5 sigma outliers: Aggressive cut out everything that is clearly due to beam motion and could have been cut by WACs # mul_prune->Draw("rcdb_sign*eigen_lagr_asym_us_avg/1e-6>>Apv(1000,-22000,22000)","(ErrorFlag&0xda7e6bff)==0 && rcdb_arm_flag==0 && rcdb_run_flag == 1 && rcdb_run_type == 1 && ( !(run_number==6563 && BurstCounter==0) && !(run_number==6564 && BurstCounter==4) && !(run_number==6567 && (BurstCounter==2 || BurstCounter==4)) && !(run_number==6571 && (BurstCounter==3 || BurstCounter==4)) && !(run_number==6593 && BurstCounter==2) && !(run_number==6629 && (BurstCounter==1 || BurstCounter==5)) && !(run_number==6976 && BurstCounter==0) && !(run_number==6983 && BurstCounter==8) && !(run_number==7013 && (BurstCounter==2 || BurstCounter==4)) && !(run_number==7014 && (BurstCounter==1 || BurstCounter==2 || BurstCounter==3)) && !(run_number==7149 && BurstCounter==6) && !(run_number==7211 && BurstCounter==4) && !(run_number==7257 && (BurstCounter==1 || BurstCounter==3 || BurstCounter==7)) && !(run_number==7657 && (BurstCounter==0 || BurstCounter==1)) && !(run_number==7865 && (BurstCounter==2 || BurstCounter==4 || BurstCounter==7 || BurstCounter==9)) && !(run_number==7866 && (BurstCounter==2 || BurstCounter==4 || BurstCounter==7)) && !(run_number==7867 && (BurstCounter==0 || BurstCounter==4)) && !(run_number==7889 && BurstCounter==0) && !(run_number==7942 && BurstCounter==5) && !(run_number==8036 && BurstCounter==2) && !(run_number==8046 && BurstCounter==2) && !(run_number==8091 && (BurstCounter==3 || BurstCounter==4)) && !(run_number==8129 && BurstCounter==2) && !(run_number==8133 && BurstCounter==1) && !(run_number==8187 && BurstCounter==3) && !(run_number==8240 && BurstCounter==1) && !(run_number==8549 && (BurstCounter==0 || BurstCounter==1 || BurstCounter==4)) )") Only cut out miniruns where something definitely should have been cut out (1.5mm or more BPM 12X motion or magnet trip). # mul_prune->Draw("rcdb_sign*eigen_lagr_asym_us_avg/1e-6>>Apv(1000,-22000,22000)","(ErrorFlag&0xda7e6bff)==0 && rcdb_slug<4000 && yield_bcm_an_us>135 && rcdb_arm_flag==0 && rcdb_run_flag == 1 && rcdb_run_type == 1 && ( !(run_number==6564 && BurstCounter==4) && !(run_number==6567 && (BurstCounter==2 || BurstCounter==4)) && !(run_number==6571 && (BurstCounter==3 || BurstCounter==4)) && !(run_number==6593 && BurstCounter==2) && !(run_number==6983 && BurstCounter==8) && !(run_number==7149 && BurstCounter==6) && !(run_number==7211 && BurstCounter==4) && !(run_number==7889 && BurstCounter==0) && !(run_number==7942 && BurstCounter==5) && !(run_number==8036 && BurstCounter==2) && !(run_number==8240 && BurstCounter==1) && !(run_number==8549 && (BurstCounter==0 || BurstCounter==1 || BurstCounter==4)) )") 5454,2 - OK - Could Cut - Has 3 outliers, coming from 3 different causes. There are 2 spikes in 12X which should have triggered WAC cuts, and one final spike that should have triggered a burp cut, but didn't. They are all in the 1 - 1.5 mm 12 X region that aren't too concerning, but they definitely show up in the USL main detector yields (but not in the us avg regressed asymmetry). 5640,3 - OK - Could Cut - There is a brief spike in all the BPMS which also shows up in the regressed us avg asymmetry. I don't think this is cause to cut the entire minirun, as it is only ~4 multiplets outside the RMS band and they are evenly distributed about 0 (as far as my eye can tell). 6256,3 - OK - Could Cut - There is a clear jump in all the BPMs at a small level, and a 1mm jump in 12X (which we had deemed as acceptable during WAC meetings). The main detector asymmetry gets broader by ~2x RMS during the 12X jump part, but it is distributed fairly evenly about zero, so I don't think this whole minirun has issues. 6440,1 - OK - Could Cut - It has 4 events with clear 12X spike and impact on PMT and asyms, but it is super short and the scale of beam motion is < 1mm 6434,2 - OK - Could Cut - FFB turn off right before a beam trip, causes wacky BPM positions and a few little outliers in 12X and PMTs, and also just barely in regressed asym, but I don't think it is pathological 6530,2 - OK - Could Cut - Large BPM 12X movement, at 1mm level (below cutting threshold) that causes PMT yields to move and a bit of hair in regressed US avg 6563,8 - Suspicious - 1mm shift in 12X and all of the regressed RMS afterwards are much larger - indicates regression failure due to shift from one slope spot to another... probably needs to be cut out ---- CUT 6564,79 - Suspicious - 1.5mm shift in 12X (more than usually allowed by WACs) and it has a non-trivial impact on regressed asymmetry and PMT yields as it recovers... looks bad enough for 1000 events that it is worth cutting out the minirun ---- CUT 6567,59 - Bad - 2 problems: First is a similar 12X jump as in the prior run, needs cuts for the same reason, and at the final 3000 events of the run there is a major problem -> LHRS Q1 tripped off and the PMT is just bad. Unnoted in the RCDB... so definitely needs to be cut ---- CUT 6571,70 - Bad - 2 instances of BPM 12X having a suddent burst of noise and then suddenly worse RMS at the 1mm scale! But no net shift in position. Basically the BPM is having instrumental problems that the ther BPMs are not seeing in the same way, the PMT yields are getting way noisier, and the corrections fail somewhat. 6585,10 - OK - No Cut - Looks OK to me. No clear problem except BPM 4eX gets a bit noisier for a few seconds. ---- CUT 6593,23 - Suspicious - 2mm scale 12X jumps and clear impact on PMT signals and asym width. The spread of messy asymmetry is normally distributed though, so it is just an inadequate slope for a few seconds I believe. 6629,2 - Suspicious - 2 spots where the BPM 12X jumps in what should get cut by a burp cut, shows up in PMT signals, but which introduces just a single spike event in regressed asymmetry. All 3 bad multiplets pull the same direction, so we could cut this out if we wanted to... 6976,5 - Suspicious - There are 2 spots, the first is just a tiny 1 event spike in 12X, and it shows up in the PMT and regressed asym signals, and the second instance is a very clear BPM 12X jump that only lasts a few events long, but it is bad and should definitely be cut (and it is most of the horrible outliers too :) ) ---- CUT 6983,71 - Bad - 2 spots again, one is a clear 3 mm jump in BPM 12X which affects the asym width badly, other appears to be a sudden introduction of eventwise noise in 12X and some spots in the X and Y BPMs. The PMT yields get noisier but actually the asymmetry RMS is not affected except 1 entry. Anyway, do cut here. 7013,2 - Suspicious - multiple spots of eventwise hair in 12X yield, which shows up in PMT yields and asym hair as well... energy control problems 7014,6 - Suspicious - even more spots, just like prior run, indicating 12X issues 7140,2 - OK - Could Cut - 1mm jump in 12X that shows up in PMTs and asym, but only 1 entry that is noticeable and not a big issue 7149,3 - Suspicious - 3mm jump in 12X that shows up everywhere 7211,2 - Suspicious - 3mm jump in 12X, again, same kind of problem as before. The BPMs are all drifting around anyway, so some issue in the accelerator? 7257,2 - Suspicious - 3 cases of 1mm jump in 12X that show up in the PMTs and regressed asyms, and they are just single outlier events. Seems like substantial instrumental problem that is very transient here? 7262,2 - OK - No Cut - There is no clear issue here other than small 12X bit, but it is not cutable 7657,3 - Suspicious - 2 cases in minirun 1 that have random jumps in 12X appearing to correspond to the BMOD coils active. The main detector is affected in RHRS yield, but not clearly at all in the asym points 7681,4 - OK - Could Cut - 2 cases in minirun 1 that have random jumps in 12X appearing to correspond to the BMOD coils active. The main detector is affected in RHRS yield, but not clearly at all in the asym points 7865,8 - Suspicious - 1mm scale 12X significant instability, definitely something wrong in the accelerator, which probably makes the slope calculations bad. 7866,5 - Suspicious - 1mm scale 12X significant instability, definitely something wrong in the accelerator, which probably makes the slope calculations bad. 7867,4 - Suspicious - 1mm scale 12X significant instability, definitely something wrong in the accelerator, which probably makes the slope calculations bad. 7889,4 - Bad - Clear 1mm scale jumps in 12X that cause PMT problems and asym hair. Cut 7902,2 - OK - No Cut - Looks fine. A bit noisy BPMs but no clear problem. 7903,22 - OK - Could Cut - 1mm scale 12X jumps that should have been burp cut I guess, but no big impact on asym hair 7904,3 - OK - Could Cut - 2mm scale 12X jumps that should have been burp cut I guess, but no big impact on asym hair 7907,6 - OK - Could Cut - 2mm scale 12X jumps that should have been burp cut I guess, but no big impact on asym hair 7908,5 - OK - Could Cut - 2mm scale 12X jumps that should have been burp cut I guess, but no big impact on asym hair 7909,9 - OK - Could Cut - 2mm scale 12X jumps that should have been burp cut I guess, but no big impact on asym hair 7910,8 - OK - Could Cut - 2mm scale 12X jumps that should have been burp cut I guess, but no big impact on asym hair 7911,13 - OK - Could Cut - 2mm scale 12X jumps that should have been burp cut I guess, but no big impact on asym hair 7912,3 - OK - Could Cut - 2mm scale 12X jumps that should have been burp cut I guess, but no big impact on asym hair 7913,10 - OK - Could Cut - 2mm scale 12X jumps that should have been burp cut I guess, but no big impact on asym hair 7914,14 - OK - Could Cut - 2mm scale 12X jumps that should have been burp cut I guess, but no big impact on asym hair 7924,3 - OK - Could Cut - 2mm scale 12X jumps that should have been burp cut I guess, but no big impact on asym hair 7932,3 - OK - Could Cut - 2mm scale 12X jumps that should have been burp cut I guess, but no big impact on asym hair 7933,7 - OK - Could Cut - 2mm scale 12X jumps that should have been burp cut I guess, but no big impact on asym hair 7935,2 - OK - Could Cut - 2mm scale 12X jumps that should have been burp cut I guess, but no big impact on asym hair 7936,3 - OK - Could Cut - 2mm scale 12X jumps that should have been burp cut I guess, but no big impact on asym hair 7937,5 - OK - Could Cut - 2mm scale 12X jumps that should have been burp cut I guess, but no big impact on asym hair ---- CUT 7942,4 - Suspicious - some of the same noise as in prior runs, but also has one bit where the 12X BPM jumps around at 1mm very aggressively and has a bit of an impact on the regressed RMS... not that bad though 7953,3 - OK - Could Cut - 2mm scale 12X jumps that should have been burp cut I guess, but no big impact on asym hair 7956,2 - OK - Could Cut - 2mm scale 12X jumps that should have been burp cut I guess, but no big impact on asym hair ---- CUT 8036,3 - Suspicious - some of the same noise as in prior runs, but also has one bit where the 12X BPM jumps around at 1mm very aggressively and has a bit of an impact on the regressed RMS... not that bad though 8046,7 - Suspicious - Brief beam on period that is noisy... basically should have been considered a trip 8091,4 - Suspicious - Two spots of BPM 12X motion at 1mm level, but not a too-bad impact on the asym width (on the later glitch) 8129,3 - Suspicious - BPM 12X motiona at the 1mm level again... again not that bad, but it has an impact 8133,3 - Suspicious - BPM 12X motiona at the 1mm level again... again not that bad, but it has an impact 8177,2 - OK - No Cut - It may be a single event 12X glitch, but it is not really visible. Twice! 8187,6 - Suspicious - BPM 12X motiona at the 1mm level again... again not that bad, but it has an impact. There are some outliers with no clear cause... so not all of it is cuttable. ---- CUT 8240,26 - Bad - Large 12X noise and 1.5 mm shift that has a noticeable effect, post beam trip recovery ---- CUT 8549,7 - Bad - Large 12X noise and 1.5 mm shift that has a noticeable effect 8551,2 - OK - No Cut - There is a clear problem with the regressed asymmetry at some point, but it is not BPM related (probably BCM?)