0.90
#... |
|---|
|
When I take lights, darks, and bias frames my image files are rectangular. However, then I take flats I get a circular image with two sides truncated. It is almost like a physical clipping of some sort, or the optics are not fully illuminating the sensor. Attached are a few .jpgs to illustrate. The set-up: 1) Spike A Flat 12" panel about 6" ahead of the scope 2) Tele Vue NP101is refractor 3) ZWO OAG 4) ZWO EFW (36mm filters) 5) ZWO ASI2600MM (APS-C) Is this normal?     |
7.76
#...
·
1
like
|
|---|
| If you can flatten the raw frames then there is nothing to worry about. This said I suspect the filters (and their position ahead of the sensor) are to blame for the circular effect you see. |
0.00
#... |
|---|
|
Jeff, how are you capturing your flats? If I accidentally connected to my guide cam using a OAG this is exactly what my flats would look like. I would cofirm you are connected to the right camera and make sure you are getting to a median 30000 ADU with your exposure time. |
4.72
#... |
|---|
|
Hi Jeff, if you have PixInsight, you may use the script FlatContourPlot to analyse your flats. It helps you to spot any irregular patterns. CS Gernot |
5.26
#...
·
2
likes
|
|---|
|
I don't see nothing wrong in particular. To me, this is only vignetting. The truncation on the bottom side of the flat frame is the shadow of the OAG prism. The sharpness of the circular pattern of the flat is most probably a stretching effect. And, as Andrea said, if calibration is good, there's nothing to worry about. Ciao. Lorenzo |
0.00
#... |
|---|
| What capture software/device are you using? My flats look similar to yours when taken with my asiair with the auto stretch preview turned on but it saves the unstretched files so my calibration is fine. Perhaps your capture software is doing an auto stretch and thus exaggerating the apparent vignetting. |
|
0.90
#... |
|---|
Jeff, how are you capturing your flats? If I accidentally connected to my guide cam using a OAG this is exactly what my flats would look like. Attached is a photo of the set-up to capture everything above excepting the light frame. I use an OAG, and as you can see the guide camera was not installed. The exposure is automatically calculated by the ASIAir. I have tried all of the filters and varying the Spike A Flat settings. The AAP adjusts the exposures accordingly.  |
|
0.90
#... |
|---|
Lorenzo Siciliano: Thanks Lorenzo! I'm going to run with this and see if I can produce some results. |
5.73
#...
·
3
likes
|
|---|
|
Judging by your comments, you may not understand what a flat frame is. Bias and Dark frame are images of your chip noise with no light from optics. A Flat frame is an image of the optics. It does, as all images do, includes the chip noise, but its goal is to show the defects of the optics. All optics will have some light fall off towards the edges. That is why the image gets darker towards the edges. It circular because your optics are circular. That should be obvious. The purpose of the Dark frame is to photograph the chips noise created by dark random electrons created by the chip, and substract that from the astro image. The purpose of the Bias frame is to photograph the random inconsistently of the pixels. Not all pixels are created equal. They have bias. Some are more sisitive than others. The Bias frame photographs this inconsistens and smooths the image. A Dark frame includes this pixel bias, and some imagers only use the Dark frame for calibration. Some Sony chips such as the ICX694 has such low dark noise that calibrating a Dark frame will add noise, so only the Bias frame is used in calibration. Unless you have extremely flat optics and very dark skies, the will be light fall off toward the outer edge of you images. There may also be dust acculation on the optical elements. The Falte frame photographs these imperfections and subtracts them from the image. It is best to bias subtract the flats. Dark subtraction is usually unecassary because the exposure is so short. Mine are usually a fraction of a second. I highly suggest you read and understand the basic concepts of image acqusition and processing. It will make the whole process easier when one starts to push software buttons and not fully understand what is actually going on. Lynn. K |
|
0.00
#... |
|---|
Jeff, how are you capturing your flats? If I accidentally connected to my guide cam using a OAG this is exactly what my flats would look like. Jeff, nothing looks too wrong in your imaging here but things to check. 1. Do your light images have the same vignetting if so this is what flats are for. If not try moving the panel closer to see if you are picking up the edges of the panel. 2. Your OAG will cast a shadow and flats will help eliminate that too so you should see that in your lights. 3. Is your exposure time at least a few seconds, you don't want your flats to be fractions of a second with CMOS cameras. If so your panes is too bright. 4. Open up the flats and make sure the median ADU is approximately 30000. Your ASI Air should be adjusting the exposure length to achieve this but that is where you may be going too short of an exposure for quality flat frames. |
|
7.76
#... |
|---|
3. Is your exposure time at least a few seconds, you don't want your flats to be fractions of a second with CMOS cameras. If so your panes is too bright. Why not? Mine are always fractions of a second to maybe few seconds (I mean 1 or 2) and I use a CMOS camera and I'm none the worse for it. |
|
0.00
#...
·
1
like
|
|---|
andrea tasselli:3. Is your exposure time at least a few seconds, you don't want your flats to be fractions of a second with CMOS cameras. If so your panes is too bright. Andrea, 1 to 2 seconds should be fine and I am referring to fractions of a second like 100ms. From my understanding CMOS chips can be a bit weird when trying capture very short frames this is why it is suggested to do dark flats instead and eliminate bias frames for CMOS cameras. The second thing to consider is the frequency of the LEDs in the light panel. Very short exposures will show banding in your frames that does not exits in your lights . If the frequency is low 1s exposures could show this banding if you stretch out your flat frame. Some higher end panels use a very high frequency to avoid this. You want an exposure length long enough to even out the LED frequency to a nice even flat frame. |
|
7.76
#... |
|---|
|
Hi, My typical exposure time is 0.1s to 0.2s for sky flats. Bias subtracted. Obviously not dark subtracted. When I use the FL panel (nearly always with DSLR i.e., a CMOS sensor, my typical exposure times are 1/60s to 1/80s. Yet not a shred of banding in my flats (see attached). Nothing special about the panel, just one cheap run-of-the-mill variety.  |
|
0.00
#... |
|---|
andrea tasselli: I am glad it works for you as you know in this hobby experiences can be very different. Here is an ASI camera at 0.45 sec exposure and aprox 30,000 ADU with a standard flat panel. The banding here is what I am talking about. It is the flickering frequency of the LEDs. If you every see this in your flats the exposure is too short for your camera relative to your flat panel.  |
|
0.90
#...
·
1
like
|
|---|
Lynn K: I do understand, read Bracken's Primer thoroughly (twice) before spending the money on this little adventure. My expectations were probably not accurate though. My thinking was the "is" version of this scope should easily (brightly) illuminate a 35mm diagonal sensor. With some edge fall-off of course. Looking at it now, the truncation may very well be normal - the short dimensions well illuminated to sensor edge with very little fall-off. (BTW, removing the prism and shooting yields the same result). If I should be "complaining" about anything, the NP is falling off quite a bit on the long dimension. Again, perhaps normal. Just need more experience to know. Tempted to take out my TMB130, but really want to get the smaller focal length "mastered" first. It's already been a handful ;-) Going out tonight and hope to get some good data, download the rental version of APP tomorrow and have some fun. |
0.00
#... |
|---|
|
They looks pretty decent to me - one thing that's worth checking/remembering is that, when stretched, the difference between the most heavily vignetted areas and the lightest can often be as little as a few percent, but the stretch tends to emphasise it (using PI or similar tools hover over the pixels to see just how little they vary). One simple test for flat frame quality is to get all of your masters together (including flats) in APP and then, having switched off Star Analysis/Registration, take some "lights" of your flat field and stack/proces them. You should get a nice even "flat" image (hence the name) - if you don't then it's a good sign your flats aren't consistent or you have a non-linear source of light in the system (aka a light leak) and/or your filter placement isn't consistent. |
|
5.26
#... |
|---|
|
I'm used to shoot 5 sec flats. Under 1 sec cmos sensors (294mm for sure, don't know others) show non linearity, so calibration is an issue. This is a well documented topic (search cloudynights or Christian Buil's website). So my advice is to take 2 sec or more flats to be safe. My two cents here. Ciao. Lorenzo |
0.00
#...
·
2
likes
|
|---|
|
Why do you put the flat panel so far away? Make sure you don't have any secondary light sources in the room that can hit the chip. Place the Scope so that it is in direct contact with the panel and make sure it is not dimmed via PWM. cheers Thomas |
