Cover plate to close access opening for tendon inspection

[ajk1]

Background:

Inspection openings have been chipped into the soffit of a number of beams to permit periodic inspection of the condition of the unbonded tendons in the beams of an above-grade open-air parking structure. The multi-span beams vary in width, but for the illustration, let us consider the 30" wide beams (most other beams are about 24" wide). For the 30" wide beams, the inspection openings are about 26" wide (perpendicular to the beam span) x 16" long (parallel to the beam span). The multi-span beam has about 16 unbonded tendons with the mid-span low point about 2.5" clear above the beam soffit.

When a tendon recently failed (all 7 wires) due to corrosion, it looped out the bottom of the beam and struck the steel closure plate. This caused the closure plate fasteners on one side of the plate to pull completely out of the concrete. The plate was left dangling by the fasteners on the opposite side of the plate.

Question:

I am concerned with the design of the plate fasteners. Is there a way to determine the force that they would have to resist when a tendon breaks and loops down and the loop strikes the plate? Any energy method of equating the energy released by the tendon to the energy absorbed by the plate? I realize the thinner I make the plate and the longer I make its span, the less the force that the plate anchors will have to take. But I am interested in finding how to determine that force.

 

[Lomarandil]

I hate to reject the premise here, because I think it is an interesting question.

But why?

I presume you want to know that a tendon has failed. This seems like a very striking (if potentially unreliable) tattle.

Without knowing more details, it seems that tendon failure is a larger life safety risk than a (non-structural) concrete anchor failure.

 

[ajk1]

To Lomarandil - you have an interesting take on this, but the problem that I see is that if the plate is dislodged, it can fall on someone and kill or injure them, or a lesser issue is that it would hit a vehicle and damage it. Do you feel that these are not concerns? We have an acoustic monitoring system (SpoundPrint) in place that identifies the particularities of a tendon wire break when it occurs, as well as its approximate location (within about 3 m), so it would not seem to me that having the dislodged plate as an additional signal is not all that useful. It may be that the probability of someone being hit by a dislodged plate is low, but that does not mean that it cannot happen...or am I worrying too much?

 

[Ingenuity]

...or am I worrying too much?

You are worrying too much.

Since the acoustic monitor system has been in place, how many tendon failures have been detected, and as a % of the total number of tendons that are being monitored on that particular floor/level, what is that %, and the number of failures per year?

But to partially respond to your initial question: Link: Dynamic Response of Unbonded Prestressing Tendons Cut During Demolition by Martin S. Williams and Peter Waldron

 

[ajk1]

To Ingenuity:

Thank you for your response.

The acoustic monitoring system has been in place about 2 years. We have had 16 "possible wire breaks", or about one every 1½ months. In that time we have had 2 tendons fail and break out of the the beam bottom. These were the 2 where we found the cover plate dislodged. The acoustic monitoring did not report these, but the acoustic monitoring company says that when they looked back at their records they found that it was basically a judgement error in not reporting this, and this error won't happen again.

Whenever I am certain that I am right, I ask myself what if I am wrong? What are the consequences if a plate does hit someone when the tendon fails? Would I be deemed to have not carried out the design that a reasonably prudent engineer would have done? What is the advantage of not carrying out at least some sort of analysis to determine an approximate design force for the fasteners holding the the plate? Which brings me back to my original question: How to determine the force applied to the plate when a failing tendon its it?

 

[graybeach]

Why not bolt up some strong netting below each plate?

 

[Lomarandil]

ajk, I wouldn't say the concern of the plate falling is worth disregarding -- you're right to consider that risk.

Especially in hindsight of the previous failure, I think it is unlikely that a thin plate anchored with a substantial number of small diameter concrete anchors actually falls due to being struck by a tendon. It seems that would require all of the anchors to fail -- if the system is relatively redundant (e.g. the factors mentioned previously), even one or two remaining anchors should be adequate against the self weight of the plate.

(I don't know your scale, but I'm thinking of something like a 1/2"x18"x36" plate, anchored with (6) 3/4" diameter post-installed concrete anchors with plenty of edge distance and spacing).

Granted, you still end up with anchors falling, but the damage from a 6"x3/4" anchor is substantially less than a 100lb plate.

 

[Ingenuity]

ajk1,

How do you address tendon/s that fail and spall a 5lb chunk
of concrete at the soffit low point?

 

[ajk1]

To Lomarandil- the plate is < 3/16" thick. The anchors were similar to Tapcon screws. No design was done to determine the screw size, apparently. You make an interesting point about 1 or 2 fasteners still being adequate to prevent the plate rom falling even if the other anchors fail.

To Ingenuity - although a number of tendons failed (before the inspection openings were chipped out of the concrete), there has never been a piece of concrete come off the bottom of the beam. The tendon failures resulted either in the tendon coming out at the end of the beam, or quietly dying within the structure. It is my experience over the last 30 years of involvement with post-tensioning that the tendon does not break out of the beam if the clear cover is 1.5" to 2" or more. The tendon breaking out is more of a problem with slab with 1" cover. When that is the case, then yes indeed it is a significant concern about whether someone will be hit by a failing tendon and/or a piece of concrete. I believe that different engineers have dealt with that in various ways. But we are straying from my question. It would be good if there were an academic on the forum who can help me with the question, even if you don't agree with my concern...just for intellectual interest.

Thanks for the help to date. Much appreciated

 

[Ingenuity]

With multi-millions of square feet of PT constructed SLABS with antiquated UNbonded PT systems (over the period from 1965 to say 1985, in North America) how many tendon failures over these two decades resulted in slab soffit spalls, and compare that to your situation where you don't have any slabs (only PT beams with greater cover) and your concern is tendon failures striking a few (isolated?) soffit cover plates - pretty small probability, IMO.

Academics like Rogowsky and MacDougall (both Canadian) have researched monostrand tendon failures, related corrosion, wire breaks, etc. It may be worth checking with Uni of Alberta and Queens (where they are/were professors) to see if either/both have any related research similar to your situation.

If you read the Williams and Waldron paper I referenced above, they did both a mathematical model and lab testing program to verify tendon failure dynamics of strand exiting live-end anchorages due to forced severing. May be worth a read for some background information.

Given the client has probably spent $100k+ on a SoundPrint[sup]®[/sup] acoustic monitoring system, it may just be prudent to increase the anchors size and move on.

 

[ajk1]

Thank you for your comments and reference papers and contacts. I know Dave Ragowsky somewhat from my CSA committee work about 25 years ago. Where is he now? University of Alberta? I will consider calling him. Much appreciated. I will also refer to Williams and Waldron as you suggest.

You do not know the structure and I was perhaps amiss in not giving more information in my original post. In the case that I am concerned with, there are several hundred of the these soffit inspection opening closure plates, located in main walking and parking areas, each plate weighing about 12 pounds.

I do not have the statistics to answer the question about how many of these tendons having broken out of slab soffits in North America, but I expect that it would be a significant number...I do not know if there has been any attempt to determine that statistic. I don't expect that it would be easy information to obtain, because owners would not want it broadcast that they had a problem with their building.

I agree with your conclusion about increasing the plate fastener size, but I thought it should be based on an engineering calculation. That is the reason for my original question. But I do appreciate your take on things and will factor that into my deliberations.