I have a lots of questions on flats

[jimwc6491]

All

 

I would like to discuss ways of taking flats.

why take them?

what do they do?

are they necessary?

can you create a library of them?

can you shoot them during the day?

if I need more than one can I take 1 exposure and copy it as many times as I need?

 

 

I know it is usually a white light source taken at the same settings as the lights.

how many exposures do you need?

I have heard you can use a white "T" shirt, a white poster board with a bright light source, LED light box. 

I am sure there are many other ways but what works best and easiest to build?

 

 

[astroman133]

>> why take them?

 

To assist with repair of uneven field illumination (UFI) in your lights during image calibration. UFI could be caused by vignetting (darkening of your image in the corners), by dust on the sensor or filter, or even by variations in quantum efficiency of the pixels across the sensor. Because the flat-field frame records the response of the entire optical system, including the telescope, they should be taken with the camera in the EXACT same orientation as was used for the light frames.

 

>> what do they do?

 

Typically brightness variations in a master flat are applied to each light frame to correct for UFI. Mathematically, it is usually a division operation. It really does not matter what the source of the UFI is, so long as the optical system has not changed.

 

>> are they necessary?

 

Not if you don't ever see vignetting or dust donuts in your images.

 

>> can you create a library of them?

 

They are usually taken after your light frames are done AND before the camera has been touched (moved or rotated).  They are different from darks because they include the entire optical system.  Therefore you can not use libraries of flat frames that were previously taken since the orientation of the camera relative to the optical tube would have likely changed.

 

>> can you shoot them during the day?

 

It depends on the source of illumination and ambient temperature.  I have, but I most often use a cooled camera and so I am able to minimize noise from the images. Day time is typically hotter which for an uncooled camera means more noise.  I may shoot more flat exposures if it was hot, to help reduce random thermal noise in the master flat. You do not want your flats to have a lot of noise that would then be incorporated into your light frames during calibration.

 

>> if I need more than one can I take 1 exposure and copy it as many times as I need?

 

No! If you did that then any noise in the image would also be repeated and not reduced.  You might as well use that 1 image as a master flat.  I typically shoot 20 flats, at standard shooting settings, and then apply dark flats with the same settings from my library of master darks.

 

Depending on how you shoot your flats (sky flats, T-shirt flats, dome flats, light box flats). You should also apply a dark or bias calibration frame to your flat frame during the creation of your master flat. Because the exposure duration of the flat frame is usually short, they don't accumulate much noise. That allows you to use a library bias frame instead of a flat dark frame, if you so desire.

 

I use an electroluminescent (EL) panel that I made for about $100. The panel itself is flexible, about like a plastic placemat at a restaurant. I clamped it between (nearly) opaque pieces of acrylic and glued some small blocks to it to allow me to hang it on my objective. I have a friend who made one out of foam board and some LEDs.

 

I hope that answers some of your questions.

[jimwc6491]

Rick

if I scribed a mark on my camera adapter so I could index it to the same place on the focuser.

then noted the focuser position (my motorized feather touch has a numeric counter) as long as I zero it before focusing, I should be able to duplicate the focus point.

(or just remove the camera from the focuser without moving the focuser) I should be able to take flats during the day correct?  inside the cool house.

 

can I use the histogram in BYEOS to set the ISO and duration if so how?

[astroman133]

Jim,

 

I don't think that a scribed mark is adequate, because even if the two halves of the mark are carefully aligned you could still be off by a few pixels. This would degrade the usefulness of the flat frame.

 

Focus is not as critical because the light source will be right near the end of the scope and will be very de-focused,as compared to the focus distance at infinity for targets.

 

You can take flats anywhere, as long as you can re-locate the telescope and camera without disturbing the orientation and focus. Why would you risk mis-aligning the camera during transport, however.

 

Still, how long would it take you to take 20 10-second images? probably about 5 minutes.  Depending on the brightness of your light source it could be much less than that. Given that, why wouldn't you just take them at the end of the session, or before you need to rotate the camera to frame a different target.

 

I may be wrong about what I am about to say and if so am more than willing to be corrected....The histogram that BYE displays is the histogram of the JPG image which is embedded inside of the CR2 image. As such it has been stretched in the camera. For this reason I am not sure whether it is a good basis for determining the correct exposure for flat frames. I would suggest a more involved process when determining your exposure for flats where you view the histogram of the RAW image in some other software that displays the histogram based on the actual raw pixel values. If you are using an artificial light source with a constant brighteness, once you have determined the correct exposure for your flats, it should not change. I would suggest saving it in a separate capture plan, if you have the Premium version of BYE/BYN.

[s3igell]

The histogram that BYE displays is the histogram of the JPG image which is embedded inside of the CR2 image. As such it has been stretched in the camera. For this reason I am not sure whether it is a good basis for determining the correct exposure for flat frames. I would suggest a more involved process when determining your exposure for flats where you view the histogram of the RAW image in some other software that displays the histogram based on the actual raw pixel values. If you are using an artificial light source with a constant brighteness, once you have determined the correct exposure for your flats, it should not change. I would suggest saving it in a separate capture plan, if you have the Premium version of BYE/BYN.

Counterpoint:

The emphasis on "Correct Exposure" for Flats was an emphasis many years ago, when most Imagers were Old-school CCDs and they had significant issues with Linearity.  One had to be concerned to take Flats in "Just the Most Linear" portion of the Histogram.

Now, both CCDs and DSLRs are very Linear throughout the bulk of their Exposure Range.  Hence, as long as you have the Exposure decently well away from either edge of the Histogram, there is no problem.

[jimwc6491]

s3igell

that means that I can setup a capture plan during the day with the correct exposer.

and I wont need to change it every time I do a batch of flats at the end of a session.

that sounds promising a little less work at night.  

[s3igell]

Only if you can get the Daytime Flats Exposure to run at a Temp equivalent to the Lights you Expose at Night...

[astroman133]

However, the exposure would not change with the temperature, only with the brightness of the light source.

 

s3igell is correct that the temperature of the flats should be close to the temperatures of the lights, darks, and dark flats (or biases).

 

I am confused about what you mean when you say "I can setup a capture plan during the day with the correct exposer. and I wont need to change it".

 

If you have the premium version of BYE/BYN you can create an offline capture plan (without being connected to a camera). Then you can load and execute that capture plan at any time you need it.  This is what I recommended in one of my previous posts to this thread. You should not use this capture plan to get flats during the daytime unless the equipment setup has not changed and the temperature is roughly the same as during the imaging session.

[AaronSmall]

I agree with Guylain.  6 LEDs won't be too much.  Vellum should work as well as mylar but I suggest 2 layers to help diffuse the light more.  Vellum also seemed to give a grainy pattern when run too fast on the old fashioned blue print machines.  Two layers should get rid of any "error pattern" created by one layer.  May not matter too much since it is for a flat and it is placed so close to the objective.

[jimwc6491]

Rick, S3igell

 

yes I have the premium edition of BYE.

 

What I meant by save a capture plan was create a capture plan with 20 - 10 second exposures at a particular exposure and save it.

save several at different exposures so I can match what I am  shooting.

for use later at night. just one step I don't have to do at night. might be a little bit cumbersome.

 

are there any plans on the internet for making a light panel for taking flats? where did you get your EL panel?

is the light panel you have 12 volt? I will need something that runs on 12V I don't have a observatory.

 

I can just see what would happen if you turned on a white light at a star party, what can I use for a black cape to cover everything so I don't disturb people around me at a dark site?

I am thinking of getting a black apron like they use in a barber shop when you get your hair cut. 

I am planning on getting a truss tube RC in the near future so I defiantly will need to cover it some how. 

[astroman133]

Because the capture settings for your flats are only dependent on the brightness of the light source and the configuration of your optical trains you may not need to "have several at different exposures so I can match what I am shooting". I have 2 flats capture plans. One for when I use a focal reducer and one for when I don't. Perhaps others would disagree, but I would not worry too much about temperature matching your flats.  The signal from the artificial light source would overwhelm any thermal noise generated by the camera, especially with such short exposures.

 

I should also note that the EL panel operates at a 60 Hz cycle, like a flourescent light. This means that the light source needs to be dim enough that your exposures are long enough that you don't see that cycle in your raw flat images. I had to put a few sheets of copier paper in front of the EL panel to darken it sufficiently.  Another option is to get a variable power supply for the panel. The exposure duration for my flats is on the order of 1/4 - 1/2 of a second, but will rise to 15-20 seconds when I am using a narrowband filter.

 

I will also say that the EL panel has a greenish color cast. The color cast is not an issue for the flat frames since only the variations in brightness are used during calibration.

 

I bought the EL Panel and power supply from GlowHut.com and I had the acrylic panels cut to my specifications by a local supplier. The power supply operates off of 12V so I can operate it from a car battery or 110VAC through an adapter. The panel is sandwiched between sheets of dark blue acrylic. But unfortunately, they leak light so when I have used the panel at star parties I throw a shroud over it to minimize light trespass on my fellow astronomers. The shroud is custom made and I also use it to enable me to read my PC screen during the rare times that I try to do solar imaging. It is black on one side with a shiny, reflective silver fabric on the other. It is about 3 ft square. I have never had any complaints from those around me about the light.

 

FYI, my friend, Joe Golias at AstroZap makes and sells shrouds for truss tube scopes.

 

I hope this helps.

[S-D]

"Depending on how you shoot your flats (sky flats, T-shirt flats, dome flats, light box flats). You should also apply a dark or bias calibration frame to your flat frame during the creation of your master flat. Because the exposure duration of the flat frame is usually short, they don't accumulate much noise. That allows you to use a library bias frame instead of a flat dark frame, if you so desire."

 

I have a few questions:

 

1.  I've read conflicting things about Bias frames.  I've read some that say they are unnecassary if you use Flats and Darks.  What is the general concensus on this?

 

2.  I've read that you only need to takenew Bais frames once or twice a year since the read noise pattern rarely changes.  Is this true?

 

Thanks.

 

S-D

[astroman133]

Fact: If you use Darks (the same temperature, ISO, and duration as your lights) and you use Dark Flats (the same temperature, ISO, and duration as your flats) then bias frames are not necessary.

 

Opinion: If the duration of your flats is short then you may not need either dark flats or bias frames to calibrate them.

 

That said, if it only takes a few minutes to capture and process dark flats or biases, why wouldn't you use them even if the effect is minimal?

 

Both dark flats and biases can be stored in a library and can be used for several months. In order to keep dark flats in a library you have to work out, in advance, the duration and ISO of your flats so that the dark flats can be used to correctly calibrate them.

[jimwc6491]

Rick

when building a light box what color white should you have (cool, warm). I was looking at using foam core poster board and 12V led lights. mainly because of size >15" for a 10" scope.

also how bright should it be, just so I sort of know how many LED's I will need. or would this be trial and error.

[astroman133]

The color temperature is not a critical choice.  Much more important is that they diffuse evenly. I have never built a light box so hopefully someone who has will chime in. It does not need to be very bright.

[S-D]

I guess this is not just a question for flats, but I've read that there is an ideal ISO for each Canon model to shoot deep space objects with.  I've seen a few tables with these settings, but none ever contain an entry for my camera (T5i). 

 

Is there a way to figure this out to ensure that your flats (and lights and darks for that matter) are at the optimum ISO?

 

Is this even a true statement, or do you generally choose a different ISO based on the target?

 

Thanks.

 

S-D

[astroman133]

Your question about ISO has a simple answer, but with a lot of underlying complexity.

 

Here is an article that discusses some of the complexities, but does not actually answer your question:

 

http://www.clarkvision.com/articles/iso/

 

It does however try to educate you. The problem is that the article is a bit dated and the author tends to evaluate the higher end cameras.

 

Unless you want to try to collect test data for your T5i as the author did, I would suggest setting the ISO on your T5i at either 800 or 1600 and leave it there. If you choose 800, then you can shoot your dark library at 800.  That means that you will have a smaller library with everything at the same exposure and ISO.

[S-D]

Thanks Rick.

 

I'm not sure I fully understand your last two sentences:

 

"If you choose 800, then you can shoot your dark library at 800.  That means that you will have a smaller library with everything at the same exposure and ISO."

 

Wouldn't this also hold true if I shot at 1600?  Is there an advantage to one over the other (other than the amount of noise)?

 

Thanks.

 

S-D

[astroman133]

In the context of my previous sentence, I suggested that you choose an ISO and stick with it.  Use it for all your imaging, regardless of frame type (light, dark, flat, etc).

 

The advantage of one of the other depends on the dynamic range curve for the T5i. You should use the ISO that is closest to the curve's breakpoint, as explained by Roger Clark in the linked article.

 

in the absence of experimental data, I would tend to suggest using 800, unless you don't get round stars at the exposure necessary for a well placed histogram. Then I would increase to 1600 and try it there. The shorter exposures may help give you better stars.

[AaronSmall]

Rick

when building a light box what color white should you have (cool, warm). I was looking at using foam core poster board and 12V led lights. mainly because of size >15" for a 10" scope.

also how bright should it be, just so I sort of know how many LED's I will need. or would this be trial and error.

Jim,

 

I built a lightbox a couple years ago and plan on building another soon for my 8".  It has 4 white incandescent bulbs and is plenty bright.  I plan on using the same number of LEDs (white of course) for the new box.  Like Rick said, diffusing is the answer.  My lights actually sit behind a baffle near the level of the corrector plate and has to reflect on the back of the box and pass through two double sided sheets of drafting mylar to diffuse the light.  I ran it off of the 12v accessory plug on my Celestron mount.  With the light box sealed up tight, there is almost no light escaping so a shroud may not be necessary 

 

Aaron

[jimwc6491]

Aaron

 

I am in the process of building a light box now. I bought 6 LED's . I tried to find a white plastic florescent light panel but couldn't find one, so I guess it will be drafting velum.

I just hope it is not too bright, if it is I can always remove one or two of the LED's. I am building for a 10" that I plan on getting in the near future. but I figure I can use it on my 8" in the meantime.  

[admin]

Vellum works very well to diffuse light.  I have 2 layers of vellum about 1 inch apart in my light box to diffuse light evenly.  I use 3 of these lights in mine.

 

http://www.homedepot.com/p/Sylvania-Dot-It-LED-Battery-Operated-Stick-On-Tap-Light-Silver-36010/100670785

 

Regards,

[jimwc6491]

All

these are the ones I got.

https://www.superbrightleds.com/moreinfo/led-license-plate-lights/bolt-beam-5mm-led-light/57/

[admin]

Nice, I don't think 6 lights is going to be too much.

[jimwc6491]

Guylain

Thanks