From: michael.clarke-gayther@stfc.ac.uk [mailto:michael.clarke-gayther@stfc.ac.uk] 
Sent: 03 May 2012 16:46
To: m.aslaninejad@imperial.ac.uk
Cc: s.jolly@ucl.ac.uk
Subject: FETS MEBT Optics

Morteza,

With reference to the discussion at the FETS meeting yesterday, regarding the input reference plane for the original scheme A, please find a copy of the CERN Linac 4 IPHI RFQ output distribution attached. This was the distribution used by Frank Gerigk in his original Tracewin version. My ‘simulation’ of this distribution in GPT was a close match using the GPT functions, with the ‘tails of the original version truncated. As you pointed out, the transverse phase space distributions are more or less ‘symmetrical’, so I agree that this is most likely to be the distribution at the RFQ vane end plane, and not as I had previously assumed, at the external face of the end flange. My mistake! Scott informs me that for our RFQ, the distance from the vane end to the external face of the end flange is 21 mm. So to correct for this error one needs to add this drift length of 21 mm, at the beginning of the GPT input distribution – effectively shifting all the downstream components with respect to the start datum. Of course, this will disturb the input match, but hopefully, only the first and second quadrupoles will need to be adjusted to correct for the mismatch. 

An alternative would be to leave all dimensions intact and ‘find’ an extra 21mm. I have done this, by making the gate valve and intensity monitor design at the beginning of the MEBT line more compact. Please find a new drawing showing these modifications attached. The reference plane is now in the correct position, and the critical dimensions are unchanged.

Another very important point that I would like to re-state is that the MEBT ‘chopping plane’ was changed over a year ago from horizontal to vertical. This was ‘officially’ introduced at a meeting at Imperial in March last year (Ajit, Ciprian, etc.). Since then, I have mentioned this a number of times at FETS meetings. Please double check with Simon that this is the case. If it is not, the end cell(s) of the RFQ must be modified to make it so.

My take on Simons talk yesterday was that it might be wise to investigate the effect of increasing the MEBT X an Y plane input emittances from 0.25 to 0.35 mm.mr  to take into account misalignment effects. 


Regards


Mike