My approach would depend on the tensile properties of the steel in your bolts or rods.  What is the yield stress?  If a low alloy steel, I think you can just detension the bolts and tighten again, don't pull so hard next time.  Tension to say 85% fy.

It would be interesting to know how you determined that some bolts yielded and others did not.

Did the bolt material yield or did they move by loss of bond with the concrete?  How were they installed?  Did you do ultrasonic testing to determine if they were cracked?

Thanks Ron

The anchors are cast in place.

The anchors have a strong bearing mechanism at their lower tips and have been greased to prevent bond.

Some hydraulic equipment used  to provide some preloading force was not adjusted correctly when some bolts were being worked on.

The machine records pressure(correlated to prestressing force) and also the elongation.

For the case where operators did not make correct adjustments, excessive pressure(hence prestress) as well as excessive elongation were recorded on the logs.

I wasnt at site and I am not familiar with this machine. This is what has been reported to me

Analysis shows that if these bolts are assumed nonexistent, remaining bolts will work fine with a few overloaded bolts closest to the weakest ones.

It would be nice to add more bolts. I was just thinking there was a way to bring them back to elastic state.

there is a fatigue issue too.

respects
ijr

Technically if the bolts have not been ruined and you are still able to unload them, they will return to a non stressed state, and then you will be able to restress to an elastic state (or merely just to unload to such state, but I think more believable and reliable the process from a fully nonstressed state. You will have used a cycle to plastification then unloaded and then loaded to a still slastic state. So it looks feasible except that per the facts the bolts are preferred be left as they are, or neglected in computations. What said above requires that the nut can be drivable on the bolt in its by now permanently deformed status. Maybe not work the try, since unnecessary if neglected.

ishvaaag

how do you unload bolts that have almost reached yield strains? That is exactly what I am looking for

respects
ijr

See my post above.  You just loosen the nut.  If the bolt has yielded, it will probably be permanently elongated, but it will return to an elastic state.  Don't tighten so much next time.

hokie66

you mean when we unload, some permanent strain will remain but later loading will produce linear strain increase? Like in the way we test materials for repeated loading?

But is the final strain after reloading acceptable?

respects
ijr

 

When the bolts are unloaded, they will return to an elastic state, albeit, with internal stresses.  Subsequent loading to the same limit should be totally elastic and unless you have failed the anchorage, they should be OK as long as they are not presently damaged.  You should not take them over the current load.  Some bolt materials don't permit you to re-tension them... A490 bolts for example.

Dik

By definition and having entered the plastic state some degree of permanent deformation has already been imparted. In spite of this we have from screw experience the fact of that we reuse damaged items, many times out of mere inconvenience of getting some new substitutes. We also know that to redrive a damaged screw in wood is a much more likely successful affair than to redrive a damaged bolt-nut pair.

Obviously, the ordinary way to detension (and only if you plan to reuse) the bolts is to turn loose the nut, with everything it implies for a yielded bolt. Hence to drive again to the wanted state may be at best feasible, but no one would call this a standard installation. You may try anyway for what would be a more uniform setup of the pretension, since, anyway, if non-existent you have found the structure to meet what required. I unfortunately am not an expert of bolt technology and must not give advice on what would depend on geometry of the engagement, the level of force, the chemistry etc. Since you do not lose but a present unruly situation for another, and both look mechanically acceptable to you, you might decide more than anything to see the feasibility of loosing the nut, then put it back again, satisfying technical curiosity and always learning something. If I would be in such mindset I would try to learn from the experts on bolts what to expect. If not available or not caring, likely would accept the present state of affairs, you know there's something not as well as it should be, but at least stays "controlled" and "clean". After touching the nuts we might not have these niceties.

dik answered for me.  Yes, there will be a permanent elongation.  Provided the bolt is a ductile type, low alloy as I previously stated, the unloading and reloading will be along the same stress-strain straight line, just displaced by the amount you stretched it inelastically.

ishvaaag, being not an expert should not prevent you from commenting. When you are not an expert you are more likely to be more creative, and getting out of the box.

Yes,the RCSC guide supports this reuse, though it recommends a maximum of  one reuse. I appreciate what hokie mentions: the ductility

My anchor bolts are indeed alloy steels.

respects
ijr

But how easy is it to loosen the nut from a fully preloaded bolt?

Ah, now, that's a practical problem.  I don't know, you will just have to try it.  My gut feeling is that the nuts will come off.

Hokie... I just use a good wench for that...  

Your wench must be stronger than mine :)

You guys are real great! Thanks for it all. Dave, dont censor this post. It is pure respect