One issue with PHASE_REFERENCE of the RFCA element

[duanz]

Hi Michael,

I was trying to modify the RFCA elements in one .ele script and dump the modified lattice with the save_lattice command. However, in the new lattice file generated, the PHASE_REFERENCEs of RFCA elements are some very large integers like 9223372036854775807. When I use this lattice for later tracking, I got an error "error: please use a small integer for the phase_reference number".

I then tried to set PHASE_REFERENCE to 0 (the default value) in the .ele script, but found this PHASE_REFERENCE = 0 didn't go into the parameter file, and thus didn't override its value of a very large integer in the lattice file generated with save_lattice. If I set PHASE_REFERENCE to some positive value, I could see its value appear in the parameter file, and also in the lattice file generated with save_lattice. But this might not be the preferred setting, since I don't assume the RF phase would be linked with some other time-dependent elements.

Thanks and best regards,

Zhe

[michael_borland]

Zhe,

I'm very surprised that the saved lattice would have those strange values for the PHASE_REFERENCE parameter. Can you send me an example that illustrates this?

Thanks--Michael

[duanz]

Hi Michael,

I use PAR ring as an illustration, see attached files. I used both a MAC and a linux server, the results are the same.

Best,

Zhe

[michael_borland]

Zhe,

The bug seems to be in save_lattice when output_seq=1. If you can set this to 0 (in most cases it is ok), that will solve it for now.

--Michael

[michael_borland]

Zhe,

I've fixed this bug. The fix will appear in the next release.

Thanks again.

--Michael

[duanz]

Hi Michael,

Thanks! This did solve the strange large value of PHASE_REFERENCE when dumping the lattice.

Best,

Zhe