View Q&A

If I recall correctly, Dr. Sicking stated that he prefers other systems to a rigid barrier for safety performance, and also, I think he made reference to some study or documentation showing that lateral offset increases the likelihood of a more direct angle of impact.

We are discussing with one of our District Offices the location of a concrete barrier in a Freeway median.  They are building a 6-lane section through an interchange area for inclusion in a future (maybe far in the future) add lanes project on the corridor.  Ultimately, the permanent median barrier will be located at the center of the median, about 8 feet off the edge of the inside lane.  Until the corridor lane addition is achieved only four lanes will be present and a barrier at the center of the median would be about 20 feet from the edge of the through lane.  Another option for them would be to place two lines of temporary concrete barrier, each about 12 from the edge of an inside lane.

We all prefer a concrete barrier in this location due to the high truck traffic (9600 multiple units, and 1600 single unit trucks out of a total ADT of 34700 (2005)). We would like the reference to documentation of the increased likelihood of more direct impacts with increasing offset to the barrier, and also any comments.

Unfortunately, it is apparent that I confused many people with the discussion of the relationship between lateral offset and impact angle and another discussion of my concerns over the safety of safety shaped concrete barrier. I will recap my comments below in an attempt to clarify the now very muddy water.

HISTORY

Through the 1970's and into the 1980's, FHWA strongly discouraged the use of concrete barriers in medians more than 30 ft wide.
The concern was that impact angles would increase as the barrier was placed further from the traffic.
As a result of a number of truck penetrations of its W-beam median barrier, the New Jersey Turnpike Authority contracted with TTI, in the mid-1980's to develop a flexible barrier that could be used in wider medians and still provide TL-5 performance. During the process of trying to design a metal barrier that could contain a tractor-trailer truck, it became obvious that the resulting, "flexible" barrier would be rigid with respect to passenger cars. Based upon the results of this study, FHWA realized that it had a choice, either continue to require W-beam and Thrie-beam median barriers in wide medians and accept that heavy trucks will sometimes penetrate the barrier, or allow rigid barriers to be used in wide medians and accept the assumed increase I risk associated with higher impact angles.
About this same time, King Mak completed a study of pole and narrow bridge crashes that indicated a modest increase in impact angle as the hazard was placed further from the travelway.
Due to the extreme severity and high visibility of some of the cross-median truck crashes, (one such crash involving a gasoline tanker killed or seriously injured 33 people) the FHWA concluded that stopping such crashes on high volume freeways was very important.
Hence, FHWA's restrictions against the use of concrete barriers in wide medians were dropped.

The safety shape concrete barrier was developed by GM to prevent sheet metal damage during low angle impacts. The GM Shape as it was called incorporated a 13" high lower curb that would allow an impacting vehicle's tires to climb the barrier without any sheet metal contact for impact angles up to 3 or 4 degrees. Because GM used this barrier only on test tracks where the potential for traffic conflicts are low, most of their crashes were caused by inattention rather than avoidance maneuvers. Hence, the barrier was able to redirect impacting vehicles without damaging the sheet metal during most crashes.
When this barrier was introduced onto the highway, it was found to cause large numbers of rollovers.
In recognition of this problem, the state of New Jersey reduced the lower curb to a height of 10" and produced the widely used New Jersey Shape. In the 1970's,
the Southwest Research Institute (SWRI) conducted a parametric study to examine changes in the NJ shape that would reduce rollovers. They studied a number of revisions to the NJ shape, the first configuration was labeled Shape A.
SWRI looked at a number of different versions, finally arriving at what they deemed to be the best, Shape F.
This is the origin of the F Shape barrier and it basically involved reducing the height of the lower curb by another 3 inches.

During the 1980's, King Mak and I studied rollovers associated with concrete safety shaped barriers and found that safety shapes are not very safe.
This study concluded that concrete safety shaped barriers are the most dangerous barriers along our roadsides.
Clearly part of this increased danger is related to the rigidity of the barrier.
However, Mak's findings appear to indicate that the greatest portion of the increased risk was related to rollover frequency.

CURRENT RECOMMENDATIONS

It is generally accepted that there is a modest increase in impact angle as a barrier is placed farther from the travelway. However, the increase in severity with higher impact angles does not outweigh the attendant reduction in crash frequency as the barrier is moved farther from traffic. Hence, agencies are urged to place barriers as far from the travel way as possible within the available geometric limitations (considering roadside and median slope limitations).
Therefore, we strongly recommend that median barriers be placed in the center of the median whenever possible.
Further, our primary concern regarding the use of concrete safety shaped barriers is that the shape was originally designed to limit sheet metal damage and as a result, doesn't provide the maximum level of safety possible for a rigid barrier.
Under a currently funded pooled fund project we are attempting to verify the findings from Mak's study of barrier shape. If that study proves Mak to be correct, we will begin recommending that agencies abandon the safety shaped barriers in favor of more vertical designs.