TGBT14A_912_99 - Compare Mode 10 (hw), Mode 10 (sw) Mode 23

Goals

• Compare spectra taken in Mode 10 (hw), Mode 10 (sw) Mode 23

Details

• 9:30am load software, check Astrid scripts
• Adam uses X-band receiver, so configure for that also (may not be at focus), even though we are using the noise source
• 11:00am, observing:
• Archivist
• noise_test_mode10_xband_nocal - lbw_gain set to 4096 (this is L8 BOF, not L1 BOF)
• adc and sb1snap powers approx -18dBm
• Run scans 1-11
• bandpass shape looks bell-shaped, L & R polarizations identical. All 11 scans look identical
• noise_test_mode23_xband_nocal - lbw_gain still set to 4096 (this is L8 BOF, not L1 BOF)
• Run scans 12 - had left nwin = 1, switch to nwin = 8 (and 8 restfreqs)
• Scan 13 - now 8 windows, wasn't setting all 8 lbw_gains, fix this. Get the syntax wrong and crash Grail, try a different syntax.
• Still can't get it right, just do it manually
• Scan 14 - now working correctly. All look identical, but below Mode 10 scan
• Run Scans 15-24
• 12:10pm. Joe switches to "no software fix" version, and latest BOF (l8_ver160_1500_2015_Dec_15)
• noise_test_mode10_xband_nocal
• Scan 25 - looks different again!
• Scan 26-35 - 10 more of the hardware version of mode 10. lbw_gain still 4096
• Scan 25 and 32 are off
• top white line (at center) = hardware mode 10, red (middle) = software mode 10, white (bottom) = mode 23
• noise_test_mode23_xband_nocal (using new BOF)
• Scan 36 - had lbw_gain of 1024 (default value)
• Scan 37 - lbw_gain = 4096, sb1snap = -9dBm
• revert to mode 10, lbw_gain = 4096 - looks correct
• Mode 23, lbw_gain = 1024 give -20.5dBm
• Scan 38,39 - Mode 23, new BOF, no fix necessary, lbw_gain = 1024. Spectra power level is way low. 1.1 10**9 c.f. 2.5 10**10
• Switch to 2**10.5 (1448) now looks ok. Histograms are thinner than mode 10, because mode 23 histogram only shows one component of power.
• Scan 40 - goes from 1.1 10**9 to 2.4 10**9
• Switch back to lbw_gain of 4096. Seems to produce spectrum power same as mode 10, although sb1snap is ~ -9dBm
• Scan 41 - 50 mode 23 (new BOF)

• Conclusion: old Mode 10 (sw, old bof) and new Mode 10 (hw, new bof) match, but new bof is bistable
• old mode 23 and new mod 23 agree, but are different from mode 10
• Joe notes they are using different coefficients.

• 1:25pm - finito