When did the cloverleaf fall out of fashion?

[MCRoads]

Every single video about the subject I have seen has talked about how awesome the interchange at CO 43@I-25 is, and how decloverleafing doesn't need to be expensive.

There is no CO 43. Did you perchance mean US 34?

Dang it. Yes, that is indeed what I had in mind. We drive on 25 all the time, but almost never 34, so I guess being close is still pretty good.

[Super Mateo]

Interesting. I didn't realize it was that high. I guess it is pretty well justified then.

I didn't know that, either.  It makes sense, though.  IL 59 runs due south, then ends and goes pretty much straight onto I-55.  I can see a lot of that traffic taking the loop ramp.

I'm also surprised at just how low the AADT is between the two legs leading away from Chicago.  I figured it would be lower, but a lot higher than 15-20% of the traffic of the busiest ramps.

[kphoger]

Interesting that all the loop ramps have lower numbers than their opposite-direction corresponding non-loop ramp.

Only one of the pairs has more than a 10% difference, though.

E→S = 1,800  /  N→W = 1,700 (5.6% lower)
N→E = 7,800  /  W→S = 6,600 (15.4% lower)
W→N = 8,800  /  S→E = 8,500 (3.4% lower)
S→W = 11,000  /  E→N = 10,100 (8.2% lower)

[webny99]

Interesting that all the loop ramps have lower numbers than their opposite-direction corresponding non-loop ramp.

Only one of the pairs has more than a 10% difference, though.

E→S = 1,800  /  N→W = 1,700 (5.6% lower)
N→E = 7,800  /  W→S = 6,600 (15.4% lower)
W→N = 8,800  /  S→E = 8,500 (3.4% lower)
S→W = 11,000  /  E→N = 10,100 (8.2% lower)

Those percentages probably correlate roughly with the percentage of trips for which an alternate route is faster than the loop... but not the non-loop.

[kphoger]

Interesting that all the loop ramps have lower numbers than their opposite-direction corresponding non-loop ramp.

Only one of the pairs has more than a 10% difference, though.

E→S = 1,800  /  N→W = 1,700 (5.6% lower)
N→E = 7,800  /  W→S = 6,600 (15.4% lower)
W→N = 8,800  /  S→E = 8,500 (3.4% lower)
S→W = 11,000  /  E→N = 10,100 (8.2% lower)

Those percentages probably correlate roughly with the percentage of trips for which an alternate route is faster than the loop... but not the non-loop.

Doubt it.

[webny99]

Why else would the volumes on the loop be lower? I thought I was just stating the obvious... probably true for every loop, not just these ones.

[kphoger]

Interesting that all the loop ramps have lower numbers than their opposite-direction corresponding non-loop ramp.

Only one of the pairs has more than a 10% difference, though.

E→S = 1,800  /  N→W = 1,700 (5.6% lower)
N→E = 7,800  /  W→S = 6,600 (15.4% lower)
W→N = 8,800  /  S→E = 8,500 (3.4% lower)
S→W = 11,000  /  E→N = 10,100 (8.2% lower)

Those percentages probably correlate roughly with the percentage of trips for which an alternate route is faster than the loop... but not the non-loop.

Doubt it.

Why else would the volumes on the loop be lower? I thought I was just stating the obvious... probably true for every loop, not just these ones.

I mean, show me such a trip–especially the pair with a 15% difference.

[webny99]

Those percentages probably correlate roughly with the percentage of trips for which an alternate route is faster than the loop... but not the non-loop.

Doubt it.

Why else would the volumes on the loop be lower? I thought I was just stating the obvious... probably true for every loop, not just these ones.

I mean, show me such a trip–especially the pair with a 15% difference.

Google is not cooperating with me today! But, here's a potential one. And another. And here's one for S/W vs. E/N where Google doesn't even present the freeway as an option in the direction that requires the loop.

There's really countless localized trips that you could come up with. That extra 60-90 seconds comes into play a lot more often than you might think.

[hobsini2]

Those percentages probably correlate roughly with the percentage of trips for which an alternate route is faster than the loop... but not the non-loop.

Doubt it.

Why else would the volumes on the loop be lower? I thought I was just stating the obvious... probably true for every loop, not just these ones.

I mean, show me such a trip–especially the pair with a 15% difference.

Google is not cooperating with me today! But, here's a potential one. And another. And here's one for S/W vs. E/N where Google doesn't even present the freeway as an option in the direction that requires the loop.

There's really countless localized trips that you could come up with. That extra 60-90 seconds comes into play a lot more often than you might think.

And if I am on 6 by Youngs Rd, I get on 80 at Houbolt Rd.

[OCGuy81]

Hoping this isn't too off topic, but what's been the best replacement for these? Stacks?

[Ned Weasel]

Hoping this isn't too off topic, but what's been the best replacement for these? Stacks?

Only if you have the space and the money.

[sparker]

Hoping this isn't too off topic, but what's been the best replacement for these? Stacks?

Only if you have the space and the money.

If one can live with 35-40 mph ramps, turbine interchanges work reasonably well.   I still think cloverleaves with C/D lanes for both facilities work fine in rural situations (the I-22/269 interchange in NW MS is an example of such -- although I would have liked to have seen them modify it with a WB>SB flyover; also, the I-95/future 87 interchange in NC will probably remain a double-CD cloverleaf).

[kphoger]

I think that, for most situations, adding one or two flyover ramps is fine.  Generally, not all four accel/decel crunch points are equally bad, and a couple of them might as well remain as-is.

[jeffandnicole]

Hoping this isn't too off topic, but what's been the best replacement for these? Stacks?

It depends on the overall interchange.  If the cross road is a local roadway and isn't that busy, a diamond or some variation thereof would work fine.  If you desire freeflow movements on both roadways, the turbine or stack are better, but they also come at a much higher cost. 

There's also the consideration of where to place the ramps, especially if you don't close down any point of the interchange.  Working with a clean slate is easy; working around existing ramps is quite a bit more challenging.

[mgk920]

Some are fine with the addition of C/D lanes, some with added direct ramps and some, if they are with surface roads, with more conventional style service interchanges.

Mike

[sparker]

It seems as if, with the presence of C/D lanes, the cloverleaf interchange -- if laid out with medium-high speed (>30-35 mph) loop ramps -- may still be a viable cost-cutting choice for rural "system" interchanges between limited-access facilities.  But I'd be willing to wager that there will be no more pure cloverleaves built as interchanges between freeways/expressways and surface roads. 

[Finrod]

I don't remember where I saw it, but one plan I saw for the new interchange between I-95 and I-42 used a cloverleaf, with C/D lanes for I-95, IIRC.  It seems that the cloverleaf kind of still is in fashion, just in a different form.

[froggie]

^ Perhaps this one? (from a Johnston County newspaper article)

[ethanhopkin14]

In Texas, it was sometime in the early 90s.  I remember a cloverleaf being built in the 80s, but since the 90s on, not only do they not build any more cloverleafs, they don't build any parclos nor do they build any more intersections at all that have any kind of loop in the exit/entrance ramp.  Plus, many old cloverleafs are being torn down for complete stack interchanges (with the exception of the new Loop 375/I-10 intersection in El Paso, where they built a stack and just left the old cloverleaf there).  Seeing a cloverleaf in Texas today is like seeing a white whale. 

[Finrod]

^ Perhaps this one? (from a Johnston County newspaper article)

That's exactly the one.  It has I-95 swerving east so they can build an intersection with future I-42 (currently Bypass 70) that's a cloverleaf in the middle of what's now an open field with a bunch of solar panels in the middle of it.  I just remember being amused when I first saw it: "they still design cloverleafs?"

[ixnay]

They're dangerous and terrible in urban areas. Weaving issues make them dangerous and cut their ability to handle high amounts of traffic. They also take up too much space in urban areas too. Parclos are so much better. Cloverleaf's are hot garbage in urban areas, but they're fine in rural areas. Not sure when we started to realize this but that's largely why they're not popular anymore.

Without reading the rest of this thread (yet) (I rarely visit the boards that aren't Northeast or Mid-Atlantic), I'll point out that I-695 in MD had three full clovers within +/- 2 miles (clockwise, MD 41, MD 147, and US 1 in the Parkville/Carney stretch).  Then the MD 147 interchange was slightly reconfigured, but you must still risk weaving on the counterclockwise loop at 147. 

At least PennDOT has CD roads on 95 at the (still) full clover at PA 420 south of the Philly airport.  Plenty of room IMO for that interchange, and the CDs have been there since day one in 1972(?).

ixnay

[ztonyg]

Arizona is a state where they never were really in fashion but have now completely fallen out of fashion.

I believe Arizona used to have 3 full cloverleafs:

I-17 / I-40 / AZ89A (former US89A) in Flagstaff
I-17 / Deer Valley Rd in Phoenix
I-17 / Bell Rd in Phoenix

All 3 are no longer full cloverleafs. I-17 / Bell Rd. is a diamond. I-17 / Deer Valley Rd is now a parclo. I-17 / I-40 / AZ89A has 2 loops, 1 flyover, and 1 right exit / left entrance ramp now.

Arizona doesn't even like parclos very much as it removed them completely at I-17 / Thunderbird Rd, I-19 / Ajo Way, I-19 / Valencia Rd, I-10 / Valencia Rd.

Arizona also never built the parclo that it intended to built at I-19 / Irvington Rd. The only recent build Arizona parclos are I-17 / AZ 74 and the "folded diamond" at US 60 / Meridian Rd

In addition, Arizona has also been steadily removing trumpet interchanges.

[jakeroot]

I've definitely thought about this for WA, but never put pen-to-paper:

The major full cloverleafs in Washington State right now are as follows:

1960s:
* I-5 @ JBLM Main & North Gates (Exit 120)
* I-5 @ S 56 St (Tacoma; Exit 130)
* I-405 @ NE 8 St (Bellevue; Exit 13B)
* WA-99 @ Des Moines Dr/14 Ave S (South Seattle)

1970s:
* I-205 @ WA-500 (Vancouver; Exit 30)
* I-405 @ NE 85 St (Kirkland; Exit 18)
* I-182/US-12 @ US-395 (Pasco; Exit 14)
* I-90 @ 148/150 Ave SE (Bellevue; Exit 11)

As far as when I believe it officially died in Washington State (since that's sort of the point of this thread), I would trace it to the late 1990s when I-5 @ WA-512 in Lakewood was reconfigured into a partial cloverleaf. The western loop ramps were conflicting with each other very heavily, so the southwest loop ramp was replaced with a triple left turn. However, no knew cloverleafs have been built since the 1970s, to my knowledge, so the late 1970s is when they fell out of fashion.

tl;dr... the negatives of cloverleafs in WA were clear by the late 1970s, and they stopped building them; by the 1990s, their issues became serious enough to warrant their reconstruction, primarily into partial cloverleafs. This practice continues today.

Following the modification at I-5 & WA-512 in the late 1990s, many cloverleafs in WA were modified to remove ramps, including...

* I-5 @ S 38 St (Tacoma; Exit 132), modified in 2001 into a Parclo B4
* I-405 @ WA-167 (Renton, Exit 2), modified in 2002 and 2018 with flyover ramps
* I-5 @ WA-18 (Federal Way, Exit 142), modified in 2012 into a cloverstack

[ethanhopkin14]

Arizona is a state where they never were really in fashion but have now completely fallen out of fashion.

I believe Arizona used to have 3 full cloverleafs:

I-17 / I-40 / AZ89A (former US89A) in Flagstaff
I-17 / Deer Valley Rd in Phoenix
I-17 / Bell Rd in Phoenix

All 3 are no longer full cloverleafs. I-17 / Bell Rd. is a diamond. I-17 / Deer Valley Rd is now a parclo. I-17 / I-40 / AZ89A has 2 loops, 1 flyover, and 1 right exit / left entrance ramp now.

Arizona doesn't even like parclos very much as it removed them completely at I-17 / Thunderbird Rd, I-19 / Ajo Way, I-19 / Valencia Rd, I-10 / Valencia Rd.

Arizona also never built the parclo that it intended to built at I-19 / Irvington Rd. The only recent build Arizona parclos are I-17 / AZ 74 and the "folded diamond" at US 60 / Meridian Rd

In addition, Arizona has also been steadily removing trumpet interchanges.

Some parclos still exist on I-10 from Benson east to the New Mexico state line, in Wilcox, Bowie and San Simon.

[doglover44]

I remember we used to have a Clover I-70 and I-75 near Dayton