Compensation failures proceed

https://www.dacm-logiciels.fr/tracewin
User avatar
FranceDidier
Administrator
Administrator
Posts: 851
Joined: Wed 26 Aug 2020 14:40
Country:
France (fr)
France

Re: Compensation failures proceed

Post by FranceDidier »

Dear Bruce,

I fixed the problem concerning solenoid with 100% field error. Please, upgrade your code.

About the RF & phase, perhaps this post could help you: viewtopic.php?f=3&t=55

Regards,

Didier
JapanBruce Yee
Competent
Competent
Posts: 83
Joined: Tue 22 Dec 2020 03:47
Country:
Japan (jp)
Japan

Re: Compensation failures proceed

Post by JapanBruce Yee »

Dear Didier,

Thank you very much for the post, it helps a lot.
One comment, as you pointed out in the post, the synch phase is computed using the following equation.
Energy_gain.PNG
Energy_gain.PNG (115.61 KiB) Viewed 2508 times
,

Using Eq.( A.2), the "Abs Phase" can be calculated. The frequency that the code used in Eq. (A.2), it the one defined in the main window by hand or by importing the parameters of the beam distribution.

This last is important if you use frequency jumped in your linac.
For instance in my case, (
Abs. Phase.PNG
Abs. Phase.PNG (23.83 KiB) Viewed 2508 times
), the original model starts with 162 MHz and increased to 324 MHz.
I use a beam distribution for the region of 324 MHz, however, the beam distribution is saved with the 162 MHz.

Therefore, even I operated with a 324 MHz, the frequency used to compute the "Abs Phase" is 162 MHz.

To compute the synch phase using Eq. (A.5), you need phis (s) [Eq (A.2)].
But for Eq. (A.2), you required the beta(s'), but the beta(s') depends on the particle energy.
How does the program estimate beta(s')? because to compute the energy gain [ Eq.(A.1)], you need to know Eq.(A.2).

Best regards,

Bruce
User avatar
FranceDidier
Administrator
Administrator
Posts: 851
Joined: Wed 26 Aug 2020 14:40
Country:
France (fr)
France

Re: Compensation failures proceed

Post by FranceDidier »

Hi Bruce,

Honestly I'm a bit lost in your questions, are they about frequency jump or how the synchronous phase is calculated or a mix of both. To clarify before continuing:
- For information, the part of the documentation concerning the synchronous phase has been recently rewritten, especially because a new definition has been introduced.
- Most importantly, all code outputs (beam distribution, envelope, 'Abs phase' data...) are referenced to the bunch frequency and not to the RF frequency. (So why this choice, I don't know exactly, it's 20 years old..., I just remember that at that time we had to work on a multi-frequency machine and it seemed simpler that way)
- But be sure, of course in the calculation of the synchronous phase, we take into account the RF frequency and its jumps.

I would like to ask you, please, to maybe re-formulate what you are worried about or want to clarify in a new topic because here we are clearly moving away from the initial topic and I would really like the different topics in this forum to be useful to the community and therefore relatively clear and short.

Thank in adavance for that,

Regards,

Didier
JapanBruce Yee
Competent
Competent
Posts: 83
Joined: Tue 22 Dec 2020 03:47
Country:
Japan (jp)
Japan

Re: Compensation failures proceed

Post by JapanBruce Yee »

Dear Didier,

Thanks for the explanation. I will continue with the discussion on a new topic.

Best regards,

Bruce
JapanBruce Yee
Competent
Competent
Posts: 83
Joined: Tue 22 Dec 2020 03:47
Country:
Japan (jp)
Japan

Re: Compensation failures proceed

Post by JapanBruce Yee »

Dear Didier,

Re-taking again the transient studies.

I am trying to apply for compensation in a transient regime.

The next figure shows the model that I am using (that we discussed in the past).
Lattice.png
Lattice.png (37.33 KiB) Viewed 2358 times
There the second cavity has a faulty behavior, the details of the Cav file and the simulations time is shown next.
Simulations.png
Simulations.png (70.54 KiB) Viewed 2358 times
.

The cavity starts the failures 1ms from the start point of the simulations, I am using "Fast_Start" option.

At 2.5 ms the compensation is applied.


However, even from the beginning(when everything must be an ideal machine) The "beam energy" did not reach the "linac energy"
Energy.png
Energy.png (31.57 KiB) Viewed 2358 times
.

The energy gap is reduced( about 0.2%) when you omitted "Fast_Start" option. However, it is remained.
I noticed that the cavities that I used for the compensation (before the adjusted command). The phase and Eacc are modified
since the beginning of the simulations. You can see the phase and Eacc from a time of 1e-5 seconds.

Cavity_parameters.png
Cavity_parameters.png (61.87 KiB) Viewed 2358 times
Those values are changed during the matching; thus, I can not achieve the same performance as the linac option.

Is there are a way to make the cavity's adjustment after the failure occurred?
Or another way to indicate which element must be used for the compensation?

I include the file for more details.

Thanks for the help,


Bruce
Attachments
Transient_Compensation.zip
(6.14 KiB) Downloaded 123 times
User avatar
FranceDidier
Administrator
Administrator
Posts: 851
Joined: Wed 26 Aug 2020 14:40
Country:
France (fr)
France

Re: Compensation failures proceed

Post by FranceDidier »

Dear Bruce,

Attached my proposal.
Energy starting point was not good because you made partially intial adjusting (see *.cal results), So the best is to remove matching option and matching results, because I think you dont' need that.
energy.png
Regards,

Didier
didier_proposal.zip
(3.71 KiB) Downloaded 118 times
Attachments
energy.png
energy.png (16.84 KiB) Viewed 2350 times
JapanBruce Yee
Competent
Competent
Posts: 83
Joined: Tue 22 Dec 2020 03:47
Country:
Japan (jp)
Japan

Re: Compensation failures proceed

Post by JapanBruce Yee »

Dear Didier,

Thank you for your help, I came again with a request.

In the past, you suggested removing the matching options and matching diagnostic. This works fine in most cases; however, in another, I need to make a match to avoid emittance growth.

The matching options were removed because at the beginning of the run, the "adjust commands" (used to compensate for failures only) were applied. Consequently, this has a negative effect on the lattice.

Could we have a new class of "adjusting commands" for the transient failure that starts according to transient settings?
Transient_command.png
Transient_command.png (18.18 KiB) Viewed 2187 times
. Subsequently, I can use again the matching options. Or Could you suggest to me something else?

One more comment about transient studies, when the phase cavity is about -90 deg (For beam matching conditions, that -90 deg cavity is needed).
The energy and phase present an unusual behavior (the first one diverge drastically and the second does not have the beam data)
Screenshot (419).png
Screenshot (419).png (72.81 KiB) Viewed 2187 times
.

Thanks for your help and support.

Bruce
User avatar
FranceDidier
Administrator
Administrator
Posts: 851
Joined: Wed 26 Aug 2020 14:40
Country:
France (fr)
France

Re: Compensation failures proceed

Post by FranceDidier »

Dear Bruce,

I'm sorry to reply so late, but I'm travelling until the 18th of October, as soon as I get back I'll try to answer you.

Regards,

Didier
JapanBruce Yee
Competent
Competent
Posts: 83
Joined: Tue 22 Dec 2020 03:47
Country:
Japan (jp)
Japan

Re: Compensation failures proceed

Post by JapanBruce Yee »

Dear Didier,

Sure, thanks for the help.

Bruce
User avatar
FranceDidier
Administrator
Administrator
Posts: 851
Joined: Wed 26 Aug 2020 14:40
Country:
France (fr)
France

Re: Compensation failures proceed

Post by FranceDidier »

Dear Bruce,

Sorry for my late answer,
For your first request, I could by default cancel intial ADJUST process at beggening of the run, when transient simulation is selected. What do you thing about this solution
For you second remark about 90° cavity, honnestly, I'm not sure to anderstand the problem, could you send me a little example ?

Regards,

Didier
Post Reply