allowable steel stresses for monorails

huangyhg

allowable steel stresses for monorails
what is the governing standard for the allowable steel stresses for the steel support structure for a monorail with trolley hoist running on the bottom flange of an i-beam track, i.e. the i-beam, end support beams, hangers, etc?
i am confused for a number of reasons.  first, it appears that the cmaa #74 standard, section 74-3, 3.4, permits the use of aisc manual allowable stresses, which generally impose a factor of safety of around 1.67 on yield (about 2.7 on ultimate) but section 1.7.1 of that document says "all other load carrying parts shall be designed so that the calculated static stress in the material, based on rated crane capacity, shall not exceed 20 percent of the published average ultimate strength of the material." which means a factor of safety of 5.0 on ultimate, nearly twice as much.
further, all the hoisting equipment such as chains, shackles, hooks, and the trolley are all provided with a factor of safety of 5.0 on ultimate strength per their manufacturers.
it doesn't make sense to me to require everything but the supporting steel structure to meet a 5.0 on ultimate?  what am i missing here?
check out our whitepaper library.
this is a great question.  i will be interested to see the replies.
keep in mind that the 5.0 for hoisting equipment is not a true "safety factor" since it is based on "static loads" without consideration of the dynamics, impact or fatigue associated with lifting equipment in general.  this is a fairly 'simplified' analysis approach that involves only static loads and static ultimate strengths.  thus not a true 5:1 safety factor.
i would suggest using aisc, lrfd approach for the supporting structure.  service loads are then not simply 'static loads' but should include dynamic/impact/ hoisting factors. loads are then factored based on lrfd principles.  you should then also include appropriate fatigue considerations within the your analysis.
the seemingly "big difference" between the safety factors is thus accounted for.  i would still expect the static 5:1 to be more conservative since it involves 'less accurate' analysis procedures and as such should be somewhat more conservative.
the following are my very general comments that may be helpfull, it seems like a crane code is applicable. if there is hazardous materials or people being moved around/or below then consider safety factors of 10 as more appropriate, and that is on yield, not ultimate. from that work backwords using the aisi design manual (not limit state) to work out fatigue loads, on bolts connections etc depending on weld types. i personally found that designing for 35mpa using a 350mpa fty steel saved me from alote of rework, redesign and steel was inexpensive compared with engineering hours costs.
thanks for the comments from jgrek and dooron.
regarding jgrek's post, i agree that the 5:1 is for a static load while the design of steel support structure per aisc code and the cmaa #74 standard are to include dynamic amplification factors on the static load.  but the cmaa dynamic amplification factors are only up to 1.2 on dead load and 1.5 on lifted load, and could be as little as 1.1 on dead load and 1.15 on lifted load.  even in the most conservative case, then, the factor of safety for the steel stress is about 2.7x1.5 = 4.05 on ultimate strength.  this is still 20% less than the 1:5 factor?
also, i think that your comments about using asd versus lrfd for steel design are not correct, as either method requires the imposition of dynamic load factors on the loads.
regarding dooron's comment, i would love to use a fat-daddy factor of safety of 10 to sleep better at night, but my client would go somewhere else for his future structural work.  i am targeting code compliance here.
i work for a group that engineers cranes and lifting and handling equipment.  here is my understanding of the different safety factors.
1.  items that will see the direct load and a failure would be catistrophic have a minimum design factor of 5 over ultimate(i.e  wire rope, shackles, hooks, lifting lugs, below the hook lifting fixtures).
2.  items that are subject to full fatigue reversal have a design factor of 5 on ultimate or require a fatigue analysis with a design factor of 1.5 over infinite cycle strength (ie. shafts, brakes, gears).  most mechanical items end up with at least the 5:1.
3.  structures that are not subject to full fatigue reversal such as the crane girders and the support structure can use the lower aisc allowable.  they are less likely to fail or there would normally be extensive yielding prior to catistrophic failure.  where you have welded joints in fatigue cmaa 74 table 3.4.7-1 and the associated tables and figures reduce the allowable stress to as low as 11 ksi.  this is a little more than 5:1 design factor for 36 ksi steel.  your crane girders and support beams are often governed more by deflection than strength.  
i hope this helps.
replies are very interesting.
scofie, i am sorry. i did not mean to imply that aisc asd excluded dynamic factors or handled them in any manner that contradicts greatly with lrfd.  i simply prefer the lrfd.  you lost me a bit on your sf numbers.
nevertheless, i believe you should apply a 'realistic' dynamic factor between the cmaa 1.15 and 1.5 based on the application.  this is now a dynamic "service load" that must be further factored.   then apply lrfd factors ll = 1.6 & dl = 1.2 to obtain "factored loads".  reduce strength for fatigue conditions based on lrfd chapter k and appendix k.
as i said, the 5:1, static stress based approach, will likely be more conservative.  there are many reasons for this. i also agree with much of what rockengineer states, especially in regard to mechanical items subject to fully reversed stresses.
i had a professor who called a safety factor an "ignorance factor" because it must account for everything we overlook or are unaware of including potential misuses and abuses.  keep in mind there is no substitute for good sound engineering judgment.  pick an analysis approach that meets this criteria and you can't go wrong.
i like dooron's cautiousness but 10:1 is quite high for standard industrial crane systems.
jg
rolled steel sections are pretty homogenious as far as physical properties are concerned so a fs on 1.67 on yield.
but the manufacture of wire ropes, chains etc. involves more variables and the qualities vary a lot. moreover these 'fabricated' items do not show a clearly marked yield point. thus these are designed with a fs of 6 on "breaking load". whether the breaking load is ultimate load, well, i don't have a clue...
further to my comments, this discussion is very interesting and has very important aspects to do with approach to design.
what i always question is basically that trade study, that if factors are a matter of conjecture, when the cost of steel is $a1200/ tonne and the engineering cost is $a100/hr, is it really worth while to add 1 tonne of steel and be certain or to analyse for 12 hours and not be sure ? further to this, having designed elevating work platforms and mollten metal lifting equipment to factors of safety of 6,7 and 10 what i found from experience, is that a little fillet weld in one detail, (that as an engineer has never given a second thought to) cracks. i will still design those fillet welds, butt fillet welds and other unprepared welds for fatigue, and according to the codes a factor of 10 still stands (35mpa for 350mpa steel).
the issues of designing on the edge of codes comes to haunt engineers now that so many insurance claims put pressure to prove that the thing was designed to a "code", etc when asked to design a balcony railing and the code says 1000mm, why not design it to 1020mm and be sure  you are covered?.
i agree with you entirely dooron, i said as much in an earlier thread - "asd allowable overstress (1.03)".
when materials are relatively cheap compared with labour costs, why try and screw every last n/mm^2 (lb/in^2) out of your steelwork. keep it simple and play it safe, that way if there is anything you have forgotten in your calculations, there is always a bit in hand to play with (if you will pardon the expression).
this applies just as much to this topic, as any others in this forum.
regards,
neilmo
    my thoughts:
    assuming the hoist beam would fail - it doesn't mean it breaks and falls down, it will start bending, deflecting, twisting, it goes through several steps that allow the people to acknowledge that something is going wrong. eventually it will end up in a position where equilibrium is reached... it still has reserves. you'll have to change the beam and figure out what went wrong.
   for a cable or a rope the failing is not so obvious and it doesn't give you time to react... and also there are no reserves there. that's why the big safety factor.