Study of TMCI

Moderators: cyao, michael_borland

arahim
Posts: 26
Joined: 08 Jun 2012, 05:28

Re: Study of TMCI

Post by arahim » 26 Jun 2019, 07:39

I am trying to make an sdds file for impedance (freq vs real part and imaginary part) using plaindata2sdds. In sdds file many of the rows were missing. I copied the data from data file and pasted in sdds file but during simulation I got error message "frequency values not equally spaced for real data(zlongit)". Please help me to understand this. Warning and sdds file are attached for the reference.
Attachments
HBSRS_booster_imp.sdds
(12.25 KiB) Downloaded 266 times
Doc2.docx
(265.2 KiB) Downloaded 269 times

Akhyani
Posts: 16
Joined: 26 Jun 2018, 08:09

Re: Study of TMCI

Post by Akhyani » 07 Aug 2019, 05:37

Dear Abdurrahim,

Your sdds file is a bit weird for me. The number which will be written in the first line of your txt file (before using the plaindata2sdds command) is the number of lines in your impedance txt file. This number must be equal to 2^n+1. If you fix this glitch, I think the elegant will run completely.

--Mina

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

Re: Study of TMCI

Post by michael_borland » 07 Aug 2019, 08:12

Something is definitely amiss with the file. When I run

Code: Select all

sddscheck HBSRS_booster_imp.sdds
it returns "corrupted".

The most obvious problem, as Mina points out, is the row count. I changed this from 257 to 400, which made the file valid at least. However, after plotting the impedance, I'd have to say something still looks off.
impedance.png
Perhaps you can post the original file (prior to running plaindata2sdds) along with the commandline used for conversion.

--Michael

arahim
Posts: 26
Joined: 08 Jun 2012, 05:28

Re: Study of TMCI

Post by arahim » 30 Sep 2019, 08:24

I am trying to find the threshold of microwave instability using the aps upgraded lattice as an example. I am using zlongit for the input of broadband impedance. I have considered the Z/n=0.180 Ohm and freq 50GHz, corresponding RA=6.6e4. But during tracking, particles are getting lost after only few 100 turns, even current is too small (0.01 mA). Please help me to find the reason of beam loss..Thanks in advance.
Attachments
aps_modified.lte
(1.05 KiB) Downloaded 274 times
run2.ele
(308 Bytes) Downloaded 237 times
bunch10kP.out
(508.76 KiB) Downloaded 254 times

lindberg
Posts: 4
Joined: 07 Nov 2018, 14:47

Re: Study of TMCI

Post by lindberg » 02 Oct 2019, 09:56

I checked things out, and the reason for the particle loss has nothing to do with the impedance; eliminating the ZLONGIT element leads to the same loss. It seems to me that the RFCA element is not set up properly for a single RF cavity. For V=4.1 MV I find that if you change the present settings of
freq=351919480.995373,volt=4.1e6,phase=144.7614434245066
to the new settings of
freq=351936980.203300,volt=4.1e6,phase=146.4219618701754
the beam is properly stored. Note that its really the frequency that needs to be fixed.

arahim
Posts: 26
Joined: 08 Jun 2012, 05:28

Re: Study of TMCI

Post by arahim » 03 Oct 2019, 04:34

Thanks a lot..It worked.
To simulate the microwave instability, I have used the elegant functions ILMATRIX, SREFFETCS, RFCA and ZLONGIT in the above files but the elegant tracking result gives an underestimated value (~0.1mA) of microwave threshold. I have doubt whether I have chosen the correct functions or have wrongly configured the ZLONGIT to simulate the MWI. Please help me to correct above elegant files (previous communication).

lindberg
Posts: 4
Joined: 07 Nov 2018, 14:47

Re: Study of TMCI

Post by lindberg » 04 Oct 2019, 15:24

I'm not sure why you think this is "underestimated." Your impedance is rather strong and the microwave instability threshold should be pretty low. If you reduce Ra to, say 20 MOhm then the threshold should increase to ~0.8 mA. One thing to note is that with only 10,000 particles the microwave instability threshold current will be less sharp, meaning that the simulation will show a small degree of energy spread growth below the "true" MWI threshold.

arahim
Posts: 26
Joined: 08 Jun 2012, 05:28

Re: Study of TMCI

Post by arahim » 05 Oct 2019, 04:52

Thanks a lot for your quick help.
As per your suggestion, I will increase the number of particles in the simulation. I have a doubt about Ra. For a given value of Z_longitudinal/n in a storage ring, what value of Ra should be fixed for the input of zlongit ?

I have calculated the Ra to be given as an input for zlongit is as follows (considering APS-upgraded as an example)
f_revolution=271.84 kHz and Z_longitudinal/n=0.18 Ohm.
The freq. considered for microwave simulation is up to 50GHz.
n for 50 GHz=1.8393e+005, now Ra=2*Rs=2*(Z_longitudinal/n)*n=6.6+04 Ohm.
Therefore, I have defined zz: zlongit,broad_band=1,freq=50e9,Ra=6.6e4,Q=1,bin_size=1e-12,n_bins=0
With above input MWI threshold is 0.1mA, however in literature APS MWI threshold is ~0.8mA (w/o HHC). Then , how should I define the Ra and freq…?

arahim
Posts: 26
Joined: 08 Jun 2012, 05:28

Re: Study of TMCI

Post by arahim » 08 Jun 2020, 04:55

Dear Dr. Borland,
I am tracking the beam (single bunch) through the Ring1 lattice without any impedance but the horizontal emittance, energy spread and bunch length is keep on increasing with the number of turns, while vertical emittance is nearly constant. I have configured the RF with the parameters of RF out file. I am not able to pin point, where the mistake is. Please help me to figure out the error. I have attached the files for the reference.
Attachments
Booster_single_bunch.out
(1016.73 KiB) Downloaded 206 times
run2.ele
(312 Bytes) Downloaded 219 times
Ring1.lte
(759 Bytes) Downloaded 235 times

arahim
Posts: 26
Joined: 08 Jun 2012, 05:28

Re: Study of TMCI

Post by arahim » 24 Jun 2020, 07:43

I could locate the error...The lattice is having non zero values of alphax,alphay and etapx. I was not assigning these parameters any values in my input files and hence by default, code is reading them as zero and hence emittance blowing..
Arahim

Post Reply