Split phase signal phasing.....let me say that I loathe these kind of phases but I understand the premise behind them (in most cases). The idea is that more people want to turn left than proceed straight.
For those here who don't know what I'm talking about, split phasing generally works like this, picture a 4 way intersection:
NB-SB get a green light together (if the phasing has protected movements doesn't matter, but as long as the straight is timed together), NB-SB cycles back to red
EB gets a green light plus a green arrow, which then cycles to yellow light/arrow, and back to red.
WB then gets a green light/arrow and the same as above for EB
Now for the obvious observations, as the intersection now has three distinct phases, this means that wait times are of course going to be longer, I have noticed that split phases generally are installed in a situation where a former T-intersection gets extended in the missing direction.....a great example of this is the now infamous Spadina/Bremner/Fort York intersection in Downtown Toronto, I'll get to this specific example later on.
I'll use three distinct examples of split phasing and the problems they cause:
#1 Bovaird Drive and Mountainash Road, Brampton, ON:
https://goo.gl/maps/AywtDGrVYT62 (https://goo.gl/maps/AywtDGrVYT62)
During sometime in the early 20-aughts, a new subdivision south of Bovaird was created. Indeed, prior to the creation of this subdivision, Bovaird and Mountainash met at a T intersection. As I recall, SB Mountainash at the intersection with Bovaird had three lanes, 2 left turn lanes and 1 right turn lane. It was controlled solely with a green ball (requiring a yield to pedestrians).
When the southern road came to be, now known as Sunnyvale, the configuration of lanes on Mountainash had to be tweaked slightly, the middle lane which was solely a left turn lane before, is now an option lane, where you can proceed left or go straight. The right turn lane is an option lane as well.
Also when this intersection was tweaked, the pedestrian crossing connecting the NE and SE corners of this intersection had to be removed to keep traffic flowing at a reasonable pace. Obviously, there would be much less demand for traffic heading NB on Sunnyvale than traffic heading SB on Mountainash, look there's a fairly large plaza here. As a pedestrian could only cross safely from NE and SE on the NB Sunnyvale split phase, creating a situation where NB Sunnyvale has a phase lasting 40 seconds or more would be unreasonable.
Especially nowadays when we want to make communities more pedestrian friendly, indeed Brampton has introduced an express bus system on Bovaird known as Zum, a pedestrian who needs to make that movement from NE to SE has to cross the intersection 3 times, or disregard the restriction and either cross on Mountainash's split phase (creating a turning conflict with the pedestrian), or crossing on Sunnyvale's split phase (which is not going to be long enough to cross the intersection safely) which is not going to be safe to do.
In this case, I'm not sure if there's a real solution that can be done here, it may be possible to turn Mountainash Road's middle option lane back into a left turn lane and have a protected only sequence, while Sunnyvale gets a permissive-protected situation, and thus eliminating the split phase, but then comes the issue of a potential extended pedestrian phase on NB-SB, as the biggest movement at this intersection is the turns from Mountainash.
#2 Airport Road and Derry Road, Mississauga, ON
https://goo.gl/maps/aE4erf8XJHJ2 (https://goo.gl/maps/aE4erf8XJHJ2)
This intersection is absolutely notorious during PM Rush, the backups caused by the intersection heading NB on Airport extend as far south as American Drive, and it can take upwards of 30-40 minutes easily to get through this intersection during PM Rush if one is as far south as American Drive.
In this case, NB and SB Airport Road get split phases, while EB-WB Derry gets a regular phase. Also complicating matters are the intersections of Airport/Hull, and Derry/Hull, which SORT OF create a quadrant roadway intersection, but it doesn't matter as all movements the Airport/Derry intersection are allowed. Indeed, the intersections with Hull are synchronized with what goes on at the Airport/Derry intersection (SB Airport faces an unusually long red light at Hull, as it only turns green roughly 10 seconds before the SB Airport split phase turns green).
My solution, DIG BABY DIG, grade separate Airport and Derry, and create a grade separated "quadrant roadway intersection", which there are many in the GTA already, like the one at Dundas/Cawthra, or Hwy 7/Yonge. In this case the quadrant roadway will be Hull Street. This is a simple fix, but will the powers that be want to do it, probably not.
#3 Spadina/Bremner/Fort York, Toronto, ON
https://goo.gl/maps/xA57AkPBBay (https://goo.gl/maps/xA57AkPBBay)
Originally this was just a T-intersection with Bremner/Spadina. What happens on Spadina, especially during PM Rush, traffic heading SB on Spadina has a huge disadvantage with two intersections that involve dual left turns coming WB onto SB Spadina, and right turns coming EB onto SB Spadina. Spadina is the last chance to get on the Gardiner WB before the Lake Shore onramp at the former Etobicoke border, the Jameson onramp is closed during rush hours.
What happens at rush hour is a nasty loop that feeds traffic onto SB Spadina, thus traffic that wants to proceed straight on Spadina, will have to wait upwards of 6 or 7 cycles to proceed through the intersection as the backups are so tight by cross street turning traffic that it's usually lucky for one or two cars per lane to get through on a green. Combine this with the awfully long split phase at Bremner/Fort York/Spadina, and you got a nasty recipe for disaster, a simple 2-3 minute trek down Spadina takes 20-30 minutes to travel a few 100 feet.
Combine this again with split phasing not being entirely friendly to pedestrians, as pedestrians will often bolt into an intersection as long as cross traffic is stopped. This creates turning conflicts if the pedestrians do not obey their signals. And like my earlier example, crossing from SE to SW has been eliminated as it would only make the sequence longer. This is a curious decision because in a downtown area like this, there are often during rush hour, more pedestrians walking around here than cars on the road. Maybe a reconfiguration of this intersection could help matters, maybe giving Bremner's left a protected only left may work. But that's just me thinking out loud.
So with a few examples, I have pointed out some very negative situations that split phases cause. Let's be honest, is there real any positives to this kind of phasing?
At major intersections split phasing is annoying, but I've seen it used when said intersection is accident-prone.
Split phasing is a godsend at intersections where a shopping center driveway meets the major street running next to it.
Quote from: MisterSG1 on July 18, 2016, 01:52:41 AM
#1 Bovaird Drive and Mountainash Road, Brampton, ON:
https://goo.gl/maps/AywtDGrVYT62 (https://goo.gl/maps/AywtDGrVYT62)
During sometime in the early 20-aughts, a new subdivision south of Bovaird was created. Indeed, prior to the creation of this subdivision, Bovaird and Mountainash met at a T intersection. As I recall, SB Mountainash at the intersection with Bovaird had three lanes, 2 left turn lanes and 1 right turn lane. It was controlled solely with a green ball (requiring a yield to pedestrians).
When the southern road came to be, now known as Sunnyvale, the configuration of lanes on Mountainash had to be tweaked slightly, the middle lane which was solely a left turn lane before, is now an option lane, where you can proceed left or go straight. The right turn lane is an option lane as well.
Also when this intersection was tweaked, the pedestrian crossing connecting the NE and SE corners of this intersection had to be removed to keep traffic flowing at a reasonable pace. Obviously, there would be much less demand for traffic heading NB on Sunnyvale than traffic heading SB on Mountainash, look there's a fairly large plaza here. As a pedestrian could only cross safely from NE and SE on the NB Sunnyvale split phase, creating a situation where NB Sunnyvale has a phase lasting 40 seconds or more would be unreasonable.
Especially nowadays when we want to make communities more pedestrian friendly, indeed Brampton has introduced an express bus system on Bovaird known as Zum, a pedestrian who needs to make that movement from NE to SE has to cross the intersection 3 times, or disregard the restriction and either cross on Mountainash's split phase (creating a turning conflict with the pedestrian), or crossing on Sunnyvale's split phase (which is not going to be long enough to cross the intersection safely) which is not going to be safe to do.
In this case, I'm not sure if there's a real solution that can be done here, it may be possible to turn Mountainash Road's middle option lane back into a left turn lane and have a protected only sequence, while Sunnyvale gets a permissive-protected situation, and thus eliminating the split phase, but then comes the issue of a potential extended pedestrian phase on NB-SB, as the biggest movement at this intersection is the turns from Mountainash.
The Fortinos plaza! This brings back memories of my childhood in Brampton :)
TBH, I'm not sure if I understand the wording in your last paragraph I quoted, so I apologize if my solution is the same as yours :-D
My solution would be to have both Mountainash and Sunnyvale have advanced greens at the same time (i.e. red bulb and green arrow). Then, have both Mountainash and Sunnyvale have regular green lights. If a pedestrian presses the crosswalk button, then this green light phase will run longer, and if there are no pedestrians, this phase would be short due to the small amount of through-traffic.
There is an obvious trade-off with my idea though. It would require left-turning vehicles to stop when the phase switches from left arrow to regular bulb,
but it is a more pedestrian-friendly idea.
I agree that this phasing doesn't seem like the best solution. I've seen a few of these in KW (I'll have to try and remember where) and they often feel unnecessary. The fact that they're not pedestrian friendly is something I haven't thought about, but this is also a good point.
EDIT: The intersection of King St N and Farmers Market Rd just north of Waterloo has split phasing along King St
Quote from: MisterSG1 on July 18, 2016, 01:52:41 AM
Split phase signal phasing.....let me say that I loathe these kind of phases but I understand the premise behind them (in most cases). The idea is that more people want to turn left than proceed straight.
...
Now for the obvious observations, as the intersection now has three distinct phases, this means that wait times are of course going to be longer, I have noticed that split phases generally are installed in a situation where a former T-intersection gets extended in the missing direction.....
...
So with a few examples, I have pointed out some very negative situations that split phases cause. Let's be honest, is there real any positives to this kind of phasing?
What necessitates a split phase is when you have an option lane on a approach (such as: Left – Left/Through – Through) because there is too much turning demand for one turn lane but there isn't enough demand to convert that lane to a dedicated turn lane and still adequately serve through vehicles. Since there is no way to predict which cars in the middle lane will do what, you have to have the whole approach protected to cover the bases with the optional lane.
So it's not necessarily a function of former T-intersections, but converting a T-intersection to a 4-way where the T leg wasn't initially planned to accommodate future through vehicles is a very likely reason why these split phases were implemented.
Split phasing is never an ideal situation. It eats up a lot of time in the signal cycle which, for the cross street, decreases efficiency and can be a detriment to coordination.
split phase with straight through and left turn at the same time may be less than ideal - but as any other tool it has to be used with some understanding of what is to be achieved, and how that is to be achieved.
Too bad when things are being used just for the sake of being used (roundabouts too often fall into that category) - there are not many "one size fit all" solutions to real world problems. I still have to see a split phase used without proper understanding behind that solution, but apparently your mileage may vary..
Delaware does a LOT of split phasing; much of which I think is unnecessary.
There's one intersection near me that was recently converted to a split phase. Traffic from all 4 directions is relatively equal. The intersection needed work due to a lot of traffic from the various shopping malls in the area, and I didn't think split phasing was going to be effective when I first saw they went that route. At first, the split phasing wasn't working well - the timing was bad with a short green on one of the approaches, causing significant delays. Additionally, the thru street's left turn cycle was too short, leaving traffic backed up into the thru lanes. However, they worked with the timing and it's gotten much better. Except in times of severe congestion, most traffic gets thru their phase of the light on a single cycle, which wasn't happening previously.
So, it's not optimal, but it can work when done properly.
What do you call leading lefts that come on together but don't terminate simultaneously? I often say the left turns split phased, but i don't think this is the proper terminology.
Quote from: tradephoric on July 18, 2016, 11:52:07 AM
What do you call leading lefts that come on together but don't terminate simultaneously? I often say the left turns split phased, but i don't think this isn't the proper terminology.
So I assume you are talking about a red ball/green arrow situation on both sides, and then one of the protected lefts ends early. (Which creates a red ball only situation on the side that ended early, and a green ball plus green arrow on the side that didn't)
There is no special name for this, as I notice around here that these kind of situations se flexible depending on the volume in the left turn lane, I honestly see this simply being called a "lead-lead left", or what I prefer to call it, a "simultaneous lead left" in which one of the protected lefts ends early.
Split phase explicitly refers to the situation I was taking about at the beginning of the thread.
Quote from: MisterSG1 on July 18, 2016, 12:02:44 PM
Quote from: tradephoric on July 18, 2016, 11:52:07 AM
What do you call leading lefts that come on together but don't terminate simultaneously? I often say the left turns split phased, but i don't think this isn't the proper terminology.
So I assume you are talking about a red ball/green arrow situation on both sides, and then one of the protected lefts ends early. (Which creates a red ball only situation on the side that ended early, and a green ball plus green arrow on the side that didn't)
There is no special name for this, as I notice around here that these kind of situations se flexible depending on the volume in the left turn lane, I honestly see this simply being called a "lead-lead left", or what I prefer to call it, a "simultaneous lead left" in which one of the protected lefts ends early.
Split phase explicitly refers to the situation I was taking about at the beginning of the thread.
Yeah, this is just a case of lead-lead lefts.
By me, split phasing is used at where Kellogg's frontage roads meet arterials (e.g. Oliver, Edgemoor, Woodlawn). A little annoying, but no better solution.
I've noticed Mexico loves split phasing, which annoys me. It makes me wonder: is it cheaper to run? When I think of Mexico's traffic solutions, I tend to assume cost was the #1 consideration.
This is interesting to me, because I've only heard it called "split phasing" when a shared through-left exists without permissive lefts. Therefore the approaches on the same side of the barrier can NEVER get concurrent greens. Unless I'm seeing something wrong (which is entirely possible since I haven't eaten dinner yet :spin: ) only your first intersection is split-phased.
I'll use this diagram for comments...
(https://www.aaroads.com/forum/proxy.php?request=http%3A%2F%2Fops.fhwa.dot.gov%2Fpublications%2Fsignal_timing%2Fimages%2F050415a_img_2.jpg&hash=192dd777f5e3dbae5a73af1118eef91b686a60e3)
Is it possible that intersection 2 is timed like:
12||34
56||87
and intersection 3:
12||34
65||78
?
Maybe the engineer decided at spot #2 that all the SB should go at the same time (because the short storage and coordination on the corridor), so she did an alt sequence for phases 7&8.
On spot #3 the EBL and WBL conflict. I'm not in controllers enough to know all the details, but I know it's possible to run lead-lag so that phases 2 and 6 can run concurrently (and therefore it's not split phased).
If that's the case, I don't THINK it's defined as split phased though, because you always have the ability to go to
12||34
56||78
etc.
Ok, I've sort of looked at ring-and-barrier diagrams, but I don't think you ever are supposed to switch the numbers around. The numbers that appear top to bottom are phases that happen concurrently as I understand, and by using your approach, this is how I see it. Indeed, Intersections 2 and 3 are split phases, here's how:
In intersection 2, if we use your method based on the ring and barrier diagram, this is what happens:
12||34
65||78
Airport Road and Derry Road are "even" in the road hierarchy, as they are both arterials, I do assume that Airport Road is more heavily travelled NB/SB, than Derry is EB/WB.
In other words, this is what happens if we start with Derry Road, which is arguably the minor road in this situation:
EB/WB Derry Road has a lead-lead protected left followed by a permissive left as the straight movement gets green.
(Red Ball and Green Arrow.....followed by Red Ball and Yellow Arrow.....then Green Ball....then Yellow Ball....then Red Ball)
NB Airport Road has phases 1&6 happening concurrently, thus a green ball and green arrow. Generally, here in Ontario, when lights use split phasing, there is no yellow arrow, it goes from Green Ball/Green Arrow to Yellow Ball ONLY then Red. Both phases end at the same time as well.
SB Airport Road has phases 2&5 happening concurrently. Both phases end at the same time as well.
Basically, the major road, or the road more traveled, in this case appears to be Airport Road is the one that gets the split phases here, usually the minor road gets the split phases.
In intersection #3, the same thing also happens:
If Spadina is said to be the major road (which it most definitely is), this is what happens:
12||43
56||78
Spadina Avenue starts with a protected only left, followed by a green for phases 2 and 6.
Bremner Blvd gets a split phase as phases 4 and 7 happen concurrently and end concurrently.
Fort York Blvd gets a split phase as phases 3 and 8 happen concurrently and end concurrently.
Intersections 2 & 3 do not have any problems regarding option lanes and it may be possible to remove the split phases. But this is just me thinking and I am not expert on this.
I was hoping someone else might respond first, but I'll try to keep this going.
Quote from: MisterSG1 on July 18, 2016, 08:52:18 PMI don't think you ever are supposed to switch the numbers around.
The numbers are commonly switched. For instance, I've used different sequences for different cycle lengths. Maybe on a 120 sec cycle along a corridor it works best to lead the NBL at intersection A, but on a 100 sec cycle it works best to lag the same NBL. It just depends on the coordinated phases and the distance between intersections.
Quote from: MisterSG1 on July 18, 2016, 08:52:18 PM
numbers that appear top to bottom are phases that happen concurrently as I understand, and by using your approach, this is how I see it.
You're right, but one thing to keep in mind is that the two rings don't have to have the same amount of green time. Here's an example:
(https://www.aaroads.com/forum/proxy.php?request=http%3A%2F%2Fops.fhwa.dot.gov%2Fpublications%2Ffhwahop08024%2Fimages%2F6_4.png&hash=6e4e35e2a042494d916707b808b661622f39917d)
So imagine a scenario where, for whatever reason, the engineer wants EBL and EBT to get flushed at the same time. Maybe like your example where there was a close intersection upstream. You would flip phases 7 and 8 so that your ring/barrier went:
12||34
56||87
In my understanding this would not technically be split phasing because, even though the side streets appear to get greens at different times, the engineer has the OPTION to switch to different timings at different times of day.
Another way to think about it is how it would treat unbalanced movements...let's imagine the above timing is in place and we're trying to go WBL (phase 7). Let's say there are 100 EBL cars (ph 3) and 1 EBT (ph 8). If it were really split phased, phases 3 and 8 would really just be one phase and it would use the full split time (even though that time going to ph 8 is really wasted after the first car goes through). If it were a normal intersection, ph 8 would "gap out", so the green indications would go from phases 3&8 to 3&7.
I'd like to hear other opinions on this though, I'm no "expert"!
Quote from: roadfro on July 18, 2016, 10:34:28 AM
What necessitates a split phase is when you have an option lane on a approach (such as: Left — Left/Through — Through) because there is too much turning demand for one turn lane but there isn't enough demand to convert that lane to a dedicated turn lane and still adequately serve through vehicles. Since there is no way to predict which cars in the middle lane will do what, you have to have the whole approach protected to cover the bases with the optional lane.
So it's not necessarily a function of former T-intersections, but converting a T-intersection to a 4-way where the T leg wasn't initially planned to accommodate future through vehicles is a very likely reason why these split phases were implemented.
Split phasing is never an ideal situation. It eats up a lot of time in the signal cycle which, for the cross street, decreases efficiency and can be a detriment to coordination.
Excellent point. This is the primary situation where you might expect to see a split-phase signal setup.
These are the 3 situations where I find a split-phase signal to be justified:
1. Split-phase intersections are a good idea when one or both of the legs on a minor street have a heavy preference toward left-turns; this is commonly the case at diagonal intersections. The junction of IL-71 and US34 is a pretty good example of a rightful split-phase intersection:
https://goo.gl/maps/3Q79ekN1WnN2
2. Another justification for a split-phase intersection applies when two opposite legs are at skew angles from each other. This avoids head-on collisions for "straight" moving traffic between the skewed minor legs. This goes with roadfro's comment on "a T-intersection [converted] to a 4-way where the T leg wasn't initially planned to accommodate future through vehicles." An example of this is the former configuration of the IL31/176 intersection before it was recently reconstructed: https://goo.gl/maps/7gZZsEzSVvM2 (refer to 2012 imagery only)
Another example: the turn in Oswego https://goo.gl/maps/diG9h2Jjfo82
3. And a third appropriate application for a split-phase signal occurs where two opposite legs of an intersection aren't colinear. This will sometimes happen with a driveway interfering with T-intersection of two streets.
Example: https://goo.gl/maps/pfbpZqRdaXs
With that said, we all know that there are plenty more split-phase signals that are NOT warranted. There are plenty of examples we can list here in this thread. Here is one example that I contest: https://goo.gl/maps/d893yQR9SPm
Neither the north or south legs of the intersection have enough traffic to warrant their own individual signal phase. The north and south legs should just have 3-section heads with no separate left turn phase (so 3 phases in all).
Quote from: paulthemapguy on July 21, 2016, 10:45:29 PM
3. And a third appropriate application for a split-phase signal occurs where two opposite legs of an intersection aren't colinear. This will sometimes happen with a driveway interfering with T-intersection of two streets.
Example: https://goo.gl/maps/pfbpZqRdaXs
There is a lot of wasted time when a minor drive has a dedicated signal phase. For these minor drives with very little traffic, I prefer drivers just find a safe gap (usually after the termination of a signal phase) and go when they feel it's safe. Here are a few examples of minor drives that aren't signalized. I've never seen issues with the fact that the drives are unsignalized.
https://www.google.com/maps/@42.7205761,-83.4183551,156m/data=!3m1!1e3
https://www.google.com/maps/@42.6816187,-83.3737803,113m/data=!3m1!1e3
Here is an example of a drive that is signalized. This intersection experiences significant backups during rush hour, partly due to the fact that the drive is signalized and split phases (this drive services a smaller business than the unsignalized examples above):
https://www.google.com/maps/@42.7805659,-83.2396458,132m/data=!3m1!1e3
Quote from: tradephoric on July 22, 2016, 10:42:56 AM
Quote from: paulthemapguy on July 21, 2016, 10:45:29 PM
3. And a third appropriate application for a split-phase signal occurs where two opposite legs of an intersection aren't colinear. This will sometimes happen with a driveway interfering with T-intersection of two streets.
Example: https://goo.gl/maps/pfbpZqRdaXs
There is a lot of wasted time when a minor drive has a dedicated signal phase. For these minor drives with very little traffic, I prefer drivers just find a safe gap (usually after the termination of a signal phase) and go when they feel it's safe.
Agreed.
This intersection (https://goo.gl/LYmRQa) in Arvada, CO has a double left against a driveway. Both the driveway and Miller St see green orbs at the same time....seems to work just fine:
(https://www.aaroads.com/forum/proxy.php?request=http%3A%2F%2Fi.imgur.com%2FaM5trw9.png&hash=bc3de8eff7e55bf2fb6fc89eeb6826e4d5520b00)
This intersection (https://goo.gl/fVHrWa) in Seattle has a double left against a severely offset minor road. Just like the intersection above, the near total-lack of traffic from the minor road means that the intersection feels like it's split phased. But the city reminds you that it isn't with the "yield on green" sign. Could this intersection use split-phasing? Sure, but it simply doesn't need to.
(https://www.aaroads.com/forum/proxy.php?request=http%3A%2F%2Fi.imgur.com%2F6M5bDak.png&hash=af4826f5d7d297b45af8d4c737633813853c39fd)
I like split phasing.
Of course some situations necessitate it, like having no left turn lane.
Also, it is good where left turn volumes are nearly equal to or greater than straight volumes. Shopping center driveways are often like this, because more vehicles turn left or right on to the street rather than continue straight across to the other shopping center.
Another benefit of it is safety, because it eliminates permissive left turns. This is also more convenient for drivers that are turning left because, at standard-phased intersections, green arrow times are often too short.
Quote from: Brian556 on July 22, 2016, 01:33:31 PM
split phasing ... eliminates permissive left turns. This is also more convenient for drivers that are turning left
I, for one, do not find it convenient to have to wait for a dedicated left-turn arrow when, were left turns permissive, I could instead turn at any gap in oncoming traffic.
Quote from: Brian556 on July 22, 2016, 01:33:31 PM
Another benefit of it is safety, because it eliminates permissive left turns. This is also more convenient for drivers that are turning left because, at standard-phased intersections, green arrow times are often too short.
The first part, I don't agree with. But the second part doesn't even make sense. Permissive turns are
more convenient for drivers because they permit left turns in gaps, which is helpful when green arrow times are short. Protected-only turns at junctions with short green arrow times are, IMO, the single worst thing ever to grace our highways.
Quote from: Brian556 on July 22, 2016, 01:33:31 PM
I like split phasing....Of course some situations necessitate it, like having no left turn lane.
Besides the fact that this sentence structure is incredibly misleading, the lack of a left turn lane does not automatically warrant split phasing. If left turn volumes are low, the lack of a dedicated lane or phase does not negatively impact traffic flow.
Quote from: Brian556 on July 22, 2016, 01:33:31 PM
Also, it is good where left turn volumes are nearly equal to or greater than straight volumes. Shopping center driveways are often like this, because more vehicles turn left or right on to the street rather than continue straight across to the other shopping center.
I agree with first sentence, in certain cases. But the second sentence, I'm not so sure. Shopping center driveways are very low-trafficked, at best. A dedicated phase simply isn't necessary...a simple "yield on green" works just fine.
Quote from KPHoger:
QuoteI, for one, do not find it convenient to have to wait for a dedicated left-turn arrow when, were left turns permissive, I could instead turn at any gap in oncoming traffic.
At at split-phased intersection, you do not have to wait for a green arrow, because it is always on with the green ball. Also, a lot of the times, gaps in oncoming traffic are very hard to come by so you often don't get a fair opportunity to turn left during the permissive phase anyway.
Quote from: Brian556 on July 22, 2016, 02:11:12 PM
At at split-phased intersection, you do not have to wait for a green arrow, because it is always on with the green ball.
This seems true on the surface. However, with split phasing, normal circumstances dictate a full four cycles (at a four-way intersection). This frequently means having to wait through two or three whole cycles before turning left. OTOH, permissive lefts, unless they are timed, only run the additional cycles if they are triggered by turning vehicles; no arrow activation may be rare at busy intersections, but it's common at less busy ones. Dedicated left-turn arrows instead of permissive lefts are quite
inconvenient at minor stoplights.
Quote from: jakeroot on July 22, 2016, 01:24:24 PM
Quote from: tradephoric on July 22, 2016, 10:42:56 AM
Quote from: paulthemapguy on July 21, 2016, 10:45:29 PM
3. And a third appropriate application for a split-phase signal occurs where two opposite legs of an intersection aren't colinear. This will sometimes happen with a driveway interfering with T-intersection of two streets.
Example: https://goo.gl/maps/pfbpZqRdaXs
There is a lot of wasted time when a minor drive has a dedicated signal phase. For these minor drives with very little traffic, I prefer drivers just find a safe gap (usually after the termination of a signal phase) and go when they feel it's safe.
Agreed.
-etc.-
Actually, in cases like this, it might be better to give the more important sidestreet a ball and left arrow indication, then when someone uses the minor driveway, only use the ball in both directions. So this would be a 4-section head for the sidestreet, with both a green ball and a green arrow. When someone leaves the driveway, the 4-section head only has the green ball illuminated, as the driveway's green ball (3-section head) is illuminated as well.
NJ is big on that in some places. It used to annoy me big time especially when no one was turning left like on weekends when the counts were low and you would have a total of five cars. Three one way and two the other all going straight where all could do it at the same time, but instead you have two separate phases with three going at one moment and the other two as the next moment.
To answer the question of the OP, yes its a curse if there is no 24/7 traffic that is very heavy and if the signal is strictly on timer. No if there is steady traffic that constantly has vehicles turning left all times of day.
What would be the term for single-direction-at-a-time phasing: starting with full-movement W (all other directions red), full-movement E (again, all other directions red), full-movement S (you know the drill), full-movement S (yada, yada). Each direction has both ball and arrow tri-colors and channelized left lanes with about a 7 vehicle capacity E/W and 5 vehicles N/S. By far, largest volume is E-W; most S turns E or continues S, while very little turning movement NB. What would you call such an intersection type: quad-phase, single-serve, or some other term that I'm not familiar with?
Intersection in question: Curtner Ave. and Almaden Road (aka Old Almaden) in San Jose, CA. I've sat up to 5 minutes waiting to pass through this intersection -- which operates this way 24/7! I avoid it whenever possible, although it's quite close to my residence.
Split phasing is common where there are two shopping centers across the street from each other. Where the two major exits meet the main street, there is much heavier turning traffic. Very few people want to drive from one shopping center to the other.
Instead of split phasing, you can have the following phasing:
1) Main street
2) Opposing simultaneous left turns from shopping center to main street
3) Simultaneous left turns from main street to shopping center (and corresponding right turns)
I think it is more efficient than split-phasing.
Quote from: sparker on July 28, 2016, 10:30:23 PM
What would be the term for single-direction-at-a-time phasing: starting with full-movement W (all other directions red), full-movement E (again, all other directions red), full-movement S (you know the drill), full-movement S (yada, yada). Each direction has both ball and arrow tri-colors and channelized left lanes with about a 7 vehicle capacity E/W and 5 vehicles N/S. By far, largest volume is E-W; most S turns E or continues S, while very little turning movement NB. What would you call such an intersection type: quad-phase, single-serve, or some other term that I'm not familiar with?
Intersection in question: Curtner Ave. and Almaden Road (aka Old Almaden) in San Jose, CA. I've sat up to 5 minutes waiting to pass through this intersection -- which operates this way 24/7! I avoid it whenever possible, although it's quite close to my residence.
I'd say that these are double split-phasing. And they are terrible.
Quote from: sparker on July 28, 2016, 10:30:23 PM
What would be the term for single-direction-at-a-time phasing: starting with full-movement W (all other directions red), full-movement E (again, all other directions red), full-movement S (you know the drill), full-movement S (yada, yada). Each direction has both ball and arrow tri-colors and channelized left lanes with about a 7 vehicle capacity E/W and 5 vehicles N/S. By far, largest volume is E-W; most S turns E or continues S, while very little turning movement NB. What would you call such an intersection type: quad-phase, single-serve, or some other term that I'm not familiar with?
Intersection in question: Curtner Ave. and Almaden Road (aka Old Almaden) in San Jose, CA. I've sat up to 5 minutes waiting to pass through this intersection -- which operates this way 24/7! I avoid it whenever possible, although it's quite close to my residence.
There's another one at Storke Road and Hollister Avenue in Goleta, California.
Double split phasing! Wow! Now I know what to call it when I complain to San Jose Traffic Planning -- which will likely fall on deaf ears, as they seem committed to citywide "traffic calming" (obviously no connection to driver calming!). Likely the notion is to keep the traffic on westbound Curtner coming in waves or bunches; the neighborhood just west of the signal in question is Willow Glen, one of the pricier (and more politically active) sections of town. Well, guess what -- as Robert Burns averred, "the best laid plans...."; I live a half-block off Curtner about a mile to the west, and traffic is not only spread out by that point, but pissed off enough to be traveling well over the 35 mph limit to make up time (it's the only multilane through E-W corridor in that part of town). And since San Jose is about 150 cops short of their normal number (many leave for higher-paying gigs in Santa Clara, Sunnyvale, and Cupertino), speed enforcement is at or near the bottom of their priority list. But the planners at least, to their peers, look like they're doing something -- even if it's just creating bottlenecks!
Quote from: sparker on July 29, 2016, 02:13:31 AM
Double split phasing! Wow! Now I know what to call it when I complain to San Jose Traffic Planning -- which will likely fall on deaf ears, as they seem committed to citywide "traffic calming" (obviously no connection to driver calming!). Likely the notion is to keep the traffic on westbound Curtner coming in waves or bunches; the neighborhood just west of the signal in question is Willow Glen, one of the pricier (and more politically active) sections of town. Well, guess what -- as Robert Burns averred, "the best laid plans...."; I live a half-block off Curtner about a mile to the west, and traffic is not only spread out by that point, but pissed off enough to be traveling well over the 35 mph limit to make up time (it's the only multilane through E-W corridor in that part of town). And since San Jose is about 150 cops short of their normal number (many leave for higher-paying gigs in Santa Clara, Sunnyvale, and Cupertino), speed enforcement is at or near the bottom of their priority list. But the planners at least, to their peers, look like they're doing something -- even if it's just creating bottlenecks!
You know the area better than I do, but looking at GSV it seems that only Curtner is split-phased. Almaden Rd appears to be single phased with protective-only left turn phasing. That is much better than a true double split-phasing situation, but it is still terrible for Curtner traffic. The main problem seems to be the narrow bridge over the Guadalupe River that can't be easily widened - so there is no way to fit a left turn lane with 2 eastbound and 2 westbound lanes.
La Rue / Hutchison in Davis, CA (on the UCD campus) used to also be double split-phased, when I lived in the area about 20 years ago. It now appears to be split phased only on La Rue. A similar set up to Curtner/Almaden Rd. I've felt that given the traffic patterns there, La Rue would be much better with 2 left turn lanes and one thru lane (each direction) than having the option lane (left or straight in the middle lane) which necessitates split-phasing.
The Sacramento area has a lot of split-phasing. Usually done where there is no room for a left turn lane, particularly where a bike lane is installed. But I do not recall any double split-phasing.
Quote from: mrsman on July 28, 2016, 10:43:40 PM
Quote from: sparker on July 28, 2016, 10:30:23 PM
What would be the term for single-direction-at-a-time phasing: starting with full-movement W (all other directions red), full-movement E (again, all other directions red), full-movement S (you know the drill), full-movement S (yada, yada). Each direction has both ball and arrow tri-colors and channelized left lanes with about a 7 vehicle capacity E/W and 5 vehicles N/S. By far, largest volume is E-W; most S turns E or continues S, while very little turning movement NB. What would you call such an intersection type: quad-phase, single-serve, or some other term that I'm not familiar with?
Intersection in question: Curtner Ave. and Almaden Road (aka Old Almaden) in San Jose, CA. I've sat up to 5 minutes waiting to pass through this intersection -- which operates this way 24/7! I avoid it whenever possible, although it's quite close to my residence.
I'd say that these are double split-phasing. And they are terrible.
I don't know I've ever heard terminology for it, but I do occasionally use this intersection in Avalon, PA that is all-way split-phase:
(GSV): https://goo.gl/maps/UWBYRkgw3tw (https://goo.gl/maps/UWBYRkgw3tw)
Split phasing has its uses and certain regions of NYSDOT make use of partial split-phasing, where one direction has a leading left and the other lagging. What I don't get is split phasing where normal phasing would work, especially if the intersection is NTOR. Take the intersection of 6th Avenue and Federal St in Troy (https://www.google.com/maps/@42.7335076,-73.6857977,3a,63.3y,22.16h,86.9t/data=!3m6!1e1!3m4!1s5IC5FA83LlQCsPbwyKzFcw!2e0!7i13312!8i6656). Goes NB-SB-Federal LT-Federal. Semi-actuated to extend if cars are still coming, but timed so all phases cycle. There could easily be left and right turn lanes in each direction on 6th, as turn volume is high and there is the space for it. Given the signal timing in Troy, you're almost guaranteed to get stopped at this light and, as it's NTOR, you're stuck there. Straight traffic isn't enough to warrant 2 through lanes and, heading SB, the right lane becomes a turn lane with no warning immediately south of the intersection, so it's a common accident spot as people are forced to change lanes.
Quote from: Mr_Northside on July 29, 2016, 03:03:44 PM
Quote from: mrsman on July 28, 2016, 10:43:40 PM
Quote from: sparker on July 28, 2016, 10:30:23 PM
What would be the term for single-direction-at-a-time phasing: starting with full-movement W (all other directions red), full-movement E (again, all other directions red), full-movement S (you know the drill), full-movement S (yada, yada). Each direction has both ball and arrow tri-colors and channelized left lanes with about a 7 vehicle capacity E/W and 5 vehicles N/S. By far, largest volume is E-W; most S turns E or continues S, while very little turning movement NB. What would you call such an intersection type: quad-phase, single-serve, or some other term that I'm not familiar with?
Intersection in question: Curtner Ave. and Almaden Road (aka Old Almaden) in San Jose, CA. I've sat up to 5 minutes waiting to pass through this intersection -- which operates this way 24/7! I avoid it whenever possible, although it's quite close to my residence.
I'd say that these are double split-phasing. And they are terrible.
I don't know I've ever heard terminology for it, but...
This is the type of timing that Wichita uses for basically the whole daylight hours for intersections where Kellogg's frontage roads meet surface arterials. It makes sense here, because the whole intersection can function as one big intersection that happens to have a freeway-sized gap in the middle.
What annoys the heck out of me is an intersection like
this one in Mexico (https://goo.gl/maps/yg78H8mLiV42)–which, incidentally, I'll probably be driving through on Monday. It's not a busy intersection, the north and south approaches are barely skewed at all, yet it operates on "double split" phasing all the time. There are several others in town that do the same thing.
Looking at the current GSV for the Curtner/Almaden intersection that was referred to by mrsman shows the previous configuration (from prior to 10 weeks ago, when that section was repaved and restriped); while the Guadalupe bridge wasn't widened, an actual left-turn lane from E. Curtner to N. Almaden was carved out by narrowing the travel lanes to 10 feet. A new massive condo complex about a half-mile north of Curtner on Almaden Rd. opened its latest phase a couple of months ago (this has been the subject of a running political/planning local controversy at least since I moved to the area four years ago); apparently, there's a move to accommodate traffic movement to and from that facility area. And there's also a planned high-rise due to be constructed in the adjacent tract now occupied by a mobile-home park (same developers), which has heated up the density-vs-existing neighborhood rhetoric in town. Since Almaden Road is, north of Curtner, a narrow 2-lane winding street (with little room for expansion), I fully expect to see more controversial traffic-pattern manipulation deployed locally in the immediate future. When the Almaden Expressway was built east of the "old" road in the '60's, it was expected that the old road would devolve into a secondary local server; the planning efforts to utilize that stretch of street for ultra-dense housing has certainly set planners against long-time residents (the local city councilperson who originally opposed the "high-rise" during their campaign but, after being elected to office, reversed their position, is likely to be the subject of a recall petition soon (they've already been by my home twice!). I think we're just at the beginning of traffic manipulation in the area -- I have no choice but to "keep tuned"; I live here!
More info on the ever-changing Curtner/Almaden intersection: Talked to SJ traffic planning; that interchange is an ever-changing "work in progress" to them -- nothing they do seems to satisfy anyone! Indeed, double-split phasing is the default setting on the signal controls, but they're out there every couple of weeks trying new configurations. Right now, what they're using is single-directional full-function on E-W Curtner (regular all-movement W followed by regular all-movement E). N-S (Almaden) starts with double lefts, but segueing to green-ball + green-left NB in the morning, followed by double-green balls N-S and red-lefts; the situation is reversed in the afternoon, with southbound being the recipient of the green ball/green left interim phase.
Seems SJ traffic planning has finally realized that that short section of (Old) Almaden serves as a shortcut to Bird Avenue, a local N-S arterial that (with a slight jog) empties out onto Old Almaden; it's a viable alternative (although some local Willow Glen residents would demur!) to get to downtown SJ when the Almaden Expressway ramp to NB CA 87 (the next access north of Curtner) backs up, as it is prone to do during morning commute times. The inverse occurs in the late afternoon. So SJ traffic planning certainly has their hands full. Latest talk -- since the EB left-turn lane on Curner has squeezed that street's bike paths down to virtually nothing (prompting very loud and consistent complaints from that quarter!), that dedicated-left lane may not be long for this world!
Dilemmas such as these -- trying to please as many parties as possible while dodging thrown tomatoes AND attempting to plan things for the foreseeable future -- has given me at least a grudging respect for what some of these folks in traffic planning have to do on an ongoing basis (although I'm not always pleased with some of their agenda-driven activities). I certainly don't envy them given the current environment in which they function!
Quote from: sparker on July 31, 2016, 05:01:00 AM
More info on the ever-changing Curtner/Almaden intersection: Talked to SJ traffic planning; that interchange is an ever-changing "work in progress" to them -- nothing they do seems to satisfy anyone! Indeed, double-split phasing is the default setting on the signal controls, but they're out there every couple of weeks trying new configurations. Right now, what they're using is single-directional full-function on E-W Curtner (regular all-movement W followed by regular all-movement E). N-S (Almaden) starts with double lefts, but segueing to green-ball + green-left NB in the morning, followed by double-green balls N-S and red-lefts; the situation is reversed in the afternoon, with southbound being the recipient of the green ball/green left interim phase.
Seems SJ traffic planning has finally realized that that short section of (Old) Almaden serves as a shortcut to Bird Avenue, a local N-S arterial that (with a slight jog) empties out onto Old Almaden; it's a viable alternative (although some local Willow Glen residents would demur!) to get to downtown SJ when the Almaden Expressway ramp to NB CA 87 (the next access north of Curtner) backs up, as it is prone to do during morning commute times. The inverse occurs in the late afternoon. So SJ traffic planning certainly has their hands full. Latest talk -- since the EB left-turn lane on Curner has squeezed that street's bike paths down to virtually nothing (prompting very loud and consistent complaints from that quarter!), that dedicated-left lane may not be long for this world!
Dilemmas such as these -- trying to please as many parties as possible while dodging thrown tomatoes AND attempting to plan things for the foreseeable future -- has given me at least a grudging respect for what some of these folks in traffic planning have to do on an ongoing basis (although I'm not always pleased with some of their agenda-driven activities). I certainly don't envy them given the current environment in which they function!
Traffic engineers, like most other professionals, often face some kind of conflict amongst their constituencies. And it is very difficult to keep everyone happy.
Hennepin Drive at Essington Road in Joliet is another signal that exclusively uses split-phasing. The north-south legs I understand, because the signal is timed in synchrony with the mall loop signal 1/2 a block to the north. The east approach has a heavy imbalance toward left-turn movements, but I think it can still be regular phase with protected left turns.
https://goo.gl/maps/1DmNbsifeno
Quote from: paulthemapguy on August 16, 2016, 08:50:46 AM
Hennepin Drive at Essington Road in Joliet is another signal that exclusively uses split-phasing. The north-south legs I understand, because the signal is timed in synchrony with the mall loop signal 1/2 a block to the north. The east approach has a heavy imbalance toward left-turn movements, but I think it can still be regular phase with protected left turns.
https://goo.gl/maps/1DmNbsifeno
It, and the signal at Ring Road and Essington Road are effectively one signal. The city (all of these are city streets) decided that it would be best to tie them together to facilitate traffic flow at a very busy entry to the mall. IMHO, it works great.
Quote from: Brandon on August 16, 2016, 11:35:48 AM
Quote from: paulthemapguy on August 16, 2016, 08:50:46 AM
Hennepin Drive at Essington Road in Joliet is another signal that exclusively uses split-phasing. The north-south legs I understand, because the signal is timed in synchrony with the mall loop signal 1/2 a block to the north. The east approach has a heavy imbalance toward left-turn movements, but I think it can still be regular phase with protected left turns.
https://goo.gl/maps/1DmNbsifeno
It, and the signal at Ring Road and Essington Road are effectively one signal. The city (all of these are city streets) decided that it would be best to tie them together to facilitate traffic flow at a very busy entry to the mall. IMHO, it works great.
Yeah I know I like it too. I don't see why they HAD to split the east and west phases though, other than the heavy preference for westbound traffic to turn left onto the south leg.
Quote from: paulthemapguy on August 16, 2016, 01:59:35 PM
Quote from: Brandon on August 16, 2016, 11:35:48 AM
Quote from: paulthemapguy on August 16, 2016, 08:50:46 AM
Hennepin Drive at Essington Road in Joliet is another signal that exclusively uses split-phasing. The north-south legs I understand, because the signal is timed in synchrony with the mall loop signal 1/2 a block to the north. The east approach has a heavy imbalance toward left-turn movements, but I think it can still be regular phase with protected left turns.
https://goo.gl/maps/1DmNbsifeno
It, and the signal at Ring Road and Essington Road are effectively one signal. The city (all of these are city streets) decided that it would be best to tie them together to facilitate traffic flow at a very busy entry to the mall. IMHO, it works great.
Yeah I know I like it too. I don't see why they HAD to split the east and west phases though, other than the heavy preference for westbound traffic to turn left onto the south leg.
It's so that you can manipulate it with the other intersection at Ring Road. Otherwise, traffic might back up between Ring Road and Hennepin Drive.
The intersection of Wis 114 and US 10 is a split phase intersection and works extremely well because US 10 eastbound turns left, and almost all of the traffic turns left. The oncoming road is a minor street and a lot of times that phase is skipped because there isn't much traffic coming from that part of the intersection. When the majority of traffic turns left from one leg of an intersection, it can be effective.
https://www.google.com/maps/@44.2121952,-88.4043398,3a,75y,195.27h,70.54t/data=!3m6!1e1!3m4!1s_FdMm8H4LnCvCPP0vtVPeQ!2e0!7i13312!8i6656