Find Reviews by Make:
Ever been stuck behind acres of traffic waiting for a left turn signal to enter a freeway? If so, you know it’s one of the more annoying traffic scenarios out there. But a crazy scheme called “diverging diamonds” might just be the fix. NPR has a widget that makes it a lot easier to understand, as well as this memorable response:
Some folks say, ‘It’s crazy. Why did they put it in? Wrecks gonna happen. Why did they do that to us?’
[Hat Tip: Richard Chen]
25 Comments on ““Diverging Diamonds”: The Solution To Onramp Congestion?...”
Read all comments
So now I would have to potentially stop to let the other side go instead of potentially stopping to let left turns go. Don’t really see the difference.
John
During peak traffic times its also a problem for cars merging onto the highway, clogging the on ramp and then everything behind it…I don’t see how this helps at all.
I get how the pattern works, and I trust that engineers can model this out to prove that it really is more efficient than a standard setup. But I’m not sure thay I buy that this setup is more efficient than two roundabouts (that’s a roundaboot to you Canadians) replacing the lane-crossing points. No lights involved with a roundabout.
That said, I know nothing about the cost of a roundabout vs. the cost of a stoplight and lane crossover.
Roundabouts are really really good when they’re used in an appropriate manner. But get 3 lanes approaching and you’re not looking at a giant mess, not an efficient roundabout.
As for costs, in a complete reconstruction like this kind of thing would be, the difference between putting in a roundabout, signal, or other appurtenances would be very very minor compared to the major reconstruction work that needs to be done all around it.
Makes sense. I’ve seen this in Mexico but is was on a 4-lane city street.
I just ran into one of there is Springfield MO. Hwy 13 and I-44. It was new to me, I didn’t see any signage. Very confusing. I thought maybe it was rerouted for construction.
And why is this superior to a standard cloverleaf?
It takes up much less land area than a cloverleaf.
Why not just bridge the diamonds and eliminate divergence and signal lights???
I thought the same thing. Bridges would work well.
Takes up a lot less real estate.
Reminds me of a Single Point Urban Interchange (SPUI), like those at Sawmill Rd and I270 in Columbus, Ohio, and Telegraph Rd and I94 in Dearborn MI. Those significantly reduce the wait for a left turn and flow a LOT of traffic with a single set of signals. This looks to eliminate the lefts. Clever.
Oh, good. When I’m on the secondary road, I’ll still get to stop twice for the convenience of those on the expressway.
Why not just put in the damned cloverleaf and be done with it? No lefts, no waiting, no stopping.
—
rocketrodeo,
We have some of those in MN. They are an improvement in traffic flow but I think it requires a more massive bridge.
—
As it is, every now and then, someone ends up headed the wrong way on the freeway… Will this interchange tend to increase such incidents?
—
Yes, it takes up less real estate but MN has built plenty of diamonds where real estate was cheap. Penny-wise and pound foolish.
First, “the damned cloverleaf” is darned expensive. Most cross streets don’t warrant that amount of investment.
The SPUI does require a much bigger bridge, and has operational difficulties as well (cars drive past you on the right side, disorienting a lot of drivers until they’re used to it, and you have to get into the right opening on the far side).
As for your safety question: this design would tend to prevent people from going the wrong way onto the freeway. There are no turns into or out of the freeway ramps, just exits. And there are no turns at the signals, just straight through. It also has operational benefits over a diamond because the signals are just two-phase.
If people can get over the weird crossings, it will be an interesting way to deal with some intersections that have high volumes but no room for a completely separated interchange.
Why not just have Michigan Lefts everywhere
It’s a different approach to the traditional interchange, and – regardless of whether or not this is the best solution – I’d like to applaud these traffic engineers for thinking outside of the box. After all, if we stuck to the status quo for the sake of status quo, we’d still be driving exclusively on two-laned roads with at-grade crossings.
I can see this helping in locations where space for cloverleafs is limited. Beltline-Delta on the north side of Eugene has various environmental and cost issues that make a full cloverleaf system unlikely to happen, but this system looks like it would fit and rationalise flow quite a bit at modest expense.
I can see the point and think it is an interesting approach where there is not enough real estate for a cloverleaf. Kudos to the engineers for thinking outside of the box. I think it will save lives and reduce wrecks.
I love interesting developments in the field of traffic design (comes with being a roadway designer). Not that this is a new idea, but the major development the weave diamond like this has is that it cuts down significantly on conflicting motions. Note that the two intersections that are left in the design are just two cross-streets. Two simple movements, much more efficient, no turns to take up extra phases or cause accidents and delays. The problems with this design are things that don’t really show up in a simple diagram like the illustration. Having enough deflection to cross the streets effectively means you need to have enough width to be able to curve the road. And like I mentioned, it’s not a new idea: The state of Maryland just removed an interchange based on this layout at I-95 and I-695 north of Baltimore.
Something that a lot of people don’t realize about roadway construction, or even land development now, is that real estate is becoming a very significant cost. It’s now frequently cheaper to build structures than to buy right-of-way or incur other costs of land impact like environmental mitigation. So we’re seeing walls, bridges, and other structures going up where we would have just bought more land before.
As for the cloverleaf, I’ll be glad to see the end of it.
The only issue that I see is that you will have traffic slowing and exiting from the left lane that is, in theory, the high speed lane. I know that there are a few left-lane exits out there in interstatedom, but in my area they are rare.
Also, it looks like the folks exiting have a very sharp turn to make, and an immediate merging as those leaving the highway from opposite directions begin to share the same exit ramp.
It would be interesting to see some traffic studies which would tell us if these are really problems.
The left lane isn’t really considered a ‘high speed lane’ in an area between two close-together signals. The sharp turn you reference isn’t really that sharp, it just looks that way on this diagram, and the immediate merge could be a join (two lanes) that merge farther down the ramp.
Not entirely convinced – you are still sitting at 2 lights to cross the bridge.
Two things that would help – make the intersections as close to 90 degrees as possible (like Highway27 said) and make the median on the bridge opaque, to lessen the perception that you are on the “wrong” side of the road – more like a narrow one-way bridge.
With all due respect to the traffic engineers out there, most of the changes I’ve seen made on my local streets don’t seem to fix anything. The above proposal, as pointed out, makes the through traffic stop twice, vice the turning traffic, and replaces T-bone collisions with head-on collisions. Yes, it saves real-estate, but only if you are building a brand new interchange. To take an existing interchange and replace it with one of these would certainly cost a lot of money. Would it be worth the expense? Who makes that kind of decision, and what factors into the decision? Recently they widened my local freeway on-ramp (westbound to southbound) from 1 to 2 lanes. Did it fix the morning traffic problem? No, because the bottleneck is on the single lane road that is residential on my side of the freeway, but commercial once it crosses the freeway and widens into three lanes. The back up happens because of all the people getting off the freeway from both directions and heading west. If I can figure that out, why can’t the engineers? What am I missing? On the other hand, a few years ago at the same intersection they widened the freeway offramp (northbound to eastbound) from 1 to 2 right turn lanes. This seems to have helped a lot in reducing the back up onto the freeway itself, but of course doesn’t do anything to help the traffic on the road as I try to get home…
Sigs, as I said above, don’t believe the diagram so much. The benefit of this kind of alignment is that there wouldn’t be head on accidents, there would be the same likelihood of t-bone accident, but because you’ve simplified the intersection – by having 2 simple movements with no turns – you’ve really helped with people able to pay attention to the signal itself. In fact, going through the signal, you only have opposing traffic from a single direction.
The NPR story shows more of it, and it does seem that they’ve used the existing bridge, which is pretty wide on Google maps. Basically, they turned the left turn lanes on the bridge into the wide median, and use that space to effect the needed alignment changes. As for ‘stopping twice’, that happened already: There was a signal at either end of the bridge for the ramps on and off I-44. So they’ve taken 2 5-phase signals and made them 2 2-phase signals.
As for your roadway, unfortunately not every problem is fixable given the funding available to roadway projects. Sometimes the funding required would be astronomical to reconstruct a roadway / buy right-of-way / provide other capacity. Maybe what they could do was prevent backups onto the freeway, so that’s what they did. There are a lot of times where you know something else is an issue, but it’s outside of the scope of what you can do.
This could be a significant improvement for many interchanges. The cloverleaf sets up a conflicting movement where the exiting traffic must merge with the entering traffic in a short distance. It’s also relatively expensive, using quite a bit of real estate as well as a big structure.
Yes, there are two traffic lights in the crisscrossed diamond, but they don’t have a phase for left turns, and the two lights could be coordinated so that you’d never (well, hardly ever) have to wait for more than one of them.