Deck mounted handrail post base

[HouseBoy]

I am looking for a detail to improve the base anchorage of a rail post on a wood deck.
Originally, we specified rim board ties such as the Dek-Lok product. Now the owner selected a rail post that mounts on top of the deck (and not onto the rim board).
Does anybody have a good resource/reference or good detail for this?
Thanks

 

[Lion06]

how wide is this base in plan?
If the spacing between the lag screws (in and out of the page) are 3" (or if you can make it 3") then you can attach a small piece of 2x10 to the joist right under where the post is. Then lag screw into the smaller pieces of 2x10. This will take that load right into your joist.

 

[HouseBoy]

It is not practical to coordinate the intermediate rail post locations with the joist layout. In fact, the joists are already in place. I suppose I could ask that they ad a joist at each rail post location.
Are there any accessories (like the Dek Lok product) that would work for this condition?

 

[ctcray]

Endurance Rail by RDI has a "flushmount" structural post base.

http://www.rdirail.com/pdf/co_structpost_instr.pdf

I guess the key with these is to get some good, solid blocking beneath.

 

[HouseBoy]

Thanks but unfortunately the specific "problem" I'm having is just how to go about getting "some good, solid blocking beneath".
In the illustration it appears that the base is being mounted to a triple header that is flush with the joists. Obviously you understand the problem but they don't seem to be concerned at all. We usually do not have that condition and it seems that the manufacturer is just glossing over that aspect.
Usually, we see dropped beams with joists spanning over and having a foot or so of cantilever. It's tough to position joists ahead of time so they will be located right at the rail post location.
additionally, it seems that there is a practical limit to the withdrawal capacity of a ¼ inch wood screw of about 600# or so (considering several inches embedment. Not only that, but the attachment of any blocking is also an issue.

Just wondering how folks get these things to “work”

 

[AlmostPE]

I dont like it at all. That rail looks like it was made for concrete floor or steel. On the top of that each post works individually because you connect them with cables. That means if you have a 90 bend on the railing, it wouldnt help you stablize it. If you must do this I wouldnt use lag bolt. I would use thru bolt with big washer trough the blocking and make sure the blocking is securely fasten onto the joists.

Not almost anymore!

 

[Lion06]

I don't know. I am looking at that, and if you have to attach through the deck without going into a joist I think you'll have a hard time getting the lag screw to work for pull-out on the tension side.
I know it might be a bit of a pain for the guy building it, but have him put the posts only at joist locations and add a short (maybe 1'-6" or 2') piece to each side of those joists. With the joists having a regular spacing, it shouldn't be much of a problem for it to look fine.... it's not like you'll have randomly placed/spaced posts.

 

[JMirisola]

Maybe I'm over-simplifying the issue here, but...
Having built many decks over the years, I don't see a problem with determining approximate post locations prior to laying the decking. By doing this, you can put in a 2x6/8/10 on the flat, flush to the top of the joists in the appropriate joist bays. If said 2x is nailed on three sides with 16d nails, or screwed in, it should provide you with more than enough anchorage. You could up the ante by using 4x material instead. So long as your lag screws get good penetration into the wood, there shouldn't be an issue.
Just my $.02, as a former framer...

Jeff Mirisola, CSWP
Certified DriveWorks AE

http://designsmarter.typepad.com/jeffs_blog

Dell M90, Core2 Duo
4GB RAM
Nvidia 3500M

 

[HouseBoy]

If you are talking about end nailing thru the joist and into the flat blocking, I don't think the fastener capacities come close to "working".
What I'm saying is that the 200# code required loading at the top of the rail is going to put considerable bending at the base. Basically the blocking is going to be twisted. A couple of nails in end penetration don't figure for the 1000 or so pounds (actually it is probably more)that would be present under the code required loading.

I'm not saying that what you describe isn't done (maybe often). I'm just saying that it doesn't figure and that (actually making it figure on paper) is a standard that someone could be held to so I am trying to find a "correct" solution.

 

[JMirisola]

Then use 4x material and lag screws. While lag screws don't have published shear strengths, the Industrial Fastener Institute states:
As an empirical guide, shear strengths of carbon steel fasteners may be assumed to be approximately 60 percent of their specified minimum tensile strengths.

Where a low or medium grade carbon steel lag bolt has a tensile strength of ~60 ksi, four on each side will provide you with more than enough shear strength (two on each side would probably suffice). By using a piece of 4x material, instead of 2x, your figures should work.

Jeff Mirisola, CSWP
Certified DriveWorks AE

http://designsmarter.typepad.com/jeffs_blog

Dell M90, Core2 Duo
4GB RAM
Nvidia 3500M

 

[HouseBoy]

Don't you suppose that bearing on the wood is the weak link (despite the relatively high material strength of the screw itself)?
Additionally, because the screws are installed thru end grain, the capacity is less that cross grain bearing of screws.
I suspect that there are not good answers posted for this question because that are not many simple and easy ways to solve this problem. (Certainly none that are common).

 

[Lion06]

houseguy,
can you do this:
It looks in your detail like (2) of the screws from the base connection can go into the rim joist. Is this correct? Can you get these screws to work in pull-out?
If so, can you add another joist (either 2x or 4x, whatever you need) that is parallel to that rim joist, spanning to the perpindicular joists with a s
Simpson hangre and use that to attach the inside (2) screws. This would put that "additional" joist in bending.

jmirisola-
A 2xanything laid flat has little to no torsional resistance. Your suggested detail isputting that member into torsion.

 

[JMirisola]

I thought I had the perfect example this morning, then the coffee kicked in.
I wonder though, could you still use a piece of 4x material and have it sandwiched between the rim and another 2x block? This would allow you to bolt through the sides versus the ends, plus give you some protection from twisting. Then, though it would probably give the framers fits, through bolt the post to the block.
Beyond that, good luck.

Jeff Mirisola, CSWP
Certified DriveWorks AE

http://designsmarter.typepad.com/jeffs_blog

Dell M90, Core2 Duo
4GB RAM
Nvidia 3500M

 

[twinnell]

I remember reading an article in either Structure Magazine or Structural Engineer that discussed wood handrail posts for decks. After numerous tests of different configurations, I think only one passed the equivalent test for the 200# load. The configuration was running the 4x4 post down into the joist framing, either inside or outside the band, thru bolt through the band and use a Simpson HD2AHDG on the bottom bolt, which then bolts into the adjacent joist.

I think this is the only configuration that past except for extending the columns up to use as handrail posts.

My parents were having a deck built and it was half done until I interviened. The contractor was toe-screwing, 1/4" screws, two in each face of the 4x4, the 4x4 post to the top of the framing. He would add flat blocking as necessary.

He said he had built hundreds this way and said the handrail meets code and would hold a 200# load. I then gave him a test to do. I asked him how much he weighed. 185 lbs. Then I told him, using the same screw pattern, screw the handrail post to the face of the 6x6 column (15' tall deck), just below the band (so that it sticks out horizontally out from the deck)...the walk out on it. If it holds, I will allow you to install them your way. Needless to say, they were installed my way.

 

[Lion06]

twinnel-
I read that article a while back when it came out. I remember the testing was done at VT. It was a good article.