I've got a wee problem at work
a customer has managed to bent one of the anchor bolts that holds up a steel structure - its a very slight bend
the bolts are cast into the concrete and can't be changed without removing the whole foundation
I know that the bolt will still be strong enough (its an M24 8.8 FFS), but they want something to prove its still strong enough 
anyone know where I can find some theory on this to keep them happy ?
-Robert
Surely if it was able to get bent in the first place it has been used beyond it's design limits ??
or did sombody do somthing stupid to it.... like swing a 20 ton RSJ into it.
If it ain't fallen down yet it's strong enough 
I very much doubt there's going to be theory on "slightly bent bolts" on the basis that any effect on the bolt will surely depend on
how the bolt got bent in the first place, the loads applied to it etc etc.
Also 8.8 graded bolts will be tested to 8.8 standards. I'm pretty sure that test won't include "how much strength is left after
it's been bent a bit"....
Is it still holding the structure? its strong enough then.
Can you still fit the nut onto the thread?
Can you "persuade" it back into shape?
Can you install another fixing/anchor point close by?
Once the structure is built the anchor bolts have nothing to do but keep the leg plumb, it's not under any more load than how tight you have
tightened it. Tell them that you will have to employ the services of an engineer to calculate the new load capacity (which will not have changed,
steel is elastic to a certain degree remember), and as they have damaged it, they will have to pay for the engineer.
Either that, or bend it back with the old 14lb sledge hammer.
Some mesurements and back of the envelope simple bending theory will give the numbers for the tensile and compressive plastic deformation strain.
But you would need tensile test result for the type of bolt in question to interpret the numbers.
What it really hinges on has the bolt been stressed beyond its yield point ----- yield point and the limit of elasticity are different on some
steels they are quite far apart, if they weren't stretch head bolts wouldn't work.
If that looks OK
They next part of the problem is to figure how the bolt material will behave when it is straighten and the best way to straighten it -- ie the
mechanical properties if the bolt is heat to red heat and cooled. For that you would need a metallurgist.
Because bolts have stress raising features ie the threads I would either apply a generous factor of safety to the results or get a finite element
analysis done by an a credited practioner.
The bolt has been plastically deformed and may have surface cracking and work hardening. The physical properties have been changed and no Qualified
Engineer would certify the strength. The other problem is that if you try to use as it is then the bolt will try to straighten leading to further
deformation etc etc.
If this is a real problem, then the simple answer is to diamond drill a core out of the concrete and fix in a new anchor bolt using a proper chemical
bonding system like Kem fix. It is relatively cheap and certifiable.
I have done this a few times when either the bolting pattern was not known before the foundation was poured, or there was a snafu like the one you
have. It is a perfectly acceptable system but casting in the anchor bolts is cheaper.
If the customer bent it and are the worried ones then surely they will be expecting to pay for it to be corrected.
I'd put together a quote for fixing it properly (ripping out foundations and re-laying with new bolts etc) and the time it will take and let them
decide if they want to take the pain.
Anything else will require a proper structural engineer to estimate the load the bolt will be taking as a worst case - it's possible the other
undamaged bolts are enough on their own, but if not then it's going to be virtually impossible to estimate the remaining strength of a bent
bolt.
Bugger - too late again
[Edited on 3/2/09 by iank]
quote:
Originally posted by yellow melos
or did sombody do somthing stupid to it.... like swing a 20 ton RSJ into it.
quote:
Originally posted by richardlee237
If this is a real problem, then the simple answer is to diamond drill a core out of the concrete and fix in a new anchor bolt using a proper chemical bonding system like Kem fix. It is relatively cheap and certifiable.
I have done this a few times when either the bolting pattern was not known before the foundation was poured, or there was a snafu like the one you have. It is a perfectly acceptable system but casting in the anchor bolts is cheaper.
quote:
Originally posted by richardlee237
The bolt has been plastically deformed and may have surface cracking and work hardening. The physical properties have been changed and no Qualified Engineer would certify the strength
quote:
Originally posted by britishtrident
.....or get a finite element analysis done by an a credited practioner.
You cannot assume that the foundation bolts are doing nothing other than locating the structure, without calculating the loadings.
It is quite possible to have an uplift on the structure due to wind loading or other dynamic loads imposed by machinery.
That said I have seen a lot of plant where the foundation bolts were not even tightened up or the nuts were missing completely and they had been
standing for years.
If you can determine the leg loadings either by calculation or by the original design calcs then you may be able to show the bolt is redundant.
quote:
Originally posted by richardlee237
If you can determine the leg loadings either by calculation or by the original design calcs then you may be able to show the bolt is redundant.
So the client is trying to pass the risk of failure onto yourselves by asking for the assurance.
I wouldn't accept that, I'd respond with - we can no longer guarantee the structural integrity, and some words around you bent it you cough
up for the rectification.
As you are aware of the design loadings and if you are confident that you have enough capacity in the remaining bolts then you can certify it. The
fact that you had surplus capacity in the system before the damage, was because your experience has shown that these things happen. Or words to that
effect. You can also point out that there is some remaining capacity in the deformed bolt as well.
This is when an Engineer has to have the courage of his convictions, training and experience.
quote:
Originally posted by richardlee237
if you are confident that you have enough capacity in the remaining bolts then you can certify it.
My attitude would be the same as when I lend someone a car:
you bend it, you mend it
The other approach of course is to straighten the bolt and get a pull out test done but the foundations might not have enough exposed area to take two
40 ton jacks and load cell.
If takes say 50% of UTS without showing more plastic deformation you have won a watch.
However if the bolt is bent through an arc of more than about ten degrees it is almost certainly beyond saving
quote:
Originally posted by BenB
My attitude would be the same as when I lend someone a car:
you bend it, you mend it
)
quote:
Originally posted by britishtrident
The other approach of course is to straighten the bolt and get a pull out test done but the foundations might not have enough exposed area to take two 40 ton jacks and load cell.
If takes say 50% of UTS without showing more plastic deformation you have won a watch.
However if the bolt is bent through an arc of more than about ten degrees it is almost certainly beyond saving
This sort of situation is exactly what Project Engineering is all about.
It is very easy to say that it is the fault of the steel erector, crane driver or whoever and that they must put it right.
Are you really going to demand they dismantle the whole plant to replace 1 bolt ?
I don't think the client would too impressed or mcerd1's employer even though it would cover his a**e.
The Engineer has got to be able to demonstrate (either by calculation or physical load test) that the plant is still safe or not, that is why he is an
Engineer and not a steel erector.
i can't follow most of this but ....
sounds like they know there is a problem and want someone else to say its ok - that way if its not ok, they get to blame someone else (and have taken
reasonable precautions if judges get involved).
be careful.
Re-thinking pullout test
simpler solution is a crane & load cell -- all thats required is some kind of special fitting to allow a pull to be applied directly to the
bolt.
[Edited on 3/2/09 by britishtrident]
quote:
Originally posted by richardlee237
This sort of situation is exactly what Project Engineering is all about.
It is very easy to say that it is the fault of the steel erector, crane driver or whoever and that they must put it right.
Are you really going to demand they dismantle the whole plant to replace 1 bolt ?
I don't think the client would too impressed or mcerd1's employer even though it would cover his a**e.
The Engineer has got to be able to demonstrate (either by calculation or physical load test) that the plant is still safe or not, that is why he is an Engineer and not a steel erector.
quote:
Originally posted by mcerd1
[I was comeing to that conclusion myself - but its not ideal as the next question will be "why were there so many bolts then if they arn't needed?"
The Construction Design Management Regulations require the designer to provide calculations of the plant loadings and design details and calculations
of the holding down arrangments. It is quite reasonable that the client asks that any deviation from the design is certified to be safe.
If they have specifically queried a bent h/d bolt then you have to assume that bolt is no longer any good. As you cannot calculate it, even by FEM as
you cannot certify the material.
If you really cannot certify the design with the remaining bolts by calculation then it is always possible to weld on an extension to the base plate
and bolt that down to the foundation. Thus restoring the original calculations.
The CDM Regulations were brought in to eliminate, amongst other things, the "she'll be right" attitude.
quote:
Originally posted by richardlee237
If you really cannot certify the design with the remaining bolts by calculation then it is always possible to weld on an extension to the base plate and bolt that down to the foundation. Thus restoring the original calculations.
quote:
Originally posted by richardlee237
The CDM Regulations were brought in to eliminate, amongst other things, the "she'll be right" attitude.
Yes, I know what you mean.
But,
Here in Yemen the concept of Health and Safety barely exists let alone the legislation and practice. Some of the accidents are appalling.
Normal workwear is a pair of sandals, even in the local steelworks !
Health + Safety. What like this??? Absolute quality
Description
Hi Robert,
I am a structural engineer, but of course I'm female so feel free to ignore me.
This isn't much help but to put things in perspective the NSSS allows for bolts to bent under load up to 3 degrees. This is not explicit, but is
implied by tapered flange members not needing tapered washers unless the taper is more than 3 degrees.
Even grade 8.8 bolts have good ductility and will probably be ok, but unfortunately the only way to know is to examine the affected bolt for cracks
before and after it is straightened; not very easy with the structure in place!
What you could do is break out the concrete under the baseplate to expose some unbent bolt. Then cut off the bolt below the bend and fit a coupler.
This could be done with the base plate in place but it's a rigmarole.
How about installing a new resin anchor in line with the two bolts on the affected side of the column and fix a small beam (100x100x15 RSA perhaps)
over the base plate, thus effectively clamping the base plate down?
The above thread does sadden me; the days of "engineering judgement" made by experienced civil engineers do seem to be fading fast. My
field, railways, seems to be the last bastion! I have to laugh when I see consultants produce reams of FEA to proudly proclaim their fixing bolts have
a safety factor of 40+, while they say nothing about the 130 year old structure they want to fix to.
Cheers R
quote:
Originally posted by rachaeljf
How about installing a new resin anchor in line with the two bolts on the affected side of the column and fix a small beam (100x100x15 RSA perhaps) over the base plate, thus effectively clamping the base plate down?
quote:
Originally posted by rachaeljf
The above thread does sadden me; the days of "engineering judgement" made by experienced civil engineers do seem to be fading fast. My field, railways, seems to be the last bastion! I have to laugh when I see consultants produce reams of FEA to proudly proclaim their fixing bolts have a safety factor of 40+, while they say nothing about the 130 year old structure they want to fix to.
quote:
Originally posted by rachaeljf
I am a structural engineer, but of course I'm female so feel free to ignore me.
Right,
You have no moments carried to the feet of those legs because the towers are lattice structures. You will only have tension and/or shear shared
equally between all the HD bolts for each leg base. Therefore it's likely you can get away with 3 of the 4 HD bolts working. If so, you can
straighten the bent bolt and put a nut on it just for looks.
If you do need the 4th HD bolt to work, what you can do is cut off the bent bolt flush to the base plate. Then diamond core the base plate hole a bit
bigger to take a sleeve nut over the cut off HD bolt. The diamond core will also need to take out some concrete so you have enough thread to screw on
to. If you get a sleeve nut turned up to look like a simple nut and washer, and screw in a dummy M24 stud sticking out the top, no one will ever
detect the "deliberate mistake!"
Cheers R
there is actually a bit of moment at the bases, but its not that much (now that I've actually looked at it properly it actually took a few
hour just to find out which structure had the problem! )
I like the idea with the sleve nut
I don't know if they'll go for it in this case, but its worth a shot
The base moment on any leg will be insignificant and results from the geometry, mainly because your truss forces don't go exactly through the
centroid of the base connection. It's not worth bothering about.
How badly bent was this bolt anyway?
quote:
Originally posted by rachaeljf
How badly bent was this bolt anyway?
I am sure some of the engineers where i work must have come across this problem ( they are ex arups, Buro happold).
I will ask them in the morning.
But if the structure is up how did the base plate fit over it,
has the top of the bolt bent after installation?