Optimizing the quadrupole strengths

[fabioluneli]

Thank you Michael for the explanation! It helped me understand some procedures and looks like this is correct. If possible i would like to share with you the paper which i am trying to repeat the results. Here is the DOI: https://doi.org/10.1038/s41598-021-93276-8.

Analyzing methods section and looking at supplementary information, i noticed that ELEGANT were used there to find the optimal values of the positions of quadrupoles and the focusing strength needed to focus the beam on a depth 20 cm inside a drift of size 30 cm. This procedure was achieved using optimization and this part about minimizing beta function that you mentioned.

In order to check this results, i repeated the simulations for the symmetric beam in order to find this focal point in 20cm inside this last drift of 30cm. To do this, i created the fodo.ele file (attached) together with lattice.lte file (attached), executed elegant fodo.ele and then using the command:

sddsplot -col=s,S[xy] fodo.sig -graph=line,vary -legend -unsuppress=y \
-col=s,Profile fodo.mag -overlay=xmode=norm,yfact=0.04

...so i got the following figure named graphics.png (attached). Looking at ther i can see that the focal point is happening on a depth of 4 cm inside the last drift and not on 20 cm as i expected. If it's not too much to ask, i would like to know if this steps seem corrected for this goal?

Thank you again for your time and disposition!

[michael_borland]

I don't see anything wrong in your simulation. Looking quickly through the paper, it wasn't clear to me what the distance was from the center of the last quadrupole to the desired focal point, but it seems to be s7=38.2cm.

It's probably best to contact the authors of the paper. They state that they are willing to share their data.

--Michael

[fabioluneli]

Hi Michael, thanks for the quick response!

I'm trying to contact the authors but without success so far.

About the s7 mentioned: For me looks like the s7 is the distance between the center of the last quadrupole and the center of the phantom. This way, to find the size of the drift spaces i had to subtract half of the sizes of the quadrupoles and for the last drift subtract half of the phantom either. Also i had to find the focusing strength, because as you can see he supllies the gradient of the quadrupoles.

One question about this part on the previous post: "using the optimizer, minimize betax and betay at the end of the beamline". I apply the changes that you suggested, and i would like, if possible, to confirm that the &optimization_term and &twiss_output parts on transport.ele (attached) seems correct.

[michael_borland]

The &twiss_output command isn't right.

Code: Select all

&twiss_output
	filename="%s.twi",
      	matched= 0,
	beta_x = 4e-3,		 !4mm rms size
	beta_y = 4e-3,		 !4mm rms size
        output_at_each_step = 1,
	statistics = 1
&end

beta_x and beta_y are not the beam sizes. You need to compute the initial beta functions from betax=sigmax/sigmax' and betay=sigmay/sigmay' assuming an upright beam ellipse.

Your &bunched_beam_moments command has the rms beam sizes. They are consistent with betax=betay=1.25m

--Michael