Need help with foundation design spreadsheet

[Fat Ryan]

I am an civil engineering intern for SAIC in MD. The PE I work under asked me to create a spread sheet for designing the foundations for our projects which are 30-50' steel speaker poles subjected to 100-140 mph winds. there are 4 foundation/pole types, standard anchor bolt, direct embed, stub up, and slip fit. all types that use bolts have circular bolt patterns. I am working on the standard anchor bolt design.

we (SAIC) apparently send out the pole dimensions and loading to a third party to calculate the forces at the base and design the specific amount of bolts, their diameters, and the loading on each one. I am assuming they use finite element analysis, which I am unfamiliar with at the moment. Is there a way I can calculate this stuff myself to save them some money? I've heard FEA is rather expensive, and this is taking quite a while to get the results as well.

Or perhaps, if I could approximate the pole calc by assuming its a cantilever beam, I could get the N, V, and M at the base. But from there Id need a method to determine the correct amount of bolts to use and their respective diameters. Assuming I desiged for a worst case M scenario, then the most heavily loaded bolt would be the furthest from the NA, correct? Also, I can assume it has an eccentric axial load. by the way, the lateral (wind) load controls by a long shot in these cases. in older hand calcs i have access to, its assumed that the shear force at the base is divided evenly amongst the bolts. given all this, it seems as though i have a method for doing this by hand. the problem comes in deciding a bolt pattern. if I could make it decide on an amount and diameter of bolts, that'd be convenient, however I am not quite sure if I can do that. furthermore, i don't know if its common practice to only have an even number of bolts. it seems that an odd number makes things a bit more complicated.

perhaps I've just overthought this so much that my brain has given up on trying to find a solution lol. that said, if you have any advice, let me know...and don't laugh at me if this is way simpler than I am making it out to be :D

 

[Fat Ryan]

anyone?

 

[sardonicus]

It's a good task for an intern I think. You mention foundations and you mention the base connection, you'll need to get that clarified.

If it's the foundation get a copy of the typical design calculations and make your spread sheet do the same thing. The IBC has formula for embedded post in ground with lateral load, that is commonly used for light posts etc around here but our wind is only 90 mph.

It sounds like you are concentrating on the anchor bolts though. It is common to divide the entire shear by the number of bolts then combine that with the highest bolt tension. Design of an anchor bolt is quite a project in itself now particularly if it's within three feet of anything but mass concrete, courtesy of the ACI.

If the question is how to find the maximum bolt tension for a given bending moment, bolt pattern, and plate thickness, that can be easy or hard, your choice. If your office does structural engineering you should have a structural analysis program, most can be used for simple FEM. A simple brute force design would do in my opinion, like design of a concrete beam, .85fc'A on one side AsFy on the other, then check the plate.

Might try eng-tips.com too. S

 

[Fat Ryan]

It's a good task for an intern I think. You mention foundations and you mention the base connection, you'll need to get that clarified.

If it's the foundation get a copy of the typical design calculations and make your spread sheet do the same thing. The IBC has formula for embedded post in ground with lateral load, that is commonly used for light posts etc around here but our wind is only 90 mph.

It sounds like you are concentrating on the anchor bolts though. It is common to divide the entire shear by the number of bolts then combine that with the highest bolt tension. Design of an anchor bolt is quite a project in itself now particularly if it's within three feet of anything but mass concrete, courtesy of the ACI.

If the question is how to find the maximum bolt tension for a given bending moment, bolt pattern, and plate thickness, that can be easy or hard, your choice. If your office does structural engineering you should have a structural analysis program, most can be used for simple FEM. A simple brute force design would do in my opinion, like design of a concrete beam, .85fc'A on one side AsFy on the other, then check the plate.

Might try eng-tips.com too. S

Thanks. What I'm REALLY supposed to do is design the embedment lengths in the spreadsheet, because the third party doing the analysis should be determining the anchor bolt diameter and number of bolts. i can only assume that their design accounts for the combined loading and any applicable limit states.

i was just hoping to create something a bit more useful and really "wow" them lol. having a spreadsheet that still relies on third party analysis input seems ridiculous to me. i was having some issues determining the correct design to use for the embedment lengths. this is a 19" pole (circular baseplate & pattern) on a 36" concrete foundation. the foundation is either flush with grade or 3.5' above grade depending on location.

i have AASHTO 2009, ASCE 7-05, and ACI 318-08. there are all kinds of different conditions having to do with the reinforcing bar development and its proximity to the anchor bolts. i think reading all that code over and over has only confused me more lol.

as for the other issue if i were to design the bolts for loading...i can get their loads/stresses like a stated before. the problem comes when calculating "I" and Mc/I...the "c" is the issue. first, the c will vary depending on the amount of bolts and the placement of the NA. how do you find the worst case scenario? for example, in a 6 bolt pattern with NA through 2 bolts leaves c1 = c2 = R*sin(theta), but if the NA goes between the middle of 2 bolts on either side c2=R and c1 = c3 = R*sin(theta). and as i said before, having an odd number of bolts just makes it even harder. there are so many different combinations of bolt patterns, NA positions, and "c's" for the MC/I calculation. I've just been racking my brain to figure out how to avoid making hundreds of different calculations lol. Please, someone give me some insight!

edit: by the way, I am using ASD (because the CE i work under uses ASD and doesn't seem to want to try LRFD). furthermore, its assumed the anchor bolts' length from foundation surface to bottom of leveling nut is <= D(anchor bolt) meaning there's no buckling checks as per AASHTO specs.

 

[sardonicus]

Finding the bolt embedment is a project in itself, will just have to plow through it.

You should be able to come up with factors for patterns with low numbers of bolts, it can be independent of bolt diameter and bolt circle radius. Plow through that.

For larger numbers of bolts you can use an equivalent ring, a ring that has a diameter equal to the diameter of the bolt circle and a thickness such that the area of the ring equals the area of all the bolts.

Most still use ASD but strength design makes sense for concrete beams and things like it.

 

[Fat Ryan]

so what your saying is, i have to do 8,000 calculations...awesome

 

[sardonicus]

Maybe focus on one thing at a time. The bolt pattern is the easiest to solve.

The lowest value for St will result using an axis perpendicular to a line bisecting the angle between two adjacent bolts. Odd number of bolts or not. (St=I/c tension side)

360 / (2 * number of bolts) = angle that first bolt is off center. From there it's just numerical integration ... summing, and a division or two.

 

[Fat Ryan]

Maybe focus on one thing at a time. The bolt pattern is the easiest to solve.

The lowest value for St will result using an axis perpendicular to a line bisecting the angle between two adjacent bolts. Odd number of bolts or not. (St=I/c tension side)

360 / (2 * number of bolts) = angle that first bolt is off center. From there it's just numerical integration ... summing, and a division or two.

for the life of me, i cannot seem to make sense of what youre saying lol. any way you could sketch this up in mspaint real quick?

 

[sardonicus]

I think you'll find that the neutral axis will be parallel to a line drawn between two adjacent bolts.

 

[sardonicus]

I'm not a mcpainter but attached a rough sketch.

 

[Fat Ryan]

I'm not quite sure how this relates to the anchor bolts. Did I fail to mention that there's no grout between pole and foundation? Thus means all strength is in the anchor bolts, doesn't it? Meaning a whiteny stress block isn't applicable...

 

[sardonicus]

Ryan, can you make a sketch? That's easier than guessing.

 

[Fat Ryan]

mines not the greatest either, but it should work...

 

[sardonicus]

That helps.

So for x number of bolts you want the lowest value of I/cA (inches).

I think there is just two calculations to do for each, one with the axis through a bolt and one with the axis half way between two bolts. So set up spreadsheet to do both and then take the lowest value.

 

[Fat Ryan]

that is assuming an even number of bolts, which was a concern as well. I would imagine it'd be a even number but you never know. Ill give this a try.

As for the embeddment depth, do I use ACI spec for deformed bar? It doesn't really seem to specify one specifically for anchor bolts. And if they're threaded bolts, that's kinda like deformed bar, right?

 

[PhanthomJay]

I'm not quite sure how this relates to the anchor bolts. Did I fail to mention that there's no grout between pole and foundation? Thus means all strength is in the anchor bolts, doesn't it? Meaning a whiteny stress block isn't applicable...

You need to use double nutted anchor bolts...one above the base plate and one below the base plate...then yes, there is not stress block on the concrete, just tension and compression in the anchor bolts...per Mc/I...once you determine the max load in the bolts, use deformed bars for the anchor bolts, with the top few inches threaded for the nuts... And then use ACI for anchor bolt embeddment depth.

 

[Fat Ryan]

Yeah o left leveling and locking nuts out of the picture, but that's what I meant.

Can you clarify your last part? I'm not sure what you're trying to say to use for embeddment depth, you mentioned both deformed bar and anchor bolt

 

[PhanthomJay]

The deformed bar IS the anchor bolt. That's one way of fabricating the anchor bolt...you take a deformed bar of the strength required per your force calculations for the anchor bolt size, and just thread the top few inches to get the base plate fastened to it with the double nuts. There are other ways... a plain partially threaded rod with a hook at the end, a swedge bolt, a Hilti anchor, etc., but I like the partially threaded deformed bar. Let's say in your example you have, to use a number (I didn't do any actual calcs), 4800 in-kips of moment (factored load) at the base of the pole. Let's ignore shear and axial loads. Try using 4 anchor bolts (always use even numbers in multiples of 4) on a 20 inch bolt circle. 2 of the bolts lie on the neutral axis for a certain wind direction, and carry no load, so just the outer 2 bolts take the load, one in tension at 4800/20 = 240 K, the other in compression at 240 K. For a 60ksi rebar, a #18S will do, so use 4 -#18S rebars, threaded top few inches, on a 20 inch bolt circle, for the anchor bolts. Look up the embeddment depth required to develop the 240 K load in the concrete based on the bond stress. I don't have my book handy, it'd probably about 10 feet. Now that may not be economical, you'll need at least a 11 foot deep concrete pier, (actual pier depth depends on soil conditions), so you might want to try 8 anchor bolts, and then your rebar (anchor bolt) size and embedment length will be less, and concrete pier will be less deep (soil permitting).

 

[Fat Ryan]

Are the anchor bolts in this example separate from the reinforcing bar for moment capacity of the pier?

In my situation, they are separate. And they have a bolt at the embedded end for anchorage. Are you sure I can assume they're the same as deformed bar or are there other common options? I don't want to go assuming they're the same as the deformed bar then they end up being some more ductile steel with no deformations or something of the like.

Also, I thought there was more to embeddment depth than just bond strength. Don't you have to check pull-out, pry-out, moment reinf development, etc?

 

[PhanthomJay]

Are the anchor bolts in this example separate from the reinforcing bar for moment capacity of the pier?

yes

In my situation, they are separate. And they have a bolt at the embedded end for anchorage. Are you sure I can assume they're the same as deformed bar or are there other common options? I don't want to go assuming they're the same as the deformed bar then they end up being some more ductile steel with no deformations or something of the like.

Also, I thought there was more to embeddment depth than just bond strength. Don't you have to check pull-out, pry-out, moment reinf development, etc?

Oh, OK, I am looking at your figure again, and see that you have both anchor bolts and rebar in your pier. This is fine, but you show the rebar and anchor bolts with the same bolt circle. This is not good practice, because often the anchor bolts are preasembled in a 'cage' so to speak, for ease of placement rather than place them individually and mess up the spacing. The rebar should have say a 26 inch bolt circle if the anchor bolts have a 20 inch bolt circle. The rebar for the pier reinforcement is usually just minimum steel required for temperature reinforcement rather than based on stress. In regard to anchor bolt embeddment depth using deformed bars as the anchor bolt, all you need to know is the ultimate bond stress for the concrete (I think its about 300 psi for 3000 psi concrete). Don't worry about the other stuff.

 

[Fat Ryan]

Yeah but I need to follow ACI, and that has dozens of checks.

The anchor bolt circle is smaller than the deformed bar circle...that is another mistake in my drawing.

I don't know if it helps, but we use a jig for laying the anchor bolts. You bolt them all onto this metal circle, lay them in the pier with the circle about the top of the concrete, then pour, and remove jig when cured.