Determining Existing Strength of 40+ yr Old Anchor Bolts 3

[christructural]

I have a project where we are adding equipment to an existing structure and need to check that the existing anchor bolts work. The structure is 40+ years old and the existing drawings show that A490 anchor bolts were used.

We don’t think that material was possible at the time, so we don’t think that high strength bolts were used, but the owner doesn't have any records on the anchor bolts installed.

We checked the existing loading conditions of the structure and high strength anchor bolts are required to resist the loading. Since we are adding load, we are not confident that the existing anchor bolts will work.

Is there a way to determine the strength (or material properties – i.e. A36, A307, etc) of an existing anchor bolt without destroying the bolt?

 

[racookpe1978]

You've got to "approach" the problem asymptotically and so you will have to be satisfied with only a partial answer, since you do not want to destroy the existing anchors.

At the same time, be ready to provide a "Oh crud, they are not strong enough (or the tested anchor bolts broke out at 50% of their assumed strength) and all have to be replaced" solution.

Thus - First visually inspect each anchor bolt (assign them a unique ID number on each anchor leg location). Surface rust probably OK, but note it anyway. Deep crusted rust, tearing, bending, obviously scrap already, deep corrosion, visible damage. (Bent, distorted, run over by twelve forklift truck wheels.)

Then, pull test every bolt at 50% "design" capacity. If you cannot get a nut and pull test adapter onto every anchor bolt (bad threads, no room, etc) that is itself an indication you need to replace bolts.) If all pass, go to 90% intended capacity.

If all pass, then they will probably accept 100% intended capacity, but you don't know reserve margin. If 10 or 25% fail at either, then you need to start over with new foundations. While looking at how to survive with the current tank or vessel.

 

[Tmoose]

I'd do a hardness test with one of those bounce testers on the exposed bolt end after gently grinding it flat. If the hardness was in the range needed, I'd sneak up on the torque or pull test as described by Racookpe1978.

Some of the anchor bolts holding down our equipment in indoor powerplants of that vintage have rusted to pencil necks. They are fairly well protected as they go thru raised, poured housekeeping pads. Don't know if the corrosion is from decades of washdowns, or just because coal is evil.

Your first tension test will expose those.

Is it worth moving right to new anchors offset by a few diameters, or via clamps?

http://www.plastics007.com/upload/Product/Image1/PlasticAuxiliaryMachine10370m55622.jpg

 

[CoryPad]

I agree with Tmoose. The attached file is an image of portable hardness testing of fasteners, and the link below is the instrument that was used:

http://www.ge-mcs.com/en/hardness-testing/uci/mic-10.html

 

[metengr]

I also concur that portable hardness testing would yield approximate tensile strength results for the embedded anchor bolts. If you go the route of hardness testing make sure those that perform the test have a procedure and can demonstrate proficiency at achieving reliable results.

 

[Ron]

Yes, A490 bolts were around 40 years ago. You can tell if you have A325 or A490 bolts by the markings on the head of the bolts.

 

[Duwe6]

". . . A490 bolts by the markings on the head of the bolts." Problem is, that head will be embedded in the concrete pour.

I refuse to use 'rebound hardness' testers for anything other than point-to-point comparisons. Never, never has one of these testers given me anything approaching a valid hardness reading in 'the field'. They just do not work, even though they calibrate just fine on the perfectly machined test coupon supplied with the tester. If you choose to use one of these devices, please just use it for direct item-to-item comparisons. Get some nonshrink grout, an A490 bolt the same diameter and similar length, and a steel bucket. Embed that bolt in a bucket of grout, with the same amount of threads above the concrete. Use that as your comparison specimen. Ron is right, it is indeed possible that those embeds are A490's. I'll bet you a virtual drink at Pat's Pub that they are not, but the possibility needs to be investigated.

 

[racookpe1978]

Notice that the proposed rebound hardness test will only help indicate whether the old anchor bolt has the same hardness as the expected A490 metal.

It will not tell you is the bolt is cut, bent, rusted, deformed or actually (strongly) embedded in the old concrete - A piece of otherwise good metal wobbling loosely in a cracked, oblonged hole that has been vibrating for 3-13-33 years will pull out as soon as it is loaded. A bent old anchor bolt with rusted off threads cannot be re-used, regardless of what steel it was originally made of.

 

[christructural]

Thank you all for your responses.

The problem gets tricky because the anchor bolts are 1 3/8" dia, 7 feet long for a dead end transmission tower - that is loaded. The new equipment weight is negligible but the original line loads have increased and we have been tasked with checking the structure. The check has revealed that high strength bolts are needed; which is why we wanted a way to check them.

 

[metengr]

The 1-3/8" diameter bolt cross section and 7 foot length for mass behind it should be suitable for portable hardness testing.

 

[Guest102023]

Rebound (bounce?) hardness testers? Scrap them all.

Remove a representative sample or two and obtain a full diagnostic profile: metallurgical condition at core and thread root (structure, Vickers hardness), a profile assessment to confirm dimensional conformance and to find any deleterious forming defects, chemical analysis (actual, NOT PMI), and examination of the surface for corrosion pitting.

Grout integrity is a whole nother can of worms.

"If you don't have time to do the job right the first time, when are you going to find time to repair it?"

 

[tbuelna]

I always scratch my head when I read questions like this. If you have any question about the suitability of these bolts, then just replace them. New bolts are cheap when you consider the cost if they fail.

Since the anchor bolts have been encased in concrete for 4 decades, I don't think a hardness check of just the head would be sufficient. There could be corrosion of the surfaces just below the head which are typically areas that are highly stressed. Since you are apparently on the hook for signing off on the engineering of this modification and there are no reliable records of the legacy installation available, I imagine you would want to ask to replace every fastener. And if the customer does not agree, get someone else to sign off on the engineering.

 

[oldrunner]

I didn't think that one could get an A490 bolt longer than about 6". It may be that you have A354 bolts which are about the same yield but obtainable in long lengths. The yield is around 120 ksi or more - depending upon the diameter and original callout. I have some drawings and specs on a 50 year tower that I helped repair (rock collision on the bracing) and I'll look tomorrow as I did check the footings - similar to yours. (about 1950 construction)