Fri, 12/20/2019 - 19:46
AAVSO Alert Notice 688 announces an observing campaign on the very bright, triple-star eclipsing system b Per beginning in early January. Please see the notice for details and observing instructions.
Many thanks, and Good observing,
Elizabeth O. Waagen, AAVSO HQ
The URL for this forum thread has been fixed. My apologies for the error!
Best wishes, and Good observing,
I am very pleased to see the good data on b Persei coming in to the AAVSO to help establish the good ellipsoidal baseline in anticipation for the upcoming primary eclipse. It is also good to see new observers to the campaign!
Let me emphasize that at this stage it is important to obtain several nights of runs of several hours a night to establish a good ellipsoidal light curve before and/or after the eclipse.
I also wish everyone good weather during the 4 days centered on the eclipse (UT Jan 17.5 and UT Jan 21.5 = N. American nights of Jan 16 to Jan 20). If we experience "partly cloudy" conditions on those nights, I encourage observing "between the clouds" to salvage any clear windows of observations.
I will soon post a predicted light curve for the primary eclipse as I had done for the last eclipse of November 2018.
Just a reminder that b Per, the target of this campaign (AN 688), is NOT beta Per (Algol). There has been some confusion, so I thought I would clarify.
Thanks, and Good observing,
I recently finished calculating the expected light curve for this month's primary eclipse of the close-orbiting A and B components of b Persei by its third star, the C component. We owe many thanks to AAVSO observers who contributed to the previous five highly time-resolved eclipses and to those observers who observed the first ever eclipse of this fascinating system in 2013. By extrapolating the ephemeri (for both the C star's orbit and the inner A-B close orbiting stars) I have fairly successfully predicted the light curve that we verified by observing the November 2018 secondary eclipse. I've now "stuck my neck out" and predicted the upcoming eclipse shown in the accompanying figure. The timings may be in error by as much as 0.1 day. The first dip is shown to begin at JD 2458866.5 (Jan 18.0 UT). That is "prime time" for European and N. American observers. This expected dip is a partial occultation of the fainter B star by the third star. The simulation then shows the ingress occultation of the brighter (and larger A star) by the third star that shows a much deeper dip in the light curve. The simulated light curve shows the results for the C star fully ingressed over the approximate center of the brighter, larger A star. The curve continues as the C star passes the CM of the AB pair and re-ingresses the A star on the other side of the A star's orbit. The large major dip is then followed by a moderate dip as the C star almost fully covers the B star (the C and B stars are assumed the same size and brightness in this model. Finally there is an expected grazing occultation at JD 2458869.5 (Jan.21.0 UT) which will be a challenge to detect.
A word of caution: These simulations are only a guess as to the solutions of the b Persei system. My model only has concentrated on the timings, the angle of intersection of the AB orbit with the sky, and the dates of mid-eclipse. The model has not experimented with the sizes and luminosities of the three stars involved so the depths of the occultations may be sizeably different. At each eclipse we learn more about the b Persei system. Many thanks to all of you for participating!
I look forward to another successful observing campaign. Clear skies, everyone.
telescope and CCD ready for V-photometry (and B if you want) in South Europe. This next week I will run some observations pre-eclipse. But forecast at this moment seems compromised, after some weeks of good weather. Hope this prediction will change to better...
Fran, Glad you're ready if the clouds break. I am doing B as well, and so are some others. B would be good also, but the primary mission is V. Thanks,
I am pleased to see a number of visual observers participating in the b Persei campaign. The quickness of making visual observations should help tremendously in the presence of frequent clouds. The maximum depth of the upcoming eclipse is expected to be on the order of half a magnitude, well within the sensitivity of the human eye - especially if each visual observer observe as much as possible throughout the nights.
Hi Observers, Our long-awaited primary eclipse of bPer is expected to begin tomorrow night (Jan 17-18 local times - about Jan 18.0 UT) with a "small" dip as the third star is expected to eclipse the faint B component. The brighter A star of the A-B pair is expected to be transited the next night. Shoot between the clouds....
We've been pretty clouded out in the southeast US this past week. I shot between the clouds last night and posted my results. Pretty poor, but consistent enough to discern that no eclipse was happening.
Good observing, everyone. Keep the data coming, were getting pretty good European and N. American coverage.
as soon as the night has begun, I started to do B and V photometry. The star is a bit dim... Maybe I'm observing the small dip. Today the sky is clear but for very few time: the forecast is terrible for the next seven days, so I'm afraid that only today I will be able to provide data.
It's raining cats and dogs this evening over south-east of France.
Tomorrow should be better, but the weather prospect is still uncertain, and the quality of the data - if any - won't be at the best, probably.
I got good data for last night (JD 865) but tonight (JD 866) it looks hopeless for the expected small dip that is expected. Take any clear sky - even if only for 10 minutes.
Good luck everyone!
Hopeless near my spot in Germany...rain even...
I found a potential eclipse, but on the night of 0104. I know that is too early. Thought I would post anyway to pass the word. Others may look back at their "ingress" more carefully as a result and get corroborating data. The data on AID is the raw data, 2 sec exposures in V, 20 inch, defocussed 3000 microns, Sierra Remote Obs. The attached plot is binned x10 to take out scintillation. The error bars are STOM for the Target and the Comp Star--should be conservative. No images were edited out. By editing out a few images, I can see that the error bars could be reduced. The night in question is the binned result of 50 images. The plot is for 5 nights. Other nights were clouded out. Your Thoughts? Questions welcome.
Gary, the time-axis on your figure you sent is confusing. The time-axis title ("20191230-20200115") looks like calendar days. The numbers along the time axis are 0 .. 45. These numbers don't make sense with your date of suggested eclipse of "0104" which I interpret with Jan 04. Can you please explain? - Thanks.
Does the "SDOM M + C Quad" your notation for Stdard Deviation of Magnitude = the std dev of object and comparison added in quadrature?
By any chance, were these were last night's data (night of Jan 17-18)?
Sorry for the confusion. The dip in question is 20200104 (YYYYMMDD) as I said in the first line of my post--0104. The calender dates of the data in the plot are for calendar dates are 20191230 thru 20200115. The actual x axis is for "JD - 2458848".
Yes the "SDOM M + C Quad" your notation for Stdard Deviation of Magnitude = the std dev of object and comparison added in quadrature? This is the output of the standard AAVSO Tool for boxing obserations. My raw observations are for 2 second exposures, on a 20 inch, so to beat down any scintillation effects, I have boxed them x10 for this plot. This is probably the most representative error bars for these observations. Most of my observations in the AID are not Boxed, so that they can be processed uniformly to the requirements of the researcher. The Poisson Errors from the raw images are typically .0001 to .0003---yes that is three zeros--not a typo. Images were defocussed 3000 microns--ie about 3 mm. Focal Length of telescope is 3450mm. So a defocus of about 0.1% of focal length. FWHM were about 20 arc seconds.
Gary, Thanks for the clarification of the x-axis on your graph. However I'm still confused. The numbers at the top (0 ... 45) cannot be days after JD2458848 because that would go into Feb. On the plot is each point (with error bars) the binned average of 10 - 2sec observations on the night of Jan 4-5? That would mean each dot is 200 sec, there are 40 dots meaning 8000 sec span or 2.2 hrs. Is this correct? Are these points on the LCG for Jan 4? What you have plotted for JD8852 (Jan 4-5) averages about the same as your plot in the forum for those 2 hours. The dip is not evident in the LCG even when zoomed-in to show only your obs that night.
I see on the LCG for Last night (Jan 18-19) that you grabbed the star very close to the minimum light in last night's big dip. Congratulations! I was clouded out both Friday and last night, but tonight looks good - but COLD!
Thanks for your work/
That is an interesting setup! What is the max ADU count you get in a pixel of b Per in this configuration? Back-of-the-envelope, one needs to manage somewhere between 10 and 20 million electons every 2 sec or so, that must be pretty challenging in terms of finding the right amount of defocus.
here are the results of the last night. The eclipse began as scheduled, but clouds appeared at 1:30 UT. Well, that's not very bad because the star was already quite low. These points are 120" exposures but I tried also to promediate every 3 points in order to reduce dispersion. I have also photometry in Johnson V (similar results).
Don, in order to upload the photometry, please tell me if you want the single points or you prefer the 3 points average.
Fran's 4 hour light curve in B clearly shows the ingress to the eclipse that we attribute to the C star beginning to transit across the orbiting B star. The time of the beginning of the dip in brightness of Fran's light curve is JD 2458866.48. That is about 0.08 d (~ 2 hr) before the calculated date in the predicted curve I recently posted. Congratulations to Fran for being the first to report the onset of the predicted transit!
Further support for detection of ingress is the fact that at the time of the beginning of the drop in brightness corresponds to close to the minimum light of the ellipsoidal light curve where the normal light curve would begin to increase in brightness.
Congratulations on these data just before impending clouds!
I am in the process of combining all observers' data from last week's eclipse - a slow process. The workflow consists of finding the offset from my current model of the ellipsoidal curve. I am standardizing the offset based on an average out of eclipse of the residuals from the two observers who transformed high quality photo electric data : Gerald Persha and Javier DeElias. When I adjust Fran's data and Richard Biernikowicz' data for the same night to fit their out-of-eclipse data, I see a surprise extra eclipse dip on day 866 that I hadn't predicted. I'm attaching a preliminary graph of the aligned data with the out-of-eclipse curve superimposed. The extra unexpected grazing eclipse is seen by several observers (more to follow which haven't been compiled yet) at HJD 1866.2 on the graph. It turns out that the "kink" that Fran reported earlier was the egress out of the extra eclipse, not the early ingress of the expected first dip. The expected first dip was observed by Gerald at HJD 1866.6, closer to the expected time of the first dip.
Never have I been "overjoyed" at being wrong in my predictions! These observations give us an opportunity to learn more about the b Persei system! Thanks again to all observers for the diligent and successful observations.
I hope all observers have looked at the AAVSO LCG for this eclipse. Very good coverage from Europe and N. America! I will keep the group updated at regular intervals in this process. Stay tuned.
this night the sky is cloudy. Before this I captured some images: the star is faint (mag. 4.8) and fading (19h UT), so we are going to a second minimum. Damm weather! Now the time has come for the US people. Good luck!
I'm on the target since 17h30 UT. I begun acquiring data only one hour after sunset in a still blue sky.
Now the night is cristal clear. Unexpected, but useful.
The star, that was easily seen to the naked eye after sunset, is now somewhat difficult to see in direct vision, maybe around mag 5.0 (I'm not a specialist of visual estimation).
Well, I would say your timing is pretty good, Don. Well done !
b Per still high up in sky and there are some breaks in the clouds coming... I might be able to get a few datapoints here... will reduce them tomorrow, I guess there is no urgency now that the ingress was confirmed.
My last word for tonight, before going to bed .... This is 1h00 UT, and the star seems to be back to the naked eye. Some clouds coming.
Looking at the light curve generator and knowing where other observers match the ellipsoidal light curve, it looks like Gerald Persha (pep observations) last night (Jan 17-18) recorded the minimum of the first mini-dip in the expected light curve at about 4.651 V. The minimim ellipsoidal LC on that date is 4.625 V. We now have 2 observers who have recorded the first contact. It looks like my prediction based on my crude model predicted the first contact LC to be deeper than the observations.
I'll try to catch up on data reduction on the AAVSO data tomorrow afternoon, concentrating on the observers that have covered within the last couple of days.
Keep it coming! The deepest eclipse is expected tonight (Jan 18-19 UT) and re-ingress of the third star back into the B star around JD 868.0 for any Pacific observers
It seems the expected large dip was to be seen last night. The deepest point, maybe, was recorded somewhere near 0h15 / 0h30 UT (867.51079 to 867.52120), as seen from my station, near 4.92, in my system.
The star was still 42deg above the horizon when I stopped recording because of high level clouds coming more and more. So, there is some dispersion in the data after 0h00 UT.
Funny to see that my visual impression were partially false. Humm, physiology, physiology !
More tonight, if the sky cooperates.
Please, see the lightcurve below.
Last night, the clouds left a gap of about 2 hours in Madrid and the data I measured reached a value of 4.95 at 23h UT. Tonight the forecasts are not very good, but let's see what can be done.
I was fortunate to get the dip also. Raw data was 0.2 sec exposures with a 20 inch--1223 images over approxiamately a 3 hour period---021530 UT to 050642 UT. I binned the data by 50x in the AID. SDOM of the Target plus the Comp of .003 mags. Supports data by chrismlt and javierdeelias.
Clear sky, but windy this evening.
Analysis of a first set of images taken between 17h30 to 19h50 UT, 19 jan, is showing that the star is regularly regaining brightness, from 4.89 mag to 4.76.
Very interesting eclipse, indeed.
Now the sky is becoming very cloudy, so things could rapidly go to a stop.
Congratulations to the several observers who observed the descending into the bottom of the big dip that we expected last night for b Per. Unfortunately I was clouded out both Friday night and last night, but things look good tonight for ascending out of the expected double dip of the big dip. Looks like we have another "Big Dipper" in the sky!!
Keep on looking! The anticipated LC egress looks just as interesting as the ingress LC. We expect a more gradual rise as the C star slowly uncovers the bright A star but will start covering the B before complete emergence. On Monday night Europeans may try to see the grazing transit of C across an edge of the bright A star.
Good conditions in Northern Germany right now, but I see a big fat cloud system appraching fast on sat images :-( . Not sure how many hours I can cover this night. Reducing last night's battle with the clouds was a nightmare and didn't result in very good photometry either.
I'm taking a quick sneak preview look at images while data is being collected, and at around 21:30 UT, b Per was already back at almost ca 4.7 , quite as predicted IIRC. Too bad, I won't be able to observe much longer, I'll miss the next predicted dip so if you haven't observed the eclipse yet and you are in the right place ==> join now , this is fun:-)
The data that have come in from last night's egress look very good and informative! A nice long egress run by ATE (Teofilo Arranz of Spain) and MCHB (Christophe Marlot of France) followed by CDK (me, US) and PGD (Gerald Persha, US).
Of special astrophysical significance from observing this eclipse is that Gerald obtained good PEP data of both conditions of near max eclipse of the bright blue star A as well as the out-of-eclipse after egress. I've attached Gerald's part of the AAVSO LCG. Notice the color index (B-V) = +0.06 at max eclipse than after egress where (B-V = .042). At maximum eclipse of the bluer A star (assumed) the color index becomes less blue. B-V = +0.06 is less blue than B-V=.04. (Astronomers love the backwards color and magnitude scales due to long historical precedents....) If anyone else (myself included) have obtained B observations as welll as B observations, it will be a good project to include those as well as the V observations for further analysis.
Keep shooting for b Persei one tonight if you have clear skies - it may be possible to detect the grazing eclipse of the starB by starC.
Also if you haven't uploaded any out of eclipse data please do so. I'll be compiling the uploaded soon with a common "baseline"
Many thanks for the many hours of observations ands processing that all of you have done!
I took some data this night but will reduce it only by Wednesday, latest.
We are still supposed to submit out-of-eclipse data in the following nights if possible, right?
Yes, please submit out-of-eclipse data - preferably both before and after the eclipse for a run of at least a couple of hours. If you have B data include those as well.
I will upload B and R (Tri-B + Tri-R) data in the late afternoon.
No observations yesterday because of clouds. Let's see what will happens tonight, but I not really optimistic about the weather.
Post eclipse observation assap.
I've posted my DSLR observations from January 19th night. That's the only one I've got due to bad weather.
Observations outside the eclipse will come soon, when the weather permits.
So far I see no evidence of the expected grazing eclipse predicted by my crude model for the night of Jan 20-21. However, my model has not trimmed the star sizes and luminosities. These values were assumed from educated guesses solely for timing and general pattern.
The first thing I plan to do with the data is to display everyone's data to one common baseline. The various observing systems all show a slight variation of the baseline. The "baseline" is the ellipsoidal model at out-of-eclipse. That is why we need everyone's out-of-eclipse data - preferably for at least a couple of hours both before and after the eclipse. I will subtract the model ellipsoidal light curve, which is well established, modified by the standard levels determined with the help of a couple of peoples' data sets that are transformed. I promise to keep this forum posted during this process.
I appreciate everyone's dedication to contributing the immense volume of data we have posted on the AAVSO database. Anyone wish to count (look-up) the observations the week surrounding the eclipse????? Let the computers do the counting...
I still have several nights of UBVRI photometry to post from BSM_NH2 and BSM_Hamren. It may be a week before I can get the analysis done. None of the observations were during the main part of the eclipse; only caught the shoulders. Dang winter weather! One nice thing is that it appears U-band photometry is possible with the CMOS cameras.
I'm pleased to see the BSM instruments participating in the b Per campaign. The shoulders of egress and ingress of eclipses to me are just as important as I'm sure you know.
Check out the VPHOT Processing Queue
b_per_bsm_nh2_b_200120.1805_hqa.fts is holding up the queue for many hours.
I changed my sequence from 2 sec/defocussed 3000 microns to 0.2 second expsoures/defocussed 1000 microns on 2020-01-18 for the b Per campaign. So the ingress observations will have one offset, and the eclipse night and the egress observations should be adjusted separately.
Gary, Thanks for the update about your processing protocol which may change the offsets. I will ask a favor of you to re-check on me if I considered the two possible offsets when I post the results to this group.
I just uploaded all data collected during the eclipse, in B and V filters.
Please, I should ignore the V-data collected in JD 8865.5, because there is a systhematic error of 0.2 mag. I checked the observations, but I didn't found the source of this desviation. Sorry. Fortunately, this error was outside of the eclipse.
I couldn't observe more: Gloria Mediterranean storm has provided very bad weather these days and serious inundations.
I'm now in the process of extracting Tricolor-B photometry from my DSLR images. I'm just a bit sceptical whether this is worthwhile, the color-change effect is so much more subtle than teh overall eclipse and the observations per night span a huge range of airmass...hmmm....
I uploaded Tb data.
Please note that some observations began less than one hour after sunset under a blue sky (I mean not a perfectly dark sky), that could possibly affect the blue data.
Derived B-V (TriB-TriV) can be seen below.
(Not time correlated ; B-V not to scale, typically 0.13 to 0.15)
I am very pleased with the high quality of much of the recent eclipse coverage by many AAVSO observers. About 25 observers reported data for a total of about 9000 observations - some observers contributing other than the V or TG bands.
My data reduction consists of the following:
1) examining each observer's data set for scatter and doing a "boxcar" average of 4 to 1 or 16 to 1. When observers have small integration times and many, many observations there is a larger scatter in the data supposedly caused by atmospheric scintillations.
2) compare the smoothed data with the "standard" ellipsoidal curve which was set to minimize residuals from the two sets of data that were transformed to V. Each data set was set to compare the out-of-eclipse residuals from the 'standard" ellipsoidal curve.
The result of combining the data processed so far is shown in an expanded graph of 3 days spanning the eclipse. Not all the posted AAVSO data have been processed yet which is attached to this note. Ignoring the west Pacific and Asian gaps, the data look very good and have some surprises compared with the eclipse predicted from the previous "fits" in an earlier post.
In the deep eclipse, the data appear to show that the second dip is significantly deeper than the first deep dip of which we observed the bottom of. Both parts of the major eclipse fit the idea that the third star passes the CM of the AB pair between these two parts of the deep eclipse and re-eclipses the brighter of the AB stars on the second half of the AB orbit. This will be more evident when I share the animated simulations - yet to come.
The first expected minor dip predicted at d. 1866.5 we have substantial evidence before the Pacific-Asian "blackout".
Finally there is ample evidence of a grazing eclipse at day 1866. I presume this occurs when the third star has an earlier chance to contact the larger, bighter A star a cycle of the AB pair before the CM passage.
More to come,,, Stay tuned.
Many, Many thanks again for the diligence in observing. A number of new observers have contributed who haven't observed b Persei before. Welcome!
Fantatstic! Thanks for sharing early results. How better to demonstrate that this kind of event calls for team work. I wish we had more team members in Asia, tho.