Why is my plot not working within my for loop?

I am trying to plot multiple helices. With the code I am using (below) the helix is plotting how I want it to, however, only one helix is not plotting. Does anyone know why this may be the case?

clc;
clear variables;
close all;

n = input('number of revolutions '); 
r = input('radius of stent ');
a = input('angle with respect to the upwards horizontal in degrees '); 
w = input('number of wires ');
R = input('radius of wire ');

h=(2*n*pi*r)/tand(90-a); %height of stent determined by other inputs
e = (2*pi)/(w/2); %theta (spacing between where each wire starts)
if mod(w,2) ==0 %number of wires is even
else
  fprintf('ERROR:number of wires must be even'); %stops code if wire number is odd
  return
end
if mod(n,1) ==0 %number of revolutions is whole number
else
  fprintf('ERROR:number of revolutions must be whole number'); 
  return
end

for i=1:w/2
  for t = (i-1)*e:2*pi/w:(n*2*pi)+e*(i-1); %t value for CCW helices 
    if mod(t,4*pi/w)==0 
      x{i} = (r-R)*sin(t); 
      y{i} = (r-R)*cos(t);
      z{i} = (h/(n*2*pi))*t(1);
      plot3(x{i},y{i},z{i},'.','MarkerSize',25,'MarkerFaceColor','red','MarkerEdgeColor','red')
      hold on
    else
      x{i} = (r+R)*sin(t); 
      y{i} = (r+R)*cos(t);
      z{i} = (h/(n*2*pi))*t(11);
      plot3(x{i},y{i},z{i},'.','MarkerSize',25,'MarkerFaceColor','black','MarkerEdgeColor','black')
      hold on 
    end
    hold on
  end
  hold on
end

[Adam Danz]

Could you provide a set of input values so your code can be run?

Off the bat I have a suggestion. You are calling 'hold on' too many times. Prior to your loop you can create the figure and axis and hold it once. This isn't the source of the problem you described, though.

figure; 
axh = axis(); 
hold(axh, 'on')

then you can get rid of all the other 'hold on' lines.

[Julia Morandi]

Thank you for your initial help. A set of input values could be 2,20,34,12,1 (in the order it asks you for).

[dpb]

A guess would be

   if mod(t,4*pi/w)==0

is never actually satisfied for most cases owing to FP rounding...

[Julia Morandi]

Do you have any suggestions on how to fix this?

[Adam Danz]

Using those inputs, your code breaks in the 'else' section of your conditional at line

z{i} = (h/(n*2*pi))*t(11);

I'm guessing that t(11) is a mistake.

Answer 1

[Adam Danz]

Change

z{i} = (h/(n*2*pi))*t(11);

to

z{i} = (h/(n*2*pi))*t(1);  %or whatever the index should be

and it works with the inputs you specified in the comment section above.

For inputs [2,20,34,12,1]

[Julia Morandi]

Yes it works for one single wire, however it should be plotting multiple wires.

[Adam Danz]

It is plotting multiple data sets (wires). It's just plotting directly over top of previous data with an occasional unique data point appearing.

I determined that in debug mode by changing the colors after the first complete iteration of the i-loop.

Another way to test that is to create a new figure at every i-loop iteration.

for i=1:w/2
    figure
    axis 
    hold on
  for t = (i-1)*e:2*pi/w:(n*2*pi)+e*(i-1); %t value for CCW helices 
    if mod(t,4*pi/w)==0 
      x{i} = (r-R)*sin(t); 
      y{i} = (r-R)*cos(t);
      z{i} = (h/(n*2*pi))*t(1);
      plot3(x{i},y{i},z{i},'.','MarkerSize',25,'MarkerFaceColor','red','MarkerEdgeColor','red')
    else
      x{i} = (r+R)*sin(t); 
      y{i} = (r+R)*cos(t);
      z{i} = (h/(n*2*pi))*t(1);
      plot3(x{i},y{i},z{i},'.','MarkerSize',25,'MarkerFaceColor','black','MarkerEdgeColor','black') 
    end
  end
end

Here are all 6 'wires'; they appear 2D because the axes aren't rotated.

[Julia Morandi]

But when the i value changes, the t values change, which means the x and y values would change, so how are they overlapping if the points should be different?

[Adam Danz]

But they mostly do not change (some do) so something's wrong with you code.

To start, you are currently only storing x,y,z values over the i-loop so you're overwriting a lot of the data making it impossible to look at all of the x,y,z values at once. I changed the code so you store ALL of the data for both i and t loops. If you go this route, you should allocate the x y z data prior to the loops.

figure
axis
hold on
for i=1:w/2
    tAll = (i-1)*e:2*pi/w:(n*2*pi)+e*(i-1); 
  for t = 1:length(tAll); %t value for CCW helices 
    if mod(tAll(t),4*pi/w)==0 
      x(i,t) = (r-R)*sin(tAll(t)); 
      y(i,t) = (r-R)*cos(tAll(t));
      z(i,t) = (h/(n*2*pi))*tAll(t);
      plot3(x(i,t),y(i,t),z(i,t),'.','MarkerSize',25,'MarkerFaceColor','red','MarkerEdgeColor','red')
    else
      x(i,t) = (r+R)*sin(tAll(t)); 
      y(i,t) = (r+R)*cos(tAll(t));
      z(i,t) = (h/(n*2*pi))*tAll(t);
      plot3(x(i,t),y(i,t),z(i,t),'.','MarkerSize',25,'MarkerFaceColor','black','MarkerEdgeColor','black') 
    end
  end
end

Now we can look at ALL of the x,y,z values and you can see that there are lots of duplicates. Here I show just the x values. The rows are iterations across the i-loop and the columns are across the t-loop. There are tiny differences between these numbers but they are close enough to be plotted on top of each other. It looks like there are about 6 near-repetitions for each (x,y,z) value and there also happens to be 6 'wires'.

If the values are not what you expect, you'll need to go through your code and figure out what went wrong.

Answer 2

[Julia Morandi]

Thank you for your help!

[Adam Danz]

No problem, come back if you get stuck.