Vehicle Cable Barriers - Short Lengths 1

[Boiler106]

Im designing cable barriers for my concrete commercial building, and am running across an issue.

It seems when you have short lengths of cable, the required number of cables increases beyond reason. I wanted to check and see if anyone had ideas or similar experiences.

Currently, i have a single 12ft run that requires 8 cables to be struck using a 5000 lbs vehicle at 5mph per PTI Tech Note 14. That's 8 cables over a 1 ft area which seems unreasonable.

Similarly, i have a single 20 ft run that requires 5 cables to be struck using the same setup.

Anyone with experience have thoughts on this?

 

[TehMightyEngineer]

Can you use an alternative barrier system?

Professional and Structural Engineer (ME, NH, MA)
American Concrete Industries

http://www.americanconcrete.com

 

[KootK]

Short cable = limited sag = more cables. That part makes sense. I've never done it but I wonder if one could create a sacrificial/flexible termination detail that would introduce greater sag into the system during an impact event

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[TehMightyEngineer]

KootK: Kind of a like a strap on a fall protection harness?

Professional and Structural Engineer (ME, NH, MA)
American Concrete Industries

http://www.americanconcrete.com

 

[KootK]

Quite like that TME. Maybe a stiff spring or something built into the anchorage that would increase sag during a vehicle stopping event.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[BUGGAR]

You are absorbing energy. The more deflection, the more energy absorbed. The longer (more flexible) the cables, the more the deflection and energy absorbed.

A car is rolling towards you. You are given the choice of stopping the car by pushing on it as hard as you can over a distance of 10 feet. Or having a distance of only 2 feet to push on it to stop it.

 

[BUGGAR]

In reflection, you can cut the distance required to absorb the energy with some plastic action like in earthquake design.

 

[Boiler106]

thanks guys but none of these follow PTI guidelines which wouldnt hold up real well in front of a judge.

Im also limiting my cable drift to 18 inches per PTI's recommended max.

 

[KootK]

Does PTI have some kind of monopoly on vehicle barrier design? I say build it out of bamboo shoots and slinkies if you can make the physics check out.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[thaidavid40]

Take a look at some of the pre-tensionable "springs" used as shock absorbers at the ends of some fall-protection lifelines. Maybe something there could be adapted for your situation?
Dave

Thaidavid

 

[TehMightyEngineer]

I think the problem that Boiler is running into is PTI likely has their barriers crashed tested to a AASHTO Test Level and thus going outside the testing limits would require retesting which PTI wont do for free.

Professional and Structural Engineer (ME, NH, MA)
American Concrete Industries

http://www.americanconcrete.com

 

[KootK]

I think the problem that Boiler is running into is PTI likely has their barriers crashed tested to a AASHTO Test Level and thus going outside the testing limits would require retesting which PTI wont do for free.

New to me. Thanks for setting me straight.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[Ingenuity]

Here is a paper/article that PTI references in their Tech Note #14 -on BC's: Link

...PTI likely has their barriers crashed tested to a AASHTO Test Level...

I am not so sure of that. A US-based manufacturer (and patent holder) of related components that I know well (and a PTI member) does NOT have such testing data and his components are used extensively in parking structures for BC applications. It was tough getting some mechanical test data/properties from him, let alone crash data.

I would be interested if PTI has indeed conducted such system/component testing to AASHTO or other.

 

[TehMightyEngineer]

Huh, interesting. I'll have to look more into this Ingenuity as I've never personally worked with a tensioned cable barrier.

Professional and Structural Engineer (ME, NH, MA)
American Concrete Industries

http://www.americanconcrete.com

 

[MacGruber22]

Interesting. Is the 12-ft just the distance between intermediate supports or is it literally only one span at 12-ft?

Assuming the single 12-ft span, I am getting (3) 1/2" 250 ksi cables with a demand capacity ratio of 0.94 (32.3 kips/90% MUTS) at 8" of deflection. TO me, the issue with your configuration is less about the cable capacity or deflection and more about the force imparted on the supports, which is obviously greater when the resisting barrier is stiffer. For your single 20-ft span, I calculate 2 cables with 14" deflection and 12" deflection for 3 cables. I hope my spreadsheet is not way off.

Also, most of your jacking force is going to go into overcoming the wedge seating loss. I see no point in pre-stressing other then to pull the slack out, and then backstress to 80% MUTS to fully seat the wedges.

Are you designing this as a retrofit or new building? Can your columns or whatever support do the work? That is often a major problem in retrofits for me.





"It is imperative Cunth doesn't get his hands on those codes."

 

[Boiler106]

MacGruber22, your results seem very different from mine. I am attaching mine which was based on Tech Note Issue 14 from PTI.

I have a single 12' span.

I get a factor of safety of 1.04. Although, the paper never mentions an acceptable factor of safety, i can only assume 1.6. Perhaps, this is where Im going wrong.

I get a tension value, T, of 32.04 kips. I have (2) thirty inch sq columns i will be anchoring into. This is a new project, and not a renovation.

I have found a presentation that gives me an alternate to backstressing by using a "Grabit" anchor that requires turn of the nut methods for tensioning. See pg 8-15 link here:

http://www.post-tensioning.org/Uploads/Conference/2009 Convention/Mon150Frederick.pdf

 

[Boiler106]

boo1, i am using a 1/2" dia gr 250 cable which is standard for parking structures in the us, the 33.4kips (not ksi) is based on an area of 0.153 sq in.

The code requires the vehicle impact force to be distributed over a 1ft section. It is typical to have 11 cables distributed over a height of 42 inches at roughly 4 inches on center, which gives you only 3 cables restraining the force.

 

[MacGruber22]

Boiler, I will dig into your spreadsheet this weekend and report back.

"It is imperative Cunth doesn't get his hands on those codes."

 

[Boiler106]

MacGruber, that'd be a huge help, thanks!

i should note that i calibrrated it with the examples in tech note 14

 

[MacGruber22]

Boiler - good news: Both of our spreadsheets produce approx. the same results, minus the F.S. I would have expected that your spreadsheet would calculate the minimum jacking force required based on some assumed seating loss (3/8"+/-). But, I have never used the Grab-it anchors so maybe that is all you use to anchor with. I assume that the Grab-it system uses something similar to a A328/A490 tension-control bolt, so that you don't have to rely on the calibration of the stressing equipment. Though, how that works when there are variable jacking forces used for the PT cable, is interesting.

Sorry for the digression...

Frankly, I haven't worried about a safety factor to yielding for this situation (my sheet compares the cable tension directly to 90% MUTS), because I tend to substitute a few cantilevered pipe bollards for super short spans. For longer stressing spans with a few intermediate supports, deflection tends to control, and my cable end reactions are usually no more than 12 kips or so.

Maybe I am flying in the face of better reasoning, but if I were to use a different F.S., I wouldn't use a safety factor to yielding of any more than 1.25 to represent 80% of MUTS. My logic for that consists of a few things:

1. We are applying a jacking force of 80% MUTS all day, every day to PT cables in concrete slabs, etc. Beyond a defect or ill-calibrated equipment, it is fairly certain that no yielding will occur during that process.

2. I am about to purposefully muddy the waters in order to prove my point....The "real" range of impact speeds that can occur in garages is quite a lot, particularly when you consider the results over that range of impact speed. My simple table below is roughly based on your 12-ft span and 3 cables resisting impact. If you consider a total F.S. (load & resistance) of 1.6 between the cable tension and the 5mph car, then the "ultimate" tension is 1.6*31800 = 50880 lbs. That "ultimate" tension is approximate the same car at 8 mph. I realize that this table is linear, but if you consider comparing between span lengths it is not.

12-FT SPAN
SPEED___IMPACT FORCE___TOTAL SINGLE CABLE TENSION (THREE CABLE GROUP)
5 mph____8350 lbs_________31800 lbs
6 mph___12000 lbs_________38200 lbs
7 mph___16400 lbs_________44600 lbs
8 mph___21400 lbs_________51000 lbs

In my mind, thinking about the difference between 5 mph and 8 mph "feels" super negligible in terms of what I imagine is the true bell curve of speeding cars in garages. I can only imagine what speed corresponds to even 1 STD deviation from the average speed. In short, we can't possibly design car barrier systems in parking structures to the naturally occurring load variations, so we provide a "load restrictions" in the form of 5 mph speed limit signs.

The above analysis is only considering the problem with cable strength at short spans. To me, there is just as much of a problem with deflection and long spans. One difference being the number of intermediate supports required to limit the deflection to 18" (or maybe some structure behind the cables). No one that I know applies a F.S. to the deflection at 5 mph, nor checks the deflection with a larger number of resisting cables at a higher speeds to consider what happens at some defined "ultimate" load. Strictly speaking, the deflection associated with cable barriers is not anything like the deflection serviceability of statically-loaded structures that we typically design, as violating it can be just as risky as exceeding the strength of the cables. Thus, I feel like the total F.S. or Reliability Index (if thinking in terms of AISC steel code) should be compatible between each limit state. In short, using as safety factor of 1.6 for cable failure as 5mph is not logically consistent to me when I think about the big picture.




"It is imperative Cunth doesn't get his hands on those codes."