Amplitude and Phase Repeatability

Software Requirements

Measurement

Measurement proceedure is here.

Additional parameters needed so analysis software can automatically window the data:
  1. Tilt table angle, and
  2. Bit/Field indentifying when WCA power is off.
Time Estimation:

From e-mail: Morgan McLeod Wed 21 Jul 2010

As the test procedure is currently written and understood, the BARE MINIMUM time for one measurement is: + 10 minutes data collection at start + 0.5 minute per tilt operation X 10 repetitions + 2 minutes after each tilt X 10 + 10 minutes data collection prior to standby band selection + 0.5 minute per standby band selection X 10 repetitions + 2 minutes after each repetition X 10 = 70 minutes.

Time to analyze and compute the relevant metrics from the measurement: + 2 minutes once automated (presently much longer.) = 72 minutes

This is multiplied by 3 LO frequencies = 216 minutes

To this is added the time to lock the LO and signal source above and then below the LO; + 2 minutes X 6 lock operations = 228 minutes

This is multiplied by 3 gross tilt angles. + 30 minutes settling after gross tilt (may not be enough) = 774 minutes

This is multiplied by 4 IF channels for most bands, and 2 for band 9: 4-band configuration (X 14) = 10836 minutes (180.6 hours, 18.1 ten-hour days) 6-band configuration (X 22) = 17028 minutes (283.8 hours, 28.4 ten-hour days)

For each band there is a fixed cost of setting up and waiting for the cartridges and signal source to stabilize. + 2 hours 4-band configuration: 8 hours 6-band configuration: 12 hours. This cost will be mitigated on future receivers by performing the test immediately after amplitude and phase stability when the source is already mounted and ready.

The foregoing assumes very high efficiency in use of time and resources and implies full automation. The present situation is that only data collection is automated and staff is very busy so there can be expected at least 30% inefficiency of of measurement time and that the data analysis is probably at least 30 minutes per data set.

Obvious proposals for time reduction: - One tilt angle rather than three: nearly 3X time savings - One LO frequency rather than three: nearly 3X time savings - Fewer than 10 repetitions

Analysis

Analysis software is a VBA script run from here and a data template is here. Copy the relevant cells from the txt and s2p files into the data template, select the data spreadsheet, then run the routine ProcessRepetabilityData from RepeatabilityCode.xlsm.

An early results document is here.

-- JohnEffland - 2010-07-29

This topic: ALMA > FEICSoftwareStatus > RepeatabilitySoftware
Topic revision: 2010-07-29, JohnEffland
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding NRAO Public Wiki? Send feedback