Logged in as: Public User

Reasonable Variance of Cable Barrier Placement

State MO
Description Text

I am requesting your concurrence on a reasonable variance in the lateral placement of a cable barrier system.  

Here's the problem:

We are currently under construction of a 10-mile segment of high-tension (Gibraltar) cable barrier in the median of I-55 in Jefferson County, MO.  As a first item of work the 3-in. thick by 6-ft. HMA vegetative barrier has been placed throughout the project limits.  The subsequent work generally consists of auguring 12-in. holes and pouring the concrete sockets.

On this particular project, when an inspector went to provide a line for the contractor to bore holes, he realized that a 2009 resurfacing project had incorrectly striped a 13-ft. passing lane at the expense of half a foot of driving lane and half a foot of shoulder.  So now we are left with the situation I've depicted below, wherein, on average, there is a 3.6-ft. inside shoulder.

We realize that there is a documented 8.6-ft. dynamic deflection when a pickup impacts the Gibraltar system, so a backside hit on our 7.6 offset would result in a 1-ft. intrusion into the lane.  Of course this is an undesirable situation, so we set about looking for an appropriate remedial option.

The following is a truncated list of potential solutions, along with the primary reason they're infeasible.

Option 1 " Install the barrier 0.5 ft. further down the 6:1 inslope.  This is undesirable because the sizeable rocks being augured out of the socket holes are tearing out chunks of the HMA mat currently and a location closer to the edge only compounds this problem.  The project budget cannot tolerate additional costs associated with repairing damage to the mat.

Option 2 " Lower deflections by decreasing post spacing.  The project budget cannot tolerate this increased cost and the 50% increase in posts and sockets would compound the previously mentioned problem of damaging the HMA mat.

Option 3 " Fog-seal the existing rumble-stripe and restripe the passing lane to the correct width.  This option is undesirable because fogging a line with oil only partially obscures it.  In fact on rainy days, at night, and in certain sunlight conditions, the covered line is still visible and confusing to the driver.  Given the thin lift (NovaChip) layer upon which the stripe is painted, milling off the stripe is out of the question.

Option 4 " Use a driven (steel sleeve) socket 0.5 ft. further down slope.  Over 50% of the posts locations are in heavy rock fill where it is impossible to drive a sleeve.  If a pilot hole were to be augured, the same problem identified in Option 1 would present itself.

Option 5 " Stockpile the cable barrier system for installation after the striping is corrected with the next overlay.  This option is undesirable because MoDOT wishes to start realizing the safety benefit immediately, not in 3 to 4 years.

After exploring as many as 6 additional remedial options, a do-nothing approach (with respect to lateral placement) seems to have merit.  The reasoning is based on the low likelihood that a vehicle would be crowding the rumble-stripe on a 13-ft. lane, and the fact that this corridor is eligible for resurfacing in as little as three years, at which time the geometry of the pavement markings would be corrected and a new rumble strip milled.

So here's the question:

Would a 7.6-ft. available deflection distance be considered a reasonable variance in the lateral placement of the three-strand Gibraltar system on a 6:1 slope?

I look forward to your prompt response.

  • Guardrail
Other Keywords Cable Barrier
Date July 13, 2012
Attachment variance of cable barrier placement.gif


With all things considered and your reasons for being unable to make significant changes, you may have to utilize the proposed layout. I know that you understand that there exists approximately 1 ft of anticipated encroachment into the adjacent lane during extreme impact events with passenger vehicles (based on crash test data only). This additional encroachment could result a very slight increase in risk for partial vehicle contact with a very, very small portion of oncoming traffic, although the majority of the vehicles would not typically be traveling in the outer 1 ft of the traveled lane. Second, we do not believe that this proposed placement would significantly degrade barrier performance below that provided by the same barrier system installed 1 ft farther away from the lane edge.


Let me know if you want to further discuss via phone.

Date July 18, 2012


The below request has been reviewed and commented upon by both the FHWA Office of Safety in Washington D.C., and the Midwest Roadside Safety Facility (MwRSF) at the University of Nebraska. Both of these offices are authorities in the roadside safety field and their prompt, thoughtful responses are appreciated.  In both cases, the do nothing approach (Option 10) was shown to be a reasonably safe, albeit less than ideal solution for the cable barrier placement issues on Jefferson I-55.  Their comments are summarized below:

  • The FHWA stated that as long as this particular segment of I-55 showed no inordinately high level of crossover accidents from the northbound toward the southbound direction (backside hits), then the 9 other options explored and rejected were "adequate justification for the status quo." The accident history has been reviewed and the rate of crossover accidents is at a normal level.
  • The MwRSF also cited the 9 infeasible options explored as justification for installing the barrier as designed.  They did note that a "very slight" increase in incidental contact could be apparent in a backside hit, although "the majority of the vehicles would not typically be traveling in the outer 1 ft of the traveled lane."

If either of these positions have been misrepresented, please advise.

A subsequent conversation with the FHWA Missouri Division Office gave verbal approval to a design exception proposing Option 10 as a temporary solution until the pavement markings can be corrected with the next resurfacing project.

In light of all this, I recommend that the cable barrier be installed as shown on the plans and the pavement markings, along with the rumble strips, be corrected with the next pavement resurfacing.  If you find this solution reasonable, please work with the SE district liaison to draft and seek approval of a design exception.

Date July 20, 2012

Contact Us:
130 Whittier Research Center
2200 Vine Street
Lincoln, NE 68583-0853
(402) 472-0965
Email: mwrsf@unl.edu
The information contained on the Midwest Roadside Safety Facility (MwRSF) website is subject to change without prior notice. The University of Nebraska and the MwRSF is not responsible or liable, directly or indirectly, for any damage or loss caused or alleged to be caused by or in connection with the use or misuse of or reliance upon any such content, goods, or services available on this site.