Should you Pretension Anchor Bolts?

One of the practices that I have gone back and forth on in my career is specifying a preload for anchor bolts. The idea is that inducing tension in an anchor rod as part of the installation process will avoid load reversal in the rod under transient loading (wind, seismic, etc.) conditions. Consider a self-supporting water tower. Even under modest wind conditions you could build up thousands of cycles of tension-detension pretty quickly which can lead to fatigue issues. The ASCE guide Anchorage Design for Petrochemical Facilities lists several advantages of pre-tensioning other than reduction in load reversals.

1. Prevents loosening of nuts during load reversals.

2. May increase dampening for pulsating or vibrating equipment.

3. Will decrease, to some extent, the drift for tall slender structures and equipment under wind or seismic load.

4. Will increase the downward force and thus the frictional resistance for process towers, other equipment, and structural base plates.

It seems pretty straightforward; however, there are some drawbacks, and, because of that, not everyone agrees that anchor should be pretentioned at all. The ASCE guide also lists several disadvantages.

1. Can be costly to install accurately.

2. No recognized code authority gives guidance on the design and installation of tensioned anchors.

3. The long-term load on the anchor is questionable because of the reduction in tension due to creep of the concrete under tension load.

4. The pre-stretch during anchor tensioning reduces the amount of inelastic stretch that may be considered effective for energy dissipation under seismic loads.

5. Typically, there is no bearing resistance to shear on the anchor because during tensioning, the sleeve around the anchor is not filled with grout.

6. There is little assurance that the anchor will be properly installed and tensioned in the field.

7. Direct damage from tensioning is possible. That is, the tensioning itself can damage the concrete if not properly designed or if the tension load is not properly regulated.

8. It Is difficult to ensure that there is consistency between the design of the anchor and the design of the vessel anchor chair; that is, to ensure that the vessel anchor chair has been designed to carry the anchor tension load.

9. The stress level is difficult to maintain because of concrete shrinkage and creep, and relaxation of the anchor material.

10. Only alloy anchor bolts can be effectively pre-tensioned. ASTM A307 and ASTM A36 bolts do not hold their pre-tensioning values and are thus ineffective in this regard ( Kulak, et al (1987) ).

More disadvantages than advantages, so why even consider it, right? Well, if you really consider the advantages v. disadvantages while also considering specific applications instead of the sort of academic exercise of preloading a generic anchor rod, I believe you will find that, for certain applications, pretensioning is still a very good option. Generally, I only consider specifying a pretension for equipment - specifically tall, slender vertical process equipment; however, I suppose there could be applications where pretensioning the anchor rods for a structure would be appealing. A piece of equipment must meet one of two criteria in order for the anchor bolts to be designed for a preload. First, the height-to-diameter ratio must exceed 10. Second, the height must exceed 150 ft. If it meets either of these two criteria, I design for a preload.

Now, considering a generic process tower with either a height-to-diameter ratio exceeding 10 or a height exceeding 150 ft, let’s look at the disadvantages listed above and decide if pretensioning the anchors is still appealing.

1. Costly to install. Absolutely. The ideal method of preloading an anchor is hydraulic jacking which could potentially involve bringing on an additional contractor to complete this work. Even if less-precise methods are used to achieve the preload, it will still be more labor than simply tightening the nut to snug-tight and then giving it one more crank for good measure. Really, I think the question to ask is, “Is the additional installation cost worth it?” Process towers like this are very expensive, and, typically, will require the complete shutdown of a facility if they are taken out of service. How long would it take to repair an anchor rod that has failed due to fatigue cracking, and how many hundreds of thousands of dollars in revenue will be lost in that time?

2. No recognized code authority. Honestly, I have never understood why this is listed because it isn’t true. PIP, ACI, even the very same ASCE guide all give guidance on pretensioned anchors. Whether it’s a formula for estimating the required nut turn to achieve a preload or recommendations on a minimum preload, there are resources available to facilitate a dependable design.

3. Reduction in pretension load over time. Yes, this absolutely happens. Some studies show a 4-10% drop in preload over time for higher-strength alloy rods up to a 30-50% drop in preload for lower-strength rods such as ASTM F1554, Gr. 36 or ASTM A307. For the type of process equipment we are considering, you are generally going to be considering a higher strength alloy as it is. Additionally, the main benefit that is driving preloading the bolts is reduction in cyclical loading which can cause fatigue cracking. Even if the rods were to experience a 50% drop in preload, this will still be enough tension to avoid cycling from tensioned to no tension under lower wind speeds which are the most common.

4. Reduction in amount of inelastic deformation available for seismic load. For the overwhelming majority of locations in the United States, wind load will control the anchorage design for this type of application; however, if your project is located in an area with the potential for large seismic loading, you should consider this before specifying a pretension for your anchor rods.

5. No shear from vessel taken by anchor. This is fine. Ideal, actually. Any sort of shear load will only decrease the tensile capacity of an anchor. The calculation set will need to include a check to ensure that for every load combination there is enough friction developed to completely transfer the vessel shear to the foundation.

6. Anchor could be installed/tensioned incorrectly. To be fair, this could be a problem with literally anything else on site. As an engineer, all you can do is make sure your design is solid and your specifications are clear.

7. Damage to concrete during tensioning. This really falls under the same category as #6. For this type of foundation, I always design the brittle (concrete) limit states for a minimum load of the tensile strength of the rod, and the minimum pretension load is going to be some fraction (usually 1/3) of that.

8. Difficult to ensure the vessel is designed for pretensioned anchors. This is important, and, I think, probably the most important consideration when considering preload for an anchor rod. There is no scenario when you should specify a preload for an anchor rod without verifying that the vessel can accommodate that preload.

9. Stress level difficult to maintain because of material creep. Sure. I think this falls in with #3 above. There will definitely be some relaxation of the anchor rods over time; however, in my opinion, this will still leave at least 50% of the design preload active which will be enough to avoid load reversal the overwhelming majority of the time.

10. Only alloy anchors can be pretensioned. I think that stating that lower strength anchors cannot be pretensioned is a little disingenuous. They lose much of their preload, sure, but as mentioned several times over they can be expected to maintain at least 50% of their original load which should be enough to avoid load reversal for the overwhelming majority of cases. For tall process vessels, we are probably considering high strength alloy anchors for the foundation anyway which increases the amount of preload that is retained.

So, if you skipped directly to the last paragraph, should you pretension your anchor bolts? I don’t know. Do what you want. I would not pretension anchor bolts all of the time, no. However, I do think for certain applications, you can create a compelling argument that the benefits of specifying a preload far outweigh the potential disadvantages. Furthermore, many of those disadvantages are things that are in your direct control and can be mitigated during the design & specification process.