I have been measuring the times of minimum (ToM) of eclipsing binary stars in both B & V. I often find the V data is a bit more scattered than the B. An example I was processing today illustrates this very well (probably the most extreme I’ve seen).
I use Tycho Tracker to do the analysis and the star is DI Peg.
I’ve looked carefully at the comp star I’m using and it looks quite good. I’ve tried using a variety of comps and I always get this same difference so I assume it is real.
Any thoughts on why there is this difference between V & B?
It also seems the ToM in V is slightly different than in B.
Can you please share what comp stars you are using? Are you transforming your data? If yes, would it be possible to share with us your transform coefficients?
What is the capture order: BVVVB, BVBV, something else?
You may have done this but I would analyse everything, starting with examining the images from the two channels and making sure both V and B images look good, looking at the raw ADUs from the two channels, checking if there is any saturation of the images, checking the calculations of the magnitudes. Are the B and V filters OK? Is the filter wheel operating normally? Essentially, check every step in preprocessing and processing.
Thanks to all that commented. I have tracked the issue done to having “Fix hot pixels” checked in the calibration settings of Tycho Tracker. I’ve been working with Daniel, the developer, and it’s not quite clear why this is happening. He says this option should only be used when images a dithered, which I don’t do.
Although I have had this setting on for a while, I think it was only apparent in this set of images because the star was brighter that usual and there were a couple of images where a pixel in the target star was brighter that 60000 ADU.
I do need to go back through past sessions and see if it looks like “Fix hot pixels” might have had an effect in any of them.
The moral of the story is turn this OFF unless you are using the synthetic tracker, when the feature is invaluable.
Hard to say if this is a bug or not in Tycho. However it is a dangerous feature IMO.
Yes, I know and I normally would reduce it. I was trying to get Time of Minumum for the eclipsing pair and wasn’t directly interested in the readings near maximum.
