Fall Protection

[ARLORD]

When designing fall protection the common load to use is 5000#. Does anyone know the basis for this value and whether it should be treated as a dead or live load. Are there any factors that should be used similar to impact.

 

[structuresguy]

I believe the 5000 lb load comes from OSHA requirements. When I did fall protection work, i always designed using it as live load, but I was also using ASD design, so it didn't really make a difference. Even if I normally used LRFD, I don't think I would factor up the load by normal LL factor.

Some have even argued that you can check ultimate stress level using that load, and forget about allowable stress and design codes altogether, with the argument that all you need is for the attachement and supporting member not to rupture in the event of a fall due to the impact load. I would tend to agree with that approach, unless the structure would be unsafe to use afterwards, or was carrying significant other loads at the same time. Kind of a judgement call here. So I would check both conditions, and use judegement from there.

 

[structuresguy]

Here is a link to the OSHA fall protection requirements.

http://www.osha.gov/pls/oshaweb/owadisp.show_document?p_table=STANDARDS&p_id=9730

 

[ChipB]

The 5000#s is a the load including impact. As far as stress goes, it's up to your client. Some clients allow the stress to go past yield, (but not to utlimate stress) and discard the element once it has been utilized. Kind of like breaking the glass to get to the fire alarm. Others allow it a 1/3 increase over ASD. You need to get their criteria in writing, and signed by them.

The main idea is for the member not to FAIL (yield is okay). If it fails, the person falls, and the client looks for someone to blame.

BTW, how many of us have seen workers tie off to a bar joist? What the??? If we see it, are we supposed to tell them it's not okay? I got thoroughly chewed out once for telling the supervisor that wasn't adequate.

 

[jberg]

It also depends if your state adopts their own standards. For example, Michigan states 5,400#. When I used to design and install fall protection systems, we used a factor of safety of 2.0, which is based on the following OSHA standard:

"(10) Anchorages to which personal fall arrest equipment is attached shall be capable of supporting at least 5,000 pounds (22.2 kN) per employee attached, or shall be designed, installed, and used as part of a complete personal fall arrest system which maintains a safety factor of at least two, under the supervision of a qualified person."

Per OSHA, the maximum force that can be imparted to the user by the lanyard is 1,800# when using a full-body harness. With that in mind, we would either used 3,600# per person (F.S.=2) or find a supplier that would guarantee the maximum force (they call it Maximum Arresting Force or M.A.F.) would be less than 900#, so you can design for 1,800#, and still maintain the factor of safety. We would provide a P.E. stamped design (our interpretation of "qualified person") and qualify how the system is to be used (i.e. only use equipment we list) to ensure the factor of safety of 2 is maintained.

With Michigan, though, I have not found this same rule. They actually state that the 5,400# is a Dead Load, not Live Load. That is interesting in my opinion.

Again, just my interpretation of the rules.

Joel Berg

 

[Bobbalott]

I've been chewing this over in my mind and I would like to have my understanding confirmed by someone with definite knowledge. Nonetheless I think I grasp what the 5,000 number means from a structural engineer's point of view.

OSHA states that for rigid installations such as eyebolts (as opposed to a horizontal cable line), the anchorage force is either 5,000 pounds or if the fall protection system is engineered (not the anchorage, the system of lanyards and shock absorbing devices), the anchorage shall be designed for twice the maximum dynamic load that will be imparted to the person attached (maximum arrest force). That maximum load a person is subject to is not allowed to exceed 1,800 pounds. So for engineered systems, the anchorage will not likely require a higher design load than 3,600 pounds. But the factor of safety is all ready include (FOS=2.0), so this can be treated as an ultimate load. Presumably, the load factor is 1.0 not unlike the value for E in seismic design.

By the way, the actual engineering of the fall protection system is an involved and tedious task and covers planning everything from the purchase of specific and proprietary equipment to detailed procedures for its use.

When we don't know what protection system the worker will employ, the 5,000 pound requirement is appropriate to use. If we assume the protection system limits the maximum force to 1,800 pounds, the effective factor of safety is 2.78 (5,000/1,800). Again, the 5,000 pound force is an ultimate load.

OSHA regulations refer to the "Strength" of the anchorage points and that seems to confirm my understanding that the forces are ultimate strength values. These days we engineers refer to strength as the maximum value that does not cause failure.

So a rigid anchorage in a non-engineered system should be designed for a 5,000 pound force with a factor of safety of 1.0 (ASD) or with a load factor of 1.0 (LRFD), but still applying the appropriate phi factors.

 

[jchi]

Here is a twist... Three guys tied off to the same beam. 5000lbs x 3 = Way Too Conservative.

How do others handle this situation: Permenant fall restraint beam for doing maintenance on the roof with (3) rolling fall restaint tie-off trolleys.

 

[jberg]

Bobbalott,

That is my understanding of the OSHA rules as well. I can only say that my fall protection engineering experience comes from working with a safety instructor with 20 years experience who has a P.E. and is a C.S.P. (Certified Safety Professional). Also, the logic always seemed to make engineering sense in my mind.

Jchi,

As far as multiple people, I would assume 2 people falling at the same time (M.A.F. of 1,800# or less applied) at midspan and then the remaining people's static weight (I used to assume 300# or so) at midspan. That way you do not get an overly conservative design. This concept is more for a "flexible" system like a cable, but I feel that it is not too conservative or unrealistic for a "rigid" system.

Finally, I would clarify your definition of "fall restraint". Please refer to:

http://www.latchways.com/default.aspx?item=34

If it is truly "fall restraint", then you do not need to apply the Arresting Force. I would use 2x the user weight, since the user cannot fall off the working surface. Now for "fall arrest", you would need to use the M.A.F.

Hope that helps.

Joel Berg

 

[MiketheEngineer]

All arguments have their merit. We use shock absorbing lanyards and shock absorbing lifelines. OSHA allows us to design to the 1800 lb load. BUT three guys on one tie is not genrally acceptable.... because as one guy falls he often pulls or knocks a second or third guy with him. You would too and I have seen it happen....

 

[Teguci]

When designing a horizontal lifeline, we stipulate that the rated Max Arresting Force of the lanyard and harness is not to exceed 900 lbs (fairly typical for shock absorbing lanyards).

We place this load at the center of the span and load the cable in order to determine the needed strength of the cable. For multiple people, the impact load from a fall will not perfectly coincide with other falls. Therefore a reduction of the loading is used (per documentation).

The final tension and angle of the cable is determined. The cable strength and anchorage is checked to make sure it is at least 2 times stronger (safety factor of 2).

For stiff tie-off connections (ex tie off to a beam) I would not use a reduction in the loading. However, there is going to be a considerable amount of energy that a beam can absorb while deforming under an impact load.

 

[stroyan]

I used to do a lot of fall protection system design. Typically with a beam and trolley anchorage system, we would use a facter of safety of 2 on the maximum arresting force of the self-retracting or shock-absorbing lanyard and add a 300# vertical point load representing another worker hanging already. Don't forget to consider a 10-degree angle of fall and the corresponding resulting horizontal load.

 

[BigH]

A colleague of mine (who is INTO fall protection) commented to me on the issue - he wrote and extracted from a reference the following (see attachment).