to check if beam crack is live or not 1

[zeeshanahmed]

There are minor cracks in the midspan of the beam, i want to check if the cracks are "LIVE" and still increasing or not. I usually paste 2mm thick 5" long glass stips exactly below the crack, so if glass breaks it means crack is still expnading ? is this correct or not ? any better suggestions ?

 

[csd72]

There are a number of ways you can mark it for this.

The simplest and most crude is to mark two spaces a certain distance apart.

another is to use 'tell tales' two tear shaped plastic items which are nail fixed either side of the crack. You then put a felt tip mark across both so that you can tell when they move.

Glass is a suprercooled liquid and as such will flow with slow movement over time, thus it may not necessarily indicate movement.

 

[jheidt2543]

There are crack monitoring devices specifically for your situation and here are two places to look. I've used the Avongard monitors in the past and they are economical, easy to install and simple to read.

http://www.avongard.com/

http://www.sy-survey.com/index/eng/index.asp?gclid=CNP9us-IspACFSBMGgodMQ_FLw

 

[bvanhiel]

Glass is a suprercooled liquid and as such will flow with slow movement over time, thus it may not necessarily indicate movement.

Glass does not flow at room temperature. This is an oft repeated jewel of "wisdom". I don't want to push this thread completely off tipic, but I do hate to see misinformation spread. Do a couple of google searches and you'll find some sites debunking it.

-b

 

[prost]

A common method to measure crack propagation in a fatigue test is to use a 'crack gage.'

http://www.vishay.com/brands/measurements_group/guide/500/lists/cpg_list.htm

OR

http://www.hbm.com/Products/SEURLF/...ORY.19/MM.3,33,89/SFE/DisplayProductTable.htm

Don't have any idea how robust these are in cold/wet/hot environments (I am guessing your beam is out in the elements someplace, not in a lab).

I would doubt putting two marks on the beam, opposite sides of the crack, would give you the precision you needed to detect cracks, unless of course the crack opens up a discernible gap, AND you are able to make the measurement when that gap appears. There is a gage made by DMI, Inc. that you could conceivably lay across the crack and use their small optical tool to measure variations. I just heard about this gage, so don't know much more than I just told you.

Kind of off topic, but none of the 'glass doesn't flow' arguments I have found on the Net mentioned the obvious waves in glass windows that only are at the bottom of old glass windows; they all talk about glass thickness variations. Certainly glass panes could have been made with the waviness; does seem rather odd that the waviness seems to be most often at the bottom of windows. An opposing argument is of course if 'glass flows' then all old glass would have this waviness, but that's not true either of course, as there are many counterexamples in the very building where I sit. Therefore the waviness must be in the pane before it was mounted in the window frame.

 

[dik]

Drill an 1/8" hole on either side and install a concrete nail in each. Use dial gauge vernier calipers to measure any movement. Holes can be 2" away from crack. Can measure at 1 day, 1 week, 1 month, 6 months, 1 year, 2 years, etc. If no movement then not necessary to check further.

Dik

 

[VoyageofDiscovery]

zeeshanahmed,

Where are these cracks located at midspan and at what orientation? What is this beam functioning as? Are there diaphragms present here?

VOD

 

[rb1957]

why so sophisticated ? an indelable ink mark at the current tip position, and another 1" away would tell you plenty ... do you need remote monitoring ? add a video head, heck add a laptop and have live monitoring over a webcam ...

 

[HgTX]

prost--dangerously off topic indeed. If you want to see how it was already discussed to death a decade ago (including the "thicker on the bottom" false reasoning), go here:

http://tafkac.org/science/glass.flow/glass_flow_the_thread.html

Hg

Eng-Tips policies: faq731-376

 

[zeeshanahmed]

let me shed some more light on the subject.

This is a building of Telenor (a cell phone company) its a MSC building which means loads and loads of equipment inside which is so dust sensitive that when you go inside the hall you need to wear shower caps, gloves and shoe covers. on the 2nd floor in this two story building all the longitudinal (major span) beams have typical bending cracks in the center of the beam starting from bottom of the section and moving up on sides -its a good textbook example of bending cracks- appernetly there are heavy aircons unit on the roof which are directly on top of these beams and it seems that these loads were not taken in account while telenor was renting this building. Anyways even the thought of some kind of retrofiting creeps the hell out of the management here as it would mean to shut down a complete sector, pack the equipment inside and then do some retrofiting work. So i performed an analysis (taking the aircon loads in account) and failure ratio roughly is 1.2. I have suggested to them that lets put some type of gauges on the crack to observe the cracks, if there is no more propagation in the cracks then lets live with this as it is, but if the cracks are expanding then there is not other option other than going through the retrofiting of building. The building is a simpe 25'x20' regular grid of 120' x140' , two story building. SO hows that ? ?

 

[corus]

I'm not sure why a crack would 'die' as such unless the loads were transmitted elsewhere in the structure, and as such the crack is live. Crack propogation will increase under some Paris law until eventually it reaches the critical crack length and then 'ping' - to use a technical term. How close is the crack to the critical length. Has the neighbourhood been evacuated?
I'd maybe drill a hole at the crack tip to arrest crack growth and then add some kind of stiffener to it so that the beam meets design requirements for the expected loads.

corus

 

[rb1957]

by "failure ratio" i figure you mean that the loads exceed the building strength, as defined by your codes, by 20%. the question should be how safe is the building ? the codes consider extreme events, but then nothing particularly extreme has happened to initiate the cracks ?

"typical" is another word that looks funny to me ... do you mean that pretty well all the beams have these cracks, or that most buildings have them (somehow i don't think so)?

are the cracks in the flanges, growing towards the free edge; or in the web ?

maybe you could calculate the intact stresses in the beam (i think you've done this) and then calculate the loss of strength due to the cracks.

is the critical loading plain bending due to the loads, or is there a vibration component ?

if the a/c units are the problem, could they be moved to the ground ? or maybe have some supporting structure (external) to help your beams ?

measuring the crack growth is something, but like corus i don't see wht the crack would "die"; maybe if it is growing towards the edge you might gain some time before it reinitiates on the opposite (web side) of the hole (i'm assuming the cracks started at a hole, and not the free edge of the flange). if the crack is growing into the web, don't think it'll slow down due to the bending stresses in the beam, it'll speed up 'cause all that moment is being pushed into a less effective section (assuming the crack growth is being driven by bending).

i think the building has bigger problems than down time (to implement a repair). as they say, "if you want to find to price of safety, try an accident".

a thought, if you tried to rework the beams, the work could be contained in small sections, and the work space could be kept at a small -ve pressure, vented to the outside.

 

[jetmaker]

Why not do the reinforcing from the outside? If it is the AC units which are the problem, new beams on the outside which support the AC units mught resolve the problem as the AC loads bypass the original roof beams.

On the crack "die" issue. If the beam is in bending, then it is possible for the crack to stop as it moves closer to the neutral axis of the beam. Now this assumes there is other structure in the beam that prevents the neutral axis from shifting significantly during crack growth (i.e. crack in web of built-up structure where chords remain in tact).

jetmaker

 

[corus]

Jetmaker, I'm not sure what you are saying. The crack doesn't die because it moves to the beam neutral axis where there is zero stress, simply because as the crack deepens the neutral axis moves too so the crack tip is effectively always at the beam surface. The section modulus of the beam reduces as the crack deepens and hence the bending stress continually increases for the same load as the crack deepens until eventually..ping.

How can you have a structure that prevents the neutral axis from moving when it's cracked half way through the depth??

corus

 

[rb1957]

corus, i had the same thoughts but check jetmaker's final comment (in brackets) ... if the beam chords are separate elements then the web crack will slow down, but the load in the chords will increase (as the load redistributes away from the cracked web). a minor (?) may be the net shear strength of the web ... we're primarily looking at the bending strength of the section, it also has a job to do in carrying shear.

 

[VoyageofDiscovery]

This is amazing, the lengths people go to appease the Owner.

Evacuate this building! Forget rehab and carry out replacement. The calcs show overstress, the midspan is cracking, now you want to watch it fall down with strain gauges!

 

[apsix]

Assuming we are talking about concrete beams (or have I missed something?) tension cracks do not mean imminent failure and don't necessarily mean that there is a problem at all.
If the calcs show 20% overstress then intervention is required, if that's what you mean by 'failure ratio roughly is 1.2'.

 

[hokie66]

Yes, apsix, I agree. Some of the responses above were talking about steel beams, when the OP was obviously about reinforced concrete beams. The cell phone company has negligently overloaded the roof slab and now must pay the price.

 

[zeeshanahmed]

ok some more clarificaions:

a- The beam is concrete beam and NOT steel. this is concrete frame structure.

VoyageofDiscovery:
you are suggesting the easy way out obviously evacuating and repairing is the best solution, but these cracks are hairline cracks, and not the kind of cracks which would cause immediate failure. shooting a bird with a heat seeker missile sounds easy and fun. anyhow i think observing cracks for a while say a year or two should be done and if there is even slightest of movement then offcoarse retrofiting would be it. If not then i presume it can go on.

 

[VoyageofDiscovery]

Good point, but you need to be more descriptive of your situation when you make a post.

I assumed this section was steel.