Back to the NIKA2 main page

Daily Reports (2026/10/6-13, NIKA2 run 81, Cryo run 93)

Observers on-site: Rémi, Giorgos, Mairi, Miren

AoD: Ioannis

Operators: Pedro, Sergio

Important

Focus offsets shall be measured on Ar1
Pointing offsets shall be measured on Ar2

Pool-Account usage: Accounting of observing times

• Calibration observations like pointing, focus, and calib_1scan (the latter in case of NIKA2-Pol only, aimed at characterizing the instrumental polarization) which are conducted before and during project observations are part of the project time. Note that all proposals include such overheads in the time and sensitivity estimates. This holds for total power and polarimetry observations.

• Other calibration observations which are important for all projects like beam maps, skydips, or calib_1scan for absolute flux calibration may be accounted under the pool account (nikaX-YY, where X = s,w and YY = year).

• When conditions are unstable and it is unclear whether project observations make sense at all, calibrations may be accounted to the pool account.

Book-keeping of important scans

We ask you to please keep track of focus sequences (as much as possible), beam maps (always), and changes of sweep (until the new sweep DAFs are permanently installed). Thanks

Focus sequences

Please fill the Table with the Ar1,2,3 focus values as follows. At best, make one table per day.

• Focus sequences
  Scan nr	Target	Ar2	Ar1	Ar3	Polarized?	Notes
  20261007s25-29	Mars	1.85 mm	1.71 mm	1.70 mm	no	with run92.ini and old DAFs
  20261007s33-37	Mars	1.76 mm	1.63 mm	1.59 mm	no	with run92.ini and old DAFs
  20261007s43-47	1156+295	2 mm	1.75 mm	1.75 mm	no	with run92.ini and old DAFs
  20261007s48-52	1156+295	2.2 mm	1.75 mm	1.8 mm	no	with run92.ini and old DAFs
  20261007s53-57	1156+295	2.05 mm	1.78 mm	1.8 mm	no	with run92.ini and old DAFs
  20261007s58-62	1156+295	2.3 mm	1.85 mm	1.83 mm	no	with run92.ini and old DAFs
  20261007s65-69	1641+399	2.35 mm	1.7 mm	1.74 mm	no	with run92.ini and old DAFs
  20261007s70-74	1641+399	1.9 mm	1.75 mm	1.75 mm	no	with run92.ini and old DAFs
  20261007s75-79	1641+399	1.9 mm	1.69 mm	1.71 mm	no	with run92.ini and old DAFs
  20261007s80-84	1641+399	2.22 mm	1.93 mm	1.96 mm	no	with run92.ini and old DAFs, with the original focus script (0.4 mm steps)
  20261007s97-101	1641+399	2mm	1.5mm	1.52mm	no	with run92.ini and old DAFs, with the original focus script (0.4 mm steps), a2 not well constrained
  20261007s162-166	2200+420	1.05 mm	0.7 mm	0.71 mm	no	with run92.ini and old DAFs

  Focus sequences
  Scan nr	Target	Ar2	Ar1	Ar3	Polarized?	Notes
  20261008s1-5	Uranus	1.36 mm	1.09 mm	1.09 mm	no	with run92.ini and old DAFs
  20261008s11-15	Uranus	1.55 mm	1.09 mm	1.10 mm	no	with run92.ini and old DAFs. Starting on A1 value. Not great at 2mm. Using peak values
  20261008s16-20	Uranus	1.55 mm	1.18 mm	1.18 mm	no	with run92.ini and old DAFs. Starting on A2 value. Using peak values
  20261008s25-29	Uranus	1.51 mm	1.09 mm	1.11 mm	no	with run92.ini and old DAFs. Starting on A1 value. Using peak values
  20261008s31-35	Uranus	1.41 mm	1.14 mm	1.14 mm	no	with run92.ini and old DAFs. Starting on A2 value. Using peak values
  20261008s40-44	Uranus	1.47 mm	1.09 mm	1.08 mm	no	with run92.ini and old DAFs. Starting on A1 value. Using peak values
  20261008s45-49	Uranus	1.51 mm	1.17 mm	1.17 mm	no	with run92.ini and old DAFs. Starting on A2 value. Using peak values
  20261008s56-60	Uranus	1.32 mm	1.11 mm	1.11 mm	no	with run92.ini and old DAFs. Starting close to A2 value. Using peak values
  20261008s67-71	Uranus	1.43 mm	1.05 mm	1.06 mm	no	with run92.ini and old DAFs. Starting close to A1 value. Using peak values
  20261008s67-71	0316+413	1.34 mm	0.99 mm	0.99 mm	no	with run92.ini and old DAFs. Starting close to A1 value. Using peak values
  20261008s75-79	Uranus	1.43 mm	1.05 mm	1.06 mm	no	with run92.ini and old DAFs. Starting close to A1 value. Using peak values
  20261008s218-222	Uranus	1.24 mm	0.89 mm	0.88 mm	no	with run93.ini new DAFs. Starting close at fz=1mm. Using peak values

  Focus sequences
  Scan nr	Target	Ar2	Ar1	Ar3	Polarized?	Notes
  20261009s5-9	Uranus	1.38 mm	1.11 mm	1.13 mm	no	with run93.ini new DAFs. Starting at 1.2
  20261009s10-14	Uranus	1.46 mm	1.13 mm	1.14 mm	no	with run93.ini new DAFs. Starting at 1.6
  20261009s21-25	Uranus	1.61 mm	1.25 mm	1.24 mm	no	with run93.ini new DAFs. Starting at 1.75 to reach Ar2 circularity. Peak value are reported but we use fz=1.95 afterwards.
  20261009s32-36	Uranus	1.66 mm	1.3 mm	1.3 mm	no	with run93.ini new DAFs. Starting at 1.4.
  20261009s218-222	3C454	1.4 mm	1.03 mm	1.07 mm	yes	with run93.ini new DAFs. Starting at 1.4 mm.
  20261009s223-227	3C454	1.4 mm	1.0 mm	1.03 mm	yes	with run93.ini new DAFs. Starting at 1.2 mm.
  20261009s229-233	3C454	1.47 mm	1.03 mm	1.06 mm	yes	with run93.ini new DAFs. Starting at 1.6 mm.
  20261009s261-265	Neptune	1.56 mm	1.08 mm	1.1 mm	yes	with run93.ini new DAFs. Starting at 1 mm.

  Focus sequences
  Scan nr	Target	Ar2	Ar1	Ar3	Polarized?	Notes
  20261010s7-11	Uranus	1.7 mm (but out)	1.23 mm	1.22 mm	yes	with run93.ini new DAFs. Starting at 1.1. A2 not reliable.
  20261010s18-22	Neptune	1.62 mm	1.16 mm	1.16 mm	yes	with run93.ini new DAFs. Starting at 1.3. Project 031-26
  20261010s43-47	Uranus	1.58 mm	1.27 mm	1.29 mm	yes	with run93.ini new DAFs. Starting at 1.3.
  20261010s53-57	Uranus	1.64 mm	1.32 mm	1.35 mm	yes	with run93.ini new DAFs. Starting at 1.5.

  Focus sequences
  Scan nr	Target	Ar2	Ar1	Ar3	Polarized?	Notes
  20261011s10-14	Neptune	1.6 mm (but out)	1.19 mm	1.19 mm	yes	with run93.ini new DAFs. Starting at 1.2. A2 not super reliable.
  20261011s25-29	Neptune	out of range / bad	1.11 mm	1.10 mm	yes	with run93.ini new DAFs. Starting at 1.23. A2 not reliable.

Beam maps

Similarly, if any beam maps (total power) are taken:

• Beam maps (99 sub-scans)
  Scan Nr	Target	Focus	Notes
  20261007s30	Mars	1.71 mm (Ar1)	run93, new sweep
  20261007s39	Mars	1.63 mm (Ar1)	run93, new sweep
  20261007s181	0316+413	0.8 mm (Ar1)	run93, new sweep, pointing on ???
  20261008s8	0316+413	1.1 mm (Ar1)	run93, new sweep, pointing on Ar1
  20261008s23	Uranus	1.18 mm (Ar1)	run93, new sweep, pointing on Ar1
  20261008s38	Uranus	1.14 mm (Ar1)	run93, new sweep, pointing on Ar1
  20261008s52	Uranus	1.51 mm (Ar2)	run93, new sweep, pointing on Ar2
  20261008s63	Uranus	1.11 mm (Ar1)	run93, new sweep, pointing on Ar1
  20261008s90	Uranus	1.05 mm (Ar1)	run93, new sweep, pointing on Ar2
  20261008s224	Uranus	0.89 mm (Ar1)	run93, new sweep, pointing on Ar2 (but Ar1/3 are now aligned)
  20261009s16	Uranus	1.46 mm (Ar2peak)	run93, new sweep, pointing on Ar2
  20261009s27	Uranus	1.95 mm (Ar2circ)	run93, new sweep, pointing on Ar2. Focus to maximize Ar2 circularity of the beam.
  20261009s38	Uranus	1.3 mm (Ar1)	run93, new sweep, pointing on Ar2 (but Ar1/3 are now aligned)
  20261010s14	Uranus	1.23 mm (Ar1)	run93, new sweep, pointing on Ar2
  20261010s50	Uranus	1.28 mm (Ar1)	run93, new sweep, pointing on Ar2

Changes in sweep

Please keep track of sweep changes: it makes our life easier when building the DAFs/Kidpars.

• Changes of sweep
  Sweep	Scan Nr	Time	Notes
  OLD	up to 20261007s29		for pointing & focus
  NEW	20261007s30		beammap
  OLD	20261007s31-38		for pointing & focus
  NEW	20261007s39		beammap
  OLD	20261007s42-		for pointing & focus
  NEW	20261007s180-		beammap
  OLD	20261007s182		for pointing & focus
  NEW	20261008s7-		beammap
  OLD	20261008s9		for pointing & focus
  NEW	20261008s22-		beammap
  OLD	20261008s24		for pointing & focus
  NEW	20261008s37-		beammap
  OLD	20261008s39		for pointing & focus
  NEW	20261008s51-		beammap
  OLD	20261008s55		for pointing & focus
  NEW	20261008s62-		beammap
  OLD	20261008s65		for pointing & focus
  NEW	20261008s89-		beammap
  OLD	20261008s91		for pointing & focus
  NEW	20261008s97-		beammap

October 6th, 2026, Tuesday

10:30-15:30 : 3/4 observers arrived at the telescope. Rémi planned to arrive later today. Weather horrible. Maintenance ongoing.

Antenna parked all day, clouds, wind, and rain.

October 7th, 2026, Wednesday

07:00LT: pointing and focus tests with old sweep (run92). Difference between Ar2 and Ar1 ~0.35 mm. Difference between Ar3 and Ar1 practically 0.

10:00 - 11:30LT: new sweep by M. Calvo. We now have new run93.ini files.

11:45LT Miren and Ioanns : starting NIKA2 observations to get beammaps for the new sweep. Back to run92 for pointing and focus on Mars. DAQ started, lost KIDs Ar1/Ar2/Ar3: 129/6/14; traces look fine. Opacity between 0.3 - 0.5. Peak-to-peak in Ar2 +/- 4-5 Jy/beam. Focus differences: Ar1 - Ar2 = -0.14 mm, Ar1 - Ar3 = 0.01 mm.

12:45LT: moving to run93 for beammaps. Lost KIDs Ar1/Ar2/Ar3: 21/8/30. To be understood if box I is lost or not. Beammap (20261007s30) focused on Ar1.

13h25LT: back to run92 for pointing and focus before another beammap. Lost KIDs Ar1/Ar2/Ar3: 140/7/15. Moving back to run93 for beammaps. Beammap (20261007s39) focused on Ar1.

14h30LT Miren and Mairi: focus tests with run92. Using the original focus script (with 0.4 mm step) starting from scan 20261007s80.

16h55LT: we try a mini pointing session making use of the sources in the PCCS catalogue provided by S. Berta (we have had to remove the "G" from the names in the end... 13 characters otherwise...) Opacity is ~ 0.35 (tau225 GHz), but a lot of clouds. We can not point, p2p fluctuations @Ar2 of +/- 20 Jy/beam.

19h00LT: We resume. Sky is stable (<5 Jy/beam) and tau around 0.3.

Summary:

October 8th, 2026, Thursday

00:30LT Miren: Stable (=/-2 Jy/beam) and good opacity (~ 0.25). Better beammaps are needed. We move to 0316+413 for first beammaps, waiting Uranus to be high.

01:00 - 07:00 LT, Rémi : stable weather, clear sky and opacity ~ 0.25 for the entire shift.

• 2 beam maps on 0316+413 with focus and pointing optimized for A1, plus several pointing and focus
• 2 beam maps on Uranus with focus and pointing optimized for A1, plus several pointing and focus
• 1 beam map on Uranus with focus and pointing optimized for A2, plus several pointing and focus
• 1 skydip (using the new sweep): scan 20261008s53
• 1 beam map on Uranus with focus and pointing optimized for A1, plus several pointing and focus

• Remarks:

  • 1) the focus offset difference between A1/3 and A2 is about 0.3. However, the A2 (FWHM_x x FWHM_y)^1/2 is alway >21" and the beam is elongated: at best we have ~20"x25". The shape is not the same as the one we use to have. Issue with the current DAFs? The dichroic? Lateral focus (but this should also affect A1/3)? Other? 2) The best focus can vary by up to 0.1mm for A1/3 depending if we start from the current A1/3 or A2 value: systematic positive shift when starting with A2 value.

07:00 - 13:00 LT, Giorgos :

• finish the beammap of Remi's shift and back to old sweep
• IM: After a short chat in the morning with SB, realized that the beam maps should be pointed on Ar2 (as we do with the science observations). I tried to optimize for Ar1 (both focus and pointing), which was not the best practice (sorry!). So, the ones taken during the night with pointing on Ar1 might not be useful. There is one from the Rémi's shift that could be used (20261008s52, see beam maps table above) and I take another two with the proper setup (poi corr on Ar2, focus corr on Ar1) in the morning (20261008s90 and 20261008s98), starting at an elevation of about 35 degrees. Data processing for new DAFs are in progress by SB.
• start test scans of project 092-26. Opacity increasing. Ar2 beam very elongated.

13:00 - 19:00 LT, Mairi : Taking over shift from Giorgos. Continuing test scans for project 092-26. One repetition completed (s112 and 113). Unstable sky (+/-10Jy/beam), waiting to see if improves before continuing.

18:00 LT, SB. I've been processing the beam maps taken during the last night/morning with good weather and on Uranus and 0316+413. The result is good for Ar1. Ar3 is still downloading and I will then process it later tonight. And about Ar2... oh well, the beam is confirmed to be very elongated, even when supposedly "on focus". The position angle (PA) of the beam ellipse is rotating in AzEl, I guess as a function of Elevation. To give you an idea, here I include the sequence of Ar2 beam maps f.o.v., in the hope that this will help the instrument/hardware experts to figure out what is happening.

CK: Beam are elongated in the central region and by ~90deg with elevation/PA. However, beams are about circular in the region around -150”/0”. Is the 2mm array offset relative to the radio axis ?

CK: As Array 2 was moved by 17mm, is it possible that Array 2 is now very far out of focus ? SB: yes I had this thought as well. It will be tested tonight

19:00 - 01:00 LT, Ioannis :

• weather improves around 19:30, so we start with pointing and focus on 2251+158
• stability gradually improves and we decide to have a closer look on the focus and the extended beam of Ar2
• we noticed with focus scans of the previous days that there is a mismatch between the peak position of the focus curve in Ar2 and the position where one would "expect" the array to be focused based on the beam symmetry. The "symmetry" best focus position seems to be about +0.4mm from the "peak" best focus position.
• note that the above cannot be concluded by the "FWHM" focus curve (white focus curve), which usually seems in good agreement with the "peak" focus curves (green and red focus curves), possibly because the "FWHM" focus curve estimates the quadratic sum of the major and minor beam sizes and cannot assess the beam ellipticity
• so to summarize the Ar2 focus situation we see: Ar2 becomes best focused "peak-wise" at a certain focus position (this is the position that is about +0.35mm from the Ar1 optimal focus position) BUT seems to become best focused "ellipticity-wise" (i.e. more circular) at another position, which is another +0.3mm to +0.4mm from the best "peak-wise" focus position (and about +0.6mm to +0.7mm from the Ar1 optimal focus position)
• another note: apart from the focus difference, there is a X-Y pointing difference between Ar2 and Ar1/Ar3. it is more evident in elevation with [Ar2_elv minus Ar1/3_elv about -11" to -12"] and less profound in azimuth with [Ar2_azi minus Ar1/3_azi about -5"]
• to investigate the focus situation across the arrays, I ran a series of focus sequences covering an extended range of focus offsets to see the beam behavior in all three arrays. below I describe the procedure and findings:

Extended focus test procedure:

• start with a focus sequence to estimate the best focus position of all three arrays
• using the Ar1 focus as "anchor" point, let's call it f_a, run 3 focus sequences (of the "old" type, i.e with the 0.4mm steps) at the following start positions: (f_a - 1.6mm), f_a and (f_a + 1.6mm)
• this will create 15 focus "stamp images" over a focus range from (f_a - 2.4mm) to (f_a + 2.4mm)
• between focus sequences, the pointing is corrected according to the Ar2 offset
• a first test of this extended focus procedure was ran for source 2251+158 and very stable sky conditions (about 1-2 Jy/beam) and low opacity (tau_225 = 0.2-0.3). the 3 focus sequences that cover the extended focus range are 20261008s143-147 (middle part), 20261008s149-153 (left side part) and 20261008s156-160 (right side part).
• another test was done again on the same target, after the installation of the new DAFs (run93, see below). The conditions further improved (tau_225 = 0.15) and the results look of much higher quality than the first test above. The 3 focus sequences that cover the extended focus range are 20261008s184-188 (middle part), 20261008s190-194 (left side part) and 20261008s196-200 (right side part). Here is an overview of the extended focus sequences for all three arrays (Ar2, Ar1 and Ar3 from top to bottom):

• Hopefully these plots and the calibration/test data we will gather tonight will help understand the issue with Ar2

23:00 LT, SB: the new NKFR and RPP DAFs for the new sweep are now installed and running at Pico. They are still preliminary (based on one scan only), but they're fine. From now on, we are not going to use the old sweep anymore. And from now on, the run number is N2R81 and cryo run 93. The first scan (permanently) obtained with the new sweep is 20261008s176. Below here comes the f.o.v. of the adopted beam maps used for the these N2R81 RPPs. Good night

• After the new DAFs installation, the pointing offset between the three arrays is eliminated (as expected). Now all arrays agree in pointing within 1-2 arcsec. A new pointing session will also help us fix the absolute pointing offset.
• From the traces point of view, box I looks normal, i.e. if someone looked only at the traces cannot imagine that box I is dead. not sure what is happening there.

October 9th, 2026, Friday

01:00 - 07:00 LT, Rémi : opacity is very good over the shift, ~0.15. The stability is great. Some (relatively low) wind.

• Moving to Uranus
• Pointing, focus optimized for A1. Beammap. calib_1scan
• Pointing, focus optimized for A2 (maximizing the peak). Beammap. calib_1scan
• Pointing, focus optimized for A2 (maximizing the circular shape). Beammap. calib_1scan
• Pointing, focus optimized for A1. Beammap. calib_1scan.
• Skydip
• Moving to ALFORI: 12 calib_1scan, which could be used for first sensitivity estimates in good conditions
• Skydip
• Starting a new pointing session with the new DAFs using Planck sources (PCCS2_Berta2026)

SB: the pointing must always be set on Ar2, no matter what focus is used for the beam maps (Ar1 or Ar2), because Ar2 is the only one with one KID exactly on the optical axis of the telescope+NIKA2 system. Please re-check what pointing has been used for these last beam maps. If it was Ar1, then they are not usable. Cfr. the notes in previous night/morning. [IM: Yes, we discussed this with Rémi and the plan was to always point on Ar2 last night. He can also confirm, if needed.]

• Remarks:
  • At midnight, the imbfits for pointing scan 20261009s1 (following a beammap, i.e. heavy) were never produced. Running another made it worked fine. This appened again after beammap 20261009s16.
  • For some reason piic is not reducing the ALFORI scans at 2mm. [IM: Scans were probably reduced. The ql of Ar2 for science data was not running yesterday night.]
  • XEPHEM crashed during the shift: segmentation fault (core dumped).

CK: At 2mm, is it possible that we are finding only a secondary maximum showing an elongated beam due to some astigmatism ? Do we need to scan the z-focus over a much wider range to find the main maximum, where the beam is 18" and circular ?

IM: I could run an even extended focus sequence for this but not sure of the system limitations (e.g. PaKo focus limits etc). I'll try to find out. CK: let's first discuss with the experts.

SB: at this point I do not think that's possible. Ioannis tested this, with a very broad focus range of +/- 2.4 mm centered on the Ar1 focus (see the PDF he produced and that is attached above). I might be wrong, but such a double peaked focus "variety" (mathematical entity) seems hard to understand to me. If you check the f.o.v. figure of Ar2 above (23:00 LT, yesterday), to me it seems that the "circular" beam is simply not in the center of the array, but it is on the left of the f.o.v. (e.g. Ar2 KID nr 440 in the figure). This tells me that some optical element is not well centered or well aligned. Two things have been "touched" in the last intervention: the dichroic and Ar2. Since there's not such an effect on Ar1 or Ar3, my guess is that it's not a dichroic issue, but rather the shims insterted below Ar2 might not be perfectly planar (e.g. they're thicker on one side that on the other). Or something happened to the Ar2 support in general. Does it make sense?

07:00 - 13:00 LT, Giorgos : Finished pointing session. Switched to polarization. Started project 090-26, 2 reps. Source 236.60+38.70 -> very weak.

IM: We tested the Planck catalog for both pointing sessions so far. Many of the targets are bright and therefore suitable for pointing calibrators. But it seems that there are several faint sources that are marginally good for pointing sources and only if the "f" option of the pointing script is used. In particular, source 236.60+38.70 reported above by Giorgos was not detected at all with a pointing. I think there should be some additional filtering of the Planck catalog, maybe based on NIKA2 observations.

CK: 236.60+38.70 is listed with 120mJy @ 1mm in Planck catalog. Obviously, we should exclude weak sources and sources which are extended relative to the 30m beam from a Planck-based pointing catalog.

---

SL: Replies to several of the observations, comments and question above.

- The 20"x25" beam mentioned by Rémi is extremely elliptic. But in the series of focus sequences, the best compromise beam is clearly much less elliptic than that, or even not at all. So I suppose the best compromise beam should be less or equal to 20" diameter circle.

- More precisely, about elongated beams at 2mm (Ar2), sizes, peak intensity, possible secondary maximum. The beam shape evolution on focus sequences show clearly an effect of astigmatism caused by an optical element which doesn't have a shape symmetric by revolution around its vertex, but has a 2 axes of symmetry. This is very well visible on the series of 3 focus sequences displayed above. Before the optimal "best compromise" focus, the beam is very elongated in one direction, and at some point at best focus for the perpendicular direction. Then it continues to improve in the 1st direction and starts slowly to degrade in the perpendicular direction. Then we reach the best compromise: circular (squarish on side lobes due to quadrupod) beam not optimal for neither directions but still good enough. Then optimal focus for the first direction, but still more degraded in the perpendicular direction, creating an elongated beam in the direction perpendicular to the one before the best compromise. So, when there's a lot of astigmatism, the best compromise beam is in all cases worse than when there's no astigmatism. In other words, with the strong astigmatism visible in the 3 series of focus, it's certain we can't have 18" or lower circular FWHM. The best compromise beam must be a bit larger than that, even with perfect RPP. Now, is 21" FWHM too big in these conditions? If we assume RPP is not perfect... I don't know...
In all cases there's no other better maximum out of the sequence I just described.

- The preliminary 2mm RPP have been computed with not so good beam maps, so the grid is possibly not regular, while in reality, given the KIDs real positions, it should be regular. Few arcsec error position on many pixels enlarge the beam on maps by few arcsec. If we want a better idea of the real optical beam size, look at the beam profile in the pointing cross scans. I bet that when we are at best focus this is 18" or even less, isn't it? This remark on the RPP is possibly not true anymore since the Oct 8 23:00 entry from SB (tbc).

- The dichroic is totally transparent for the 1 mm beams, so any optical effect it has affects only Ar2, not Ar1 and Ar3!
- But, the dichroic is closer to the cold pupil than to an image surface, which means that even a local distortions on it will affect a very large fraction of the image. In that context, it looks quite complicated that it creates distorted beams on some part of the image and not distorted beams on other parts. That's why I think that the beam shape inhomogeneities visible on the series of beam maps reduced and posted by SB, with elliptic beams on a portion of the Array 2 and more circular beams elsewhere, is probably not a dichroic issue!
- Now, regarding the hypothesis of asymmetries induced by shims not perfectly planar, this is even more unlikely to me. Indeed, the shims have been checked, at most their planarity error is few 10s microns (an probably less in reality), which is totally negligible in terms of beam distortion, by far!

- With a downward residual distortion of the dichroic as we have now, and the fact that the dichroic is tilted, Zemax shows that this creates astigmatism for the reflected beam. So, the hypothesis that the astigmatism visible on Ar2 is caused by the dichroic, doesn't seem totally ruled out yet. But there's several elements that can help figuring out whether it's the dichroic or something else that creates the distortion we see these days on the 2mm beams:
-- The dichroic induced astigmatism is invariant with elevation. So in Nasmyth coordinates, whatever the elevation of the source, the directions of the ellipses before and after the best compromise focus are always the same. Visibly it's not the case on the beam maps...
-- The telescope astigmatism, created by the distortion of M1 with gravity, depends on elevation. It is almost 0 near 45-50° elevation, it appears with a major axis in one direction at higher elevation, and perpendicular to this direction at lower elevation, stronger and stronger when you get further away from 45-50° elevation. So, in Nasmyth the directions of the ellipses vary with elevation, and they are inverted when you pass the 45-50° limit. In other words, the 2 directions of ellipses you see on each side of the best compromise focus, are inverted at lower elevation with respect to the high elevation. On the beam maps we see the rotation with elevation, but do we see the inversion of ellipses direction when we pass the 45-50° limit? It's not clear to me yet...
-- According to my simulation, the astigmatic aberration spot created by the dichroic distortion should stay significantly smaller than the diffraction spot, which means the ellipticity shouldn't be very strong. Besides, the best compromise beam in the simulation is still smaller than the ellipse minor axis at the best focus of each astigmatic direction. In other words, the max beam intensity is at the best compromise focus, not a max a bit before the best compromise and another max a bit after the best compromise.
-- This dichroic astigmatism shouldn't be localized on the image ; it should be everywhere. Though, the direction of the ellipses is not totally uniform, there's should be an axis of symmetry passing by the center of the array, and ellipses slightly tilted on both parts of this axis. But that doesn't correspond to what we see on the maps.
- I don't know yet if, with a dichroic distortion upward there's the same kind of astigmatism. It's possibly not symmetrical. If it is symmetrical, we should have seen this effect before the intervention, but from my recollection this was not the case. I will check with simulations of course.

- We have a focus difference between array 2 and the 1 mm arrays; this means the dichroic distortion is not 1.5 mm amplitude at the center as I targeted when I put the 17 mm shims. So, first of all for the simulations, I need to figure out the distortion we have. Assuming that, based on the shape of the beam we estimate for now Ar1-Ar2 ~ -0.8 mm (for the distance between best 1mm focus, and best compromise round 2 mm beam), I will compute it. Once it's done, I'll try to quantify better everything I said above.

- When the dichroic is distorted, the position of the center of the membrane along the optical axis is not at the designed position. So with the tilt, this shifts the optical axis on the array by the same value as the distortion amplitude (to 1st order). in other word, the projection of the center of the array on sky should be different than the position of the center of the 1mm arrays (for an illustration of this effect, see the Zemax screenshot on https://publicwiki.iram.es/DailyReportsNika2Run81#Friday.2C_2nd_October_2026). That's why I'm a bit surprised that after the good beam maps reduction, Stefano finds the same pointing for all 3 arrays within 1-2 arcsec...

- Another thing that seems a bit neglected now, and might seem disconnected, but is possibly not: the radio alignment. As I wrote in the the Oct 7 NIKA2 meeting minutes, 9 mm radio axis offset offset with respect to the cryostat window center is really a lot. We should double check that. If this is confirmed, that means the "tirant" of the cold M7-M8 assembly are not at the optimal elongation. Since we can't open the cryostat to correct that (for now), the only other way then to avoid spillover on M1, is to put closer the optical/laser axis to the radio axis. To do that, one needs to move the M6 back screws accordingly.
- But first of all the radio alignment must be double checked. Doing it with a hot load in front of a very good, stable and low opacity sky is better than with a cold load on vertex closed 300K background. Sorry, this will take a couple of hours of good sky, but this is important.
- If you find out that optical axis has to be moved to match better the radio axis, and so if you do move M6 screws indeed, the consequence of course is that all the pointing parameters with change... as well as the geometry...

Sorry for the slightly scarring information/reflections...

---

13:00 - 19:00 LT, Mairi : Finished rep for 90-26. Switching to TP and performing repetition for project 092-26. 18:00 LT, switching to polarimetry. Switching to project 084-25, calib scan on 3C286 for IP characterisation. Setting off rep of AFGL2591 and handing over to Miren.

19:00 - 01:00 LT, Miren: Using 3C454 for IP characterisation calib_1scan. Opacity ~ 0.25.

Changed to nikas-26 for a few focus tests. When focus optimised for A1, A2 beam elongated: ~ 17.5" x 33".

22 LT: Back to project 084-25 for two more repetitions on AFGL2591. Opacity ~0.2 and stable (+/- 2 Jy/beam @ A2). Stability improving reaching +/- 0.5 Jy/beam @ A2.

00:35 LT: a skydip under project nikas-26 (20261009s259). Neptune is now available for IP calib_1scans. A fourth repetition on AFGL2591.

October 10th, 2026, Saturday

01:00 - 07:00, Rémi: stable conditions. Reasonable opacity ~0.2-0.3. A bit of reasonable wind.

• Continue project 084-25. Starting with Miren the repetition on one of the targets. Pointing and calib_1scan using Neptune.
• Switch to project nikas-26. Move to Uranus. Pointing, focus, calib_1scan and beam map (optimized for A1).
• Move to Neptune (closer and more similar elevation than 031-26).
• Project 031-26. neptune, pointing, focus, pointing, calib_1scan, nearby source, pointing, science script, mars (I would prefer keeping neptune, but the elevation is now too low), pointing, calib 1scan.
• Switch to project nikas-26. Move to Uranus. Pointing, focus, calib_1scan and beam map (optimized for A1), skydip, focus, calib_1scan

• Remarks:

  • Neptune does not appear on XEPHEM -> SB: to make it appear, click on the lightbulb to make the view brighter

  • We had to deal with a bug in the script of project 031-26 (@Stefano: you may want to check the README) -> SB: thank you for fixing this. I've added a note in reply in the README

  • Between 4am and 6am UT, there are no polarisation project available in good conditions. Those shown in the visibility plots are getting too low and cannot be properly calibrated (only mars may be used but maybe not ideal), plus available sources are going down quickly in elevation (hence polar systematics changing quickly). CK: Isn't this something for our database of all NIKA2Pol Uranus observations in prep. ? -> SB if Uranus and Neptune are up, please take many calib_1scan in polarimetry at different elevations (but well focused).

  • The two beam maps taken this night were done before and after the maximum elevation of Uranus (3h appart) so that hopefully it should be possible to check the rotation of the beam to better understand the issue at 2mm. And the focus value is almost the same (by chance) so maybe usefull for checking how polarisation systematics behave as well?

  • Project folders 015-17 (B-FUN: NGC1333 & NGC2924 scheduled), 046-26 (CloudG & CloudA), 083-26 and 100-26 are active in nikas-26 (to be checked from there if observers have a while; thanks so much for that). The pools-db has been updated accordingly.

07:00 - 13:00, Giorgos:Weather stayed stable, opacity around 0.3. Continued with the calibration from Uranus to ALFORI and then Mars.Did a skydip. Started project 090-26.

13:00 - 20:30, Mairi: Bracketing 090-26 IP observations with Mars and 3C286. Switching to TP. Tau increasing. Starting project 093-26, tau around 0.5. Onset of fog 15:30 LT, stopping to see if weather improves. Restarting at 17:00 LT, again with 093-26.

20:30 - 01:00, Ioannis: Continue with 093-26 observations while we wait for the opacity to drop and sky to stabilize better. Around midnight, sky stability looks good (+/- 1.5 Jy/beam) and opacity is relatively low, so I start with 084-25. IP scan on Neptune and another repetition of AFGL2591.

October 11th, 2026, Sunday

01:00 - 07:00, Rémi: opacity ~0.35. Very stable: starting shift with +/- 3 Jy/beam over a ~7mn scan at 45deg elevation, with slow variations. No wind at all.

• Continue with 084-25: finish the current repetition (5/6), pointing and calib 1scan on Neptune. Run another repetition and complete the current source (6/6). Pointing and calib 1scan on Neptune.
• Project nikas-26. Pointing and focus on Neptune.
• Project 031-26. Pointing and calib_1scan on Neptune. Pointing on NGC7538. Science scripts C and D. Pointing and calib_1scan on Neptune.
• Keep project 031-26. focus on Neptune. Pointing and calib_1scan on Neptune. Move to Mars, pointing, calib_1scan. Move to science (A and B). Pointing and calib_1scan on Mars. This way we can cross check polarization on Neptune and Mars that are at the same elevation when doing consecutive calib_1scan, with the same focus. Then we bracket the science scans with Mars, altought might not be ideal, but Neptune is getting too low.
• Project nikas-26. Pointing and focus on Uranus. Calib 1scan.

• Remarks:
  • About the spike in tau225 at 2:30am: the data are great and stable, the sky is fully clear, with no wind at all. So maybe some earwig (it's full of these everywhere) went messing up the taumeter??
  • It was extremely tempting to go to project 015-17: the stability is excellent, the source to be observed is at 80 deg (before and after max elevation to check rotation), very close to Uranus so that systematics can be very well controled. But the opacity is a bit higher than required (0.3 max vs 0.35)... And the decision had to be taken at the moment when the spike in tau happened, so the time to make sure it was fine...
  • Project folder 026-26 is now active in nikas-26. The pools-db has been updated accordingly.

07:00 - 13:00, Giorgos: opacity 0.35-0.5. Stable sky. Continued from Uranus to ALFORI and then mars to finish calibration and then started project 090-26.Finished project 090-26, switched to TP and started project 093-26.

13:00 - 19:00, Mairi: continuing with 093-26. Patchy cloud, observing in regions where the sky is clear. Opacity increasing, rising to 0.6-0.7, stop at 16:00LT. Remi takes over at 17:30LT to run some tests to make use of the poor weather conditions.

19:00 - 01:00, Ioannis: Bad weather, high opacity and unstable atmosphere throughout the shift. Remi did a test for project 094-26 at the beginning but other than that shift was lost due to adverse weather.

October 12th, 2026, Monday

01:00 - 07:00, Rémi: opacity ~ 0.7-0.8. Only EMIR observations this night.

07:00 - 13:00, Giorgos:

13:00 - 19:00, Mairi: Ioannis starts the shift by carrying out some focus tests. I take over at 14:30LT. Focussing on Mars at low elevation, Remi points out that the shape of the beam in array 1 looks similar to what we are seeing in array 2 with the cross like pattern when in focus (focus sequence s158-162). SB: it looks good and normal to me

For the remainder of the shift continue with 093-26.

19:00 - 01:00, Miren: Opacity around 0.65, trying several pointings on nikas-26 before continuing with 093-26 or moving to EMIR. Large pointing corrections.

20h30 LT: Opacity not decreasing, so we move to EMIR 3 mm backup projects. Starting with 053-26.

October 13th, 2026, Tuesday