base plate thickness and bolt preload

huangyhg

base plate thickness and bolt preload
i am checking base plate thickness calculations produced by a software. i see that bolt preload is affecting the calculated thickness quite much. it seems to me right. however, formulas in my aonthor reference book(pressure vessel handbook by eugene f. megyesy) doesn't show anywhere the term bolt preload. so i came to wonder the software is correct or not. the foundation is concrete and materials of anchor bolts and base plate are steel. second question i have now is what the typical range of adequate bolt preload would be, any way. any feedback would be appreciated. thanks.
check out our whitepaper library.
i can see where preload would affect thickness, like pre-stressing. however, in my 21 years of experience, i never designed a base plate with preloading. i have designed, like many other fellow engineers in here,  lots of base plates over the past 21 years including base plates with moment connections.
you seem to be smart engineer for questioning the software. i am totally against engineers blindly using software without knowing the theory behind it plus being able to verify the results by manual or other proven software.
i would contact the software developer and ask them. keep us posted.
hi, lutfi,
thank you. your comment is valuable to me. can you tell a little more why you don't need to consider bolt preload even though you think preload affects?
i used to get hung up on the preload in anchor bolts until you consider the following:
1.  concrete creeps under load, so if anything over time the pretension in the bolts would relax a bit.
2.  when checking the design of the bolts themselves, i ignore the preload in the bolt and only consider the actual tensile load in the bolt applied from wind, etc.  if the applied tension load is less than the preload, the plate stays clamped, if it exceeds the preload the plate separates slightly from the concrete and the load in your bolt equals the applied tension load.
3.  when checking base plate thickness it is usually pretty easy to check the moment resistances by hand by making some basic assumptions and considering the forces in the bolts and the compressive stresses in the plate.  i stay away from fancy black box programs for stuff like that.
kyong,
ditto lufti's comments about questioning the software.  
i too have never desgined a baseplate with preloaded anchor bolts.  
my experience is that its not done.  i have run across preloaded anchor bolts to hold down large compressors and things like that.  that is a very specialized design, with sleeves filled with grout, epoxy grout under baseplate and such.  this is done to prevent vibration.  also, if you were to cast an anchor bolt into concrete without sleeves, and tried to preload it, i think you would snap the bolt before it got to specified pretension.  those i refered to had  an empty sleeve to allow the deformation needed to pretension.  and they were not a307/a36 bolts either, they were a much higher strengh steel.  
the ones you are checking, look closely at them.  look at material and the details.  my first guess is they should not be pretensioned.  
keep on questioning the answers from your software.  you're a lot smarter than, and have much better judgement than any software product ever written.
regards,
chichuck
kyong,
pre-stressing anchor bolts are required in some industrial facilities (particularly in petrochemical field), e.g. for tall process towers sensitive to wind, for pumps/compressors or other pulsating or vibrating equipment.
but i didn't quite understand why the preload is affecting the base plate thickness quite much. the stress introduced to the plate is only some self-balanced local compression (the area under the bolt nut or washer). aisc column base plate design guide didn't consider this as one of the checking items.
karlt, i think you can consider the preload in the bolt tension calculation rather than ignore it. just treat it as prestress.
regards,
preload: the tension created in a fastener when first tightened. reduces after a period of time due to embedding and other factors
embedment: localized plastic deformation which occurs in the vicinity of clamped fasteners or in the fastener threads.
joint material: when a bolt is tightened, very high local pressures can exist in the contact areas on the threads and under the nut/bolt. local plastic deformation can occur at these interfaces by flattening of surface roughness.  this is a function of the material young modulus, load, and thickness. this plastic deformation as the effect of reducing a bolt's preload. research as been completed to establish guide values for the amount of embedding that typically occurs within joints. the amount of embedding determined is a loss of joint deformation.
range of  bolt preload: preload varies with the type of fastener, joint design, material clamped.  typical values for preload as low as 25% for structural applications, to 50% to 90% for machines with high vibrational loading.
i will post internet links tomorrow, when i am back in my office.  cheers
as karlit and j1d mentioned, preload is necessary to keep gripping. if the load is alternating like wind on tower, without preload, the joint become loose or gap between base plate and foundation will form. in that regard, i think the preload should be near design bolt load. on the other hand, as chichuck hinted, concrete around top portion of embeded thread may be damaged. that concern would limit amount of preload. this is simply my thinking, though.
any way, as for the reason why preload affects baseplate thickness, the bigger preload is, the more bending at off-bolt area will occur. even though addition of load(wind) will reduce bolt tension and thereby reduce the bending contributed by bolt tension, because bending contributed by load(wind) will always overweigh the decrease, net amount of bending with bigger preload is bigger than net amount of bending with smaller preload. bigger bending requires, of course, thicker plate.
but i am not saying that preload should be considered when calculating baseplate thickness. if many engineers designed baseplate without considering preload, i think, there must be good reason.
my assumed reason is : usually the most-bent area in base plate is outer edge of the plate between bolts, whereas bolt preload affects this area very liitle. it seems to me that bolt preload affects mainly only the area near bolts.
any comments will be welcomed.
boo1, your data about bolt preload range is valueable to me. thank you.
structural connections see
guide to design criteria for bolted and riveted joints at:
as with lutfi and others, would not consider pre-load in the baseplate design.  if pre-load required for other reasons, that's fine...it won't adversely affect the baseplate design at that point.