Replies

1. 
   Hokuto | Jun 25, 2014 09:10pm | #1

   I'm just a homeowner/amateur DIY, so YMMV, but

   1. Dunno how code is interpreted re #200 pounds vertical force; I've only heard about #200 lateral force.

   2. 2x10 seem a bit narrow to be used alone, and too wide to be used doubled. The carpenters who built our first deck stairway

   used single 2x12 as treads and they worked well for sixteen years (had to rebuild it for unrelated reasons).  For the next deck stairs, tho, I used doubled 2x6 as treads.

   3.  Not sure about the issue with PT wood. Are you personally opposed or does it have to do with code? Our entire deck as well as main front stairs are all PT and it's done us well. I would opine that the decision has to do with local climate, kind and amount of wood-eating bugs locally, and aesthetics. Since arsenic is no longer an ingredient, I don't read about serious environmental issues, although I guess it depends on what one reads. The main construction issue I've read about re PT is need for all SS fasteners, special hangers, and care to insulate any other metalwork from contact with the wood.

1. 
   clarkster75 | Jul 17, 2014 12:14am | #18

   Awesome

   I think I will only reply to the treaded wood issue.  There were some awesome answers on the other stuff though....

   Obviously, you are going to pay more for treated, and need to use special fasteners.

   Treated is weaker cause it's baked in chemicals.

   If the deck is going to breathe (off the ground, or in a dry climate) that will obviously aid standard lumber. 

   If you are close to the ground, or sprinklers, or water, treated is better.  If you have bugs, the treated will obviously help.

   You can paint the tops of non-treated wood/joists with wood preserative or cover them with flashing paper, etc to keep the water and/or dry rot away.  When you cover it with decking you really wont see the preservative then.

   Hope this helped.

2. 
   AndyEngel | Jul 01, 2014 11:07am | #2

   Railings must resist 200 lbs of force in any direction. Of course, resisting an outward push is most important. It's measured in testing with a lateral pull (or push, I suppose) using something like a come-along with a big fish scale. It's a static load, not a dynamic one.

   Here's where it gets weird. The wood structural details in the IRC all incorporate by reference the American Wood Council's National Design Standard for Wood Construction. The NDS requires a 250% safety factor, so in that perspective, railings should technically withstand a 500 lb. load.

   If you use an HDTT2 or similar lateral anchor on each post and at least 2x8 joists, you should make the 500 lb. mark handily.

   Depending on your local code, 2x10s might not be wide enough. A lot of the country requires 11 in. minimum treads.

1. 
   deadnuts | Jul 02, 2014 12:28pm | #4

   another tall tale....

   It only gets weird if you don't know how to read the code. The International Residential Code Table R301.5 defines the minimum concentrated live load for guardrails and handrails as 200lbf (pounds of force) in any direction along the top of the rail.There are other LL stipulated in this IRC table (such as for stairs, attics, etc.). There are no safety factors added to these LL design criteria. Your must meet these miniumums--period.

   The 500 lb requirement is just something Andy dreamed up. A Safety factor of 4 is only applied when glazing is used in guardrails. That is rare....and manufacturers that provide glazed railing systems will almost invariably provide testing information that meet code miniums. In other words, you don't have to rig up a come a-long and a fish scale like Andy does. BTW, I would love to see a photo of that contraption off the back of your deck Andy.

3. 
   gbaune | Jul 02, 2014 10:38am | #3

   Deck Design

   carpguy,

   Here is a file with some excellent information regarding deck construction, design and safety.  One piece treads (2x10/12) develop a cupping problem and are prone to splitting. Most builders use 2-2x6's, also give a little better traction.

   
4. 
   patrick_mccombe | Jul 02, 2014 01:37pm | #5

   I don't know why you're so rude deadnuts. But in fact,  Andy is correct. The 500 pounds comes from the safety factor added when lab testing. It accounts for the variability of wood and site conditions. The quote below comes for a JLC article written by Frank Woeste. Frank and his students tested rail post connections in 2005 for the article. The approved connections in the DCA6-09 document cited by others come from Frank's research.

   "Safety factor. The code requirement says that the post must be able to withstand a 200-pound load. But when a structural assembly is tested in a lab, the load gets multiplied by an appropriate safety factor, which is intended to allow for the uncertainties of field installation and the fact that the connections may degrade in service from repeated loading and weathering (but not rot).

   We used a safety factor of 2.5, a number that has been in the model codes for decades for testing structural assemblies. So, for our testing, we needed to apply a 500-pound load to determine whether the post connection could be considered "code-conforming."

1. 
   deadnuts | Jul 02, 2014 02:40pm | #6

   You're talking about two different animals:

   Testing vs Code.

   Your talking about testing for a code approved connector that is intended to be sold on the open market. Of course that component would NOT be designed to meet miniumum code standards. I am not surprised that it has a safety factor of 2.5 to cover the discrepecies of each application or installer.

   However, that does not change the IRC code requirement one iota. The code is 200#, not 500#. In other words, If an architects wants to design a non presciptive code rail assembley, then a professional structural engineer will have to show a plan reviewer or field inspector that his/her design meets the IRC criteria of 200#-not 500#. Of course he/she can exceed this minimium requirement (and may evern use the "connections in the DCA6-09 document while doing so), but he/she is NOT required to.

1. 
   AndyEngel | Jul 03, 2014 08:49am | #7

   I wish you were right.

   Before coming back to FHB in 2013, I was the editor of Professional Deck Builder magazine for 6 years. In that role, I spent a lot of time researching this issue, including many conversations with the folks who write the code. The fact is that a strict interpretation requires a 250% safety factor. Personally, I think that's excessive and based on some arbitrary assumptions, but we all know what opinions are like. It's true, although not understood or enforced in all jurisdictions. In fact, in many places, the inspector gives the railing a shake at final, and if it seems more or less solid, he passes it. In other places, building departments require engineered plans for any railing system that lacks an ICC ES report.

   One interesting note is that research found (not sure if it was Woeste or Bender) that even using DTT2s, it's essentially impossible to meet the 500 lb. benchmark with 2x6 joists. A 36 in. post simply exerts too much leverage when the bottom of the post is that close to the fulcrum (the top of the joist).

1. 
   deadnuts | Jul 03, 2014 10:35am | #8

   try again

   I believe the IRC code is written objectively. I don't see any evidence of a footnote in the IRC that backs your claim that an APA safety factor of 2.5 needs to be applied to IRC table R301.5 under "Guardrails and handrails" in order to pass code or an inspection. The only footnote attached to this table item is a note that qualifies the load and reads:

   " A single concentrated load applied in any direction at any point along the top."

   If the intention of the IRC were to require it to 500#', then I believe it would clearly state that in the R301.5 table. It doesn't.  I am not as cynical as you in believing code and safety officials who draft the IRC are engaged in a game of "gotcha" where it takes layer upon layer of subjective interpretation in order to pass an inspection. That would be simply ridiculous. I also think that it is unprofessional to lead other builders to misinterpret the code and waste valuable time and energy overdesigning or overbuilding a guardrail or handrail assembly for no good reason.

   I agree that inspectors and plan reviewers can be subjective in their interpretation of the code. It happens every day. Again, that doesn't change what the code is. We designers and builders are not required to guage our efforts based on whether the code will be "strictly" interpreted, or "regularly" interpreted.

   If you choose to continue to state your case, Andy, for the requirement of 500# in order to meet code (which I welcome) then please keep in mind my prior point : testing to show a product or an assembly surpasses code is totally different from the code requirement.

   Cheers!

1. 
   AndyEngel | Jul 03, 2014 01:30pm | #9

   Seven or 8 years ago when I first heard about saftey factor, I was outraged. The code means what it says, dammit! But the more people I talked to on the code committee level, the more I came around to agreeing that the code incorporates the NDC by reference (actually, it does so through the IBC, which the IRC is just a refinement of). In one sense though, it comes down to what the local guy will pass. Personally, I'd just throw an $8 lateral anchor on each post and be happy knowing that what I'd done met the higher hurdle. I think that's the professional approach, and it's the one I'd rather defend in court if it ever came to that. 

2. 
   deadnuts | Jul 03, 2014 03:44pm | #10

   YOu're still barking up the wrong tree

    That's because you're still talking about testing requirements established to meet min. code. Testing is done with the 2.5 x min. code requirment in effect. A company who wants to market their product or assembly as meeting code will want to incorporate passable test results via an ICC report that allows thier product to succesfully go to market and more likely prove to code enforcing officials that the product or assembly is code compliant. Otherwise, the designer will most likely have to prove code compliance via calculations of a professional engineer.

   Below is the IBC regulations that I believe you, Andy, are referring to. Professional Engineers are trained to study tributary loads, apply them to materials and assemblies, and prove "pass" or "failure" using approved engineering analysis. That's what we architects and builders pay them for.

   Read 1712.1 (particularly the first sentence) carefully and you will understand why 500# is NOT  the code requirement for guardrails, but rather the requirement for testing of ICC approved components or assemblies. This ICC report is what ultimatlye assures a plan reviewers and/or inspectors (in advance of site testing measures) that the guardrail meets min. IRC uniformly distrubuted live loading requirements contained with IRC table R301.5 . This would be in leiu of individual documentation (and seal) from a professional engineer assuring so.

   From the IBC

   1712.1 Where required.

   "Where proposed construction is not capable of being designed by approved engineering analysis, or where proposed construction design method does not comply with the applicable material design standard, the system of construction or the structural unit and the connections shall be subjected to the tests prescribed in Section 1714."

   As Glenn Mathewson writes in DECK BUILDER Mag (Sept. 15, 2012) "There are really two ways of looking at the required loads for guards: before they’re installed and after. When guards are tested prior to installation, such as what a manufacturer of a guard system would have done, they must indeed be tested to a 2.5 times safety factor. This is required by Section 1715 of the IBC, and results in the 200-pound load in the code becoming the 500-pound load we hear about."

   BTW, its not my policy to just "throw" anything at a project and think it's gonna fly. It's just not professional. Figure out what is economically required to meet or exceed code and just do it. Your code official, your customer, and your bottom line will appreciate that approach.

3. 
   AndyEngel | Jul 08, 2014 11:15am | #13

   I'm still shooting for 500 lbs.

   Why wouldn't you? It's pretty easy to achieve, and the hardware isn't really expensive. Frankly, I'd sleep better at night. 

   Glenn did not outright say that a site built railing can be built to a lower standard than a manufactured product. I'll check with him. 

4. 
   deadnuts | Jul 08, 2014 07:10pm | #15

   personal choice vs. code

   YOu're free to build to whatever standard you want to...above code, which is 200 lbs. You don't even need to give such an excuse as it makes you sleep better at night. Go for it! My only issue with your posts, Andy, is when you deviate from expressing personal choice to telling folks something code is not. You're not free to do that.

    

5. 
   AndyEngel | Jul 10, 2014 07:07am | #16

   Here's what the horse said

   I exchanged some emails with Glenn. You were right in that the 250% safety factor applies only to laboratory tested components. However, he went on to say that the code does require a 200% safety factor for non-engineered site-built structrual components. The only way to determine that is with a long-duration (24 hours) in situ test. Glenn also acknowledged how utterly impractical that is for guards, and went on to say that the code body is leaning toward laboratory testing of common assemblies for incorporation in future editions of the IRC. Being a lab test, he said that would be subject to the 250% safety factor. The higher safety factor with the lab test is because it can only test a facsimile of the component, not the actual, as-built structure. I wonder how the compares to the current "standard field test" of the inspector hip-checking the rail?

   It's all fairly confusing. One take-away is that the code is a lot more nuanced than we'd like to believe it is.

6. 
   deadnuts | Jul 10, 2014 07:45am | #17

   please site Glenn's source

   AndyEngel wrote:

   ... he went on to say that the code does require a 200% safety factor for non-engineered site-built structrual components.

   Where, in the code, does it say this?

5. 
   sapwood | Jul 05, 2014 12:15pm | #11

   Back to your original question: does it matter if the 200# guy is moving at the time of incident? No, I don't think it does, but I haven't read the code in a long time and never to the detail required by the rest of the respondents. But, I do know that if a dynamic load limit was imposed then one would have to know just how fast that 200 pounder was moving when he hit the rail. F=ma and all that stuff, you know. 

1. 
   deadnuts | Jul 05, 2014 01:48pm | #12

   I think sapwood is correct. The code only refers to live loading condtions and, I believe, is meant to address typical human activity with regard to guardrails. Thus, the railing is expected to resist the force of an average human weight when applied anywhere along the top rail. I believe that would include tripping. Bear in mind that the ability of a person to exert their total body weight against a railing is rare. When someone catches their balance against a guardrail, chances are their lower body is still simultaneously touching the deck surface and is being partially resolved into the deck floor structure. Any additional dynamic loading element of forward momentum will still (most likely) be offset by this fact and end up being under to the 200# LL min. requirement.

   It's not like the rail is expected to take the force of average human body weight being shot at it like that of a circus cannon.

1. 
   AndyEngel | Jul 08, 2014 11:21am | #14

   They load is definitely static

   Saying that there's a 200lb. load requirement without including any information on acceleration implies that it's a static load, and that's how manufactured assemblies are tested. There's a big differnce in force between 200 lbs of static load and the load imposed by 200 lbs moving at any speed.