Sat, 04/26/2014 - 07:37
Hello! I'm having trouble getting the error range of my T PYX observations down. They seem to run 0.3 to 0.5 mags. Last night, I averaged ten 60 second images through my ST-402 with BVIC and my 8 inch LX200 classic scope. The error range was still over 0.4 magnitudes. MPO Connections to control the run; AIP4Win to calibrate and stack, and Canopus, to extract the magnitudes and error ranges.
My error ranges for the targets in evening's run were typically under 0.1 magnitude.
What should I look for in order to determine the cause for the high error range? Thank you and best regards.
Mike
Hi Mike,
I need more information. Which filter are you using for your 10x60 stack? That comparison star were you using? What was your typical seeing, and how high in the sky was T Pyx? 0.3-0.4 normally means that the object was very, very faint, barely visible in your image (if you didn't know where to look, you wouldn't see it).
Arne
Hello! Thank you for your reply and consideration for help. I am puzzled by the error ranges. Mine are much higher than other magnitudes that I see in the light curve. Additionally, for other variables of 15 to 16th magnitude, my error range is typically under 0.1 magnitudes (often 0.05) and at 18th magnitude, often about 0.1.
I took 10 images through each filter through my ST-402 and 8 inch LX200 classic at 1x1 binning and -5 degrees: I = 20 sec, B = 60 sec, V = 60 sec . The final image magnitudes and errors were I = 18.834 (0.553), B = 15.888 (0.466) and V = 15.406 (0.133). The altitude was about 77 degrees and the airmass 1.024. I am located in Australia and the imags were obtain in the local morning of 4/24/2014.
The images were calibrated with a master flat composed of 10 raw flats taken through each filter at the beginning of the evening's run at the end of nautical twilight and master flat darks created seveal months ago with the median of 100 raw flat darks. The master darks were prepared a several months ago by taking the average of ten 25 minute raw darks and the master bias was prepared using the median of 100 bias frames.
The images were stacked and calibrated with AIP4Win using the advanced protocol. The magnitudes were found in Canopus. The values were reduced to a standard system from transformation coefficients found from Landolt standards about 9 months ago.
The seeing started off at 3 arc-minutes but the FWHM was measured at 5.7 in the images, not unusual with my setup with the slop in the focuser during the night with an automated run.
In Canopus, I had T PYX located at 09:04:41.50 and -32:22:47.5.
I used ensemble photometry. The check star was 000-BBQ-230.
Comps were 000-BBQ-238, 240, 216, 223, 241, 218, 222, 226, 220, 225, 232, 229, 235, 231, and 227 downloaded from AAVSO through Canopus in a 20 arc-minute field arond T PYX.
Thank you for your help in identifying the source of my error for T PYX. Best regards.
Mike
Hello! I've looked at my equipment, and I still get an error range of 1.1 to 1.3 magnitudes for T PYX in my I filter. Other variables about this magnitude have a much narrower error range for my system.
could one of the comps be off? The curious thing is that WW PYX is in the same field and using the same comps, I have good error range for that variable.
Thank you for insights. Best regards.
Mike
Mike,
I think that the problem with your I band meaurements of T Pyx is that only one of your comp stars (000-BBQ-238) has an I magnitude for it. The rest have only have B and V magnitudes. This situation is probably throwing off the software.
In quiescence, T Pyx has colors of B-V=0.1 and V-I=0.4 to 0.5. Using your V mag of 15.4, that would give B=15.5 and I=15.0.
By the way, when I examined the comp star magnitudes for the T Pyx field in the Variable Star Database, I did find a problem with one of the comp stars. The "15.5" comp has the wrong mags. From Arne's .dat file for T Pyx, I get the following positions and mags for this star:
9:04:41.49, -32:23:00.1
15.797V +/- 0.045, B-V: 0.709 +/- 0.036
So the position is correct but the magnitudes are not. This star is "15.8" instead of 15.5. I think the incorrect magnitudes are of T Pyx itself from the same .dat file. Someone should correct the Photometry Table for T Pyx.
Bob
Bob,
You should report this as a chart error to get it corrected.
Barbara,
I wasn't sure if CHET was working, so I mentioned the chart error on this forum instead. I did submit a report using CHET a few minutes ago and it seems to have been submitted O.K.
Bob
Hello! Thank you for your note.
I adjusted star 15.5 and tried different check stars. Only 11.5 and 11.3 have I data. Using either as a check gives an error range of 1.0 or more. Since these are 4 magnitudes brighter than T PYX, the paucity of stars with I and the brightness of the two that have I must be contributing to the problem.
Interetingly, my B values are on the order of 0.3 magnitudes. Could this be because the B-V of the comps is two to three times larger than the value of T PYX?
Thank you and best regards.
Mike
Hi Mike,
Send me off-line one of your I-band fits images that seem to have the 1-magnitude offset. I'll be happy to look at it and see if there is something obvious.
Arne
Hello! Well, I feel sheepish! I started to check my parameters and review my images. It turns out that my I filter image integration got changed from 60 seconds to 5 seconds without my being aware of it (or not paying attention!) and T Pyx barely showed up in the images, if at all. As a result, when I reviewed my images, I was visually checking the wrong star in the images. I was being lazy. Me bad!
I did a run last night wiht 20 stacked images of 60 second integrations each. The error magnitude of filter went from 1.3 to 0.3, with B and V about 0.25 mags. Still not close to what others are obtaining, but much better for I.
Arne, I appreciate your offer to review one of my images. Let me check everything again a few times to make sure that my technique has been corrected. Best regards.
Mike
hi Mike,
I'm glad you found the first problem with your Ic images. However, I think you still have a remaining problem.
T Pyx is currently about Ic=15, V=15.5, B=16.0. You indicate that you are taking a total of 20 minutes of integration per filter and getting errors around 0.25mag. With APASS, which is also an 8-inch telescope, I get 0.1mag errors at V=15.5 in 90 seconds total exposure.
Perhaps you can tell us how your error estimate is determined? That is, perhaps the photometric measure is ok, but the method of calculating the error is incorrect.
Arne
I noticed that someone recently entered some R and I mags from APASS into the VSP comp star list for T Pyx. There are also some R and I mags for the brighter comps from BSM. While the R and I mags from BSM look good, those from APASS appear to have some problems. Many of these mags have errors of 0.1 to 0.2 mag and, in some cases, over 1 magnitude. Some of the R and I mags for comps in the range of mag 10 to 11V may be saturated. As for the other fainter comps, I'm not sure what is going on. The V and B mags from APASS look fine when compared to BSM and Henden photometry. According to Seqplot, there are 26 measures for most of the APASS comps, which would suggest that the mags should be of good quality. Can someone at HQ look into this and figure out why the R and I mags for this field appear to have problems?
Thanks,
Bob
Hi Bob,
The current data release of APASS has been quality checked for the V-magnitude data, but not for the other photometric bands. When you see errors in the 0.2-1.3mag range as you do for T Pyx, then most likely some discrepant night of data for those paasbands has been included in the mean determination. This will be corrected in later releases of APASS. For this specific field, I'll find the errant night and remove it before creating DR8 (in a couple of weeks). Thanks for pointing this out!
Arne
Hello! I just reduced last night's run for T PYX. I am using ensemble photometry with Canopus/Photored.
I used the added the new I mag comp data and excluded those with an error range of greater than one. My error ranges are less than 0.1 - 0.08 for B and V and 0.06 for I. I am not sure why I am getting suh a different error range. Perhaps an internal processing error when the comps were downloaded originally?
In any case, I'll keep and eye on things and see how the mags and error range turn out in the coming weeks. Best regards.
Mike