lag screw withdrawal penetration

[cdowney4]

I have two questions regarding penetration for lag screws in withdrawal only (not laterally loaded) in side grain per NDS. Forgive me if these are answered elsewhere in the forums -- I have seen my first question discussed, but not specifically answered the way I'm asking.

1. The requirement for 4D minimum penetration specifically says its for "...shear connections...", so I think it does not apply to withdrawal. For example, if I had a 1/2" lag screw where the tip penetrates through and beyond the other side of a 1.5" thick main member, I think I can use the 1.5" of penetration even though I only have 3D penetration. Agree or disagree?

2. Is "thread penetration" the length of thread that is physically engaged with the main member, or is it the depth of penetration of the threads. Or, put another way, can thread penetrations include any unthreaded portion of the the lag screw? For example, in my example above, with the tip penetrating though and beyond the 1.5" thick main member, imagine there is also 1/2" of smooth shank into the main member -- is my thread penetration still considered to be 1.5" or is it the 1" of thread that is actually engaged with the member?

Thanks for any opinions.

 

[driftLimiter]

1. Disagree. See below.

2. Thread penetration is the length of the threaded portion of the screw minus the length of the tapered tip that is embedded into the member.

Conceptually the smooth area of the screw doesn't contribute to the tension strength because it has no bonding, the tapered tip as well has less bonding than the main threaded area so we disclude it.

 

[Aesur]

12.2.1.2 of 2018 NDS states: "For calculation of fastener reference withdrawal design value in pounds, the unit reference withdrawal design value in lbs/in of thread penetration from 12.2.1.1. shall be multiplied by the length of thread penetration, pt, into a wood member, excluding the length of tapered tip." Therefore it is my opinion that if the lag screw goes through the full member and you are in the active threads (ie tapered tip is fully outside the member) then you would use the 1.5 inch for your calculation. Withdrawal, to my knowledge, does not have a minimum embedment depth and is calculated in lbs/in.

 

[SWComposites]

Is "thread penetration" the length of thread that is physically engaged with the main member, or is it the depth of penetration of the threads. Or, put another way, can thread penetrations include any unthreaded portion of the the lag screw? For example, in my example above, with the tip penetrating though and beyond the 1.5" thick main member, imagine there is also 1/2" of smooth shank into the main member -- is my thread penetration still considered to be 1.5" or is it the 1" of thread that is actually engaged with the member? You can only use the actual thread length, so 1.0 inch in this example.

 

[Aesur]

Good Catch SWComposites, I didn't catch the question about smooth shank, you are correct, it's threads only, the smooth portion does not contribute to withdrawal.

 

[SWComposites]

Also, re The requirement for 4D minimum penetration specifically says its for "...shear connections...", so I think it does not apply to withdrawal. For example, if I had a 1/2" lag screw where the tip penetrates through and beyond the other side of a 1.5" thick main member, I think I can use the 1.5" of penetration even though I only have 3D penetration. Agree or disagree? it depends on how the allowable withdraw design value is given; if it is given in "lbs" per fastener then the required amount of threads must be engaged (presumably 4D); if it is given in terms of "lbs/in" where "inch" is the length of thread engagement, then you could use the 1.5 inch penetration (assuming fully engaged and the 1.5" does not include the tip or the smooth shank).

 

[cdowney4]

Thanks for the responses so far. In general, consensus seems to match what I thought:

1. No minimum required penetration for withdrawal values given in lbs/in.

2. Exclude any smooth shank in the main member and count only the engaged threads, excluding the tip.

 

[dik]

Check the site... methodology is shown and for Canadian codes:

https://www.jabacus.com/engineering/timber2014/timberconnection_lagw_2014.php

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[dik]

and for lateral loadings...

https://www.jabacus.com/engineering/timber2014/timberconnection_lag_lateral_2014.php

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[cdowney4]

Anyone up for a follow-up (and perhaps more complicated) question?

Since the consensus on #2 seems to be excluding the smooth shank, that implies mode of failure is "slip" vs. say some sort of cone pull-out like you might see with an embedment in concrete. I agree with this.

So, what if my main member was a double 2x? Assume the 2x's are nailed/screwed/bolted together sufficiently. And then imagine the effective thread crosses the joint between the two, so that there is thread engagement in both. I simply use the total thread engagement to determine withdrawal? No penalty for crossing the joint? Again, assuming they are attached together sufficiently.

My gut tells me this is problematic. Even if I think they are connected together sufficiently, there is risk that the first 2x starts to pull away from the second 2x and overloads the threads in the second 2x. Or, maybe if you ensure that there is more thread engagement in the second 2x than in the first 2x, there is more assurance that you can develop the combined thread engagement?

 

[driftLimiter]

I would start by first looking at the proportion of threads in each ply, then say each ply resists that proportion of the applied tension load. If the 'inside' ply has ample connection to the 'outside' ply that is in close proximity to the lag screw to transfer it's proportion of the tension then I would just assume it is one solid member for this calc.

If the plys are connected poorly or not enough close to the lag screw then I would just ignore the 'inside' ply.

 

[phamENG]

Even if I think they are connected together sufficiently, there is risk that the first 2x starts to pull away from the second 2x and overloads the threads in the second 2x.

If this were the case, they would not be adequately connected together. Design the connection to do what you want it to do and you can reasonably eliminate that failure mode.

But if they aren't fastened together, you're right. You'd probably want to limit the connection capacity to twice the lesser engagement to keep the load "balanced" between the plies.