Aluminum Pipe Rail Question 1

[jheidt2543]

I’m checking the structural design of a standard pipe handrail, sounds simple enough but… The 1-1/2” pipe railing is made of Schedule 40 6061-T6 Aluminum. O.D. = 1.900”, I.D. = 1.610” Wall thickness = .145”.

From the 1960 edition of the Alcoa Structural Handbook (OK, maybe I should update my reference, but humor me) there is a supplement in the back with two ASCE papers. The pertinent one is titled “Suggested Specifications for Structures of Aluminum Alloys 6061-T6 and 6062-T6, Report of the Task Committee on Lightweight Alloys, Committee on Metals, Structural Division”, ASCE Paper 3341, December 1962. (How it got into a book published in 1960 must be an interesting story).

The paper covers aluminum specifications for allowable stresses, design rules and fabrication procedures for structures built of aluminum alloys… Two sets of allowable stresses are provided; one for bridges… the other for buildings; I’m using the building stresses, which gives the following stresses:

I-18b Allowable Tension in bending = 11.7 ksi

I-18b Allowable Compression in bending = 16.2 ksi

Allowable Shear = 12 ksi

The aluminum rails check out OK, but the standard 3’-6” post, with a required 200 lb. horizontal load fails in bending. I would have to change to a 2” Schedule 80 pipe to make the post work. This just doesn’t seem reasonable.

Am I missing something here? It doesn’t seem that I’ve seen aluminum handrails with size horizontal pipe and a different size vertical pipe.

 

[boo1]

The reduction in welded material strength occurs only in the one inch haz area. Consider looking at design guide:

http://www.hollaender.com/files/1/Tech_Data/handraildesign.pdf

 

[Ron]

jheidt2543...no, you're not missing anything (other than your info is outdated, as you noted)

For a 1-1/2" Sch40 pipe, the bending moment would be 8400 in-lb and the bending stress would be 25.8 ksi. Allowable bending stress is 21.2 ksi, so it is overstressed. If you have a weld at the bottom of the post, then your allowable stress goes to 12.1 ksi, so way overstressed.

Since handrails have gone to 42", it has been a problem getting aluminum rails to pass without large sections. Most often, we go to a 2" square tube with a 0.095 or 0.120 wall thickness.

The values I'm using are from the Aluminum Design Manual, 2005 edition (the latest one). This gives a 35 ksi yield with a safety factor of 1.65 on unwelded section and a 20 ksi yield with 1.65 safety factor on welded section.

 

[jheidt2543]

Thanks Ron, you confirmed what I was finding, but I was sure I've seen the 1-1/2" Schedule 40 posts. The information from boo1 seems to find a way around this by using "Load Proportion Factors" for the interior posts. Then the Schedule 40 seems to work, but the corner and end posts still need to be Schedule 80 as I read the paper.

WOW! You guys are scream'n fast!

Thanks again.

 

[JStephen]

Note that Hollaendar's data is presumably for their bolted connections, not for welded joints.

OSHA specifies minimum handrail member sizes, but also specifies loads, and it can be difficult to show that using those member sizes will support those loads. It also doesn't help that it is not obvious if a factor of safety is to be added to those loads, or included in them, or what. Anyway, it can be helpful to consider multiple posts rather than a single post, etc.

 

[DaveAtkins]

You are not missing anything. I just went through the same thing this week. A 1-1/2" diameter Schedule 40 aluminum pipe does not work as a guardrail vertical.

DaveAtkins

 

[jheidt2543]

JStephen:

You are correct and allowable stresses for bolted/rivited construction are different (higher) than welded construction, but consideration of proportional loading to other verticals seems to make sense. Anyone have either of the two references cited in Hollaendar's paper?

 

[Ron]

jh2543...I reviewed, a few months ago, a "design" by a licensed professional engineer. It was for a guardrail design on a balcony. It's requirements were the same as yours. I rejected the design because the materials were overstressed. The "engineer of record" on the project stated that in his "professional opinion" the handrail was sufficient for the application, even though it was "theoretically" overstressed. The building official bought his argument. So we have a non-engineer accepting deficient engineering and it becomes acceptable under the code....this is just WRONG!

 

[boo1]

Check the stresses at the base (unwelded condition, 24 ksi) and the intermediate support (welded condition, 14 ksi)

 

[zcp]

Even better is when they want that same section embedded in concrete with no coating / treatment / etc.

ZCP

http://www.phoenix-engineer.com

 

[Ron]

jheidt2543...I agree that load sharing helps the intermediate posts but not the ends. Sometimes the configuration and attachment at the ends or stair returns can get you by this as well.

 

[jheidt2543]

Guys,

Even with load sharing, if the pipe is welded to an anchor plate such as to anchor it to the top of a concrete wall, the weld will not work @ 14 ksi with Sch. 40 pipe.

Now, I heard an argument around this from another engineer in which he states "since this is an OSHA safety requirement, not a building code requirement, you can use the ultimate strengh not the allowable strength, then it works." He is looking for the reference for this. But, wow!, any one else heard this? Are you willing to put your seal on it?

 

[DaveAtkins]

The 200 # point load IS a building code requirement.

DaveAtkins

 

[boo1]

Piping System Manual from NAAMM can be found at:

http://www.naamm.org/landing_pages/AMP_521-01.pdf

jheidt2543, are not OSHA, building code, ASCE, ANSI, and ASTM goals to provide safe products. Just increase the section till it works.

 

[Ron]

jh2543...no, the weld does not work at 14ksi, but if you use a weld plate, you can provide gussets. Otherwise, core the concrete and grout the post in place, then your 14ksi is OK.

Agree with Dave...it is a building code requirement (200 lb point load or 50 plf in any direction).

Ron

 

[jheidt2543]

Thanks again guys for the input and boo1 that reference is just the thing!

I'm doing the calcs for the fabricator, who is locked into a bid. The EOR is the one that specified the pipe and connection detail. That's the reason for dancing around this, the fabricator doesn't want to eat the change and neither does the EOR. The sample details in boo1's reference may offer a graceful and inexpensive way out.

 

[swm25]

I did the same design recently and the 1-1/2" pipe didn't work for a guardrail post. The strength deduction for the weld occurs at the same place as the max moment of the vertical posts. Unfortunately the one I was checking was existing as the client was ambitious I guess. I had to make them weld gusset plates at the bottom.

 

[jheidt2543]

OK, guys, I think I've found the solution from the 2006 IBC, Section 1607.7.1.3 Stress Increase.

"Where handrails and guards are designed in accordance with the provisions for allowable stress design (working stress design) exclusively for the loads specified in Section 1607.7.1, the allowable stress for the members and their attachments are permitted to be increased by one-third."

So, 14 ksi x 1.33 = 18.62 ksi allowable. Now it works!

Wow, what a long way around this has been! But, I guess it pays to read all the codes. Thanks for all the help and suggestions.

 

[jberg]

I thought there was talk that the IBC 2009 removed that stress increase provision for handrails.

JWB