Thu, 10/01/2020 - 21:13
AAVSO Alert Notice 721 announces an observing campaign on the bright, eclipsing, triple-star system b Per. Please see the Notice for details and observing instructions.
To stay informed about this campaign, particularly regarding any modifications to the schedule or other instructions from the astronomers, subscribe to this forum thread (option 1 under Subscribe below) - see feedback from the PI, comments and questions from observers, and notes from AAVSO HQ. Add a post yourself - join in the discussion about this interesting target!
Many thanks, and Good observing,
Elizabeth O. Waagen, AAVSO HQ
The AAVSO runs the AAVSOnet robotic telescope network, available for free for AAVSO members. A subset of that network are the Bright Star Monitors, four of which are in the northern hemisphere. The BSM systems are ideal to observe b Per, and are equipped with a variety of filters and diffraction gratings. The proposal process is simple. Go to:
and click on the proposal form link to submit a small informal proposal.
Welcome b Persei observers,
I anticipate that the mid eclipse of this month's passage of the C star of b Per behind the inner A-B pair will be next week from tonight. That means that we can expect the eclipse beginning 1-1/2 to 2 days before then. Now is the time to begin observing b Per on a nightly basis (weather permitting) to enable a good baseline out-of-eclipse determination. I'm pleased to see that several observers have contibuted data iln the past several days.
I've been observing it during clear nights for October and have produced the following phase plot ofl the out-of-eclipse ellipsoidal fluctuations. I haven't uploaded much of the data yet due to the needs to update my processing software. The updated results from my observatpry are on the phase plot link here.
I've also taken up Arne's invitation to apply for time on a recently upgraded Bright Star Monitor with approval for nightly (all night) coverage on either the Texas or New Mexico site between Oct 20 and Oct 30. I've requested a cadence time of hourly observations of 30 B and V images, which will allow stistics to average out scintillation variations that plague short exposures.
Good observing everyone.
Don - CDK
The phase plot has been updated with a better annotation. Link here.
All my data until now have been uploaded to AAVSO.
October in N. Carolina is the best time of year for long time-series.
oh well, it's not looking good, weather wise, in Germany at the moment :-(
I'll see what I can do but I'm not optimistic.
taking first dslr frames, iso 200 and 10 seconds... B per has around .7000 ADU of 14000 possible, 4,5 mag... it is my first, most luminous observed star so far...
Hi Bernard, Welcome to the b Persei observers. I'm pleased to see your first data. I notice that the data are in two groups. The second group fits the slope of the ellipsoidal variation at that phase. It is parallel and will need to be offset, but please use the same protocol regarding ref star (or strars) for all your dslr data. What concerns me is that your data appear to have a "large" offset (~0.09V) between the two groups. Did you have a pier-flip between those groups? Did you have to de-focus your image to prevent saturation at that rather large aperture (900mm?) Doing these bright stars is tricky! There was a recent discussion about people having "unexplained" offsets at pier-flip with highly de-focused images (often necessary to prevent saturation) especially when the de-focused images are not Gausian in star profile. When that is the case, one observer discovered that enlarging the photometry aperture wide enough to encompass the "tails" of the profile (that resembles a top-hat) has removed that offset. Is there a lot of vignetting in your system that isn't removed by vignetting? Or perhaps clouds??
Best of luck! These are loooong nights!
the 3 upper data points were at the corner of the first frames. The later points, who are better fitting...you've said, i had put them to the center of the FOV. (The first session had a soon end..) Mhhh.. at first i let them in, because other TG measurements had also some scatter...? But maybe i can delete them, because maybe the flatfield process had not catch the outer corners properly... ?
Also my first shots were at ISO200 and 10s. Just now there are gaps in the clouds...(there seems to be bad weather in central Europe everywhere... )-: ) and just now, for the data i'am collecting, i had to reduce to ISO100 and 5, than 3 and 2 seconds... to get a proper ADU under 14 000 ...for a save cutoff...(14 BIT = 16381 minus a Canon offset..) So there must have been high clouds, at my first data... otherwise ISO200 and 10s must have been also saturated...
(I remember, once tried out Polaris, but with a second, no proper compare stars were there. Luminous stars are difficult...)
Yes, I always take the same compare star 55 an d the check star 74 with the mentoined HIP 20370.
The Newton from the picture is a 8 inch type. A=200 F=900 mm.
CS, It looks like you have found some of the problems. If you ever get a chance to get more data before this event is over, remember that with short exposures you will have a lot more scatter due to the Earth's atmosphere scintillation (the "seeing"). You will need to take a number of images to eliminate the noise of short exposures. I think these notes are in the AAVSO DSLR manual and also in Bob Bucheim's article that is made available from the alert notice. Don't try to change too much protocol too much until after the eclipse. The better time to experiment is out-of-eclipse.
Hope your skies clear before the eclipse ends. My skies in southeast US are terrible this weekend.
Depending on the accuracy or inaccuracy of my prediction, there is a slight chance of a first contact of a partial ingress between JD 146.8 and 147.1 Unfortunately that is in the "Asian gap" and I know of no observers from that large block of the world.
For Europe and N. America we should see some activity tomorrow - around JD 147.5 (night of Oct. 24-25). Try to "shoot between the clouds". If you can set automatic un-tended acquisition, grab all the images then sort throu them and eliminate the clouds and outliers.
Good observing! Don (CDK)
I was able to collect images past 1am (local) and it looks like my data shows a 0.1 mag drop starting at
JD 145.82. stayed very clear last night (Az.) but tonight I have clouds.
I will try to collect images if it clears.
Just to say, that the sky was as bad here, since early october, that I could not even get a clear evening to get a decent pre-eclipse time-serie.
Possibly, could the sky be clear tonight, but the weather forecast is bad for the coming week too. Not sure about post-eclipse time-series.
Christophe, That's too bad. I'm also clouded out for the whole weekend in North Carolina after nearly 2 weeks of excellent weather. Keep watching for cloud break-ups!
I have measurements from some hours last night (with some interruptions by the clouds) and the forecast for tonight is uncertain, mainly because of the fog, but I will try to take measurements
now I hope it is clear all night (24-25. Oct) in Germany and also tomorrow. Because my balcony faces south, I cannot take pictures with tracking. So I use a normal tripod with Sony alpha 6300 camera and 85 mm lens. In one session I take always 23 pictures of 8 sec at f 4.5 (takes around 4 minutes). After dark and flat correction, the pictures are split in R, G and B channels. The 23 (fits) images in G are stacked with median function. I did the same thing (with another lens) at the last primary eclipse of b Per. If I look at my results (and the results of the check star magnitude), I’m now quite sure, that the resulting measured TG magnitude has an error of only +/- 0.01 mag. I think, it is better to stack several images (e.g. all images taken in five minutes) to one measuring point than to produce hundreds of results of single photos and report each of it. I now try to take every two hours such an image series all night long.
Hello Hans, Your technique is very similar to the technique I used back in 2013 when I did not have a tracking telescope. I used a DSLR on a stationary tripod. With multiple images and co-adding the results were very successful. The later addition of a tracking telescope and a CCD camera led to much less work on my part. I could sleep at night! The smaller sensitivity of DSLR cameras is the biggest advantage of DSLR's over CCD's for bright stars to avoid saturation and to require fairly long exposures to average out the scintillations.
Thank you for contributing! Keep observing!
Hans, Your technique of stacking multiple DSLR images from a non-tracking tripod has worked very well. You have captured the all-night decrease in b Per brightness as the third star gradually slipped behind the brighter of the inner A-B pair. (last night mJD 147) Your data are corroborated with at least two other observers. I'll eventually process all the data and adjust everyone's observations to a common base-line.
Hi, as Don had predicted, with iso100 + 2s, my TG has too big scatter now..
I'll do stacking with Muniwin.
PS: I do observe with the A=200 F=900 mm Newtonian. Is it better to obs. with the smaller A=102 F=500 mm Achromat-Refraktor? Maybe F=1000mm with an Barlow-Lens?
with A 102 and F 1000, at ISO 100, maybe you can expose your shots for 10-30s (rough estimation - you have to try). Be shure to avoid saturation! 2/3 saturation of the chip for a 4 mag star would be fine. If you take 10 shots (dark, flat corrected) and stack them (Median function), scatter should be much less.
I've submitted DSLR data to WebObs. Nothing exciting in this time interval, but the inner pair's shape-modulation is visible. Cloudy here on 10-24 and 10-25 (UT date), and forecast is for two more nights of clouds.
Bob B. (BHU)
Several observers observed an ingress into eclipse last (JD 147.2 to 147.75. Eastern Europe recorded full brightness at mjd:147.2 and western Europe recorded descending brightness throughout the night with ingress beginning at about mjd 147.35. An Arizona observer observed continued dim through the night. I expect b Persei to be dim again tonight (JD 2459148) as we pass the expected midpoint of the eclipse. Good observing for those who could see between clouds! Eastern US was cloudy.
Thank you for confirmation of my technique.
From evening 24th to morning 25th it was perfectly clear in Bavaria. I think, all measurements from 147.45 on show a first dip in magnitude of b Per!
I took photos every two hours all night long. My last two points are quite low (the last at 147.5722 is 4.7mag). At 147.69 I made a last series, but unfortunatly because of tiredness I aimed at the wrong side of Perseus, so b Per was not on the frame (Aaaargh!).
Up to now, I was clouded out, but some minutes ago in a cloud gap I could take some photos.
b Per is at 4.726 at 148.3553
The Muniwin stacking function was not working good enough. I've selected: use 5 frames, and at least two frames, to merge into one frame. I had the same scatter but with less data points. And 26 frames were rejected from the merge function?
Now i've calculated the mean of five data points, from my original measurements. This reduced the scatter (of JD 147.24 to 147.32). My results are darker than from VOL + SIV ? but I let it be, for that way for the moment...
Yesterday night, no photons... clouds were everywhere..
Hi Bernhard, If you cannot easily stack your images I can easily do "boxcar" averaging to lower the scatter when I process all the AAVSO data for the eclipse. I also need to offset all the measurements so the out-of-eclipse observations have fit the ellipsoidal curve. Most of the data are not transformed, so all observers with different telescopes, mounts, etc have diffferent zero points. We assume that the offsets for each observer remain the same during eclipse as out-of-eclipse. That is why procedures have to be identical from night to night as much as possible (ref star, optics, filters, telescope aperture, etc). It also helps to have multiple nights of out-of-eclipse data for anyone who observes eclipse data to examine consistency.
Good luck with observing! The eclipse isn't over yet.
ok, i've re-uploaded my original data (without means..) and let you do the analysis.
There is a small chance that I'll have a clear sky in a few hours, for a few hours....fingers crossed.
I'm on it. Looks somewhat uneclipsed on quick-look-photometry
I got a break in the clouds in N. Carolina starting about 0300 UT on Oct 27 (JD ~149.62). It looks like it is out-of-eclipse, but may dip down again or not depending on the actual mid eclipse midpoint which was mJD 149.8 +- 1 d.
Keep shooting as long as possible tonight.
I got several hours observing time with intermittent clouds, so I'll need to weed out the "bad" frames...will take some time. I hope to have the reduced data submitted tomorrow.
Hi, i got a gap in the clouds and collected ~30 minutes of unstacked data. Canon 200 D, ISO100, 3s and 4s exposures.
Hmmm.. for now i do not do stacking. If Muniwin cannot perform the task reliable.. Can i do stacking (in future times) with Deepskystacker eg. some frames to one, and process these with Muniwin?
PS: This time, I performed exposures every 3-4 seconds without pause, continously... i got not that big scatter this time.. Is it better to do:e.g. 4-6 exposures after another, and than pause..e.g.2 minutes, and than do exposures again? Until now i did one exposure, that 2-3 min pause and do the exposure again. Maybe this also lead to the scatter?
I would think that if you keep a steady cadence (exposure-pause-exposure-pause ...) , where the pause could also be zero seconds, would result in a more stable temperature of the sensor compared to something like N exposures - pause- N exposures ... you might be able to even see the sensor temperature in the EXIF data embedded in the camera images. Would it matter? That is a nice experiment to try :-) I doubt it makes any huge difference but OTOH I see no downside to a steady, constant exposure sequence with identical pauses before each exposure so why not do it that way.
Yesterday I was able to take measurements of B PER ... 3 shots of 15 images of 4 seconds of exposure stacked (adding). I hope they are helpful in characterizing this eclipse. The measurements are already uploaded to the server.
Soft Astroart 7., Refractor 66 / 400mm ED APO + Nickon D3200.
Davidjau, Could you please tell us your AAVSO observer code so I can find your data in the AAVSO light curve generator. Your techniwue of coadding about 15 rapid succession DSLR images taken with a small telescope should work very well.
Thanks, Don (CDK)
My AAVSO observer code is RDAE.
Hello persistent observers,
It looks like the eclipse is finished. I thank everyone for their persistence in hunting down breaks in the clouds during the event. As can be seen in the light curve generator we observed the ingress and the tail end of the egress and close to the center of the eclipse.
All of N. America and it sounds like most of Europe was clouded out most of the time. I might have jinxed the weather by my stating that "October is the best time for astronomy"... in a recent post. Sorry!
In the next few weeks I will be downloading the data from AAVSO and processing - doing any needed boxcar averaging to reduce scatter attributible to short exposures and to offset each observer's data to a common baseline - the out-of-eclipse light curve. That is why it is very important that we have out-of-eclipse data for each observer. If your data show eclipse and you don't have any out-of-eclipse data, please get a few hours' data over at least one night in out-of-eclipse data within the next several days. I should complete the processing in a couple of weeks. Keep listening...
As with last January's eclipse, this eclipse supports that my predicted date of mid-eclipse (based on my prior determination of the C star's period of 704.9 d ) the mid time of this eclipse came up to one day ahead of the predicted date which was Oct. 26.4 +- 1d UT (JD:2459148.4 +- 1d)
Again, Thanks to everyone for the observations!
at 2020/10/28 I was shooting through clouds, only at the end, there was a nice gap. The data is scattering all over the y axis...
I have deleted the 28.10. because of I had not the 5.5 mag comparison star at the FOV, and I had to take ohter comparison + check stars... labelled: 74 + 92, but the data is that much different from all the other, the honest thing would be to delete it.
Today 29.10.2020, it was clear sky, in which for the first time, my scatter seems reliable.
Don, please allow some time for making post eclipse images...we are having lots of rain and clouds here at the moment, it will be many days before I have clear skies again :-(
Thanks. Be sure to let me know when you get the post eclipse data posted. My first good night since before the eclipse started was last night. We've had a tropical storm Zeta drenching the s. e. USA These tropical storms must be responsible for the rainy European weather the past two weeks.
Indeed, currently ex-Epsilon is driving weather patterns around here :-( It's all connected.
ok , my post eclipse data is uploaded now. It's not exactly matching conditions earlier (e.g. Moon), but hopefully good enough together with shorter pre-eclipse data.
Thanks - looks good!
I have some out of eclipse TG data:
06.11.2020 Mean: 4.511
07.11.2020 Mean: 4.547
Again, many thanks to all the AAVSO observers who acquired observations during last month's b Peresei secondary eclipse campaign in spite of the wretched weather in both N. America and Europe. We did gather a number of observations during the eclipse which I have now processed and made available in the following link: (b Persei October 2020 eclipse - aavso observers). The processing involves finding an offset for each observers' dataset to fit the out-of-eclipse data to the ellipsoidal light curve of b Persei when out-of-eclipse. The OOE zero-point level was obtained from an average level for the data in the Jan 2020 campaign that were transformed. There are eclipse data this eclipse from two of the AAVSO Bright Star Monitors (BSM-TX adn BSM-NH2b) - compliments of Ken Menzies and Arne Henden.
I am also sharing a sketch of the anticipated model solution for the October 2020 eclipse showed with a hand-drawn light curve based on past experiences of the nature of the light curves during these eclipses. Here is the link:(b Per with sketch of the approximate light curve.) In a couple of weeks I plan to make a simplified model curve based on the assumed parametere such as stellar sizes, luminosities, and the kinematics of the motions of each of the three stars involved.