Return to the Cloudy Nights Telescope Reviews home pageAstronomics discounts for Cloudy Nights members
· Get a Cloudy Nights T-Shirt · Submit a Review / Article

Click here if you are having trouble logging into the forums

Privacy Policy | Please read our Terms of Service | Signup and Troubleshooting FAQ | Problems? PM a Red or a Green Gu… uh, User

Observing >> Variable Star Observing and Radio Astronomy

Pages: 1
sound chaser
member
*****

Reged: 01/13/13

Loc: France
ED80 for Photometry?
      #5898081 - 06/02/13 03:36 PM

Hi all,
I'm putting together a set up for variable star photometry, for mainly Mira's and SR stars.
I'd like a wide a field as possible so I've got an Orion ED80 with a SBIG ST7 (nabg) and I'll be using a 'V' photometric filter.
I've been reliably informed that in this set up I will be suffering with being under-sampled as I will be ~3.0"/pixel.
If i were to increase the magnification by adding a barlow (an apo 1.5x) this increases the scope to F11.25 and brings it to 2.0"/pixel, which is better, but right on the limit but with good seeing I should get reliable data.
My question is, will this extra piece of glass hamper the photometry?
Any advice will be greatly received.

Doug.


Post Extras: Print Post   Remind Me!   Notify Moderator  
jgraham
Postmaster
*****

Reged: 12/02/04

Loc: Miami Valley Astronomical Soci...
Re: ED80 for Photometry? new [Re: sound chaser]
      #5898320 - 06/02/13 06:32 PM

You should be okay, but I would want to avoid under-sampling. Purposely setting the focus a tad soft might help. Having a field of view that is too wide could also cause a challenge clearly resolving the variable and reference stars from other nearby stars. I generally shoot for a field size that is similar to an AAVSO E chart.

Have fun!


Post Extras: Print Post   Remind Me!   Notify Moderator  
NJScope
sage
*****

Reged: 03/08/04

Loc: NJ
Re: ED80 for Photometry? new [Re: sound chaser]
      #5898597 - 06/02/13 10:12 PM

I think the relatively slow optics (f/11.5) and small aperture will limit how deep you will be able to image without fairly long exposures. You might need a guide scope unless your mount is up to the task.

Post Extras: Print Post   Remind Me!   Notify Moderator  
sound chaser
member
*****

Reged: 01/13/13

Loc: France
Re: ED80 for Photometry? new [Re: NJScope]
      #5899664 - 06/03/13 01:13 PM

Cheer's guy's.
Great advice.
I've decided to ditch the idea of struggling with a small scope. I'm going to use my 6 inch refractor instead.
I'll be getting 1.55"/pixel and ~19.7"x 13" fov, might have to upgrade to a better mount though...

Thanks John, Kevin.
Doug.


Post Extras: Print Post   Remind Me!   Notify Moderator  
btieman
scholastic sledgehammer


Reged: 07/24/08

Re: ED80 for Photometry? new [Re: sound chaser]
      #5900384 - 06/03/13 07:11 PM

I fail to see why image scale matters. You're counting photons not making pretty pictures. Photometry is still sometimes done with scintillation detectors = one big fat pixel.

Case in point?

http://www.superwasp.org/technical.htm

SuperWasp uses Cannon lenses that are 200mm at f/1.8 for a nearly 8 square degree field of view with a plate scale of 13+"/pixel. SuperWasp performs photometry to mmag precision and is responsible for a number of exoplanet discoveries.

If you're interested in variable stars, you're more interested in accuracy than precision but star size still doesn't matter much provided:

1) You can resolve the individual stars from nearby contaminants (not strictly required for exoplanets, by the way!)

2) You can collect enough photons for the precision you require (taking into account the limits imposed by scintillation, read noise, etc...)

So, for example, if the pixel scale is so large that light from multiple stars falls into the same pixels, your accuracy will suffer since you're counting light from more than one source. Similarly, if you can only count 10 photons per exposure, you aren't likely to have the precision you need to see a change in photon rate unless the change is large.

FWIW: I've attempted one or two exoplanet transits with an ED80 riding piggyback on my CPC1100 which was also measuring the transits for comparison (I'm working up to simultaneous photometry/spectroscopy data collection, but that's another story!) The resulting precision from the ED80 was in the neighborhood of .05 mag...not good enough for the ~0.01 dip from the exoplanet (as measured through the CPC1100) but certainly good enough for many, many variables.


Post Extras: Print Post   Remind Me!   Notify Moderator  
gavinm
Carpal Tunnel


Reged: 08/26/05

Loc: Auckland New Zealand
Re: ED80 for Photometry? new [Re: btieman]
      #5900615 - 06/03/13 09:32 PM

AAVSO use small refractors so I can't see why not. We've seen exoplanet transits done with camera lenses a whole lot smaller. I guess it depends on what your focus will be.

Post Extras: Print Post   Remind Me!   Notify Moderator  
jgraham
Postmaster
*****

Reged: 12/02/04

Loc: Miami Valley Astronomical Soci...
Re: ED80 for Photometry? new [Re: gavinm]
      #5900642 - 06/03/13 09:49 PM

The problem with significant undersampling is that the tiny star images may not fully illuminate a pixel and could even fall between pixels. I observe this first hand with my all-sky camera. I once considered using it to watch bright varuabkes like Algol. But nope, the star brightness vary greatly as the flow from one pixel to the next or flow between pixels. The fix? Combine enough images to blur the stars out a bit. It is not nearly good enough for photometry, but it lets me get fairly good images from a simpe camera.

Post Extras: Print Post   Remind Me!   Notify Moderator  
btieman
scholastic sledgehammer


Reged: 07/24/08

Re: ED80 for Photometry? new [Re: jgraham]
      #5900808 - 06/03/13 11:24 PM

Quote:

The problem with significant undersampling is that the tiny star images may not fully illuminate a pixel and could even fall between pixels. I observe this first hand with my all-sky camera. I once considered using it to watch bright varuabkes like Algol. But nope, the star brightness vary greatly as the flow from one pixel to the next or flow between pixels. The fix? Combine enough images to blur the stars out a bit. It is not nearly good enough for photometry, but it lets me get fairly good images from a simpe camera.




John, is your all-sky cam CCD or CMOS? I'm guessing CMOS since not many are CCD and those that are are expensive CMOS can be problematic when under sampled since it's not 100% fill factor. Some fraction of a pixel is not light sensitive because that's where the readout electronics are. Also, CMOS has multiple amplifiers that can different gains between pixels causing different pixels to behave differently. That may be the effect you are seeing.

CCDs aren't the same. The more typical problem is charge bleeding from one pixel into another when approaching saturation...thus the development of anti-blooming electronics.

With a CCD, the problem from under sampling has more to do with well depth and the number of photons you can detect. The well depth of Si is related to the volume. Since thickness is fixed, it's related to area as well. Small pixels hold less charge due to less area/volume. Bigger pixels hold more. But the charge per unit area is constant. Assuming the CCD has an anti-blooming gate to prevent charge from spilling into neighboring pixels, if your star spot is 5um but your pixels are 10 um (severely under sampled), the photons can fill up the entire 10um pixel even though the star is smaller. If the pixels are 5um as well, the charge is trapped in a smaller area/volume thanks to the gate and you're limited to less photons before saturation.

Photometry is counting photons. Spatial resolution doesn't matter as long as the detector can accurately count photons. Granted, we may adjust spatial resolution for better statistics or what-have-ya, but the goal is to sample photons so looking at things like plate scale and spatial sampling are really looking at the problem the wrong way.


Post Extras: Print Post   Remind Me!   Notify Moderator  
jgraham
Postmaster
*****

Reged: 12/02/04

Loc: Miami Valley Astronomical Soci...
Re: ED80 for Photometry? new [Re: btieman]
      #5900837 - 06/03/13 11:41 PM

Sorry, but it is based on a small CCD. I built it from a spare DSI Pro, which oddly enough I used to use for photometry.


Edited by jgraham (06/03/13 11:42 PM)


Post Extras: Print Post   Remind Me!   Notify Moderator  
sound chaser
member
*****

Reged: 01/13/13

Loc: France
Re: ED80 for Photometry? new [Re: jgraham]
      #5901314 - 06/04/13 09:00 AM

Well, thanks for your replies. It's certainly food for thought.
Probably a dumb question, I was wondering if stacking 3 or 4 images would improve sampling or impede it? (using the ED80)
The stars I want to image are miras and sr's, if I can get to about ~0.03 that could be accurate enough I think.
Visually my estimates of red stars are never better than 0.1 and probably closer to 0.2 or worse.
The reason I was intent on using the ED80 as opposed to the 6 inch is for the greater field of view to encompass all the required comps.

Doug.


Post Extras: Print Post   Remind Me!   Notify Moderator  
jgraham
Postmaster
*****

Reged: 12/02/04

Loc: Miami Valley Astronomical Soci...
Re: ED80 for Photometry? new [Re: sound chaser]
      #5901704 - 06/04/13 12:36 PM

Couldn't hurt. I usually average 16-32 images for photometry. As for the ED80, if you've got it I'd go ahead and give it a try, what could possibly go wrong?

Post Extras: Print Post   Remind Me!   Notify Moderator  
Andean Slave
member


Reged: 01/08/11

Re: ED80 for Photometry? new [Re: jgraham]
      #5902218 - 06/04/13 05:19 PM

If undersampling for photometry, even with 100% fill factor, there're sensitivity variations across the pixel, and worse, they are wavelength dependent. 'We've' (lol) seen WFPC2 errors on single shots >20mmag as an e.g.
So for precision just defocus until get to a sweetspot. Get two nice stars on chip and see how the SDs vary with changing PSF radius


Post Extras: Print Post   Remind Me!   Notify Moderator  
btieman
scholastic sledgehammer


Reged: 07/24/08

Re: ED80 for Photometry? new [Re: Andean Slave]
      #5904759 - 06/05/13 09:33 PM

Quote:

'We've' (lol) seen WFPC2 errors on single shots >20mmag as an e.g.
So for precision just defocus until get to a sweetspot.




Interesting! Do you know what the nature of the errors were? And why they were wavelength dependent?

I would expect anything in the charge shifting to look like readout noise and not be dependent on where the photons strike or wavelength. Likewise, I can't see where variations in the silicon would cause variations base on photon location or wavelength...the charge shifts across the chip for readout. I can see the variations arising from variations in the phosphor coatings or cosmic ray hits, etc...but maybe it's from something else?

I'm curious to know how universal this effect may be across cameras. I don't think I can measure it on my current equipment. That said, because of this conversation and given Surplus Shed's current sale on these:

http://www.surplusshed.com/pages/item/l13001s.html?utm_source=Surplus+Shed+Of...

I've gone ahead and bought one of these 155mm f/1.25 triplets with the intention of trying to make an exoplanet hunting camera Might be challenging/undoable, but for $75 and some of my time, worth the risk, me thinks!


Post Extras: Print Post   Remind Me!   Notify Moderator  
Andean Slave
member


Reged: 01/08/11

Re: ED80 for Photometry? new [Re: btieman]
      #5905079 - 06/06/13 12:25 AM

Quote:

Quote:

'We've' (lol) seen WFPC2 errors on single shots >20mmag as an e.g.
So for precision just defocus until get to a sweetspot.




I would expect anything in the charge shifting to look like readout noise and not be dependent on where the photons strike or wavelength. Likewise, I can't see where variations in the silicon would cause variations base on photon location or wavelength...the charge shifts across the chip for readout. I can see the variations arising from variations in the phosphor coatings or cosmic ray hits, etc...but maybe it's from something else?

I'm curious to know how universal this effect may be across cameras. I don't think I can measure it on my current equipment.




Haha I really like that website

The variations are entirely optical they all seem to say, and the wavelength dep. is by reason of the depth at which photons penetrate, blue shallow and red deep. So if a backside chip then the greater variations are in the red, because the photons transit the full depth, and there's some interaction with the gates. I'm not sure but get the idea a frontside device would be around other way, i.e. blue more affected.

Testing is going to be difficult, will need to project a spot much less than pixel dimensions, and then dither accurately, ouch. Some detector labs can do routinely. I think Kepler people looked at actually doing tests in orbit, but not sure they actually did, but the telescope has sufficient pointing..?


Post Extras: Print Post   Remind Me!   Notify Moderator  
btieman
scholastic sledgehammer


Reged: 07/24/08

Re: ED80 for Photometry? new [Re: Andean Slave]
      #5908843 - 06/07/13 09:46 PM

Andean, I didn't think WFCP2 was backside illuminated? WFCP1 was, I believe. Not surprised that backside illuminated chips would show position related variances due, as you mention, to interference from the gates and such...but how many of us mere amateurs have backside illuminated CCDs I've worked with a few back in the day, but nothing mountable on a telescope!

Back at my previous career, I probably had the equipment to make this measurement. These days, not so much. Pointing accuracy is irrelevant since scintillation alone would be enough to ruin the results at this level. Would have to be a lab setup.

Anyway, the point is more that spatial resolution should not be the primary concern when performing photometry since we're not measuring along that axis. Many light curves have been generated with 0 spatial resolution what-so-ever (scintillation detectors) More important to understand how the detector reacts to the light and what can be done to make that as precise as possible.


Post Extras: Print Post   Remind Me!   Notify Moderator  
Andean Slave
member


Reged: 01/08/11

Re: ED80 for Photometry? new [Re: btieman]
      #5928237 - 06/18/13 08:47 PM

Going back to the original question of effect of undersampling of the PSF on photometric precision, and whether, if faced by an inadequate FPS, the use of Barlow to get up to a good number of pixels under the PSF would help or hinder, my view is that defocussing will tend to be easier, because the photometry is likely just going to be aperture photometry. More glass means more sources of calib errors, normally through skylight ghosting off surfaces making hell on the flats.
When the precision starts getting high, i.e. at the mmag level, then no matter what one does, things become complicated. Defocussed annuli change radius with temperature (tube length), etc etc. But this is the fun.
The main thing is to have an easy way to do an experiment on the experiment. The simplest and generally most satisfying test is just doing some high cadence photometry of a quick eclipsing binary (some of which may be in reach of an 80mm). There's a cute lightcurve at the end, though of course the real goal is measuring the SDs of differences between a few bright comparison stars, which then show whether a strategum moved us 'warmer or cooler'.
PMT photometry was most often single channel certainly, but, as in CCD work, incoming light was spread over the detector to average out sensitivity variations, for a PMT a Fabry lens was normally employed.


Post Extras: Print Post   Remind Me!   Notify Moderator  
sound chaser
member
*****

Reged: 01/13/13

Loc: France
Re: ED80 for Photometry? new [Re: Andean Slave]
      #5931580 - 06/20/13 04:06 PM

Well the 6 inch is sitting on top of a shiny new EQ6 Synscan now in the observatory. I've printed off some photometry tables from AAVSO to go with my charts and now just waiting for some clear nights.
Can't wait!
Thanks for helping me make up my mind!

Doug.


Post Extras: Print Post   Remind Me!   Notify Moderator  
Hubert
sage


Reged: 09/26/06

Loc: Belgium
Re: ED80 for Photometry? new [Re: sound chaser]
      #5934280 - 06/22/13 09:34 AM

Hi Doug, I wish you succes!

cheers, Hubert


Post Extras: Print Post   Remind Me!   Notify Moderator  
sound chaser
member
*****

Reged: 01/13/13

Loc: France
Re: ED80 for Photometry? new [Re: Hubert]
      #5937909 - 06/24/13 12:08 PM

Thanks Hubert!

Post Extras: Print Post   Remind Me!   Notify Moderator  
Pages: 1


Extra information
0 registered and 2 anonymous users are browsing this forum.

Moderator:  Rich (RLTYS) 

Print Thread

Forum Permissions
      You cannot start new topics
      You cannot reply to topics
      HTML is disabled
      UBBCode is enabled


Thread views: 2034

Jump to

CN Forums Home


Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics