FLAC3D: pile under a vertical pull out load
Current time: 11-14-2018, 02:34 PM
Users browsing this thread: 1 Guest(s)
Author: chan372
Last Post: tino27
Replies 7
Views 4989

FLAC3D: pile under a vertical pull out load
#1
Hello,

Is there anyone here who is familiar with FLAC3D?

I'm trying to model a pile (anchor) under a vertical pull out load using this software. Basically, I modified the example 8.1 "Axial & Lateral Loading of a Concrete Pile" from FLAC3D manual. I modified the direction of applied velocity to get the pull out mechanism.
Somehow I can't make the soil attached to the pile tip (i.e: can't have pile point resistance), thus I get lower pull out capacity when compared to hand calculation.
Can anyone please help me? Any comments, hints, suggestions would be very much appreciated.
Thank you.

NB: I'm not sure if I should put the project file that I've written here, but I can put it if it's necessary.
#2
Can you put the project code file here?
#3
hi nvmhumg,

ofc, here's the project code file that i've made.
The plot between axial stress vs axial displacement indicates that the maximum capacity is 4e5 N/m2 (i.e. 314kN, considering the cross section area of the pile is 0,785m2), while my calculation gives a pull out capacity of 550kN (considering self weight, friction resistance, and point resistance). Observing the contour of displacement for the soil, my suspicion is that the point resistance is not well generated on the tip of the pile.

I think I have built a right model since I wrote it based on the example from FLAC3D manual. I would like to hear any suggestion to improve / correct it.

Thank you.

<PRE>
new

gen zone radcyl p0 (0,0,0) p1 (8,0,0) p2 (0,0,-3) p3 (0,8,0) p4 (8,0,-3) p5(0,8,-3) &
p6 (8,8,0) p7 (8,8,-3) p8 (0.5,0,0) p9 (0,0.5,0) p10 (0.5,0,-3) &
p11(0,0.5,-3) size 3 3 6 8 rat 1 1 1 1.15
gen zone radcyl p0 (0,0,-3) p1 (8,0,-3) p2 (0,0,-9) p3 (0,8,-3) p4 (8,0,-9) p5(0,8,-9) &
p6 (8,8,-3) p7 (8,8,-9) p8 (0.5,0,-3) p9 (0,0.5,-3) p10 (0.5,0,-9) &
p11(0,0.5,-9) size 3 6 6 8 rat 1 1 1 1.15 fill
gen zone reflect dip 90 dd 270
group zone clay

interface 1 face range cyl end1 (0,0,0) end2 (0,0,-3.1) rad 0.51 &
cyl end1 (0,0,0) end2 (0,0,-3.1) rad 0.49 not
interface 2 face range cyl end1 (0,0,-2.9) end2 (0,0,-3.1)rad 0.51


gen zone cyl p0 (0,0,4.1) p1 (0.5,0,4.1) p2 (0,0,4) p3 (0,0.5,4.1) size 3 1 6
gen zone cyl p0 (0,0,4) p1 (0.5,0,4) p2 (0,0,1) p3 (0,0.5,4) size 3 3 6
gen zone reflect dip 90 dd 270 range z 1 4.1
group zone pile range z 1 4.1

save geo0

ini zpos add -4.0 range group pile

save geo1

model mech mohr range group clay
prop bulk 8.3e7 shear 3.8e7 coh 30000 fric 0 range group clay
model mech elas range group pile
prop bulk 8.3e7 shear 3.8e7 range group pile
interface 1 prop kn 1e8 ks 1e8 fric 20 coh 30000
interface 2 prop kn 1e8 ks 1e8 fric 20 coh 30000


ini dens 1230 range group clay
ini dens 1230 range group pile

fix z range z -9.1 -8.9
fix x range x -8.1 -7.9
fix x range x 7.9 8.1
fix y range y -0.1 0.1
fix y range y 7.9 8.1
set grav 0 0 -10
ini szz 0 grad 0 0 12300 range z -9 0
ini sxx 0. grad 0 0 5271.4 range z -9 0
ini syy 0. grad 0 0 5271.4 range z -9 0

solve rat 1e-6
save ini_cond

model mech elas range group pile
prop bulk 13.9e9 she 10.4e9 range group pile
ini dens 2500 range group pile

;find grid points at pile cap
def find_add
top_head = null
gp_pnt = gp_head
loop while gp_pnt # null
if gp_zpos(gp_pnt) > 0.05 then
mem_head = get_mem(2)
mem(mem_head) = top_head
mem(mem_head+1) = gp_pnt
top_head = mem_head
endif
gp_pnt = gp_next(gp_pnt)
endloop
end
@find_add

solve rat 1e-6
save pile_set

ini state 0
ini xdis 0 ydis 0 zdis 0

;monitor vertical loading (pull out) at pile cap
def pull_out
ad = top_head
zftot = 0.0
loop while ad # null
gp_pnt = mem(ad+1)
local zf = gp_zfunbal(gp_pnt)
zftot = zftot + zf
ad = mem(ad)
endloop
pull_out = zftot/0.3927
end
@pull_out

;apply velocity of 1e-8 using a ramp function to the pile top gridpoints
def ramp
nvel = 1e-7
while_stepping
if step < 30000
udapp = (float(step)/30000)*nvel
else
udapp = nvel
endif
ad = top_head
loop while ad # null
gp_pnt = mem(ad+1)
gp_zvel(gp_pnt) = udapp
ad = mem(ad)
endloop
end
@ramp

hist add gp zdisp 0,0,0
hist add fish @pull_out

set mech damp comb
step 300000
save pile1


plot create view capacity
plot add hist -2 vs 1

plot create view displacement
plot add contour displacement range group clay
plot add displacement colorbymag
</PRE>
[-] The following 2 users say Thank You to chan372 for this post:
  • tino27, drim_w3
#4
did you able to run ... i could not .. it shows unrecognized parameter at many locations... though i use cracked one but still it is running fine with all my files and tested with various other files too..

[/align]
#5
FLAC3D?
Just in case you may want to consider something more graphical and revolutionary...
Try midas GTS (en.midasuser.com)
#6
on the base of my experience with Flac3D software, I think you must modify the tip interface's coupled springs. In fact, making a good estimation of interface Kn and Ks are main problems in soil-structure interaction modelling. I suggest you to firstly refer to Dr.Coduto's book , "Foundation Desgin: Principles and Practices" - Chapter 14 - page 527 , for an analytical method in uplift capacity of piles. then estimate tip bearing capacity and then modify Flac3D's kn interface property from it. Side Bearing capacity must have good estimation. On the basis of "Geotechnical Engineering" book by V.N.S.Murthy, the side capacity in uplift is equal to somewhat about (2/3) side capacity in pressure. So because of a good design of pile in pressure for side bearing capacity , I think the file must show a fairly estimation of side capacity in uplift situation.
[-] The following 2 users say Thank You to viva_60_ae for this post:
  • Dell_Brett, chan372
#7
jacs127,
yes i can run the code very well.

viva_60_ae,
thank you very much for your reply. it helps a lot.
and i still have difficulties in estimating the value for kn & ks. could you give the "rule of thumb" in estimating these values? I've followed the formula
kn= max[ (K + 4/3 G)/delta(z)min], as given in the manual, and even try to set it up to 1000times higher, but I still can't get satisfactory results
#8
Hello,
I am a beginner of Flac3D user. Now I am facing a problem to create Flac 3D model. The problem is that I do not understand how to determine origin point and how to calculate normal vector and distance when using command 'plane' for interface. Can anyone help me please? Thank you so much .




Users browsing this thread: 1 Guest(s)