Exoplanet Transit - GPX-1 b (Per)

[7170]

Tonight was clear and timing was ideal to try and record the transit of exoplanet GPX-1 b (a transiting brown dwarf) planet. It transits every 1.744579 days and lasts 125.28 minutes. Below is the light curve result.

It is not perfect and looking the subs in AstroImageJ I can see there was of course some cloud passing by just after the center time, which I think explains the dip from the expected pattern. I am also measuring TG and not V which will also explain a variance from published numbers which are in V.

I also wonder if the published start/centre/end time (red lines) I used are out of date as the graph would perhaps suggest a slightly later start/centre/finish (the blue line). I think it is a little large to be a JD vs HJD difference.

It is a little noisy, so I may play with the data a bit more, but I am pretty sure I have captured the transit. It is certainly fun, though perhaps more of an exercise in maths than imaging, but I don't mind that.

Edited January 30, 2022 by 7170

[7170]

I've had another play with the data today - mainly recalibrating the frames and tweaked the running average. The resulting curve is below.

I've also indicated the time period where clouds were passing which I think resulted in the magnitude rise and fall again in that period when it should have been rising only. I'm a little less convinced my estimated center time is correct given the impact of the clouds around then. It is definitely a target I will have another go at, probably with the SCT next time though as there are actually plenty of nearby comparison stars. 

 

[DaveNicko]

Really interesting!

Is there anything you can do to improve the noise issue? Averaging the data from every 10 images perhaps (by stacking, darks, flats, biases)? Not sure what effect this has on photometry.

What equipment are you using (apologies if this is in your signature - I don't see signatures on mobile)?

I'm not a CCD/CMOS observer yet - I have just started my journey of variable star observations with binoculars. But interested to know if you can get more out of this data!

Dave

Edited February 2, 2022 by DaveNicko
Typo and added a bit more info

[7170]

Hi Dave ( @DaveNicko), yes it is a little noisy. The subs have darks/flats/bias and I have done some trend averaging in excel when making the graph first graph.

Below is the same data with short term trends removed and the theoretical light curve plotted. The starting point is raw data and not the averaging I did in excel. Despite looking noisy, the residuals from the light curve magnitude fit are actually pretty consistent, which validates my original graph and belief that I did capture the transit.

The subs were 30s each taken on an ED80 using an ASI224MC. which was guided. The noise comes from not having enough focal length for the target. At mag 12 it is not too dim but it is a small star, so even with defocus it doesn't really cover enough pixels on the sensor.  I also think there was some very thin cloud up there.

To give you an idea of what you can easily achieve, below is a small extract of some data using the same scope/camera from a light curve of exoplanet TOI 00624.01 b that I took on 2022-01-31 which I've been playing with (the data got cut short as some clouds rolled over and stopped things!). 

The average is mag 10.438, and all raw readings are +- 0.03 mag from the average, and the variance between max and min is mag 0.059802, so not bad. Doing a 10 point running average gets the variance down to mag 0.21, and a 20 point average gets it down to 0.01 mag. 

The depth of magnitude dip for TOI 00624.01 b is mag 0.0111 over a period of 205 minutes. Given each sub was 20s you could do a 30 point plus running average, or more, and also apply some trend averaging to remove short term trends. Based on this I would have hoped to have seen the dip in the data if the clouds had not stopped play.

It is worth pointing out this data is from just putting the raw subs straight into AstroImageJ, I've not applied any darks/flats or bias, or split out just the green channel, so I would expect the variance to get a bit smaller if the frames were calibrated first.

Are you submitting your magnitude estimates anywhere out of interest? Also if you have a DSLR with a zoom that gets between 200-300 mm you can make very nice variable star estimates and light curves for brighter targets, even without a tracking mount. 

Edited February 2, 2022 by 7170

[DaveNicko]

Hi James,

Thank you for your detailed explanation! My thought process was that averaging a number of frames would help with the noise, but now I realise that the moving average taken of each value performs the same function. It is obviously better to do it the way you have so as not to "hide" any results. It is very impressive that you have detected the drop in magnitude for such a faint star with relatively modest equipment! This bodes well for me when I start to look into variable star photometry with a camera.

I am submitting my observations to the BAA VSS database, I'm currently looking at binocular variables and submitting those with help from one of the BAA members.

I do have a camera (Canon 7d) and a lens (150-600mm) so in principle I could start using that (with my un-tracked tripod) with some of the brighter variables. I then will think about what direction I take, be it mount first to use with the camera/lens combination, then a small refractor or newtonian for use with the 7d (or Fuji X-T1), before investing in a CCD, or another route. I want to first have some experience with visual variable stars first - starting with binoculars and maybe also looking at a few telescope objects (to look at with my 200mm dob).

In the meantime I find posts like yours very interesting, so thank you! I love a bit of data analysis (and stats) so appreciate all this info!

Best wishes,

Dave

[7170]

3 hours ago, DaveNicko said:

I do have a camera (Canon 7d) and a lens (150-600mm) so in principle I could start using that (with my un-tracked tripod) with some of the brighter variables.

Hi Dave, you should be able to take some great readings with your 7d and lens.  This is a great read The AAVSO DSLR Observing Manual, and so is this DSLR PHOTOMETRY and should hopefully inspire you to give your DSLR a try even without a tracking mount. 

I think it is a good idea to get started with visual/binocular and try your DSLR before investing further in case you decide variables aren't your cup of tea in the long run. In terms of a scope I would personally recommend a Mak127 or better still a C8 for variables (and doubles too).

Also if you like light curves and taking measurements then plotting/analysing I would also recommend you keep an eye out for bright nova, there have been two in the last year which have been good for measuring over time (V1405 Cas being one of the most interesting).

[DaveNicko]

Thanks!

I am aware of the links you sent but thank you for forwarding them on 😃

I do intend to try some photometry with the DSLR, I'm interested in how far I can push the equipment I already have. This will also make me understand what I want to study in the future. It is possible that if I can use the 150-600 then that can be my "refractor" and I can use it firstly untracked, then add a mount with a view to again getting the most out of it (if I can work with any focus and zoom creep, and they don't become too frustrating). Maybe then I can dream of putting a c8 on the mount.

But that's a way off yet - time is my issue at the moment!

Dave