Additional Dead Loads - Bolts, Connections, Paint, Plates...etc. 1

[Superstar123]

Hello,

I am just curious as to what percent other engineers are using to account for things such as bolts, connections, paint, plates..etc. for their designs.

For context I am designing a monorail and there is some discrepancy in my office regarding whether or not 15% for connections is too much or if 5% is too little.

What are other people using / what are your thoughts?

Thanks,

 

[kostast88]

Whenever I find a case where I am conscious I should do something like that, I usually factor the dead load by 10%.

There is no general rule for that.

Just be conscious whether 5% or 15% is an actual representation of the additional elements you have.

 

[skeletron]

IIRC it was something in the magnitude of 10-15% for connections or, as one of my first supervisor's would say, "jewelry". Welds can actually add quite a bit of weight.

 

[GeorgeTheCivilEngineer]

Network Rail (owner of British railway infrastructure) has 12% in their standard for riveted girders - these were frequently heavy on the stiffeners and extra angles.

For rolled and mainly welded sections it'd be closer to 5%, no more than 10%. Well below dead load safety factor

 

[bugbus]

The consensus in my previous office was about 10% extra to account for all the above, then apply the usual load factor on top of that.

 

[canwesteng]

In a typical structural model, beams are modeled to column centers but would be cut back a bit to frame into the flange. In heavy industrial, I did a sanity check once and found 5% would be lots. I usually include zero allowance, since I'm heavy elsewhere, and having extra dead load is unconservative for uplift calcs.

 

[human909]

As an experienced structural detailer and fabricator it is generally 5%-10%. Depending on the nature of the job.

As a structural engineer it is 0%. Dead weight of the structural steel accessories becomes a rounding error for almost all applications.

 

[canwesteng]

For detailing, I think you are adding 5%-10% to the true weight of the steel - but analytically we run beams as fictional 1-D lines to nodes, so the beams are always longer for the case of software estimating dead load.

 

[Aesur]

Wait, people don't actually add up all the bolts, figure out the paint thickness etc. to calculate the weight?

Typically I have seen a MISC category in my dead load that accounts for this as well as design as canwesteng said, so rarely even think about it. The Misc loading typically is something along the lines of 1 to 4 psf (could be higher if needed) to get to a nice single digit dead load number and account for all the other misc items.

 

[MotorCity]

0%. Thats what load factors, allowable stresses, safety factors, etc, are intended for....to account for variability in loads and materials. I argue that very few loads can be accurately calculated to within 5%, nor do they need to be calculated with such accuracy.

 

[bridgebuster]

Calculating the weight of paint is taking things too far. However, on some DOT projects the steel is paid by the pound, the weight of welds and bolt heads, washers, & nuts has to be accounted for in the quantity. The specs provide weights per hundred bolts and weights for a foot of weld.

For straight estimating, I typically use 5% for those items. When I was young and ambitious, I actually checked the 5%; it was closer to 3%. As we say, "close enough for government work."

A little off topic, Paying steel removal by the pound is often be a headache for the field staff when they're dealing with built up riveted sections. I always tell younger engineers to pay removals as lump sum or each wherever possible. When I worked a construction inspector on steel bridge projects, the two worst tasks were torquing bolts and calculating weights of built up members that were removed.

 

[Lomarandil]

5% on members and 5-10% on connections is my practice for new steel. And like bridgebuster, I've done the detailed comparison to find those are typically conservative by a few percent. Increasingly so with modern mill tolerances (where most steel comes in at minimum dimensions).

Existing built-up members can get much higher, over 20% of raw weight on a built-up truss "knuckle".

I understand when permanent designers gloss over those as included in the load factors. When we are talking about pick weights (and CG) on a crane hook, sometimes the precision is needed.

----
just call me Lo.

 

[BridgeSmith]

For design of bridge superstructures, we typically add 40 plf for interior girders and 20 plf for exterior girders to account for stiffeners, crossframes/diaphragms, bolts, etc.

For the structural steel quantities, we calculate weights for all the plates, angles, etc. and add 2% to the total for bolts, nuts washers, and welds. I guess the paint would also be covered in that 2%, but we don't adjust anything for girders we don't paint.

Rod Smith, P.E., The artist formerly known as HotRod10