High-Volume Cloverleafs

[Rothman]

I-70 at I-81, Hagerstown, MD.

Traffic going to Baltimore/Philadelphia from the Great Lakes collides with traffic going to Florida from the Northeast and it's all forced onto one outdated cloverleaf.  This one frequently causes delays during rush hours.

https://www.google.com/maps/@39.6166939,-77.7883613,16z?hl=en&authuser=0

I-295 and I-195 in Trenton, NJ

https://www.google.com/maps/@40.1868096,-74.7161359,15z?hl=en&authuser=0

I have been traveling through the I-70/I-81 cloverleaf for nearly 40 years and I have yet to experience a bad backup there.

[MNHighwayMan]

Wouldn't it be possible to install some sort of in-ground sprinkler or heating tube system on overpasses so that constantly salting them wouldn't be necessary? I seem to recall this being a thing in Japan, but I'm not sure the tech has been utilised in the US.

The new I-35W bridge in Minneapolis sort of has a similar type anti-icing system.  But the technology hasn't matured in the U.S. and would be expensive to retrofit.

I can certainly see why old bridges haven't been retrofitted. That's a PITA and very disruptive. But on new overpasses, I think it would more cost-effective long term given salt's corrosive nature. I guess my hope is that MnDOT will start using (and pioneering) this technology more often. After all, they can't rely on cloverleafs forever.

EDIT: I just read that the original bridge had a de-icing system that might have contributed to corrosion. Is this true? Was this old tech?

It might have, in a minor way, but the NTSB report concludes that inadequately thick gusset plates (they were not thick enough to bear the load of the original bridge, never mind the additional load added over the years), and a lack of redundancy were the main reasons that the original bridge catastrophically failed. The anti-icing systems installed on the bridge were a relatively recent addition, so it's not terribly likely that corrosion from such systems had much to do with the bridge's failure.

[Beltway]

I have been traveling through the I-70/I-81 cloverleaf for nearly 40 years and I have yet to experience a bad backup there.

While it is a 1960s cloverleaf design, both highways have CD roadways that all the ramps and loops connect to.  That greatly improves merging conditions over a cloverleaf without CD roadways.

[Beltway]

It might have, in a minor way, but the NTSB report concludes that inadequately thick gusset plates (they were not thick enough to bear the load of the original bridge, never mind the additional load added over the years), and a lack of redundancy were the main reasons that the original bridge catastrophically failed. The anti-icing systems installed on the bridge were a relatively recent addition, so it's not terribly likely that corrosion from such systems had much to do with the bridge's failure.

Combination of a design error, major increases in structural weight from a previous modification project, and heavy loads of construction equipment and material loads on the bridge the day of the collapse.  Given that the alleged design error was in the original design, was it really a design error given that the original designers couldn't have been expected to design for the latter two factors above?
....

The National Transportation Safety Board determines that the probable cause of the collapse of the I-35W bridge in Minneapolis, Minnesota, was the inadequate load capacity, due to a design error by Sverdrup & Parcel and Associates, Inc., of the gusset plates at the U10 nodes, which failed under a combination of  (1) substantial increases in the weight of the bridge, which resulted from previous bridge modifications, and (2) the traffic and concentrated construction loads on the bridge on the day of the collapse. Contributing to the design error was the failure of Sverdrup & Parcel's quality control procedures to ensure that the appropriate main truss gusset plate calculations were performed for the I-35W bridge and the inadequate design review by Federal and State transportation officials. Contributing to the accident was the generally accepted practice among Federal and State transportation officials of giving inadequate attention to gusset plates during inspections for conditions of distortion, such as bowing, and of excluding gusset plates in load rating analyses.

https://www.ntsb.gov/investigations/AccidentReports/Reports/HAR0803.pdf

[hbelkins]

I have been traveling through the I-70/I-81 cloverleaf for nearly 40 years and I have yet to experience a bad backup there.

While it is a 1960s cloverleaf design, both highways have CD roadways that all the ramps and loops connect to.  That greatly improves merging conditions over a cloverleaf without CD roadways.

And a lot of "no merge area" signs as well. Weren't there supposed to be some big improvements to that interchange a few years ago? Last time I remember being through there was in 2015, and I don't recall that any significant work had been done.

My most frequent movements through that interchange have been 70 east to 81 north (requires use of a loop ramp) and 81 south to 70 west (no loop).

[kkt]

In Northern CA, the most obvious outdated/underpowered cloverleaf is San Jose's own US 101/I-880, which still features a surface-street jaunt from SB 880 to NB (compass WNW at that point) 101.  But there are others:  I-80/I-780 in Vallejo, despite C/D roadways along I-80, still backs up into the main lanes.  Originally that was the north end of I-680; that route was detoured onto the then-CA 21 freeway (Benicia-Cordelia) back in 1974, and the renumbered I-780 didn't see a lot of traffic until dense suburban development occurred along its length starting in the late '80's; now, it's a typical Bay Area mess, and that spills over onto I-80.  Actually, I-680 hosts a number of now-packed cloverleaves; the I-580/I-680 Dublin interchange -- albeit improved some time ago with a direct SB 680-EB 580 ramp -- still has 3 remaining 25 mph loops, including the very heavily used WB 580-SB 680 one (part of the Central Valley-Silicon Valley commute corridor) -- major modification is long overdue.  And I-680/CA 4 remains, with minor changes, as it was when opened circa 1963-64, although IIRC there was a local bond issue in Contra Costa County a few years back specifically addressing funding for revamping of that interchange.  Finally -- I-5/I-80 (with silent CA 99 overlapping I-5) in the Natomas area north of central Sacramento -- a cloverleaf with CD lanes and a single direct (WB 80>SB 5) ramp deployed in "turbine" fashion.  It was adequate until about a decade ago, when suburban development north of town overwhelmed the design capacity -- again, overdue for reconfiguration.

Why didn't they complete the us 101/i-808 movement? because it would have made sense ?

The lore surrounding the construction of the interchange has it that the property where the direct 880S (then CA 17/I-680) to 101N ramp would have been was the site of a major appliance warehouse whose owner had strong political connections and didn't want to move his business (and who had threatened to fight eminent domain proceedings tooth & nail).  The Division of Highways figured that SB 17 traffic going to Sunnyvale or Mountain View, northward via US 101, would simply use SSR 9 (now CA 237) from Milpitas over to Sunnyvale in any case, and that a direct ramp wasn't worth the deployment trouble.  So it was decided to simply use a half-diamond ramp to the old highway, Bayshore Blvd., and then require a right turn and about three blocks' worth of street travel before the ramp to NB 101 was reached.  That would have been easy if not for the SP industrial spur parallel to Bayshore and crossing it about halfway between the SSR 17 offramp and the US 101 on-ramp; the tracks are still there but only sporadically used today, while back in the '60's it served several major warehouses and a lumber yard -- and trains regularly blocked the freeway-to-freeway movement.  The appliance warehouse is no more; the property is occupied by the regional Coca-Cola distributor (who isn't going anywhere either!). 

Maybe, but the south 880 to north 101 and vice versa movements would be very low volume anyway, because it's a U-turn and most traffic needed to do that would have used 237 just a little farther north.  That's only a minor part of what's wrong -- the cloverleaf has short weaves with too much weaving traffic, the north 101 to north 880 (and vice versa) movements should have at least one more lane, the volume would easily justify a turbine instead of a cloverleaf. 

The real estate in the immediate area would certainly cost money for takings, however by San Jose standards it's not all that expensive.  I recall car junkyards, car servicing businesses, as well as the Coke distributor.  It's not like it was a cathedral or a hospital or a landmarked building; they're places that would accept a check for market value and moving expenses and go with it.

This is just leftover highway 17 that was built for San Jose when it was a farm town among orchards in the 1950s and hasn't been rebuilt yet.  Most of 17 got widened and the bad interchanges rebuilt when it was changed to I-880 in the 1980s, but this one and a few miles in Oakland were missed.

[WillWeaverRVA]

The cloverleaf interchanges on VA 150 at VA 10 and US 1/301 are excessively tight considering the high traffic volume in those areas. The latter needs some kind of reconfiguration considering it's just a couple thousand feet away from the complex interchange with I-95 and VA 895 - the loop ramps have an advisory speed of 15mph - but I'm sure that's impossible at this point.

Sure, VA 150 wasn't built to interstate standards (the RIRO intersections really suck when a truck decides to try to use them), but these interchanges are pretty rough in any case.

Also, don't get me started on the VA 288/US 360 interchange, which is a relic from when VA 288 began/ended here at one point. There's talk of reconfiguring this interchange to eliminate the pseudo-C/D setup in the northbound lanes and extend SR 2055/Commonwealth Centre Parkway to SR 654/Bailey Bridge Road, which would be quite nice because I hate having to risk my life exiting onto US 360 WB from VA 288 NB every day. 

[froggie]

I have been traveling through the I-70/I-81 cloverleaf for nearly 40 years and I have yet to experience a bad backup there.

While it is a 1960s cloverleaf design, both highways have CD roadways that all the ramps and loops connect to.  That greatly improves merging conditions over a cloverleaf without CD roadways.

And a lot of "no merge area" signs as well. Weren't there supposed to be some big improvements to that interchange a few years ago? Last time I remember being through there was in 2015, and I don't recall that any significant work had been done.

My most frequent movements through that interchange have been 70 east to 81 north (requires use of a loop ramp) and 81 south to 70 west (no loop).

There was significant work done ca. 2012, but it only covered one leg of the interchange....one of the legs you actually mention using.  The merge lanes onto westbound I-70 were lengthened, as were the exit lanes on eastbound I-70.

[adventurernumber1]

I think the I-20/I-26 interchange in Columbia, South Carolina may be a contender for this thread.

I am not certain of the exact traffic counts on these interstates at the interchange, but I would imagine that they are pretty high, because these are two significant interstates (at least for the state of South Carolina), and this interchange is very close to the downtown of the city of Columbia (which is somewhat large). This interchange is obviously a cloverleaf one. I would second other people saying that this interchange probably needs some work (which I believe I've heard before here on the forum), as it is quite outdated.

[Beltway]

Also, don't get me started on the VA 288/US 360 interchange, which is a relic from when VA 288 began/ended here at one point. There's talk of reconfiguring this interchange to eliminate the pseudo-C/D setup in the northbound lanes and extend SR 2055/Commonwealth Centre Parkway to SR 654/Bailey Bridge Road, which would be quite nice because I hate having to risk my life exiting onto US 360 WB from VA 288 NB every day. 

It is a design from the 1980s when the local area had almost no development.  VA-288 was completed between VA-76 and I-95 in 1990.  The interchange study is looking at a variety of schemes, such as the above, and/or building a semi-directional ramp from US-360 EB to VA-288 NB which would also eliminate the current loop and make the movement from VA-288 NB to US-360 WB much easier.

[WillWeaverRVA]

Also, don't get me started on the VA 288/US 360 interchange, which is a relic from when VA 288 began/ended here at one point. There's talk of reconfiguring this interchange to eliminate the pseudo-C/D setup in the northbound lanes and extend SR 2055/Commonwealth Centre Parkway to SR 654/Bailey Bridge Road, which would be quite nice because I hate having to risk my life exiting onto US 360 WB from VA 288 NB every day. 

It is a design from the 1980s when the local area had almost no development.  VA-288 was completed between VA-76 and I-95 in 1990.  The interchange study is looking at a variety of schemes, such as the above, and/or building a semi-directional ramp from US-360 EB to VA-288 NB which would also eliminate the current loop and make the movement from VA-288 NB to US-360 WB much easier.

Ah, gotcha. Hopefully the study actually leads to a conclusion, because the movement I mentioned is pretty dicey as it currently stands.

[roadman]

It might have, in a minor way, but the NTSB report concludes that inadequately thick gusset plates (they were not thick enough to bear the load of the original bridge, never mind the additional load added over the years), and a lack of redundancy were the main reasons that the original bridge catastrophically failed. The anti-icing systems installed on the bridge were a relatively recent addition, so it's not terribly likely that corrosion from such systems had much to do with the bridge's failure.

Combination of a design error, major increases in structural weight from a previous modification project, and heavy loads of construction equipment and material loads on the bridge the day of the collapse.  Given that the alleged design error was in the original design, was it really a design error given that the original designers couldn't have been expected to design for the latter two factors above?
....

The National Transportation Safety Board determines that the probable cause of the collapse of the I-35W bridge in Minneapolis, Minnesota, was the inadequate load capacity, due to a design error by Sverdrup & Parcel and Associates, Inc., of the gusset plates at the U10 nodes, which failed under a combination of  (1) substantial increases in the weight of the bridge, which resulted from previous bridge modifications, and (2) the traffic and concentrated construction loads on the bridge on the day of the collapse. Contributing to the design error was the failure of Sverdrup & Parcel’s quality control procedures to ensure that the appropriate main truss gusset plate calculations were performed for the I-35W bridge and the inadequate design review by Federal and State transportation officials. Contributing to the accident was the generally accepted practice among Federal and State transportation officials of giving inadequate attention to gusset plates during inspections for conditions of distortion, such as bowing, and of excluding gusset plates in load rating analyses.

https://www.ntsb.gov/investigations/AccidentReports/Reports/HAR0803.pdf

I have always found it interesting that the only portion of the original design records for the I-35W bridge that were never found were the gusset plate calculations.  Not only that, but while the bridge was supposedly designed for the eventual eight lanes of traffic, the original loading tests only accounted for seven lanes.

[Beltway]

I have always found it interesting that the only portion of the original design records for the I-35W bridge that were never found were the gusset plate calculations.  Not only that, but while the bridge was supposedly designed for the eventual eight lanes of traffic, the original loading tests only accounted for seven lanes.

That is odd, as the gusset plate calculations would be an important and vital part of the design plans for a truss span bridge.  The design for 8 lanes would (should) involve forecasts for a certain level of AADT and large truck AADT, and many Interstates carry far more AADT and TAADT than was originally envisioned.  After upgrades added weight to the bridge, again that is something that the original designers would not be accountable to.

[Beltway]

It is a design from the 1980s when the local area had almost no development.  VA-288 was completed between VA-76 and I-95 in 1990.  The interchange study is looking at a variety of schemes, such as the above, and/or building a semi-directional ramp from US-360 EB to VA-288 NB which would also eliminate the current loop and make the movement from VA-288 NB to US-360 WB much easier.

Ah, gotcha. Hopefully the study actually leads to a conclusion, because the movement I mentioned is pretty dicey as it currently stands.

It will be expensive, the various alternatives are in the 50 to 100 million dollar range.

[NWI_Irish96]

I-265/IN 265/I-65 in Southern Indiana is a pretty high volume interchange with three cloverleafs.  NB I-65 to WB I-265 gets particularly backed up in the afternoon rush.