Factor of Safety

[dik]

I have a coffer dam to design that is retaining about 10' of water. If it fails, it will flood a confined space with a likelihood of loss of life. Possibly as many as 20 people.

What magnitude of safety factor would you use for either service loading or limit states load factor would you normally use and are there any authorities or mandated factors of safety?

I'm leaning towards a SF on service loads of approx 5...

Dik

 

[BAretired]

A safety factor of 5 sounds extremely conservative. Water pressure is precisely known, so it is more like a dead load than a live load.

BA

 

[dik]

BA:
I was thinking of the life safety issues; there is a high likelihood of fatalities for failure and although the loads are relatively known (a little wave set up or whatever) they are real loads.

You would think a safety factor in the order of 3 would be more appropriate?

Dik

 

[vandede427]

go with your first instinct. don't comprimise. you like to sleep at night don't you?

FS = 5

 

[dik]

Are there any authorities or industrial codes that recommend a load factor or safety factor for this type of construction?

Dik

 

[miecz]

Around here, cofferdams are normally designed by the contractor and submitted for review. I've reviewed a few of them over the years and they generally follow the USS Manual.

http://www.mcipin.com/publications/sheetpiles/USSteel_1984-SteelSheetPiles.pdf

 

[BAretired]

dik,

I am not aware of any code stipulating a special factor of safety for your particular case. All structures, in the event of failure can jeopardize the lives of people. In the case of auditoriums or arenas, that would be considerably more than 20.

A safety factor of 5 is sometimes used for cranes, lift hooks, cables and chains because they are loaded dynamically, used over and over again, moved from place to place and used in all kinds of weather.

You should go with your instinct on this and use whatever SF makes you comfortable. Personally, I would use the same safety factors as I use in building design.

In addition, you should observe the ACI requirements for water retaining structures.

BA

 

[DRC1]

Typically the walls are designed on AISC ASD rules. Passive soil capacity is generally reduced by a factor of safety of 1.5. Generally a factor of safety of 1.5 to 2 is used against a bottom blow. There is an unmeasured factor of safey based on what values are selected for soil properties. This is probably the controlling factor- How well do you know the soil conditions and is your model accurate. Also some dams are designed to flood in a controlled manner in the event of a problem.
Designing cofferdams is considerably different than typical strucural or foundation engineering, and is based on considerable part on experience, contractors construction methods, and local conditions. If you are the least bit unsure, seek out a local designer experienced in cofferdams.

 

[dik]

DRC1:
The retained material is not soil, but water and a failure of this will cause flooding of a confined space where people cannot escape.

Dik

 

[SlideRuleEra]

DRC1's recommendations for the soil are important. For a typical cofferdam design (without a concrete seal), the soil (that the sheet piling are driven into) carries a load equal to, or higher than any of the structural wales.

The shape, size, and connection details of a cofferdam are more important than a large safety factor for sizing the structural members. Attention should also be directed toward disassembly after completion. Assuming some type of permanent structure was constructed inside the cofferdam, disassembly may be different than assembly.

http://www.SlideRuleEra.net

http://www.VacuumTubeEra.net

 

[dik]

The coffer dam will be secured and gasketted to a concrete base and walls. It is a steel frame with a metal 'skin' that is anchored to existing concrete.

Dik

 

[JAE]

dik

I think your general concept of higher risk = higher desired SF is correct. In fact, this concept does occur within current codes, although not to a SF = 5 extent.

ACI, for instance, sets the [φ] factor for shear at a lower level than that for flexure due to the fact that shear failure in concrete beams is abrupt, with little warning while flexural failure is more gradual.

Also, I'd remember that codes are a minimum standard and you as the EOR should certainly consider risk of failure in the design.

"When in doubt make it stout" applies when there is a higher than normal level of uncertainty in the capacity or loads. In your case, you probably have a good level of certainty of the loads, but might feel the capacity has a lot of variables that the basic code capacity calculation might not include.

Whether a SF of 5, 10 or 1.6 is appropriate is probably more per what BAretired suggests - what makes you comfortable and yet is relatively reasonable.

 

[JohnWylie]

Go with your gut...just remember, you can't over-engineer it!!

 

[JedClampett]

I've had to design some scary structures, including ones where if they failed, I hoped the workers in the vicinity could swim.
But before I would talk in terms of Factor of Safety, you need to define what it's being applied to. Yield, allowable design stress and a guillotine failure are vastly different limits and all could be compared as a factor of safety.
Make sure you know where your uncertainties (loading, materials, workmanship, etc.) are, and use good judgement. In other words, as an example, if you're certain of the loads, there's no reason not to design steel members to 21.6 ksi.

 

[Zambo]

10' of water isn't that high for a cofferdam. I've only seen one cofferdam fail that was when some dredging had been done out of sequence and too deep and replaced the soil used in the design. The partial failure happened as soon as the sheetpile cofferdam was pumped out and as it was holding back 8m of water it was a bit frightening.

But in general I would have thought normal factors of safety would be fine. In fact hopefully normal factors of safety could already be considered to be conservative.

The detailing of the cofferdam as a water retaining structure will be critical.

 

[GregLocock]

How much money do you really save by going from SF 5 to SF 3? If you ever need to reuse it in deeper water then you have paid for the more robust design already.



Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[JStephen]

I would consider also whether some other approach was appropriate rather than just increasing the factor of safety.

By analogy with the auditorium mentioned above, you could increase the factor of safety on the design of the auditorium, but that isn't ever considered a substitute for having proper exits, fire protection, lighting, etc.

 

[dik]

Jed... should have mentioned that it was the anchorage of structural steel supporting frame. If the coffer dam fails, the workers will not be able to swim to safety... they will be trapped

Dik

 

[VoyageofDiscovery]

Hi dik,

10' of water is not unusual. Design it as any other structural system, with ring beams allowing for inner strut removals where required. Get loadings and soil resistance from a geotechnical engineer. My only area of caution is the tremie plug wrt its resistance to flooding when in the dry.

HTH