Transverse impedance element

Moderators: cyao, michael_borland

Post Reply
li.chao
Posts: 45
Joined: 18 Aug 2021, 08:59

Transverse impedance element

Post by li.chao » 18 Aug 2021, 18:36

Hi,

I am trying to use the ztransverse element to simulate the interaction between beam and transverse impedance.
would you mind explain me more about the data printed out with command
"wakes="%s.wake" in the %s.wake file?

Is the data Wx in file corresponds to wake potential obtained by Ifft(Zx(\omega) * \lambda(\omega) )?
And what is the <I_x> and <I_y> corresponding to ?

How the setting n_bins and max_n_bins are used in the code in simulation process?

your Chao

michael_borland
Posts: 1927
Joined: 19 May 2008, 09:33
Location: Argonne National Laboratory
Contact:

Re: Transverse impedance element

Post by michael_borland » 20 Aug 2021, 16:11

Chao,

The Wx and Wy columns in the wake file are not the wake potential (IFFT of the impedance) but the wake field (IFFT of impedance convolved with the time-dependent moment of the current).

The Ix and Iy columns are the time-dependent transverse current-weighted moments of the bunch. More specifically, Ix is a histogram of Qi/dt*xi for all the particles i, using either the user-specified number of bins (N_BINS) or the number of bins implied by the number of points in the impedance file. dt is the bin spacing, which is derived from the impedance file or else the user-supplied BIN_SIZE parameter.

--Michael

li.chao
Posts: 45
Joined: 18 Aug 2021, 08:59

Re: Transverse impedance element

Post by li.chao » 22 Aug 2021, 17:55

Dear Michael

Thanks for your kindly reply. It is very clear.

However, I am doing a test with elegant to get the wake field with a resistive wall impedance.
The lattice is a very simple which is drift + Trans_imp + drift.

If you do not mind, would you please take a look at the script in attached files.
In the Sbatch.sh file you can see what did I do.
It is found that the "wakes="%s.wtdrw" command does not give me the results expected.
Did I do something wrong in with my elegant files.


Great thanks.

yours Chao

---------------------------------------------------------------------------------------
#!/bin/bash
N_MP=(1000 10000 100000 1000000)
N_MP=(1000)

ele_file=USSRTracking.ele

Nscan=${#N_MP[@]};

for ((i=0;i<$Nscan;i++))
do
ncharge=1
nMP=${N_MP[$i]}
charge=`rpnl $ncharge 1.e-10 \*`

echo $i $nMP
elegant $ele_file -macro=root=${nMP}_MP,qin=${charge},Nparticle=${nMP}

done


sddsplot -layout=1,2 -leg=filename -graph=l,v -scale=0,100e-12,0,0 \
-col=Deltat,'(Wy)' *.wtdrw -mode=y=normalize -end \
-col=Deltat,'(Iy)' *.wtdrw -mode=y=normalize -end #\
#-devices=png -output=power16_wydrw_0.09ps.png

sddsplot -layout=1,2 -leg=filename -graph=l,v -scale=0,100e-12,0,0 \
-col=Deltat,'(Wx)' *.wtdrw -mode=y=normalize -end \
-col=Deltat,'(Ix)' *.wtdrw -mode=y=normalize -end #\
#-devices=png -output=power16_wxdrw_0.09ps.png
Attachments
USSRTracking.ele
(659 Bytes) Downloaded 175 times
RWIMP_AL_Power16_DCModifty_1.sdds
(2.75 MiB) Downloaded 218 times
LatticeUSSRILMatrix.lte
(681 Bytes) Downloaded 179 times

michael_borland
Posts: 1927
Joined: 19 May 2008, 09:33
Location: Argonne National Laboratory
Contact:

Re: Transverse impedance element

Post by michael_borland » 23 Aug 2021, 01:29

Chao,

I'm not sure what you are expecting to see, but I wouldn't expect to see much because you are computing a transverse wake but your beam is centered on x=y=0. The only wake you'll see will be due to random offsets in <x> and <y> in the binning of the particle distribution.

Also, the frequency resolution of your impedance is very high, so you'll end up with a very long time span in the wake data, which may make the plots confusing.

Try adding some centroid offsets to you beam definition, e.g.,

Code: Select all

&bunched_beam
 ...
 centroid[0] = 1e-3
 centroid[2] = 1e-3
&end
When plotting, limit the time interval shown, e.g.,

Code: Select all

sddsplot -col=Deltat,Wx test1.wtdrw -filter=col,Deltat,0,1e-9
--Michael

li.chao
Posts: 45
Joined: 18 Aug 2021, 08:59

Re: Transverse impedance element

Post by li.chao » 23 Aug 2021, 05:59

Dear Michael

It is very clear, I got that.

Thanks very much.

yours sincerely Chao

li.chao
Posts: 45
Joined: 18 Aug 2021, 08:59

Re: Transverse impedance element

Post by li.chao » 24 Aug 2021, 07:50

Dear Michael,

Sorry to bother you again. I met another problem to print out the transverse wakefield.

As shown in the *.lte file, I am trying to print out the longitudinal and transverse wake field in files "wakes="%s.wzrw" and "wakes="%s.wtdrw".

Whereas I found only the longitudinal wake field data "wakes="%s.wzrw" can be printed out. There is not data file "wakes="%s.wtdrw" given out by the elegant.

Did I make mistakes in the transverse impedance print out settings?

your sincerely Chao
Attachments
test.tar
(2.15 MiB) Downloaded 155 times

li.chao
Posts: 45
Joined: 18 Aug 2021, 08:59

Re: Transverse impedance element

Post by li.chao » 30 Aug 2021, 11:43

michael_borland wrote:
23 Aug 2021, 01:29
Chao,

I'm not sure what you are expecting to see, but I wouldn't expect to see much because you are computing a transverse wake but your beam is centered on x=y=0. The only wake you'll see will be due to random offsets in <x> and <y> in the binning of the particle distribution.

Also, the frequency resolution of your impedance is very high, so you'll end up with a very long time span in the wake data, which may make the plots confusing.

Try adding some centroid offsets to you beam definition, e.g.,

Code: Select all

&bunched_beam
 ...
 centroid[0] = 1e-3
 centroid[2] = 1e-3
&end

When plotting, limit the time interval shown, e.g.,

Code: Select all

sddsplot -col=Deltat,Wx test1.wtdrw -filter=col,Deltat,0,1e-9
--Michael
Michael, by the way, how can I translate the wake potential data to the transverse momentum kick (for example in rad unit). Assume the beam is with <y> central offset?

michael_borland
Posts: 1927
Joined: 19 May 2008, 09:33
Location: Argonne National Laboratory
Contact:

Re: Transverse impedance element

Post by michael_borland » 08 Sep 2021, 14:06

li.chao wrote:
30 Aug 2021, 11:43
Michael, by the way, how can I translate the wake potential data to the transverse momentum kick (for example in rad unit). Assume the beam is with <y> central offset?
If you have a wakefield from the ZTRANSVERSE element with the drive exponent = 1, then you just divide the value in the Wx or Wy column by the beam energy in eV to get the kick (change in slope).

--Michael

Post Reply