From: =?UTF-8?B?Sm9zw6kgU8OhbmNoZXotR2FsbGVnbw==?= Date: Sat, 25 Jun 2022 14:54:48 -0700 Subject: Re: [sloan-25mlog 6150] Re: [infrastructure 21382] Re: Re: 2.5m --0000000000005b6e8c05e24cbc71 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Hi Dan, Conor, This is informative. However, I'm not sure this definitively points to something coming from the FPS. If this were external light being reflected on some features of the FPS, would they not rotate with the FPS as well? But assuming that the light comes directly from the FPS, which seems reasonable, then yes, it would be useful to know the location of that light on the FPS. I think the easiest way to do that would be, if you see the light, just take a short exposure (5 seconds or so if that's enough to detect the light). As long as we know the position of the rotator, then we can determine the real position on wok coordinates. Even if they are a bit "arc-y", we can probably pinpoint the location quite well. As Conor says, it would be useful to see if these are close to some of the holes in the FPS, although the block of lights at the lower left of your image shows too many sources to be that. Initially I thought the two big blocks could be light being reflected on the fixtures for the RTDs on the wok; I think the locations are roughly right. The backboard LED should never be on after booting, but I agree it's worth thinking about what LEDs may be in the back of the FPS. We also have a boresight camera which had a blue LED, but I think that camera is physically powered off. I cannot think of anything else in the lower side of the FPS that has a source of light. There's a command that can be issued to manually turn on the backboard LED of a single robot (or all of them). I don't see how that could be issued by mistake, especially if sometimes is on and sometimes not. If you see this light again, and after taking some FVC exposures, you can do jaeger led off and take another FVC image to see if something changes. Now that I'm looking at that syntax, I guess I could see somebody issuing that command manually if you were trying to turn on the FBI ... Thanks, Jos=C3=A9 On Jun 25, 2022 at 14:32:27, Conor Sayres wrote: > I just had a thought on this=E2=80=A6are the LEDs on the robot electronic= s boards > off during the night? > > There are several unplugged holes in the Wok and I wonder if those are > letting through robot backboard LED light? > > On Jun 25, 2022, at 4:29 PM, Daniel Oravetz wrote= : > > > To follow-up on this email, we took FVC images regularly last night when > > we were on sky and noted the same pattern of light features in those FVC > > images that were noted in the 59752 night log (the night when the light > > feature was seen in the r1 BOSS images during those cloudy-night > > exposures). > > > To recap the last several nights: during the 59752 cloudy-night exposures > > we saw these light features in the FVC images and also saw the scattered > > light in the r1 images, then on 59753 and 59754 we did NOT see any light > > features in either the FVC images or the r1 images, and then last night w= e > > saw the light features in all the FVC images we took while we were > > exposing BOSS. > > > Throughout the night last night, the pattern of the light features in the > > FVC images stayed the same, but the absolute positions in the FVC images > > changed with rotator angle. Attached is a screenshot of a 900s FVC > > exposure that shows the features moving with the rotator while the > > telescope was tracking the field. > > > Dan > > > > Hi Jos=C3=A9, > > > > > > Unfortunately, we couldn't come up with anything different in last > night's > > > setup compared to the night before. In both cases the enclosure was on, > > > telescope at stow, doors closed, the IEB cover was off. > > > > > > We reproduced a couple of the sequences from the previous night, too -- > > > toggling the FBI LEDs on and off, configuring the positioners, etc. -- > but > > > still nothing. > > > > > > If we get an opportunity to open tonight we'll take sequences of FVC > > > images during BOSS exposures as you and Tom suggested. I like the idea = of > > > taking a higher number of shorter FVC exposures to help constrain a > > > narrower time range to investigate should the feature reappear. We were > > > able to see the FVC "blobs" very easily with 5-minute exposures, and I > > > believe even at 100s we were seeing them fairly easily night before las= t, > > > too. > > > > > > Dan > > > > > > > > > > > >>> > > >>> Scattered-light tests > > >>> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D > > >>> See the Log for exposure-by-exposure details. > > >>> > > >>> We did NOT see the light feature in any of the test exposures tonight= . > > >>> The > > >>> only feature we saw in 900s open-shutter BOSS exposures was very fain= t > > >>> red > > >>> light from the E-stops, but this is considerably fainter (~20x) than > > >>> the > > >>> light feature we've been trying to track down. And when we pulled the > > >>> E-stops later on, that feature disappeared altogether. > > >>> > > >>> We also did NOT see the reflection/light feature in any test FVC imag= es > > >>> tonight, so at least that appears to be correlated with the lack of t= he > > >>> light feature in the BOSS r1 test images tonight. > > >>> > > >>> > > >> This is quite interesting and more definitively seems to point to an > > >> outside source. I guess in both days (when you say the signal and > > >> yesterday) the telescope was parked at the same position and the > > >> enclosure > > >> closed? Can you think of any difference in the setup? Was the IEB lid > > >> installed both nights (I remember Ryan mentioning that with the new > > >> chiller > > >> working he was going to reinstall it, but not sure when and if that > > >> happened). > > >> > > >> Jos=C3=83=C2=A9 > > >> ____________________________________________________________ > > >> Logs of Nightly APO Activities > > >> > https://trac.sdss.org/tracmailman/browser/private/sloan-25mlog/index.html > > >> > > > > > > > > > ____________________________________________________________ > > > Common APO Hardware and Software > > > > https://trac.sdss.org/tracmailman/browser/private/infrastructure/index.ht= ml > > > > > PM.png>____________________________________________________________ > > Logs of Nightly APO Activities > > https://trac.sdss.org/tracmailman/browser/private/sloan-25mlog/index.html > > > --0000000000005b6e8c05e24cbc71 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Hi Dan, Conor,This is= informative. However, I'm not sure this definitively points to somethi= ng coming from the FPS. If this were external light being reflected on some= features of the FPS, would they not rotate with the FPS as well?But assuming that the light comes d= irectly from the FPS, which seems reasonable, then yes, it would be useful = to know the location of that light on the FPS. I think the easiest way to d= o that would be, if you see the light, just take a short exposure (5 second= s or so if that's enough to detect the light). As long as we know the p= osition of the rotator, then we can determine the real position on wok coor= dinates. Even if they are a bit "arc-y", we can probably pinpoint= the location quite well.= As Conor says, it would be useful to see if these are close to some of the = holes in the FPS, although the block of lights at the lower left of your im= age shows too many sources to be that. Initially I thought the two big bloc= ks could be light being reflected on the fixtures for the RTDs on the wok; = I think the locations are roughly right.The backboard LED should never be on after booting, but I ag= ree it's worth thinking about what LEDs may be in the back of the FPS. = We also have a boresight camera which had a blue LED, but I think that came= ra is physically powered off. I cannot think of anything else in the lower = side of the FPS that has a source of light.There's a command that can be issued to manually turn= on the backboard LED of a single robot (or all of them). I don't see h= ow that could be issued by mistake, especially if sometimes is on and somet= imes not. If you see this light again, and after taking some FVC exposures,= you can dojaeger led off= and take another FVC imag= e to see if something changes. Now that I'm looking at that syntax, I g= uess I could see somebody issuing that command manually if you were trying = to turn on the FBI ...Thanks,Jos=C3=A9 On Jun 25, 2022 at 14:32:27, Cono= r Sayres wrote= : =20 I just had a thought on this=E2=80=A6are the LEDs on the robot electron= ics boards off during the night?There are several unplugged holes i= n the Wok and I wonder if those are letting through robot backboard LED lig= ht? On Jun 25, 2022, at 4:29 PM, Daniel O= ravetz wrote: To follow-up on this email, we took FVC images re= gularly last night when we were = on sky and noted the same pattern of light features in those FVC images that were noted in the 59752 night = log (the night when the light fe= ature was seen in the r1 BOSS images during those cloudy-night exposures). To recap the last se= veral nights: during the 59752 cloudy-night exposures we saw these light features in the FVC images and als= o saw the scattered light in the= r1 images, then on 59753 and 59754 we did NOT see any light features in either the FVC images or the r1 im= ages, and then last night we saw= the light features in all the FVC images we took while we were exposing BOSS. Throughout the n= ight last night, the pattern of the light features in the FVC images stayed the same, but the absolute posi= tions in the FVC images changed = with rotator angle. Attached is a screenshot of a 900s FVC= exposure that shows the features moving with the= rotator while the telescope was= tracking the field. Dan > Hi Jos=C3=A9,> > Unfortunately, we couldn't come up with anything different = in last night's> setup co= mpared to the night before. In both cases the enclosure was on,> telescope at stow, doors closed, the IE= B cover was off.> > We reproduced a couple of the sequence= s from the previous night, too --> toggling the FBI LEDs on and off, configuring the positioners, etc. -= - but> still nothing.> > If we get an opportunity to open tonight we'll take sequ= ences of FVC> images during B= OSS exposures as you and Tom suggested. I like the idea of= > taking a higher number of shorter FVC exposu= res to help constrain a> narr= ower time range to investigate should the feature reappear. We were> able to see the FVC "blobs&quo= t; very easily with 5-minute exposures, and I> believe even at 100s we were seeing them fairly easily ni= ght before last,> too.> > Dan> > > >>> >>> Scattered-light tests>>> =3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D>>> See the Log for exposure-by-exposure details.>>> >>> We did NOT see the light feature in any of the t= est exposures tonight.>>&g= t; The>>> only feature = we saw in 900s open-shutter BOSS exposures was very faint>>> red>>> light from the E-stops, but this is considerably fai= nter (~20x) than>>> the= >>> light feature we= 9;ve been trying to track down. And when we pulled the>>> E-stops later on, that feature disappear= ed altogether.>>> >>> We also did NOT see the = reflection/light feature in any test FVC images>>> tonight, so at least that appears to be correla= ted with the lack of the>>= > light feature in the BOSS r1 test images tonight.>>> >>> >> This is= quite interesting and more definitively seems to point to an>> outside source. I guess in both days = (when you say the signal and>= > yesterday) the telescope was parked at the same position and the>> enclosure>> closed? Can you think of any difference in = the setup? Was the IEB lid>&g= t; installed both nights (I remember Ryan mentioning that with the new>> chiller>> working he was going to reinstall it, but no= t sure when and if that>> = happened).>> >> Jos=C3=83=C2=A9>> ____________________________________________= ________________>> Logs of= Nightly APO Activities >> = https://trac.sdss.org/tracmailman/browser/private/sloan-25mlog/i= ndex.html>> > > > ________________= ____________________________________________> Common APO Hardware and Software > https://trac.sdss.org/tracmailman/br= owser/private/infrastructure/index.html> __________________________________________= __________________ Logs of Night= ly APO Activities htt= ps://trac.sdss.org/tracmailman/browser/private/sloan-25mlog/index.html --0000000000005b6e8c05e24cbc71--