can you do? Call up the town engineer and tell him you almost had an accident because of the sign. Tell him you're epileptic, or one of your passengers, and nearly had a seizure. Tell him that, if someone gets hurt because of the sign, the town is liable.
|
|
#
?
Nov 15, 2010 17:37
|
|
|
 Click here for the full 814x725 image. Canning Bridge Station in Perth, Western Australia. Look at the merging on this thing! Also if you're going west on Canning highway and want to go north on the freeway, you're out of luck. How would you redesign this thing?
|
|
#
?
Nov 15, 2010 20:11
|
|
|
Spookydonut posted:
I'd recommend a SPUI. The complicated factor seems to be the rail(?) station there. If it's something that needs access to Canning Highway, a SPUI wouldn't work, because it would create a massive 5-way intersection. In that case, I'd recommend a diamond with a couple median ramps, similar to this one.
|
|
#
?
Nov 15, 2010 22:29
|
|
|
Cichlidae posted:I'd recommend a SPUI. The complicated factor seems to be the rail(?) station there. If it's something that needs access to Canning Highway, a SPUI wouldn't work, because it would create a massive 5-way intersection. In that case, I'd recommend a diamond with a couple median ramps, similar to this one. Besides the rail station (most trains skip it due to mediocre bus service and lack of park&ride), there's a bus lane on the freeway too. There's two merge zones due to how manning road links into it all and can't see a SPUI doing much about it. The river doesn't really help either. Edit: My attempt to fix it  Red is turning it into a diamond like you suggested, and getting rid of the merging zones due to manning road. Green is a parking lot/bus station/pick-up & drop-off. The existing station is in serious need of this, bus stops are on the bridge between the two sets of lights and cars are always stopping on the bridge to pick up/drop off passengers. Blue is a ped-bridge to the station. Spookydonut fucked around with this message at 13:36 on Nov 16, 2010 |
|
#
?
Nov 16, 2010 08:49
|
|
|
Spookydonut posted:Besides the rail station (most trains skip it due to mediocre bus service and lack of park&ride), there's a bus lane on the freeway too. There's two merge zones due to how manning road links into it all and can't see a SPUI doing much about it. The river doesn't really help either. That does look pretty good. Normally, I wouldn't worry too much about weave zones on collector-distributor roads like the one on the right side of the freeway there. Of course, a reduction in weaving anywhere is helpful. Re-arranging the ramps in a basket weave would also work: 
|
|
#
?
Nov 17, 2010 00:50
|
|
|
Today, I got to experience a traffic engineer's version of giving birth: turning on a new signal. You might remember my project in Killingly. Delayed for months by utilities, everything was finally in place when I arrived this morning to get things rolling. Figuratively speaking, that is, since this signal's purpose is making people stop. How many engineers does it take to turn on a light?  Ten is a pretty good start.  The previously nude signal cabinet gains an intersection number...  ...and an awful lot of expensive equipment, considering it has the simplest phasing possible.  If you build it, they will come...  And at this exact moment, the signal turns on for the first time, showing the first green indication of what will hopefully be millions.  The previously reigning traffic control device looks on jealously; his battered visage will soon be removed from the site.  Number 51-2009 in your sign catalog, but number one in any environmentalist's heart, the Sabin Brook sign looks over the rest of the site with an air of dignity.  This tiny sign nearby is not for traffic, but rather for maintenance, to show where the passing zone begins and ends.  This bridge is over 40 years old, and still in relatively good shape for its age. In fact, for bridges in Connecticut, this one is relatively young.  This sign is only a couple days old, but will probably tip over as soon as something decides to burrow there.  Although I designed it in 2008, this turning lane uses the cat tracks recommended in the 2009 MUTCD. That's right, gentlemen, I'm psychic. See that sign on the right? That's our flashiest sign, "NEW SIGNAL IN OPERATION".  And yes, an engineer's best friend is duct tape!  Finally, even in this cold weather, thing are still alive and growing. Somewhere in here, there's a metaphor about metamorphosis or possibly a snow joke.  I humbled myself before the other engineers to get this photo for you. I hope you appreciate it.
|
|
#
?
Nov 18, 2010 00:25
|
|
|
I think we have the same wedding ring 
|
|
#
?
Nov 18, 2010 00:51
|
|
|
IOwnCalculus posted:I think we have the same wedding ring Tungsten carbide. My wife's got a matching one with carbon fiber inlay. I was contemplating asphalt-inlaid stainless steel rings, but engineering is just my job, not my life.  Edit: VVVVVV It's ok, so long as only cool people have it. (It was about the only one I liked that would fit my freakishly small size 5 fingers) Cichlidae fucked around with this message at 03:35 on Nov 18, 2010 |
|
#
?
Nov 18, 2010 00:59
|
|
|
IOwnCalculus posted:I think we have the same wedding ring Oh man I have that one too! Now it doesn't seem cool if everyone has it 
|
|
#
?
Nov 18, 2010 02:31
|
|
|
Cichlidae posted:
 kudos.
|
|
#
?
Nov 18, 2010 06:31
|
|
|
What is your professional opinion on bad-rear end traffic lights like these? 
|
|
#
?
Nov 18, 2010 11:15
|
|
|
jeoh-kun posted:What is your professional opinion on bad-rear end traffic lights like these? Countdown timers only work on pre-timed signals. Once they're actuated, the red and green intervals have varying lengths which can't be predicted beforehand. As long as you don't mind going with a pre-timed signal, having a countdown timer on the red would reduce the start-up delay at the beginning of each phase. A countdown on the yellow interval might help people gauge whether they can stop or not before red. Buuuut! That second display with the hourglass would be tough to read from a distance, as well as going against the existing standards, so I doubt it would be implemented. Drawing too much attention to the light's details and away from the road could be dangerous, as well. And finally, due to the quicker start-up times, we'd probably want to extend the red interval by another second or so to provide some additional clearance for light-runners, which would offset the saved time.
|
|
#
?
Nov 18, 2010 13:30
|
|
|
Cichlidae posted:Countdown timers only work on pre-timed signals. Once they're actuated, the red and green intervals have varying lengths which can't be predicted beforehand. As long as you don't mind going with a pre-timed signal, having a countdown timer on the red would reduce the start-up delay at the beginning of each phase. Is there any reason why you can't leave it without a timer until the light is sure it will change? Surely you could set up a signal so that its next action is determined (and cannot be changed) 3-10 seconds before the light changes. I agree though that both of those concepts fail miserably from a usability standpoint though. Using yellow to indicate both "about to turn red" and "about to turn green" is a horrible idea. Maybe just do something simple like a one digit countdown timer below the other three lights. It stays off until the controller determines the time remaining definitively. Is there anyone considering switching to traffic signals with only one light, that has different colored LEDs? Or is the benefit of orientation and color considered great enough to warrant triple the hardware on every light?
|
|
#
?
Nov 18, 2010 19:35
|
|
|
I doubt they'd do that for safety reasons. My uncle is so colorblind he can only tell signals by the position of the light, and I'm sure he's not the only one.
|
|
#
?
Nov 18, 2010 19:50
|
|
|
kapinga posted:Is there any reason why you can't leave it without a timer until the light is sure it will change? Surely you could set up a signal so that its next action is determined (and cannot be changed) 3-10 seconds before the light changes. Great hilarity would ensue for colorblind people. Edit: beaten
|
|
#
?
Nov 18, 2010 19:52
|
|
|
Grand Fromage posted:I doubt they'd do that for safety reasons. My uncle is so colorblind he can only tell signals by the position of the light, and I'm sure he's not the only one. How does he drive at night, when you can't necessarily see the other lights? My understanding was the the hues are chosen so that a red light does not appear the same shade as a green light, even if you're red-green colorblind. My granddad was colorblind as well, but he did just fine driving at night or driving in Texas where all the lights were sideways.
|
|
#
?
Nov 18, 2010 20:17
|
|
|
kapinga posted:Is there any reason why you can't leave it without a timer until the light is sure it will change? Surely you could set up a signal so that its next action is determined (and cannot be changed) 3-10 seconds before the light changes. When a maxed-out phase is terminated by a call on another phase or is pre-empted, it skips immediately to yellow with zero warning. If people were used to seeing a countdown before the end of green, when it came suddenly, they would be prone to make mistakes. Less so for the end of red, when they're already stopped. Just remember that signal controllers are incredibly stupid. Their only "sense" is whether or not there is a car occupying a detector at that moment. A signal can't tell if there's one more car about to extend the signal just as it's about to change to yellow, nor can it tell that a car is approaching on the side street to call a phase that almost never gets called. quote:I agree though that both of those concepts fail miserably from a usability standpoint though. Using yellow to indicate both "about to turn red" and "about to turn green" is a horrible idea. Maybe just do something simple like a one digit countdown timer below the other three lights. It stays off until the controller determines the time remaining definitively. Many countries do use yellow to indicate the impending end of the red clearance, but they usually show red and yellow at the same time to make it obvious what's going on. quote:Is there anyone considering switching to traffic signals with only one light, that has different colored LEDs? Or is the benefit of orientation and color considered great enough to warrant triple the hardware on every light? In addition to what other people say about the colorblind, having three separate signal faces provides some redundancy in the case of malfunction. Different signal faces often have different louvers on them, as well, and I believe different LEDs fail at different rates. The main reason, of course, is that incandescent lights couldn't share a face, and there's so much momentum in something as fundamental as traffic signals that it's unlikely to change in the next fifty years, even if it's theoretically possible to put them all in one face and save quite a bit of money. And for you, kapinga, even horizontally-aligned signal heads have a standard positioning. It's only confusing when they're very rarely used, like they are up here in New England. I'm sure the colorblind learn very quickly. Edit: Here's the awful signal plan I spent all day reviewing. This is about a quarter of it; the rest is similarly filled with ridiculous crap. 
Cichlidae fucked around with this message at 23:24 on Nov 18, 2010 |
|
#
?
Nov 18, 2010 22:53
|
|
|
Speaking of horizontally-aligned signal heads, is there a reason that some areas choose them over the regular vertically-aligned signal heads? With everything as standardized as it is, I would assume that there would be one set standard. Calgary, Alberta for example, has all horizontally-aligned signal heads. They sure confuse you when you're not used seeing them!
|
|
#
?
Nov 18, 2010 23:32
|
|
|
Cichlidae posted:When a maxed-out phase is terminated by a call on another phase or is pre-empted, it skips immediately to yellow with zero warning. If people were used to seeing a countdown before the end of green, when it came suddenly, they would be prone to make mistakes. Less so for the end of red, when they're already stopped. Fair enough. I still think those drat pedestrians can wait for 3 more seconds while the cars get a countdown on green/red. (And I don't even own a car...). The biggest benefit for drivers would be the yellow+red warning on startup, but as you've said before American's are too stupid for that. Yellow+Red makes fine sense, but without the red light it would be impossible for a driver approaching an intersection to know if their yellow light meant they should gun it or wait three more seconds. And I do want to say, awesome thread. As an engineer in another field it's great to see and think about all the factors that go into traffic engineering.
|
|
#
?
Nov 18, 2010 23:39
|
|
|
Swanson posted:Speaking of horizontally-aligned signal heads, is there a reason that some areas choose them over the regular vertically-aligned signal heads? With everything as standardized as it is, I would assume that there would be one set standard. Calgary, Alberta for example, has all horizontally-aligned signal heads. They sure confuse you when you're not used seeing them! I think it's just local preference. Texas seems to prefer horizontal heads, too, as well as doubled-up red lights, which are against the new MUTCD. So much of this stuff was decided by confused engineers 70 years ago and taken as canon ever since. kapinga posted:Fair enough. I still think those drat pedestrians can wait for 3 more seconds while the cars get a countdown on green/red. (And I don't even own a car...). The biggest benefit for drivers would be the yellow+red warning on startup, but as you've said before American's are too stupid for that. It's that all-important "driver expectancy" again. Just like driving on a certain side of the road, there's more than one way to do things. Problems arise when you switch back and forth. kapinga posted:And I do want to say, awesome thread. As an engineer in another field it's great to see and think about all the factors that go into traffic engineering. Would you agree that traffic engineering is probably among the least mathematically-oriented of the engineering disciplines? It seems like a lot of what I do is more psychology than engineering, but it could just be that I'm so used to all the math that it's second nature.
|
|
#
?
Nov 19, 2010 00:01
|
|
|
Cichlidae posted:Would you agree that traffic engineering is probably among the least mathematically-oriented of the engineering disciplines? It seems like a lot of what I do is more psychology than engineering, but it could just be that I'm so used to all the math that it's second nature. It definitely seems to be less math intensive than chemical engineering, as least as a student. I'm doing grad school though, so I can't speak what life is like when we get real jobs. Calculus and differential equations are comforting at this point for me - equations are a lot more reliable than experiments are. But that's not to say you're not doing engineering. You're designing around people instead of chemicals or electrons, but fundamentally you're still trying to optimize a system by adjusting the parameters you have available. Edit: Cichlidae posted:It's that all-important "driver expectancy" again. Just like driving on a certain side of the road, there's more than one way to do things. Problems arise when you switch back and forth. kapinga fucked around with this message at 00:34 on Nov 19, 2010 |
|
#
?
Nov 19, 2010 00:29
|
|
|
kapinga posted:How does he drive at night, when you can't necessarily see the other lights? My understanding was the the hues are chosen so that a red light does not appear the same shade as a green light, even if you're red-green colorblind. He basically can't see colors at all, so it doesn't matter what hue they are. I don't honestly know how he deals with that situation, I imagine you'd get a feel for what the different positions look like if you have to pay attention to that instead of the color. I'm somewhat colorblind myself but not nearly to that extent, so I don't have to use any tricks. For instance, the red and yellow look like that to me, but green lights often look basically white. I don't know if they're a really light green to people with normal color vision, I assume so. It's not normally an issue since it's obviously different, but at a distance sometimes green lights and streetlights look like the same color so I can't really tell if there's a signal coming up at an unfamiliar intersection until I get close enough to see the actual object.
|
|
#
?
Nov 19, 2010 00:47
|
|
|
kapinga posted:Is this in reference to the simple confusion of a Red + Yellow light configuration (which is solvable), or more with respect to American drivers not safely using the new information. I thought I recalled a discussion in this thread about how American drivers would start running reds a few seconds early if we gave them information about when the light was about to change. I think it's not exclusive to Americans, just that we'd tend to deal with it worse since drivers tend to be less educated here. What we want is consistent reactions to a given stimulus. If everyone jumps the light as soon as it turns green, we bump up the red clearance and things flow smoothly. If everyone waits 2 seconds after turning green to step on the gas, then we can knock down the red clearance times, and things flow smoothly. If Granny McOld ignores the red-yellow interval (assuming she even knows what it means), she's going to wait those 2+ seconds to start up, while the guy behind her is already stepping on the gas. The extended red clearance is wasted on her, and the signal's level of service suffers as a result. So yes, some would start running the light early. It would happen anywhere in the world. That's ok, so long as everyone does it, and we plan for it. In America, people are used to waiting for the green (and we even design signals so they can't tell when that's coming). We don't train people enough to get consistent reactions out of them. That's the problem.
|
|
#
?
Nov 19, 2010 00:55
|
|
|
Cichlidae posted:
|
|
#
?
Nov 19, 2010 01:57
|
|
|
GWBBQ posted:So did you guys find the Higgs boson yet? No, but I get the feeling I'm halfway toward making a new genetic code consisting entirely of the base pairs G, Y, and R. When I marked up the plan, it was actually easier to just grab a blank plan and copy over the things that WERE correct, rather than trying to fix the things that weren't.
|
|
#
?
Nov 19, 2010 02:56
|
|
|
Whats your opinion of LED vs incandescent for signals? I've found LEDs to be much clearer and distinct, however at night in poorly lit intersections they are TOO bright.
|
|
#
?
Nov 19, 2010 08:11
|
|
|
Spookydonut posted:Whats your opinion of LED vs incandescent for signals? LEDs are the clear winner. They last much longer, use less power, and are easier to maintain since they probably won't all fail at once. Brightness is something that can be adjusted easily, though I don't think we'd really consider any signals to be TOO bright unless they were blinding people. Even then, in isolated areas, they probably aren't anywhere near as bright as an approaching car using high-beams.
|
|
#
?
Nov 19, 2010 13:28
|
|
|
Cichlidae posted:LEDs are the clear winner. They last much longer, use less power, and are easier to maintain since they probably won't all fail at once. Brightness is something that can be adjusted easily, though I don't think we'd really consider any signals to be TOO bright unless they were blinding people. Even then, in isolated areas, they probably aren't anywhere near as bright as an approaching car using high-beams. I'm loving the relatively new incandescent look LEDs - those are cool  I have a couple of signals in my room - all but one I've retrofitted with LEDs I found on eBay. The remaining signal is a 3M Model 131 programmed visibility light - definitely an eye-catcher. Horse Pepsi fucked around with this message at 14:18 on Nov 19, 2010 |
|
#
?
Nov 19, 2010 14:15
|
|
|
Speaking of LED we have new LED streetlight on some streets here and they are really great. You can see actual colours, the contrast seems much better as well. Cichlidae, do you know if LED streetlighting makes sense from a economic point of view? Also any safety effects because of the improved lighting conditions? I really want all HIDs to be replaced by LED tommorow.
|
|
#
?
Nov 19, 2010 20:06
|
|
|
Horse Pepsi posted:I'm loving the relatively new incandescent look LEDs - those are cool That must be great to keep indoors. The optically programmed signals we have tend to get knocked out of alignment after a while. I tried calling Maintenance to report it once, and they just said, "you're not supposed to be able to see it." We'll stick to using louvers from now on! NihilismNow posted:Speaking of LED we have new LED streetlight on some streets here and they are really great. You can see actual colours, the contrast seems much better as well. I'm sure they make sense to buy. the initial cost might be higher (or might not, I've not looked into it), but the bulb life is significantly higher and the energy cost is significantly lower than other lighting forms. The main disadvantage until recently was the difficulty producing a variety of colors, but science has helped solve that!
|
|
#
?
Nov 19, 2010 22:57
|
|
|
NihilismNow posted:Speaking of LED we have new LED streetlight on some streets here and they are really great. You can see actual colours, the contrast seems much better as well. LED lighting works in this application for a few reasons: It's long lasting. Very, very long lasting. The less times you need to service the install, the more you save, period. Labor is your number one cost. They use less electricity for a given volume of light. Ongoing cost savings outweighs the cost of initial install, both the lighting array and the power conversion/regulation. They are almost indestructible. All sorts of physical trauma, water, etc, just won't bother LED's. Conditions which would shatter (or explode) other bulb technologies won't even make LED's flicker. LED's also tends to be installed in clusters, so a manufacturing or minor wiring defect that shuts down a few of them doesn't stop the operation of the light (streetlight, stoplight, etc). You stop paying overtime, because the issue simply isn't critical. The actual light quality tends to be more blue white, which people perceive as brighter and more appealing, but the lights themselves don't need to be any brighter. The can be, it's up to the municipality to decide (cost vs life, etc). EoRaptor fucked around with this message at 23:01 on Nov 19, 2010 |
|
#
?
Nov 19, 2010 22:58
|
|
|
I bought an LED reflector base lamp for $15 to try it out figuring if it lasts 3x longer than CFLs and uses less energy, it's win-win! Except it burnt out in 6 months of occasional usage. Yeah, LEDs last virtually forever, but lovely electronics don't.
|
|
#
?
Nov 19, 2010 23:09
|
|
|
grover posted:I bought an LED reflector base lamp for $15 to try it out figuring if it lasts 3x longer than CFLs and uses less energy, it's win-win! Except it burnt out in 6 months of occasional usage. Yeah, LEDs last virtually forever, but lovely electronics don't. If we're paying a couple grand for a good signal head, we'd better hope that it's going to last more than a couple years! The industry has very few manufacturers, so reputation means a lot. The State may be constrained by low bid, but a contractor won't buy a signal head that's going to burn out during the warranty period.
|
|
#
?
Nov 19, 2010 23:14
|
|
|
Cichlidae posted:That must be great to keep indoors. The optically programmed signals we have tend to get knocked out of alignment after a while. I tried calling Maintenance to report it once, and they just said, "you're not supposed to be able to see it." We'll stick to using louvers from now on! No wonder they're pretty much discontinued - upkeep is a bitch and the only company still making them is McCain (yech). Although it's not too bad in my room - the only minus is the weight of the signal, it's twice that of a regular signal. Still a loving cool signal though, it just looks so different that pretty much everything else...
|
|
#
?
Nov 20, 2010 04:39
|
|
|
Cichlidae posted:LEDs are the clear winner. They last much longer, use less power, and are easier to maintain since they probably won't all fail at once. Brightness is something that can be adjusted easily, though I don't think we'd really consider any signals to be TOO bright unless they were blinding people. Even then, in isolated areas, they probably aren't anywhere near as bright as an approaching car using high-beams. Don't they suffer when ice cakes up on the lens since they don't produce enough heat to melt the ice?
|
|
#
?
Nov 20, 2010 04:47
|
|
|
ijustam posted:Don't they suffer when ice cakes up on the lens since they don't produce enough heat to melt the ice? I haven't seen that happen yet. They're protected from above by the visor hood, and backspatter from below is minimal considering they're 16 feet above the road. I suppose that in a driving ice storm, they could get coated, or if someone didn't align the slot in the bottom of a tunnel visor properly they could get snow-covered, so yes, that could be a possible drawback for maintenance.
|
|
#
?
Nov 20, 2010 04:53
|
|
|
|
| # ? May 12, 2024 03:23 |
|
|
Cichlidae posted:I haven't seen that happen yet. They're protected from above by the visor hood, and backspatter from below is minimal considering they're 16 feet above the road. I suppose that in a driving ice storm, they could get coated, or if someone didn't align the slot in the bottom of a tunnel visor properly they could get snow-covered, so yes, that could be a possible drawback for maintenance. We've had LED array lights in Massachusetts for a long time now, and I can never remember one getting iced or snowed over ever. The design of the light louver makes it an impossibility.
|
|
#
?
Nov 20, 2010 12:53
|
|
