How to animate 3d plot_surface in matplotlib

I have created a 3D plot surface from a file and I'm trying to animate the plot. I have read the examples in the matplotlib webpage and other examples in SO, and notice that I need to create an update function to loop through the values in the file and then create a matplotlib.animation object but I don't understand how to do it.

I would appreciate very much if someone could explain me the syntax of the update function and how to use it in the matplotlib.animation object.

My data is a multidimensional array with 498 lines and for each line i have an array with 64x128 values. Data is organized in the following way:

Data is a time series from a force plate and each one of the 500 lines is a frame, which means that this trial lasts 10 seconds. For each frame I have another array with 64x128 values.

This is my code until now:

from mpl_toolkits.mplot3d import *
import matplotlib.pyplot as plt
import numpy as np
from random import random, seed
from matplotlib import cm
from mpl_toolkits.mplot3d import axes3d
import matplotlib.animation as animation

source_path = "c:\\Projecto\\"
destination_path = "c:\\Projecto\\EntirePlate\\"
#fid = np.loadtxt(source_path + "rolloff_xls.txt",dtype=str)

fid_MP = open(source_path + "101mp - Entire plate roll off.xls","Ur")
lines_MP = fid_MP.readlines()
fid_MP.close()

values_MP = []

for i in lines_MP:
      if i[0].isdigit():
          values_MP.append(i)

values = np.loadtxt(values_MP,dtype=float)

new_values_MP =[]

for i in range(0,(len(values_MP)/64)):
    for j in range(0,64):
        new_values_MP.append([[i],[j],values[j]])

new_values_MP = np.asarray(new_values_MP)

fig = plt.figure()
ax = fig.gca(projection='3d')               # to work in 3d
plt.hold(True)

x_surf = np.arange(0,128)                # generate a mesh
y_surf = np.arange(0,64)
x_surf, y_surf = np.meshgrid(x_surf, y_surf)
z_surf = []

for i in r