Traffic signal

[jakeroot]

Also, has anybody noticed these lime-green lights? Kirkwood MO had some before they were torn out in Jan 2020.

Those are older incandescent displays. They usually had a lime green appearance, as opposed to today's blueish green LED signals.

[STLmapboy]

You never seen anything quite like this. Jakeroot, this one's near you in Mukilteo.

And nearby at the Boeing factory. Why not just use ground mounted poles?

[jakeroot]

You never seen anything quite like this. Jakeroot, this one's near you in Mukilteo.

And nearby at the Boeing factory. Why not just use ground mounted poles?

That's a little strange, yeah. Why it's not mounted directly to the mast, I couldn't tell you.

The signal next to the Boeing Everett facility is odd only in that signals that low would normally be post-mounted. Having a mast arm across nothing, opposite a road that ends at a T-junction, is not unusual around here (exhibit A; exhibit B)

[fwydriver405]

And nearby at the Boeing factory. Why not just use ground mounted poles?

I'm going to ask the same question about this approach leaving the Bangor Mall in Bangor ME... span wire mounted very low compared to the other lights on Stillwater Ave.

[mrsman]

I can surmise that in many cases a DOT does things that are normal and common for it.   It is normal and common for these DOTs to use mast arms, and they have a lot in stock, so they use them even when not really needed, like at a T-intersection.

It is also common where a wide street ends at a T-intersection in California:

https://www.google.com/maps/@33.9165237,-118.3715981,3a,75y,189.81h,94.79t/data=!3m7!1e1!3m5!1smFcZKB5QwmY4me2Mu9RS0w!2e0!5s20180501T000000!7i16384!8i8192

[JoePCool14]

Also, has anybody noticed these lime-green lights? Kirkwood MO had some before they were torn out in Jan 2020.

Those just look like your typical classic incandescent green to me. They look very different from LED greens used today.

Here's one just like it in Kenilworth, IL. Still standing as of a couple months ago, at least.



Street View link.

[roadman65]

Never saw a signal back like these two signal heads in Florence, SC on Cashua Drive at Second Loop Drive.

Who makes these odd looking signal backs?

[STLmapboy]

This is a very unusual style for Rhode Island. Curved mast arms on RI-114 in Middletown.

[paulthemapguy]

This is a very unusual style for Rhode Island. Curved mast arms on RI-114 in Middletown.

An interesting strategy for trying to maneuver the mast arm around those aerial lines!  And they couldn't just install a new mast closer to the current mast because the sidewalk is in the way.

[jeffandnicole]

This is a very unusual style for Rhode Island. Curved mast arms on RI-114 in Middletown.

An interesting strategy for trying to maneuver the mast arm around those aerial lines!  And they couldn't just install a new mast closer to the current mast because the sidewalk is in the way.

The telephone poles are also located within the sidewalk, so the traffic light pole would need to go right against the curb line to avoid the wires.

[STLmapboy]

This is a very unusual style for Rhode Island. Curved mast arms on RI-114 in Middletown.

An interesting strategy for trying to maneuver the mast arm around those aerial lines!  And they couldn't just install a new mast closer to the current mast because the sidewalk is in the way.

The telephone poles are also located within the sidewalk, so the traffic light pole would need to go right against the curb line to avoid the wires.

There used to be two straight mast arms coming from a pole planted where the sidewalk is now located. The sidewalk was constructed in 2009-10.

[roadfro]

And nearby at the Boeing factory. Why not just use ground mounted poles?

I'm going to ask the same question about this approach leaving the Bangor Mall in Bangor ME... span wire mounted very low compared to the other lights on Stillwater Ave.

For the location near Boeing, it's obvious you need signal heads to be lower due to the bridge. Post-mounted poles would seem to make more sense, but it may actually have taken less cable to do it this way...so a cheaper install cost perhaps?

The Bangor, ME location looks like it was done that way to avoid having the signal heads obscured by, or to avoid potential collision with, the overhead utility lines (I'm assuming the utility lines predate the signal).

This is a very unusual style for Rhode Island. Curved mast arms on RI-114 in Middletown.

An interesting strategy for trying to maneuver the mast arm around those aerial lines!  And they couldn't just install a new mast closer to the current mast because the sidewalk is in the way.

The telephone poles are also located within the sidewalk, so the traffic light pole would need to go right against the curb line to avoid the wires.

There used to be two straight mast arms coming from a pole planted where the sidewalk is now located. The sidewalk was constructed in 2009-10.

Using the curved mast arm here actually makes a bit more sense. The old straight mast arm was positioned between sets of utility lines, whereas the curved mast arm connects to the mast below all the utility lines. In the case of a utility pole falling or some other major issue, it's less likely that any utility lines will get tangled with the traffic signal, reducing complications in affecting repairs.

[fwydriver405]

In this MaineDOT Mobility Report under the Signal Design and Operations section, there are two things that caught my eye regarding phasing (page 30):

No new protected left turns from a shared lane. This is very inefficient and does little to increase capacity or safety. Existing locations may remain.

I want to know how using permissive/protected phasing with a thru/left lane (or a single lane approach) is "very inefficient" and "does little to increase capacity or safety". Unless I'm overthinking this, my concern (example) is that when one approch has a protected phase with the left turn lane is allowed to phase skip, or where preemption is initated, the direction with the shared lane could be yellow trapped if there is no all-red clear and/or required barrier crossing to prevent such trap.

My other concern is what happens if there is a heavy left turning queue and there is no lagging left turn to clear the left turning traffic. Some people, on permissive left turns, in my area don't pull into the intersection, opting to wait behind the line, and some in the intersection don't complete their turn when the light changes from yellow to red.

When DOT says that above quote, does that exempt split phasing, because split phasing usually does involve "protected left turns from a shared lane"?

No protected/permitted left turns across two through lanes. History has shown that this type of movement develops into a high crash location over time.

This has been discussed a lot previously, but my question is what "history" justifies that "this type of movement develops into a high crash location over time"? I see a lot of existing retrofits and new FYA's being installed in some locations across two thru lanes in New Hampshire, Massachusetts, Rhode Island and I'm not sure about Connecticut as well. Not sure about the crash history of these locations in the mentioned states.

If they really wanted to restrict PPLT phasing because it may be difficult to turn during the peak travel periods, why don't they implement FYA TOD phasing? It could be protected only during the day (6am-9pm), then during the night, it could be permissive/protected, so cars don't have to come to a full stop and wait 5 seconds when there's no oncoming traffic in the middle of the night.

[Amtrakprod]

In this MaineDOT Mobility Report under the Signal Design and Operations section, there are two things that caught my eye regarding phasing (page 30):

No new protected left turns from a shared lane. This is very inefficient and does little to increase capacity or safety. Existing locations may remain.

I want to know how using permissive/protected phasing with a thru/left lane (or a single lane approach) is "very inefficient" and "does little to increase capacity or safety". Unless I'm overthinking this, my concern (example) is that when one approch has a protected phase with the left turn lane is allowed to phase skip, or where preemption is initated, the direction with the shared lane could be yellow trapped if there is no all-red clear and/or required barrier crossing to prevent such trap.

My other concern is what happens if there is a heavy left turning queue and there is no lagging left turn to clear the left turning traffic. Some people, on permissive left turns, in my area don't pull into the intersection, opting to wait behind the line, and some in the intersection don't complete their turn when the light changes from yellow to red.

When DOT says that above quote, does that exempt split phasing, because split phasing usually does involve "protected left turns from a shared lane"?

No protected/permitted left turns across two through lanes. History has shown that this type of movement develops into a high crash location over time.

This has been discussed a lot previously, but my question is what "history" justifies that "this type of movement develops into a high crash location over time"? I see a lot of existing retrofits and new FYA's being installed in some locations across two thru lanes in New Hampshire, Massachusetts, Rhode Island and I'm not sure about Connecticut as well. Not sure about the crash history of these locations in the mentioned states.

If they really wanted to restrict PPLT phasing because it may be difficult to turn during the peak travel periods, why don't they implement FYA TOD phasing? It could be protected only during the day (6am-9pm), then during the night, it could be permissive/protected, so cars don't have to come to a full stop and wait 5 seconds when there's no oncoming traffic in the middle of the night.

By protected lefts I think MaineDOT means protected only signals (Red Arrow; Yellow Arrow; Green Arrow)


iPhone

[fwydriver405]

In this MaineDOT Mobility Report under the Signal Design and Operations section, there are two things that caught my eye regarding phasing (page 30):

No new protected left turns from a shared lane. This is very inefficient and does little to increase capacity or safety. Existing locations may remain.

I want to know how using permissive/protected phasing with a thru/left lane (or a single lane approach) is "very inefficient" and "does little to increase capacity or safety". Unless I'm overthinking this, my concern (example) is that when one approch has a protected phase with the left turn lane is allowed to phase skip, or where preemption is initated, the direction with the shared lane could be yellow trapped if there is no all-red clear and/or required barrier crossing to prevent such trap.

My other concern is what happens if there is a heavy left turning queue and there is no lagging left turn to clear the left turning traffic. Some people, on permissive left turns, in my area don't pull into the intersection, opting to wait behind the line, and some in the intersection don't complete their turn when the light changes from yellow to red.

When DOT says that above quote, does that exempt split phasing, because split phasing usually does involve "protected left turns from a shared lane"?

No protected/permitted left turns across two through lanes. History has shown that this type of movement develops into a high crash location over time.

This has been discussed a lot previously, but my question is what "history" justifies that "this type of movement develops into a high crash location over time"? I see a lot of existing retrofits and new FYA's being installed in some locations across two thru lanes in New Hampshire, Massachusetts, Rhode Island and I'm not sure about Connecticut as well. Not sure about the crash history of these locations in the mentioned states.

If they really wanted to restrict PPLT phasing because it may be difficult to turn during the peak travel periods, why don't they implement FYA TOD phasing? It could be protected only during the day (6am-9pm), then during the night, it could be permissive/protected, so cars don't have to come to a full stop and wait 5 seconds when there's no oncoming traffic in the middle of the night.

By protected lefts I think MaineDOT means protected only signals (Red Arrow; Yellow Arrow; Green Arrow)


iPhone

I believe this is referring to approches like this. Shared left turn lane, but the phasing is protected/permissive. Correct me on this, but you can't really use a regular RA-YA-GA signal unless that approch is split phased...

[roadfro]

In this MaineDOT Mobility Report under the Signal Design and Operations section, there are two things that caught my eye regarding phasing (page 30):

No new protected left turns from a shared lane. This is very inefficient and does little to increase capacity or safety. Existing locations may remain.

I want to know how using permissive/protected phasing with a thru/left lane (or a single lane approach) is "very inefficient" and "does little to increase capacity or safety". Unless I'm overthinking this, my concern (example) is that when one approch has a protected phase with the left turn lane is allowed to phase skip, or where preemption is initated, the direction with the shared lane could be yellow trapped if there is no all-red clear and/or required barrier crossing to prevent such trap.

My other concern is what happens if there is a heavy left turning queue and there is no lagging left turn to clear the left turning traffic. Some people, on permissive left turns, in my area don't pull into the intersection, opting to wait behind the line, and some in the intersection don't complete their turn when the light changes from yellow to red.

When DOT says that above quote, does that exempt split phasing, because split phasing usually does involve "protected left turns from a shared lane"?

No protected/permitted left turns across two through lanes. History has shown that this type of movement develops into a high crash location over time.

This has been discussed a lot previously, but my question is what "history" justifies that "this type of movement develops into a high crash location over time"? I see a lot of existing retrofits and new FYA's being installed in some locations across two thru lanes in New Hampshire, Massachusetts, Rhode Island and I'm not sure about Connecticut as well. Not sure about the crash history of these locations in the mentioned states.

If they really wanted to restrict PPLT phasing because it may be difficult to turn during the peak travel periods, why don't they implement FYA TOD phasing? It could be protected only during the day (6am-9pm), then during the night, it could be permissive/protected, so cars don't have to come to a full stop and wait 5 seconds when there's no oncoming traffic in the middle of the night.

Say there's a two lane approach with #1 lane shared left/through lane and #2 is a through lane, and the left turn is protected only left turns (no permitted movement). If the first car in queue in the #1 shared lane wants to turn left, they are stuck waiting for an arrow–meanwhile, the 10 cars behind that want to go straight are unable to move (or they are darting into the #2 lane when there's a gap). There's your inefficiency.

I would imagine the split phasing scenario would be different (I think a lot of engineers don't really view that as a "protected" left, even though it technically is). In a lot of cases, split phasing is not desirable because of its inflexibility.


No PPLT across two through lanes is really limiting. I don't know what kind of historical data they're citing...I'm assuming they have historical crash rates that they've analyzed. If that rule applied in Nevada, I'd guess there would be less than 30 PPLT locations in the state... Many of the Vegas-area agencies will allow a single-lane PPLT across up to three opposing through lanes if the speed limit is 45mph or less–northern Nevada is a bit more conservative than that on the speed limit, but will still do up to three lanes.

Time of day on the FYA would be a better compromise. Some agencies are better at implementing more progressive traffic signal timing and phasing methodologies than others...

[DrSmith]

No new protected left turns from a shared lane. This is very inefficient and does little to increase capacity or safety. Existing locations may remain.

I think what it is getting at is if you have an intersection where there is a lot of left turns and those turns create a queue of traffic behind the left turners that what to go straight, that this is not a sufficient remedy. In those situations, rarely is it only the first car that wants to make the left turn. There are a whole mixture of cars that want to turn left and also go straight (or potentially turn right depending on number of lanes of travel).

I am not sure this means protected left only signals; rather my thought is that includes any amount of protected left turn is not to be used (protected only or protected/permissive).

My guess is the intention is about design and remediation of issues. If the concern is that left turns are really causing delays and increased accidents that a new intersection design is needed. Small measures that only makes for minor efficiency increases at best are considered not appropriate. As accidents may be a major driver of the problem, these need to be taken into account, which means the solution is to provide actual overall improvement in the issues.

[Revive 755]

No protected/permitted left turns across two through lanes. History has shown that this type of movement develops into a high crash location over time.

I really have to wonder whether there is valid data to back up that statement, or if this is based on a few intersections where other issues affected the crash rate, such as sight distance, negative offset between opposing left turn lanes, and/or a lack of gaps in opposing traffic to make the turn.  I think are a lot of counter examples across the country with permissive left turns across two through lanes without major crash issues.  Or does Maine have a low bar for what is considered "a high crash location"?

Semi-related side rant:  Is Maine of those places where left turns are protected only but then are unrestricted left turns to/from other unsignalized side roads and driveways on either side of that intersection?  I am really started to get annoyed with a few locations where I have to wait forever for a green arrow to turn left whereas I could turn right away into a different driveway fifty feet away from the signal.

[US 89]

Say there's a two lane approach with #1 lane shared left/through lane and #2 is a through lane, and the left turn is protected only left turns (no permitted movement). If the first car in queue in the #1 shared lane wants to turn left, they are stuck waiting for an arrow–meanwhile, the 10 cars behind that want to go straight are unable to move (or they are darting into the #2 lane when there's a gap). There's your inefficiency.

Atlanta is full of these...and it's very frustrating as someone who grew up in the west where there are very few (if any) of these, and never on major roads.

[jakeroot]

Individual cities in WA are pretty liberal with permissive phasing; there are more than a few with permissive lefts across four lanes. I'm sure there's one across five somewhere.

WSDOT (the state highway department) has often avoided protected lefts across more than one lane, but there is no strict policy that doesn't allow it. Good news is that retrofits have become common (WA-9 being a good example, plus numerous on-ramps) and new permissive lefts across two lanes are being installed. It's just slow, since WSDOT has this tendency to jump straight to protected lefts, so tons of existing FYA or 5-section 'yield on green' signals have been retrofits of protected lefts. I don't know what happened to the good old days of progressively advancing to more protected over time, but WSDOT didn't get the memo.

I'm hoping Maine doesn't fall into the same trap that WSDOT sort of fell into, by installing dozens of protected lefts, only to then rip them out later when it becomes clear that the policy has no clear basis in reality. At least when looking at all left turns as a whole (you could cherry pick data easily enough).

Semi-related side rant:  Is Maine of those places where left turns are protected only but then are unrestricted left turns to/from other unsignalized side roads and driveways on either side of that intersection?  I am really started to get annoyed with a few locations where I have to wait forever for a green arrow to turn left whereas I could turn right away into a different driveway fifty feet away from the signal.

I think it's very related, and a great question. Canyon Road in Pierce County, WA is a six to eight lane arterial with fully protected lefts south of WA-512. The protected lefts were installed because the county followed the state in not using permissive lefts across more than two lanes. However, the arterial is full of designated left and U turn points that are just regular yields. You can see both in this GSV still. Some would say, "signalized and non-signalized intersections have different warrants for left turn operations". To which I would say: why? It's the same traffic, and the same number of lanes being crossed (often enough), etc, etc. Saying that left turns across three or more lanes should not be permissive, but then installing dozens of yield-only left turns along the same corridor that have no signal at all, just screams "policy alone didn't allow us to install permissive lefts despite no obvious difference to the average driver".

[jakeroot]

Say there's a two lane approach with #1 lane shared left/through lane and #2 is a through lane, and the left turn is protected only left turns (no permitted movement). If the first car in queue in the #1 shared lane wants to turn left, they are stuck waiting for an arrow–meanwhile, the 10 cars behind that want to go straight are unable to move (or they are darting into the #2 lane when there's a gap). There's your inefficiency.

Atlanta is full of these...and it's very frustrating as someone who grew up in the west where there are very few (if any) of these, and never on major roads.

I would echo your sentiments that these are exceptionally rare (if non-existenst) out west, although I have seen them in DC.

Do you have some links to those in Atlanta? I'm curious how they operate. 5-section PPLT signals are not unusual along corridors without turn lanes anywhere in the country, but protected-only left turns along corridors without turn lanes seems really weird to me. Even in DC, where things are already pretty weird as-is, they strike me as a bit odd and perhaps not a good idea.

[jeffandnicole]

No protected/permitted left turns across two through lanes. History has shown that this type of movement develops into a high crash location over time.

I really have to wonder whether there is valid data to back up that statement, or if this is based on a few intersections where other issues affected the crash rate, such as sight distance, negative offset between opposing left turn lanes, and/or a lack of gaps in opposing traffic to make the turn.  I think are a lot of counter examples across the country with permissive left turns across two through lanes without major crash issues.  Or does Maine have a low bar for what is considered "a high crash location"?

Shouldn't you have valid data yourself to support your opinion, rather than "I think there are a lot of counter examples..."

It doesn't matter what other examples there are across the country.  Likewise, should I counter with there shouldn't be any permissive 2 lane left turns elsewhere because Maine doesn't allow it?

[jakeroot]

Likewise, should I counter with there shouldn't be any permissive 2 lane left turns elsewhere because Maine doesn't allow it?

No, because that wouldn't make sense. Maine's policy is being discussed because Maine's new policy would be counter to most other agencies, and thus any data that other said agencies use to support their own policies. The onus is on Maine to prove their more restrictive policy has some basis in reality. Based on the policy of most other places in the US, the answer would appear to be "no".

I would counter: Virtually all other states allow permissive left turns across more than one lane, thus Maine's new policy must be misguided. Unless Maine drivers are just really. that. bad.

[jeffandnicole]

Likewise, should I counter with there shouldn't be any permissive 2 lane left turns elsewhere because Maine doesn't allow it?

No, because that wouldn't make sense. Maine's policy is being discussed because Maine's new policy would be counter to most other agencies, and thus any data that other said agencies use to support their own policies. The onus is on Maine to prove their more restrictive policy has some basis in reality. Based on the policy of most other places in the US, the answer would appear to be "no".

I would counter: Virtually all other states allow permissive left turns across more than one lane, thus Maine's new policy must be misguided. Unless Maine drivers are just really. that. bad.

Actually,  Maine doesn't have to prove anything to anyone. If Maine wants to mandate a certain traffic control function, its well within their right to do so, regardless what any other state does. If they want federal funding, they would need to make certain they don't run afoul of any federal regulations or guidance, which this does not.

[US 89]

Say there's a two lane approach with #1 lane shared left/through lane and #2 is a through lane, and the left turn is protected only left turns (no permitted movement). If the first car in queue in the #1 shared lane wants to turn left, they are stuck waiting for an arrow—meanwhile, the 10 cars behind that want to go straight are unable to move (or they are darting into the #2 lane when there's a gap). There's your inefficiency.

Atlanta is full of these...and it's very frustrating as someone who grew up in the west where there are very few (if any) of these, and never on major roads.

I would echo your sentiments that these are exceptionally rare (if non-existent) out west, although I have seen them in DC.

Do you have some links to those in Atlanta? I'm curious how they operate. 5-section PPLT signals are not unusual along corridors without turn lanes anywhere in the country, but protected-only left turns along corridors without turn lanes seems really weird to me. Even in DC, where things are already pretty weird as-is, they strike me as a bit odd and perhaps not a good idea.

I guess I didn't read closely enough - these generally aren't protected-only, but they might as well be given the typical very heavy oncoming traffic. Traditionally these used regular doghouse PPLT signals, but most of them now use bimodal FYAs - many in a doghouse configuration (never seen that anywhere else). Here's one example along Piedmont Road, a corridor full of them. Turn around 180 degrees and you'll see how much backup it regularly causes.

I'm not sure I can think of any intersections at all in Utah with a shared left/straight lane that don't use split phasing. Just about every signalized intersection, if there are at least two lanes approaching, has a dedicated left turn lane.