Design for fall protection 2

[BENDOG]

I am designing a fall protection system for an existing (1960's) truck loading station at a chemical plant (see attached sketch). A person will be walking ontop of a tanker truck attached to an overhead monorail system. I am designing this for the OSHA 5000# requirement as a static force. Using ASD the existing rafter, bracing, column and anchor bolts are all overstressed.

I feel like I am too conservative on the design approach because the 5000# force is only going to be present for a short time, and after the fall you will just have the weight of the person plus any tools.

Has anyone else been involved with dsigning for fall protection? How have you considered the 5000# load?

 

[nuche1973]

Yup and you gotta design for the 5000lbs. Can you add steel or change the section modulus of the rafter? How much overstressed: 5%?

 

[BENDOG]

The stress ratio on the column is 1.5, rafter 1.25. I can reinforce them but the anchor bolts are going to be the biggest problem. I might be better off trying to drop another post at the end of the rafter on the right side. This probably would eliminate overstress in the column and anchor bolts.

 

[oneintheeye]

I share your pain as I have had a similar problem in the past. One proposal we put was that we could design so that in the event (and in our case it was unlikely) that a person fell then the members would buckle but would not fail entirely and so would halt the fall.

 

[DTGT2002]

Are you comparing your loads to 2:1 against yield or ultimate?

 

[BENDOG]

I'm treating the load as a static load and comparing the stresses to the allowable ASD stresses in AISC 1989 green book.

 

[JStephen]

I think the 5,000 lb value includes a factor of safety which doesn't necessarily need to be duplicated in your allowable stresses. Check the wording on the load requirement.

Also, you can use lanyards that limit the force and design for that force rather than design for the 5,000 lbs- should be addressed in OSHA as well.

 

[CTW]

I agree with JStephen. Also, search the site as there have been other fall protection threads that address loading and stress requirements.

 

[nevsam12]

Per OSHA, the anchor point is to be designed to 5000# at failure. So, I calculate the ultimate capacity of the connection and compare that to 5000#.

As far as members are concerned, OSHA does not say 5000#. So, I would design the members for the load recommended by the Lanyard Vendor - if I remember, it was something like 1900#.

I would like to see others respond, as to how they do it.

 

[BAretired]

The only fall protection system I remember designing was one in which several workers could clip onto the monorail simultaneously, so I designed for 5,000# spaced at 4'-0" centers.

Make sure that only one worker will be fastened to the monorail at a time.

BA

 

[JSA2]

My experience has been to design for 1000# at a working stress level and then use the 5,000# as a "breaking strength" of the system. I have designed window washing tie-offs for this criteria.

 

[jberg]

There are some states, Michigan for example, that actually require the fall protection anchorages to be designed to a 5,400# dead load.

Many states reference the National OSHA which requires either the 5,000# requirement (without certification) or a factor of safety of 2.0 (with certification).

In the past, I have used the factor of safety of 2.0 requirement that is in OSHA and ANSI requirements. Per OSHA, the maximum arresting force (MAF) can not be larger than 900# when using a body belt or 1,800# when using a full-body harness. The MAF is the maximum force that will be imparted to both the user and the anchorage or lifeline.

With that information in mind, I typically design for 1,800# (assume full-body harness) with a factor of safety of 2.0. With the 2005 AISC spec, the factor of safety against yield for ASD is the omega factor. It maybe sound a bit convoluted, but I use the actual MAF and then determine what the maximum capacity ratio of the steel can be, assuming a factor of safety of 2.0, since the omega factor in flexure is typically 1.67.

In my opinion, you can either place the factor of safety on the load or the resisting elements, but I feel it is overly conservative to use a "breaking strength" load as JSA2 stated with an additional factor of safety per AISC or the governing design criteria.

Sorry for the lengthy rant, but I have had many discussions with other engineers in the past about this and I always marvel at how everyone has a different take on the design. I do wish that OSHA would rewrite these requirements to better clarify the intent of the regulation.

JWB

 

[JSA2]

jberg,

Thanks for the contribution. Just for clarification, when I used the 5,000# load there was no safety factor associated with it. It truly was a breaking strength. In other words, the first element in the support system load path will physically break at 5,000#. I would even consider strain hardening of a ductile steel element to get to justify this 5,000# number.

 

[BENDOG]

Thanks for everyone's input. I felt that I was being too conservative. For my particular situation I am concerned about the overstress in the anchor bolts and I do not know any detail about embedment, they might even be hook bolts. The client is OK with dropping columns to the right to eliminate the cantilevered bent.

 

[Teguci]

Recommended reading - "Introduction to Fall Protection" - J. Nigel Ellis

For an engineered system use twice the MAF. MAFs vary based on the safety equipment being used and can be as low as 600 lbf

 

[jberg]

Teguci,

I forgot about Ellis' book, which really helped me when I was first learning about fall protection. You are correct that MAFs do vary from vendor to vendor, I have seen 900# in many applications. In the end, MAF times 2.0 is much more efficient for design than using 5,000#. It is great to have other engineers sharing the same approach that I have been using.

JWB

 

[BENDOG]

Teguci,
I just obtained a copy of the book from our construction department. Chapter 7 covers anchorages. Thanks for the tip.