Headed Bolts vs Rod with a Nut 1

Their test did not include a washer. Apples and oranges. Don't worry about it and keep using what you have as their test is irrelevant to your conditions.

I know...I know...the nut below the washer is under their condition....well, not really.

They are still failing across the same cross sectional area, even if it is a different locations.

So what? Differentially speaking, where is the difference to get a different value to give a difference in strength?

Mike McCann
MMC Engineering

Our typical anchor rod detail gives the contractor the option of either. I have never found anything that really shows a substantial difference between the two.

My reply as the followings

>> our office generally specifies an all thread rod with a nut-washer-nut at the bottom as opposed to a headed bolt.

The anchor rod shall NOT be all threaded and this will create a big problem. The correct way is to thread both end tip as needed, see Page-1 of attached PDF files.

Ideally you shall allow the anchor rod free to stretch in between and only two ends get engaged. If you thread the rod full length, it will increase the bond between rod and concrete and this bond is not desirable as it will relax with time, and gradually your anchor bolt will lose the pretension. The correct way is allow the anchor rod free to stretch in between and only two ends get engaged. We normally use sleeve or tape wrapped to debond the center part. See Page-2 of attached PDF file. This is also the ductility requirement as per ACI 318-11 D.3.3.4.3 (a). We call it stretch length or gauge length in ASCE 7-10, see Page-3 & 4 of attached PDF file.

>>Portland Bolt has an article on their website stating that their testing indicates that an all thread rod with a nut is >>not as strong as a headed bolt.

I would say that mostly applies to steel connection case, not anchor bolt. Anchor bolt is different as it’s fully contact with concrete in cast-in-place anchor bolt. The rod, head and nut are fully contact with the surrounding concrete. There would be no reduction, on the contrary, it gets local stress hardening and the concrete bearing get 8 times more, see the anchor rod head pullout resistance formula ACI 318-11 D.5.3.4 Np=8*Abrg*fc'
Plus, when we design anchor rod material capacity, we consider thread reduction up to 30% and use effective shank area Ase, so there is no point for you top worry about this, right ?

>> I was wondering what other engineers think about this and how others specify these types of attachments.

Just go for the detail of Heavy Hex Jam Nut – Head Plate- Heavy Hex Jam Nut as shown on Page-1 of attached PDF files. In this way we have the flexibility to use Hex, Heavy Hex, Square, Hardened Washer and Custom Head Plate to create different Abrg to suit our design needs.

anchor bolt design per ACI 318-11 crane beam design

I haven't used all thread rod for anchor bolts, and amec1004 has persuaded me that it is not a good idea.

>> I haven't used all thread rod for anchor bolts, and amec1004 has persuaded me that it is not a good idea.

persuaded me --> convinced me better ?

anchor bolt design per ACI 318-11 crane beam design

Don't get picky on me, or I'll change my mind.

amec2004,

We are still on the 2006 IBC here in northern Arizona which means we are using the 2005 ACI.

You made the statement, "The anchor rod shall NOT be all threaded and this will create a big problem." Your arguments supporting this statement appear to be based on a loss of pretension and the ductility requirement of ACI 318-11 D3.3.4.3(a).

1. The vast majority of the anchors we specify are untorqued so loss of pretension is not a concern. ACI 318-05 D3.3 gives additional requirements when the anchor design includes seismic loads. Assuming that ACI 318-11 D3.3 also only applies for seismic loads, would you qualify your statement I quoted above by saying it applies only when the anchors are specified with pretension and/or when resisting seismic loads? Or do you think that it should always apply?

2. Fig. RD.1.3(a) appears to show a condition where an all thread rod could be used and still meet the requirements of ACI 318-11 D3.3.4.3(a). Do you agree?

3. ACI 318-11 D3.3.4.3 says, "Anchors and their attachments shall satisfy one of the options (a) through (d). As I do not have a copy of the ACI318-11, do options (b), (c), or (d) allow another attachment methods that would be acceptable using all-thread?

Thanks.

1. The vast majority of the anchors we specify are untorqued so loss of pretension is not a concern. ACI 318-05 D3.3 gives additional requirements when the anchor design includes seismic loads. Assuming that ACI 318-11 D3.3 also only applies for seismic loads, would you qualify your statement I quoted above by saying it applies only when the anchors are specified with pretension and/or when resisting seismic loads? Or do you think that it should always apply?

Yes my above statement is founded based on fact that the anchor bolt needs pre-tensioned.
I recommend the same thing for the untorqued anchor bolt as well for the following reasons.

1) The bond between anchor rod and concrete is not desirable and nowhere in the ACI code counts this bond stress to develop the resistance. Look at anchor rod head pullout resistance formula ACI 318-11 D.5.3.4 Np=8*Abrg*fc' , it relies on the head bearing, not the bond stress to develop pullout resistance.

If you thread the full length of rod, it roughen the rod surface and increase the bond which is opposite to the desirable debond process. For a the untorqued anchor bolt the pullout resistance check uses the same formula as the torqued bolt. This create a scenario that your anchor bolt calculation assumption is different from the actual condition. When the anchor bolt gets tightened up, the rod/concrete bond stress will mobilize first and the head bearing may never get engaged if you have a long embedment. Just like you design a 30m long pile considering both end bearing and friction resistance, in the reality the end bearing resistance may never get mobilized until the shaft friction reaches the ultimate value and the pile move downward in a significant settlement.

2) I couldn’t find a good reason to use the full threaded length. You thread more and cost more, plus around 20%~30% rod shank area reduction everywhere.

2. Fig. RD.1.3(a) appears to show a condition where an all thread rod could be used and still meet the requirements of ACI 318-11 D3.3.4.3(a). Do you agree?

I didn’t see the anchor rod shown in ACI 318-11 Fig. RD.1.3(a) get all threaded. For this case if you already provide 8d stretch length by means of anchor chair or sleeve, you meet the requirement. Where the rod get full length threaded or not is not important, but it’s not recommended.


3. ACI 318-11 D3.3.4.3 says, "Anchors and their attachments shall satisfy one of the options (a) through (d). As I do not have a copy of the ACI318-11, do options (b), (c), or (d) allow another attachment methods that would be acceptable using all-thread?

Again, the use of full length threaded is not specifically prohibited by the code but is not recommended.
ACI 318-11 D3.3.4.3 options (b), (c), or (d) provide other options to satisfy seismic requirements. I upload attached PDF file to show you what these options are for both ACI 318-11 and ACI 318-08 codes
Nowhere in the code prohibits the use of full length threaded anchor rod.

Attached PDF File : ACI 318-11-Seismic-Requirements.pdf

anchor bolt design per ACI 318-11 crane beam design

Rod shank area reduction for threaded rod and partially threaded rod are the same. Failure will occur in threaded portion as this is the smaller cross section whether fully threaded or partially threaded.

Not a good idea to use all thread in my opinion.
I have always used anchor rod material threaded as needed on each end with a nut and washer plate at the base.
Be careful not to use too large a anchor plate if you need one at all.

I believe all thread will bond to the concrete which I cant see being a good thing when snugging up the nuts.

Ron
The loading with the bevelled washers puts a moment into the thread.
It would be interesting to see if the results were the same if the fastener were flipped around.

With the heavy hex head embedded in concrete there should little moment.

My biggest concern is the concrete being green and the heavy hex head nut and anchor rod rotating.

Dik

Thanks to all for their great comments.

amec2004,

1. I am having a difficult time understanding why having additional resistance is a bad thing. It seems like it would be conservative. Now if your concern is that under load the bond between the threads and the concrete will suddenly break apart and put an impact load on the head and/or, as ToadJones pointed out, leave the nuts loose on the base plate, I would offer the following:

Rod Diameter = 3/4"
Min Tensile Strength = 55 ksi (F1554 grade 55)
Pretension is assumed to be 70% of the specified minimum tensile strength.
(Per the Specification for Structural Joints Using ASTM A325 or A490 Bolts, June 30, 2004) This may not be the best reference but we are just ball parking it here.
Embedment = 12"
E = 29,000 ksi

A = pi*(3/8)^2 = 0.44in^2
Min Pretension = .7*55,000 psi*0.44in^2 = 17k
Elongation = 12"*(17k/0.44 in^2)/29,000ksi = 0.0159in (I got this equation from not the most reliable source so sorry in advance if it is incorrect.)

So, not including a rod shank area reduction, assuming a pretensioned rod, and assuming that the rod is instantly fully bonded to the concrete and no elongation occurs at all as the bond develops between the threads and the concrete, we are talking about 1/64" of an inch. Under the real world conditions we are talking about with an untorqued anchor, I really don't see an impact load or loose nuts being an issue.

2. It has been my experience that an all thread rod is preferred by contractors in our area as they are kept in stock and is less expensive for them as custom fabrication is not required than it is to buy custom made smooth rods with threads specified at the ends only. Simpson Strong-Tie offers what they call a PAB or Pre-Assembled Anchor Bolt. Based on their catalog and a phone call to custom service, it appears that they are also using an all thread rod which would seem to indicate that an all thread rod is cheaper than a custom rod with just the ends threaded. We mostly do residential and light commercial with less than 20 of any one anchor rod specified on the job. If we were specifying hundreds or thousands of the same rods for a project I wouldn't be surprised if rods with only the ends threaded could be fabricated at a lower cost.

Concerning ACI 318-11 Fig. RD.1.3(a), I could not think of any reason to use an anchor chair or sleeve unless I was unable to provide the 8d stretch length in the concrete. The only reason I could see for being unable to provide the stretch length in the concrete is an all thread rod is being used. So my point was that this detail makes no sense to me unless an all thread rod is being used. Can you explain another reason to use this detail?

3. Thanks for uploading the additional info. You repeatedly state that the use of an all thread rod is "not recommended" in the code but failed to provide a reference.

I would have to agree with Ron's comment that the effect of the rod shank area reduction is the same if the whole rod is threaded or is only partially threaded.

I think based on everyone's comments so far, I am going to change our call out from all thread rods to threaded rods, but am not convinced that all thread rods do not have many acceptable anchor rod applications.

>> I am having a difficult time understanding why having additional resistance is a bad thing

It's a bad thing if your anchor bolt calculation counts on head bearing while the bond stress always mobilize ahead of head bearing.

What's the point if you calculate your anchor bolt capacity using head bearing as per ACI code while the head bearing may only partial mobilize, or not mobilize at all if you have a long embedment and the bond stress takes up all the load ?

The bond stress is unreliable and it will shed off with time being, your anchor bolt will gradually loose the grip, for both torqued and untorqued case, especially when your anchor bolt are subject to cyclic load or has fatigue issue, such as high rise structure under wind load and your anchor bolt gets cyclic up and down tensile load.

In another thread we talk about the column base PIN or FIX issue. Someone may raise the similar question, can we model the base as PIN and detail it as FIX, it seems to be conservative ? Actually it's not conservative as you change the distribution of moment and moment at some point may be under estimated due to the re-distribution of moment.

anchor bolt design per ACI 318-11 crane beam design

amec2004

I understand and agree on the column base Pin or Fix issue.

You said, "It's a bad thing if your anchor bolt calculation counts on head bearing while the bond stress always mobilize ahead of head bearing."

My question is why?

You said, "What's the point if you calculate your anchor bolt capacity using head bearing as per ACI code while the head bearing may only partial mobilize, or not mobilize at all if you have a long embedment and the bond stress takes up all the load?"

You answered your own question when you said, "The bond stress is unreliable...your anchor bolt will gradually loose the grip." At which point, the anchor will function as designed.

The question is for an untorqued anchor not resisting seismic loads, what is harm in using an all thread anchor?

>>The question is for an untorqued anchor not resisting seismic loads, what is harm in using an all thread anchor?

I already addressed this in my previous post: it's not prohibited by the code, but it's not recommended by me, my 2 cents for your reference only.

It will be your own engineering judgment.

anchor bolt design per ACI 318-11 crane beam design

amec2004, I appreciate your comments.