For police, not wearing seat belts can be fatal mistake

[Beltway]

What I was getting at by my snarky snippity is that you seem to maintain both that the Texas example is sufficient for a freeway and also that it is substandard for a freeway.  Also, I keep saying that crossovers are fine assuming sufficient accel/decel lanes, yet your main defense in saying the Texas example is a different beast is that it has sufficient accel/decel lanes.

No, you seem to be using a highly unusual Interstate interchange setup, to say that there is nothing wrong with making median crossovers common on the Interstate system.  That is what I meant by the fallacy of "argument by exception", that something is ok to do if there can be found even one possible case of it being done.

[cpzilliacus]

What I've been trying to drive home is the fact that crossovers do exist on freeways, and that many of them function just fine–not as well as a flyover, perhaps, but just fine for the price tag.

As I have suggested up-thread, I assert that the potential for a serious wreck is unacceptably high to allow the general motoring public to use at-grade crossovers on freeway-class roads. 

The speeds (and frequently the traffic volumes) are just too high to allow such use. 

This is also redundant, but I am saying it again because I feel so strongly about it - I have used them sometimes (while on official business related to my job) with a bright amber strobe beacon on the roof of my truck, and I always dislike them intensely, for the drivers behind me (in spite of precautions taken) still don't seem to anticipate that a vehicle might be slowing to pull off on the fast shoulder.  And after making the turn at the crossover, then there is the matter of getting enough of a break in traffic to accelerate up to speed (and execute a safe merge with the fastest freeway traffic, which is usually in the left lane).

State DOTs and similar (toll road) agencies do plenty of other "expensive" things on freeways (that they might not do on roads of a lower functional classification) to keep them moving safely.  Even though a "U-Turn" flyover is rather expensive, I strongly prefer that for public use instead of an at-grade crossover.  There's also the matter of "expectations."  Drivers do not "expect" vehicles to be slowing in the left lane to turn onto a crossover, nor do they expect a vehicle to come out of a crossover to merge with traffic.

Years ago, the Ohio Turnpike used the best signing at its crossovers (of which there are many, probably especially for winter maintenance), which read something like this:

"No "U" Turn - Crossing Prohibited - Violators Prosecuted.

The last time I drove the Turnpike (from the Pennsylvania border to Toledo, a few years ago), those signs had been replaced by something that was probably more compliant with the MUTCD.

[kphoger]

No, you seem to be using a highly unusual Interstate interchange setup, to say that there is nothing wrong with making median crossovers common on the Interstate system.  That is what I meant by the fallacy of "argument by exception", that something is ok to do if there can be found even one possible case of it being done.

If those crossovers had accel/decel room and could handle a truck, then I'd be all for them.  However, that's a far cry from simply removing "Authorized vehicles only" signs.  So, yes, I'm in favor of crossovers; but, no, I don't think most of the ones our Interstates have are safe for common use.  In fact, much of our Interstate system really doesn't have sufficient space in the median to easily accomodate turning around.

I always dislike them intensely, for the drivers behind me (in spite of precautions taken) still don't seem to anticipate that a vehicle might be slowing to pull off on the fast shoulder.  And after making the turn at the crossover, then there is the matter of getting enough of a break in traffic to accelerate up to speed (and execute a safe merge with the fastest freeway traffic, which is usually in the left lane).

State DOTs and similar (toll road) agencies do plenty of other "expensive" things on freeways (that they might not do on roads of a lower functional classification) to keep them moving safely.  Even though a "U-Turn" flyover is rather expensive, I strongly prefer that for public use instead of an at-grade crossover.  There's also the matter of "expectations."  Drivers do not "expect" vehicles to be slowing in the left lane to turn onto a crossover, nor do they expect a vehicle to come out of a crossover to merge with traffic.

As for slowing down and speeding up on the shoulders, I totally understand and agree.  I've only had to do that a few times, and it's risky business.  Even slowing to a stop on the right shoulder freaks me out a little bit.  This is why I would only advocate crossovers with proper acceleration and deceleration lanes; having a full lane would ensure there's plenty of pavement, and would also increase drivers' expectation of seeing a turning vehicle (since it would function basically as a tight left exit).

As for people not expecting to see people using crossovers, I'd say that is a cultural phenomenon based on the fact that turning around on freeways is generally prohibited in this country.  It takes me by surprise when I see someone illegally use an Interstate crossover.  When driving in México, OTOH (where turning around is permitted), I fully expect to see it, and just include that as part of defensive driving.  Similarly, on non-Interstate expressways in the US where turning around is permitted (for example, US-54/US-400 between Andover and Augusta, KS; or US-60 between Springfield and Mountain Grove, MO), I fully expect to see it.  Most of the crossovers along those two US highways lack what it would take to make them safe for a freeway (not to mention the cross traffic), but my point is that our expectations are not concrete.  Just because American freeway drivers aren't expecting cars to turn around, that doesn't mean that no drivers on any freeway either don't or wouldn't.

[Alps]

This post modification stuff isn't really good for the group, IMHO.  Granted, we have some fights here, it's really silly and don't help anyone or anything.

I can do it as well:

I married John Lansford as soon as it became legal in DC

See how I got fucked by a goat.

Rather easy, yet rather juvenile.  That said, it isn't for the mods to do, but rather for the individuals to act like the mature adults they are supposed to be.  The only thing I would ask of the mods, and I've seen it from time to time, is a chastising of those who continue to act like asses here.

We do chastise. Keep in mind this is often done via private message, or even offline if we know the perpetrator, so most of you may not be aware of it. We also have frequent discussions about the more regular "problem" customers. And while many people may feel that there are about a dozen people worth getting rid of in order to make the forum less obnoxious, we would be losing a lot of knowledge by doing so. We've had this discussion and we're erring on the side of inclusiveness. If you're obnoxious AND don't contribute anything, then we take action.

[Beltway]

This post modification stuff isn't really good for the group, IMHO.  Granted, we have some fights here, it's really silly and don't help anyone or anything.

I can do it as well:

I married John Lansford as soon as it became legal in DC

See how I got fucked by a goat.

Rather easy, yet rather juvenile.  That said, it isn't for the mods to do, but rather for the individuals to act like the mature adults they are supposed to be.  The only thing I would ask of the mods, and I've seen it from time to time, is a chastising of those who continue to act like asses here.

It is nothing to be proud about, but I learned enough even while a teenager that I could be as nasty and abusive as most anybody on Usenet or anywhere else if I chose to.  But why?  It's useless and wouldn't prove anything of value, IMHO.

Thie group has a lot of potential, IMHO.  Unmoderated groups such as on Usenet obviously have a problem in that there is no moderation.  Moderated groups can be a very productive and stable environment, but there is always the possibility that the moderators themselves can create issues that could make things as bad or worse than when there is no moderation.  So far in the two years that this group has had very high numbers of postings, I would say that things have gone pretty well overall.  I hope that it stays that way.  The moderators do a lot of work and should be commended.

[Beltway]

No, you seem to be using a highly unusual Interstate interchange setup, to say that there is nothing wrong with making median crossovers common on the Interstate system.  That is what I meant by the fallacy of "argument by exception", that something is ok to do if there can be found even one possible case of it being done.

If those crossovers had accel/decel room and could handle a truck, then I'd be all for them.  However, that's a far cry from simply removing "Authorized vehicles only" signs.  So, yes, I'm in favor of crossovers; but, no, I don't think most of the ones our Interstates have are safe for common use.  In fact, much of our Interstate system really doesn't have sufficient space in the median to easily accomodate turning around.

To design a median crossover for 70 mph design speed mainline --

Accell and decell lanes each at least 1,500 feet length.  If large trucks are allowed, at least 2,500 feet.  This would be from and to the left lanes of the mainline, let's keep in mind.

Curve between decell and accell lane --
Design Speed 15 mph
e= 4% [superelevation]
Minimum Design Radius (ft) = 45
http://www.mhd.state.ma.us/downloads/designGuide/CH_4.pdf

Looks like it would take at minimum 110 feet of median width to accomodate even a fairly sharp curve on the crossover.

A 60 foot wide median is plenty wide enough for safety, as a general feature on a rural Interstate.

For a u-turn, it looks like in many cases it would be cheaper to build an overpass bridge and a ramp with right-hand terminals.

[Kacie Jane]

For a u-turn, it looks like in many cases it would be cheaper to build an overpass bridge and a ramp with right-hand terminals.

I'm baffled by the logic that it would be cheaper to build an overpass than to build the lanes 50 feet further apart.

[Beltway]

For a u-turn, it looks like in many cases it would be cheaper to build an overpass bridge and a ramp with right-hand terminals.

I'm baffled by the logic that it would be cheaper to build an overpass than to build the lanes 50 feet further apart.

With a 70 mph design speed, the transition to the widened area of the median would be about 1/2 mile long, and the cost would be for extra right-of-way.  That is for a very minimal 15 mph curve, it really should be more than that, IMO.

Also, include the cost of the accell and decell lanes, about 2,500 feet of each if large trucks are allowed.

And STILL that would mean left-hand terminals on the accell and decell lanes. 

[kphoger]

For a u-turn, it looks like in many cases it would be cheaper to build an overpass bridge and a ramp with right-hand terminals.

I'm baffled by the logic that it would be cheaper to build an overpass than to build the lanes 50 feet further apart.

With a 70 mph design speed, the transition to the widened area of the median would be about 1/2 mile long, and the cost would be for extra right-of-way.  That is for a very minimal 15 mph curve, it really should be more than that, IMO.

Also, include the cost of the accell and decell lanes, about 2,500 feet of each if large trucks are allowed.

And STILL that would mean left-hand terminals on the accell and decell lanes. 

Wouldn't new ROW need to be purchased anyway for an overpass with right-hand points of access?  How much are looking at in each case?

In my experience, I don't think I can say I've seen a crossover that was safe for large trucks to use.  The amount of pavement needed for a loaded truck to reach safe merging speed is much more than for a passenger car.  Having said that, I think many of our existing entrance ramps are also insufficient in that regard (Texas, I'm looking at you); I mean, there are some at which a passenger car struggles to reach safe merging speed.  In the I-10/I-20 example, the situation is helped by the acceleration lane's becoming part of the I-20 East mainline–and therefore a lane that doesn't even need to merge until traffic is clear, unless the driver is taking I-10.

PS:  For the youtube junkies, here's one of an Australian road train doing a U turn on a two-lane road.