Beam extraction line: quadrupole offset

[Bruce Yee]

Dear Didier,

Thanks for your help and support.

I am currently designing a beam extraction line consisting of a pulse bending followed by a doublet quadrupole.
As shown in the next figure:

Picture1.png (73.39 KiB) Viewed 119 times

The quadrupoles belong to the main line; therefore, the extracted beam will present a relative offset as a result of the bend of the pulse magnet (which is simulated using a dipole, the bend is on the horizontal plane).

Picture2.png (36.3 KiB) Viewed 119 times

When I simulated that in TraceWin, I noticed that the beam centroid and the quadrupoles' centroid are the same.

Picture3.png (55.56 KiB) Viewed 119 times

Thus, the offset kick is not simulated.

I tried to simulate the offset using "shift dx dy", where "dx" is the relative offset produced by the pulse magnet.
For this case, as the horizontal offset is 32 mm (computed analytically and confirmed by Synoptic plot in TraceWin), I shift the first quadrupole to -32 mm

Picture4.png (45.28 KiB) Viewed 119 times

However, when I plotted and compared the Synoptic plot in TraceWin with and without a shift. The beam centroid
is smaller at the entrance of the quadrupole for the case with shift, which is strange for me since the shift is only applied at the quadrupole not the previous drift space.

Picture5.png (670.94 KiB) Viewed 119 times

I was expecting to have the same centroid and only be different from the quadrupole to the end.
Do I simulate the quadrupole offset correct?

I have attached the TraceWin files in case you would like to cross-check.

Best regards,

Bruce

[Didier]

Dear Bruce,

I suggest you to replace EDGE+BEND+EDGE by "MULTIPOLE 1 500 100 0.11 100"

Regards,

Didier

[Bruce Yee]

Dear Didier,

Thank you very much for your advice.

Sorry for taking advantage of your kindness, but may I ask something else?

After the quadrupoles, I have a septum magnet to extract the beam (and I am planning to simulate the full transport line from the septum)
As shown in the figure.

Picture6.png (149.79 KiB) Viewed 106 times

In this case, I want that the septum magnet (simulated as dipole in this case) to have the same centroid as the beam.
When I added that element in the simulation the envelope shows the beam offset (as I expected)

Picture7.png (57.88 KiB) Viewed 106 times

But the synoptic plot shows that the centroids are align

Picture8.png (75.57 KiB) Viewed 106 times

I was thinking to use the command "shift_beam" to remove the offset at the entrance of the septum or use 'shift" to add the offset to the elements .

Do you have some advice?

Best regards,

Bruce

[Bruce Yee]

Dear Didier,

I attached the files.

Best regards,

Bruce

[Didier]

Dear Bruce,

Here it's going to be more complicated, you'll have to use a field map for your septum and make a reference change using the command "SUPERPOSE_MAP_OUT" & "SUPERPOSE_MAP", I suggest you look at the example "Field_map_dipole".
So use "MULTIPOLE" for your pulse magnet, then your 2 quads and then insert something similar to the example I suggested.

Regards,

Didier

[Didier]

Dear Bruce,

I forgot a simpler solution before your septum: use the command CHFRAME and then you can use whatever you want behind it.

Regards,

Didier

[Bruce Yee]

Dear Didier,

Thank you very much for your advice.

CHFRAM makes the work.

Picture9.png (56.88 KiB) Viewed 81 times

Best regards,

Bruce