Back to the NIKA2 Run 75 main page

Contents

  1. NIKA2 Re-Commissioning (4th. attempt)
      1. NIKA2 run 75, Cryo run 87
    2. Oct. 13, 2025
    3. Oct. 14, 2025
      1. End of the day+night summary
      2. Tables (from 2025/10/14 ~13:00 UT to 2025/10/15 ~4:00 UT)
    4. Oct. 15, 2025
      1. Summary
      2. Tables (from 2025/10/15 ~04:00 UT to 2025/10/16 ~2:30 UT)
    5. Oct. 16, 2025
      1. Summary
      2. Tables (from 2025/10/16 ~02:30 UT to 2025/10/17 ~2:30 UT)
    6. Oct. 17, 2025
      1. Tables (from 2025/10/17 ~02:30 UT to 2025/10/18 ~03:00 UT)
    7. Oct. 18, 2025
      1. Tables (from 2025/10/18 ~03:00 UT to 2025/10/19 ~02:00 UT)
    8. Oct. 19, 2025
      1. Tables (from 2025/10/19 ~02:00 UT to 2025/10/19 ... UT)

NIKA2 Re-Commissioning (4th. attempt)

NIKA2 run 75, Cryo run 87

Oct. 13, 2025

14:00-17:00; Martino, Alessandro, Dave, Angel: a new sweep was performed. Previous initialization file: run87_old.ini. Bad/total KIDs counts (cryo window closed) after sweep (Ar1, Ar2, Ar3): 9/1000, 8/574, 4/1013. New file in force: run87_new.ini. Files run87_new_tp.ini and run87_new_pol.ini were also created. See nika2-a:~/nika_V3_IRAM.ini.

Oct. 14, 2025

Angel, Miguel, Stefano, and maaaany others arrive at the telescope.

We rename the ini files:
run87.ini (active file)
run87_old.ini (OLD active file, not used in August)
run87_newSweep_tp.ini
run87_newSweep_pol.ini
run87_oldSweep_tp.ini
run87_oldSweep_pol.ini

@15:00 LT (13:00 UT) we start, after preparing the whole "NIKA2 week environment architecture: 15/8/55 lost KIDs in Ar 1,2,3. Unstable sky and thick clouds coming from the N. We're entering in the peak of the tau forecast. It will be better later this evening. Meanwhile we try to focus. The old make-imbf script does not produce the imbfits files. Angel contacts Albrecht to solve the issue. The new make-imbf produces the file regularly.

On the very variable thick clouds, we notice a very different response of Ar1 and Ar3 to sky noise (variations), the latter having a bkg few times higher than the former (compare the y-axis scales). Is it real? Is there a flux calibration problem in the old DAFs? The a posteriori (19:00 LT) check does not reveal any calibration issue: Ar1 and Ar3 give similar fluxes for the targeted sources.

Ar2 bad sky Ar1 bad sky Ar3 bad sky

While we wait, some checks to some Antenna temperature control cables are being carried out by the telescope team. We start again 18:45 LT. The old make-imbf is now producing files correctly (thanks Albrecht!).

We work on the focus. Here follow our observations/notes. It is hard to get the focus value for all three arrays at once. The Ar2 flux criterion shows big jumps. The Ar2 FWHM and flux criteria do not agree. It almost seems like the flux is maximized not when the FWHM is minimized. Strange behavious, almost as if there were more than one "best" focus position. We enlarge the focus range of the 5 scans sequence. It was +/- 0.8 mm in steps of 0.4 mm. Now we changed it to +/- 1.0 mm, in steps of 0.5mm. The aim is to see if we can have a more complete sampling of the focus curve (for all criteria). The new script is called focusOTF-Z_2025.pako. We also try with a total excursion of +/-1.2 mm and steps of 0.6 mm. It seems that this latter solution is the best.

Is it possible that the difference in focus between Ar2 and Ar1 is changing? The first time it was ~0.3 mm. The second time it was ~0.6 mm. Now it is ~1.0mm??? At the same time, the focus difference between Ar1 and Ar3 seems stable at ~0.2 mm. It's early to be worried, but let's keep an eye on this! Nope, using the +/-1.2 mm excursion, we get back to 0.6mm difference between Ar2 and Ar1. Fiuuuu.

We alternate pointing/focus obtained with the old sweep and beam maps taken with the new one. The goal is to have at least two beam maps per array, taken with the optimal focus for that same given array. In the tables below, we list: focus sequences, beam maps, calib_1scan, other relevant scans, and a book keeping of sweep changes (to be used when defining the new calibration files - PIIC DAFs, for example - for this recommissioning run).

End of the day+night summary

At the end of the day+night, we have:

In this way, we cover the whole range from Ar2-0.3 to Ar3+0.3 with 6 focus positions: Ar2-0.3, Ar2, Ar2+0.3, Ar1, Ar3, Ar3+0.3. It might turn out that we need more positions in order to determine well the three focal surfaces.

Note that the latter 2 items (in the list above) have been performed when the taumeter opacity started to increase. Nevertheless the l.o.s. peak-to-peak sky noise variations kept being ~1 Jy/beam the whole time. If these last three beam maps have issues, we'll need to perform a whole sequence of 6-7 beam-maps at different focal positions, during a very stable night, such that the reference focus does not vary across the sequence (it takes few hours, hence the very stable need.)

Note also that the best way to carry out a beam maps focus sequence qould be to observe them one after the other on the same target, during a very stable night, such that the "reference" focus would not change along the whole sequence. If we'll get a very stable night, we shall do it!

Impressions about the difference in focus of the three arrays. The focus difference between Ar2 and Ar1 seems to be between 0.6 and 0.7 mm. The focus difference between Ar1 and Ar3 seems to be further 0.2-0.3 mm. The total focus difference between Ar2 and Ar3 reaches up to 1.0 mm. A decent compromise might be to use the focus of Ar1 as reference, but a quantitative analysis is needed to say the last word.

Tables (from 2025/10/14 ~13:00 UT to 2025/10/15 ~4:00 UT)

Dinner Time at Pico Veleta...
Our friendly goat having dinner... Our friendly fox waiting for dinner... They're waiting for dinner too!
Meet our new friend. Her name is Tau. She wants dinner too. And this is C/2025 R2 Swan... it's evidently dinner time up there too. And this is C/2025 A6 Lemmon... yammi!

Oct. 15, 2025

Summary

HLS091828 Ar1 HLS091828 Ar2 HLS091828 Ar3

Tables (from 2025/10/15 ~04:00 UT to 2025/10/16 ~2:30 UT)

Oct. 16, 2025

Summary

Stefano takes over at ~2:15 UT (4:15 LT). I let the observations of the science field finish and then I move to calibrations etc. See Tables.

Meanwhile, by 10:30 UT, I complete the (preliminary) analysis of the best beam-maps and I produce (preliminary) DAFs. They are now installed at Pico and they work fine. Below here follow a couple of figures for the three arrays, obtained by combining the best (on focus) beam maps of each array.

At 11:55 UT I stop observing, because the sky degraded rapidly and dramatically (peak to peak >100 Jy/beam in Ar2, up to 500 Jy/beam in Ar1 and up to 2000 Jy/beam in Ar3!)

Ar2 beam map geometry, rotated Ar1 beam map geometry, rotated Ar3 beam map geometry, rotated
Ar2 flat field Ar1 flat field Ar3 flat field

Tables (from 2025/10/16 ~02:30 UT to 2025/10/17 ~2:30 UT)

Oct. 17, 2025

UMa, meteor, satellite, and comet Lemmon Comet C/2025 A6 Lemmon again

Stefano takes over at ~2:15 UT (4:15 LT). NIKA2 is observing since 23:55 UT. The taumeter tau is spiky and >0.5. I decide not to switch to polarimetry and to cotinue in total-power. Since yesterday ~10:30 UT we are using only the new sweep and we do not switch anymore to the old one. From now on, I won't specify "new sweep" anymore. The DAFs installed yesterday in the late afternoon are in good shape.

At 9:20 UT we start observing Comet C/2025 A6 Lemmon.

We notice that Box C in Ar2 (the Box that had the amplifier changed; the one that includes the reference KID) is still more noisy than the others! This effect seems to be there since the beginning of the run, i.e. it was so also with the old sweep. With poor tau (like today) the issue is more prominent and easier to see.

@11:15 UT we switch to the 47 Hz sampling, keeping NIKA2 in total power. KIDs lost: 23/5/77. This special setup is aimed at testing very high scan speeds. In order to keep a Nyquist sampling of the beam along the timeline, we need the faster sampling frequency. We observe M51 with 200 arcsec/s at El>75 deg, but bad taumeter tau (the sky is becoming cloudy). Wind speed 2-4 m/s (i.e. nothing). Elevation up to 80 deg. Between 6% and 9% (depending on scan) of records are lost because of too high tracking deviations (so says the PIIC monitors). Supposedly this is due to the combination of high El and high scanning speed. We could increase the time between sub-scans to try to alleviate these losees (currently it is 5 seconds). Moreover, with 200 arcsec/s scan speed, a map of 15x10 arcmin, and 5s between sub-scans, the time spent in subscans in ~55% of the total time spent (so says the PIIC monitor again). Stefano thinks that this is probably a factor to keep in mind, when computing the pro's and con's of science observations performed with high-speed scanning.

@12:30 UT ca. we stop the observations because the sky is now completely thick. The telescope is now parked. Rain is expected to start soon.

@22:30 UT: Resuming observations after hail, rain and electric storm. Opacity ~0.6 at the beginning with quite variable atmosphere stability. Even so, forecasts are promising. DAQ restarted in TP, bad KIDs Ar1/2/3: 26/2/76.

@23:50 UT: Large fluctuations from the taumeter readings. Joaquin went outside for wiping the taumeter box, just in case. But timeline variations in consecutive scans (e.g. peak-to-peak vars. from ~1 to 10 Jy/beam in a few minutes) are consistent with unstable atmosphere. We obtained a set of calib_1scan on Uranus, Neptune, MWC349, CRL2688 and NGC7027 in the 25 < El.[deg] < ~70 range, with focus optimized to Ar1 + two rounds on Field.

Tables (from 2025/10/17 ~02:30 UT to 2025/10/18 ~03:00 UT)

Oct. 18, 2025

Stefano takes over at 3:00 UT (5:00 LT) and continues the observations. Taumeter tau ~0.55; peak to peak sky noise variations <1 Jy/beam at 2mm on the Az sub-scan of a Uranus pointing. New NIKA2 KIDs tuning: 35/7/97 KIDs lost. It snowed on the mountain, during the night. I perform the usual round of science targets for building up a small deep field to the aim of determining the NEFD, as in the past days.

New M51 test, this time with scan speed 180 arcsec/s at El>56 deg, so to catch some decent sky, because later (at culmination) the taumeter predictions worsen again. If possible, the test will be repeated also later, though. With 180 arcsec/s, 5s between subscans, a 15x10 arcmin scan size, 20arcsec step between sub-scans, and 52 subscans in total, the PIIC monitor says that 61% to 63% of the time is spent in the (science) subscans. Sorry, I forgot to switch to the 47 Hz sampling.

C/2025 A6 observations up to El=88.7. The vertex shuts, but no alarm rings. Moreover, two out of three scans have too many records with too high tracking deviations (or incorrect slow traces) and the PIIC monitor does not process them.

@17:30 UT: Monica resumes the observations. Opacity ~0.4 at the beginning, quite stable after some light rain and thick clouds. DAQ restarted with bad KIDs Ar1,2,3: 16,7,41. Source HFLS3 was observed during 2 hrs (8 rounds). Source MM18423 was observed 1 hr (4 rounds). Calib_1scan in MWC349, NGC7027 and CRL2688 + skydip. Then move to source Field for 1 hr (4 rounds). Ending the shift with calib_1scan in Neptune and Uranus.

Tables (from 2025/10/18 ~03:00 UT to 2025/10/19 ~02:00 UT)

Oct. 19, 2025

Stefano joins at 3:40 LT (1:40 UT). We switch to polarimetry. Restarting the DAQ is fine, but starting the HWP takes 3 attempts. Each time we have to start from scratch. The most critical step is the "famous" "stop - start - header - reconnect" sequence to reach the good mppsync dt, that keeps being obscure to basically everyone. Imho, NIKONA will need to have a different system, if polarimetry will be included, please. NIKA2-Pol is finally ready at 4:20 LT (2:20 UT): 12/6/21 lost KIDs. tau(225)~0.2. During the W24 and S24 semesters, switching to polarimetry was smooth. Hopefully it will be again in S25 and W25 (polarimetry starts in the second S25 week).

I observe M1, alternating science and Uranus as follows:

I then continue on a QSO, but @~05:30 UT, while preparing the pointing, the PIIC monitor protests that the SWEEP has changed. Uhm. It was working fine until the last pointing. I check the expiration of the DAFs and it's all good. Besides, I extended them a couple of days ago. I repeat the pointing three times, but it won;t work. I check that the ini file is still the same. Of course it is: nobody has touched it! I restart the PIIC sessions, nothing. I re-start the acquisition. Now the probles seems to be "corrupted data" and/or "time errors". ARGH! The data taken with the old make-IMBF finally arrive @5:56 UT and all is fine. This means that the problem is in the new make-imbf. Something happened in there. I'll go on with the old one, but the problem is that it is slow. Sigh. Starting from scan 20251019s50 I use the old make-imbf, but I set the focus as Ar2+0.65, otherwise I gotta wait too long for the Ar1/3 to be ready. For example, for the sequence of scans 50-54 I set focus +1.15 (i.e. Ar2+0.65) without waiting for the old make-imbf to produce the Ar1/3 files (it needs to chatch up all the M1 observations!). In the focus-sequences table below, instead, I fill the Ar1/3 focus values a posteriori, when the old make-imbf has produced the files.

A little less than 1h lost.

I now go for the QSO with NIKA2-Pol. I follow the same procedure as for the Crab, although now Uranus is at lower Elevation. Actually, it is in an ideal configuration: Uranus and 3C84 have the same Elevation!

Checking the files produced by the new make-imbf, it seems that the problem started at scan 20251019s44.

@07:50 UT I switch back to total power (9/7/21 KIDs lost). I keep the sampling frequency to 47 Hz, so that I can finally run the high scan-speed experiment properly. I do it so early (far from culmination), because the tau prediction is bad for later today. Starting from scan 20251019s69 we're back to total power.

Tables (from 2025/10/19 ~02:00 UT to 2025/10/19 ... UT)