Skip to Content.
Sympa Menu

opal - Re: [Opal] benchmark with ASTRA

opal AT lists.psi.ch

Subject: The OPAL Discussion Forum

List archive

Re: [Opal] benchmark with ASTRA


Chronological Thread 
  • From: christof kraus <christof.j.kraus AT gmail.com>
  • To: "Power, John" <JP AT anl.gov>
  • Cc: opal <Opal AT lists.psi.ch>
  • Subject: Re: [Opal] benchmark with ASTRA
  • Date: Sun, 2 Oct 2016 10:23:11 +0200

Hi John,

what I actually meant is that we store the time when the centroid passes the cathode (t_{centroid}). With this information we can shift the phases of the cavities relative to t=0:
phi(t) = nu * (t - t_{centroid}) + phi_0.
From the perspective of the user the phases should behave just as if the centroid was emitted at t=0.

christof

On Sat, Oct 1, 2016 at 10:38 PM, Power, John <JP AT anl.gov> wrote:

Christof Kraus answered these two questions to me offline so I'll summarize them here since I think it was valuable enough so that other readers of this list will enjoy it.  Also, I think the first answer, [A1], could even be expanded and added to the OPAL-T documentation.

 

"[A1] We store the time difference and compute the phase shifts.  The location of the reference particle is at the centroid of the bunch. The time when it crosses the cathode depends on the length of the bunch. "

 

"[A2] with horizontal striation you mean this strange pattern in the initial distribution as plotted in the file opal_t_pz.pdf? This I have fixed in Opal3D (see file opal3D_t_pz.pdf), but I haven't fixed it in the current version of Opal. I'll do that as soon as possible."

 

As a follow-up to [A1], when you say "store the time difference and compute the phase shifts" could explain more precisely what you mean?   Is it the time difference between the ith and ith+1 'time step' Or is it the difference between the ith particle with the reference? Similarily, the phase shift from what to what? 

 

thanks,

John

 


From: opal-request AT lists.psi.ch [opal-request AT lists.psi.ch] on behalf of Adelmann Andreas (PSI) [andreas.adelmann AT psi.ch]
Sent: Thursday, September 29, 2016 8:48 AM
To: opal
Subject: [Opal] Fwd: benchmark with ASTRA



From: "Power, John" <JP AT anl.gov>
Subject: benchmark with ASTRA
Date: 29 Sep 2016 15:45:29 CEST
To: "Adelmann Andreas (PSI) [andreas.adelmann AT psi.ch]" <andreas.adelmann AT psi.ch>
Cc: Nicole Neveu <nneveu AT hawk.iit.edu>

Hi Andreas, 
We are trying to benchmark OPAL to ASTRA by exporting the initial distribution that is emitted by OPAL into ASTRA and two questions came up.  ----John

 

[Q1]  I see the opal file defines the zero as the head of the bunch (I assume the head and not the tail). Do you know how the code then defines its phases internally, does it place a reference particle at the barycenter of the bunch or is the first particle taken as reference. In Astra I usually define the reference as the bunch center (I have to physically put the reference there) and therefore all the phase are reference are with respect to the bunch center. IMPACT-T does the same thing (1st particle is t=0 and when we compare the phase we have to properly tweak the phase in IMPACT taking the total emission time. 
 
[Q2]  As a side not I also notice the t,pz phase space is a bit awkward, it has some strange horizontal striation -- this is not a big deal for what we are trying to do for now and might just be coming from the way the random generation is done (maybe it is not fully 2D MC generator).





Archive powered by MHonArc 2.6.19.

Top of Page