Place sq1.pro

The following describes an obsolete procedure or component.  See auto setup for a more recent alternative.

The script place_sq1.pro is intended to help with 'row selection', the procedure whereby a single row in each column is identified as being centrally positioned (relative to other rows) on the SQ2 V-phi curve, and is thenceforth used in the sq1servo step of auto-tuning to choose the SQ2 lock-point feedback.

Since the SQ2 feedback on each column MCE (in normal configuration) is locked at a SQ2 feedback signal for each column, you have to pick it carefully so that all 33 squids in that row fall on the nice, shallow, constant slope of the sq2 V-phi curve. This is done by selecting a row whose squid1 position on the squid2 curve is central relative to the other rows. Then you servo it (in sq1servo) to chose the sq2fb lock point.

There are two steps to the process:

1. Acquire a set of sq2servo ramps for every row in each column
2. Plot the ramps using place_sq1 and choose a central row

Prior to this, the normal sq2servo plots should be inspected to ensure the error signal is low (<1000); otherwise the servo gains should be adjusted.

Acquiring the sq2servo ramp set

We run auto-tuning in a special mode where it loops over all rows. The files will be in the usual auto-tuning output sort-of-a place, namely /data/cryo/current_data/<ctime>/.

Tune once with the usual experiment.cfg. Then set the auto-tuning to do the SQ1 servo on all rows, and tune again:

$ cd $MAS_DATA
$ mas_param set sq1_servo_all_rows 1
$ auto_setup_squids_sh

Suppose this most recent tuning has gone into ctime directory 1222233334. There will be a tonne of data files in this tuning folder, with names like:

1222233350_RC1_row0_sq1servo
1222233350_RC1_row1_sq1servo
1222233350_RC1_row2_sq1servo
...

Plotting and positioning the data with place_sq1

The place_sq1 script requires the filename of an sq2servo data file, and a structure containing basic results of the sq1servo analysis. You can plot to the screen, or to postscript by passing an output filename. You can plot a single column, or analyze all.

Go to the $MAS_IDL/engineering directory, open IDL. Then load the sq1 data run the place_sq1 script:

IDL> .r place_sq1
% Compiled module: VPHI_LIMITS.
% Compiled module: LOAD_SQ1SERVO_SET.
% Compiled module: PLACE_SQ1.
IDL> tuning_folder = '/data/cryo/current_data/1222233334/'
IDL> sq1_data = load_sq1servo_set(tuning_folder+'*RC1')
IDL> sq2_file = tuning_folder+'*_RC1_sq2servo'
IDL> place_sq1,sq2_file,sq1_data,0

Here, you can just drop these lines into idl:

sq1_data = load_sq1servo_set(tuning_folder+'*RC1')
sq2_file = tuning_folder+'*_RC1_sq2servo'
place_sq1,sq2_file,sq1_data,0

The 0, there, is the column you want to analyze. That form is good for interactive operation. To plot all columns to postscript, do, e.g.

IDL> place_sq1,sq2_file,sq1_data,0,plot_file=tuning_folder+'sq1_RC1.ps',/all_cols