OSHA Fall Arrest System 2

[jeffhed]

I have been asked to design a fall arrest system based on something a contractor has used in the past or maybe he just found this picture online. I have not been provided any information as far as why type of lanyards will be used etc. so I was going to design for the minimum 5000 lb load. However, this particular device was attached to the a web member of the wood roof trusses. I am concerned that during a fall event that the gang nail plates will not be sufficient to resist the lateral force at the ridge. On my specific project, the roof trusses at the peak have a single king post in the center, rather than two sloping web members. Using the bracket in the supplied picture, my king post gang nail plate would be put in shear, which I am uncertain how it would perform. I tried doing a google search for fall arrest systems thinking this was a picture of a proprietary system, if it is, I was unable to find it. Am I the only one that thinks this looks a little questionable? I am thinking it would be better to move the anchorage to the top chord, then if a fall occured, the bracket would be putting the truss top chord into compression like it is designed for already. I just finished reviewing the truss shop drawings and there were some revisions to be made so the trusses are not fabricated yet, so getting some of the trusses beefed up to support this falling load could still be done. Has anyone else designed some fall arrest anchorage like is shown in the picture?

 

[racookpe1978]

Gotta think about this one.

it doesn't look right. 5000 lbs is a 'bureaucratic" rule i know, but what ever yo are hooking a man tyo hav=s to follow the "law' even if that "law" is too high. Regardless, this s=doesn't look right. A nailed in-place split angle iron wrapped around a 2x4 holding a 1-1/2 pipe welded to an angle iron nailed to the next 2x4?

 

[KootK]

I doubt this will work by the numbers. The steel is surely fine but I don't see the nailed connection or the truss web working. A truss might be designed to carry 2500 lbs total. Asking one web member to deal with 2500 lbs shear is a lot. That's a Toyota Corolla worth of weight. The load has to be assumed to act in any direction, right?

I will concede that, given wood trusses as a starting point, this is a pretty robust system, relatively speaking.[pre][tt][/tt][/pre]

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[dhengr]

Jeffhed:
Actually, I think the device you show in the picture is installed up-side-down. The stl. end flgs. should be on top of the wood diag. members, so that you are not counting only on the shear/bearing of the 10 nails in the side grain of the wooden member, thus significantly improving its bearing and support cap’y. Furthermore, the wire rope cable should not be bent over the stl. locating/centering loop when the cable is under load, in a fall. That is, the locating loop should be on the high side of the cross pipe, which will happen when it is properly installed. Obviously, the wooden members must be able to support the design load, and that may be in question here. The system can fail at 5001 lbs., or you can use a lanyard which has been tested and proven to apply a lesser load in a fall.

 

[moltenmetal]

Whatever you do here, it is bound to be better than the fall arrest systems I see in use on most wood frame residential construction sites- that would be NOTHING.

 

[Archie264]

Racook, let me quibble a bit. If by "bureaucratic" you meant arbitrary then I don't think it is, though I also don't think that's what you meant.

My understanding is that the 5,000# number is based off of stopping a 200# man who had fallen 6' and stopping him over a distance of 6" using energy-dissipating fall-arrest gear. Oh, and with a factor of safety of 2 embedded within it.

I agree with dhengr that it looks like it might be supposed to be installed in the other direction.

jeffhed, I'm certainly no expert on the topic but I do think that the 5,000# requirement is from OSHA. So, if you're being asked to sign off on it you may want to make sure that it satisfies that criteria...and I have my doubts as to whether the 2x chords or that nailer plate will...

But...again, I'm not sure what the regulations actually are, or if there are different ones governing residential work. Good luck with it.

 

[jayrod12]

The 5000 lb load was determined I feel somewhat arbitrarily however I do believe it is a FOS of 3. A person dies when they are subjected to an arresting force of around 1600 lbs. Multiply that number by 3 and boom, basically 5000lbs.

If he is going to use special shock absorbing lanyards I believe the load can be reduced a certain amount.

 

[Duwe6]

OSHA = 5K lbs without engineering calc's. It is indeed a 'rule of thumb' and this number is not

 

[Duwe6]

OSHA = 5K lbs without engineering calc's. It is indeed a 'rule of thumb' and this number is not an 'engineering ' number. This is the 'Withstand' rating, thus the anchor is allowed to fail at 5100 lbs. For design purposes, use something like 1.5 to 2kips for the engineered value. Besides the fact that we die at something around 1.6 kips, all these systems are required to have some king of shock arrestor or deceleration device that limits the instantaneous loading to 900(?) lbs. Thus your anchor(s) will only see the resultant force from that 900 lbs.

 

[chrislaope]

As for the OSHA 5000 lb fall arrest load, it caused serious argument/debate in agricultural industry. Basically, it is determined within agricultural industry that it will be mandatory for all grain storage silos to have fall arrest system installed in future. As I understand, after a series of meetings/conferences, it is decided/determined that agricultural industry will setup its own standard of a 1800 lb fall arrest load (most probably it will appear in near future ASABE Standard), and seems like agricultural industry will not adopt OSHA 5000 lb fall arrest load in silo fall arrest system design.

 

[CANPRO]

I have seen written in OSHA that the 5000 lbs is for non-engineered applications, which I find odd because determining the capacity usually requires some engineering....unless its based purely on trial and error and testing. My personal opinion on the matter is that providing a design for 5000 lbs also adds robustness to a permanent anchor which will see wear and corrosion and may not be readily accessible for annual inspections (if the building manager keeps up with the annual inspection requirement at all). For that reason I would never design or specify a permanent anchor for less than 5000 lbs. Note that Canadian codes require that the anchor must be designed so that there is no permanent deformation at 2500lbs (yield) and no failure at 5000lbs (ultimate).

If it is engineered, in Canada, I believe the requirement is to design to 2xMAF (maximum arrest force). The typical lanyard will reduce the MAF to 900 lbs. In your configuration, I would feel comfortable designing to 1800 lbs and specifying that the anchor point must be used with a lanyard that reduces the MAF to 900lbs.

I understand your concerns about the lateral force on the gusset plate, but looking at your picture it looks like the web is bearing on the adjacent web....so the gusset plate may not be acting alone in resisting the fall arrest force.

 

[racookpe1978]

31 Oct 14 10:23
Racook, let me quibble a bit. If by "bureaucratic" you meant arbitrary then I don't think it is, though I also don't think that's what you meant.

My understanding is that the 5,000# number is based off of stopping a 200# man who had fallen 6' and stopping him over a distance of 6" using energy-dissipating fall-arrest gear. Oh, and with a factor of safety of 2 embedded within it.

Yes, your summary - and the additional comments that followed from everybody else! - are why I used the "bureaucratic" term in my very-poorly typed reply....

 

[Archie264]

Well, as Daffy Duck said, it’s not the fall that gets you, it’s the sudden stop.

Duwe6, I wouldn’t be too quick to think that the number doesn’t have calculations behind it. OSHA may claim that the 5,000# is arbitrary but it strangely correlates to the 200#/6’/6” parameters that I’m told they also require. Maybe they claim it’s arbitrary so no one will key on to the implicit factor of safety being only 2.

From physics class a falling object obeys the following:

a = g

v = gt

d = .5gt^2

therefore t = sqrt(2d/g)

So, @ d = 6’, t = sqrt[(2)(6)/32.2] = .610467 sec

v @ t = .610467 sec = (32.2)(.610467) = 19.657 ft/sec

using that…

E = .5mv^2 = (.5)(200/32.2)(19.657 ^2) = 1,200 ft-lbs

Setting energy = work:

E = Fd

Solving for d = 6” = .5’:

F = 1,200/.5 = 2,400#

With a safety factor of 2 that becomes 4,800# ~ 5,000#

But, if da dadburn gob’ment says it’s arbitrary I believe ‘em. Just like IRS audits…

 

[msquared48]

Irrespective of the fastener design for fall arrest, the thing that gets me in the picture is the worn hole through the OSB sheathing above. How much damage was done to the rope restraint during construction?

The systems I have dealt with in the past are permanent and through the roof with a sealed penetration, not temporary systems as this one obviously is, although it is still nailed in place.

Mike McCann, PE, SE (WA)

 

[jeffhed]

I understand the 5000 lbs is a minimum design force unless your are using other fall equipment that limits the force imparted to the person to 900 lbs or 1800 lbs. Then you can design the anchorage for 900X2=1800 lbs or 1800X2=3600 lbs. However, this information has not been provided to me, so that's why I am going with the 5000 lb force.

Msquared,
This will NOT be a permanent fixture to be used multiple times, it is a sacrificial anchorage that will be for construction only and then will be left in place. Looking online I have seen the systems you mention that bolt to the truss top chord and stick through the roof. I was thinking along those same lines by maybe taking the pictured anchorage device and moving it to the top chord so the fall will direct forces through the top chord of the truss. I am still not sure how that would affect the truss gang nail plates at the ridge. If there is sheathing nailed to the truss, maybe it doesn't affect the gang nail plates at all. But since this could save someones life, I want to be sure.

 

[KootK]

I was a truss designer in a previous life. Here's my take on your concerns regarding the connection plates:

1) The likely loading will be a force parallel to the truss chord. The top chord peak splices are generally fairly week because designers count on the two top chords butting together in compression. The plate there mostly just tension ties the chords down to the king post. Heel connections tend to be very strong however. A probable load path would be the top chord dragging the load down into the heel joint.

2) I was under the impression that the OSHA load needed to be applied in any direction. In this case, your main concern would be a load applied perpendicular to the truss top chord. 5 kip is a big load for that. The issue may have less to do with the strength of the plate than the tendency of the plate to tear a chunk of wood out of the chords in tension perpendicular to grain style.

3) I don't see sheathing helping much.

4) I like your idea of attaching to the top chord, for exactly the reasons that you outlined. It's the most capable place for the expected load.

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[chrislaope]

After I re-read the calculation of Archie264's post, I found that before we discuss this issue we need clarify one thing, i.e. what this 5000 lb means? If this 5000 lb means the ultimate strength of the fall arrest system, i.e. to design the fall arrest system fail at just over 5000 lb(say 5001 lb), then I would say the agricultural industry standard is not very far from 5000 lb standard because the agricultural industry requirement of 1800 lb is the required design force, i.e. with a safety factor of 2 (same safety factor as Archie264's post), the fall arrest system can be design to fail at just over 3600 lb(say 3601 lb), Apparently, 5000 lb in Archie264's post has different meaning with jeffhed (Structural) (OP)'s meaning, in which, jeffhed (Structural) (OP)’s 5000 lb means a minimum design force, i.e. with a same safety factor of 2, the fall arrest system has to be designed to fail at just over 10000 lb(say 10001 lb).

I think it is critical to clarify what this 5000 lb exactly means.

 

[steellion]

What is the height of the structure? I typically do not see fall arrest systems until the building is 3+ stories. Below that, contractors can use portable ladders for access and windows can be washed from the ground.

 

[Archie264]

Chris, I don't know; I'm sure no expert - or even novice - on the topic. But, as it was told to me OSHA requires the 5,000# capacity for the anchorage of the fall arrest system. I suppose then on top of that are the safety factors embedded within various proprietary anchorage systems' published capacities. I suppose the philosophy is that if the anchorage can take the 5,000# force then I suppose there is a degree of confidence that the anchorage could support a 200#/6'/6" fall arrest system. That's all supposition on my part and not one on which anyone should bet his life. The parameters of any life safety system are best verified for one's self.

 

[jeffhed]

KootK
Yes, the force is supposed to be in any direction. I was thinking of providing anchorage on each side, but that doesn't mean that guys will tie off to the correct side. The best would be to locate the anchorage at the ridge with the attachment protruding through the ridge. That way I could provide sufficient anchorage for all directions. The top chords resist the fall force parallel to the roof and the king post directs forces through the rest of the truss. I could design a large plate that could tie into the king post and both top chords and anchor it to the trusses with Simpson SDS screws.

chrislaope,
I think the 5000 lb force includes the factor of safety, and I believe it is ultimate. However, regardless of the level of force, there is still the question of what happens to the truss.

steellion,
This building is 3 and 4 stories tall. That is why the fall arrest anchorage is needed.