Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Big favor: can someone upload their raw (or stacked) 294MC Pro flats?

  • Please log in to reply
19 replies to this topic

#1 cuivienor

cuivienor

    Surveyor 1

  • -----
  • topic starter
  • Posts: 1929
  • Joined: 06 Oct 2010
  • Loc: Tokyo, Japan

Posted 25 March 2019 - 07:55 PM

Hello,

I am noticing a lot of mottling introduced by my flats on the ASI294MC Pro I take (flat box, t-shirt, Dawn sky provide the same results. Different gains/offsets. Cooled, uncooled. Same.). I'm not sure whether it is simply normal and a good SNR simply makes the mottling invisible.

So I want to get someone else's ASI294MC Pro flats, raw or stacked. If someone could upload that to a Dropbox or Google Drive or equivalent, it would be awesome...

Thank you!

Yannick

#2 kenkolen

kenkolen

    Vostok 1

  • *****
  • Posts: 121
  • Joined: 12 Oct 2006
  • Loc: Miami

Posted 25 March 2019 - 08:37 PM

Sorry what is mottling?



#3 ac4lt

ac4lt

    Messenger

  • -----
  • Posts: 433
  • Joined: 16 Aug 2017
  • Loc: Northern Virginia

Posted 25 March 2019 - 09:45 PM

Yannick, it seems to be a problem for many with this camera. Three suggestions:

1. Don’t use bias but instead take flat darks.
2. Use exposures of 3 seconds or longer
3. Try using a much larger rejection on your dark integrations of 7 to 8 sigma

Those will help but solve the issue (at least for me). I can get you some data tomorrow but mine has the same issue as yours.

#4 cuivienor

cuivienor

    Surveyor 1

  • -----
  • topic starter
  • Posts: 1929
  • Joined: 06 Oct 2010
  • Loc: Tokyo, Japan

Posted 25 March 2019 - 11:04 PM

Thanks!

 

@kenkolen, something like the attached, when I investigated my flats after noticing they introduced the mottling in the final light. To see the pattern, I have to do Debayer + ABE on the master flat (or an individual flat), and the pattern becomes very visible. I took flats of 5 seconds as well with the same result. It's as if the camera doesn't like light :)

 

@ac4lt: thank you so much! Getting hold of a raw flat file would be great, so I can compare! I've uploaded a sample here: https://drive.google...QnK?usp=sharing

 

Thanks again!

 

Yannick

Attached Thumbnails

  • 2019-03-26 12_58_22-Window.png


#5 ac4lt

ac4lt

    Messenger

  • -----
  • Posts: 433
  • Joined: 16 Aug 2017
  • Loc: Northern Virginia

Posted 26 March 2019 - 07:13 PM

Yannick, here is a link to my most recent flats. It's a zip file containing a raw flat, the corresponding calibrated flat and then the resulting master flat from integrating it and its companions. Take a look at that and see if it helps. I can provide a master light integration if you want also.

 

Let me know if you have any trouble accessing the link. I've had some issues with sharing google drive links in the past.

 

https://drive.google...iew?usp=sharing



#6 cuivienor

cuivienor

    Surveyor 1

  • -----
  • topic starter
  • Posts: 1929
  • Joined: 06 Oct 2010
  • Loc: Tokyo, Japan

Posted 26 March 2019 - 11:44 PM

Awesome Linda, thank you so much for that!!!

 

Yep, I can tell that those flats also exhibit the same type of mottling, so at least my "lemon anxiety" is much lower now :) So now it's just a problem of SNR, really....

 

Thanks again!!

 

Yannick



#7 wesnikon

wesnikon

    Vostok 1

  • -----
  • Posts: 141
  • Joined: 14 Nov 2018
  • Loc: Germany

Posted 27 March 2019 - 12:01 AM

Hello,

I am noticing a lot of mottling introduced by my flats on the ASI294MC Pro I take (flat box, t-shirt, Dawn sky provide the same results. Different gains/offsets. Cooled, uncooled. Same.). I'm not sure whether it is simply normal and a good SNR simply makes the mottling invisible.

So I want to get someone else's ASI294MC Pro flats, raw or stacked. If someone could upload that to a Dropbox or Google Drive or equivalent, it would be awesome...

Thank you!

Yannick

There you go - https://westheider.c...aster_flat.xisf



#8 ippiu

ippiu

    Vostok 1

  • -----
  • Posts: 182
  • Joined: 06 Dec 2010
  • Loc: Italy

Posted 27 March 2019 - 08:58 AM

Last time i used Sharpcap Pro i did a normal live stacking session, i took a 20 flat frames to have the final master flat: then i started live session with flat subtraction enabled, and my final stacked image was like cuivenior showed, horrible, mottling pattern

 

stacked_flat1.jpg

 

stacked_flat.jpg

 

So i decided to try a live stacking without flat frame: the final stacked image was perfect, with no artifacts (apart noise caused by having stretched the histogram)

 

stacked_no_flat.jpg

 

I took my flats at: bias enabled, monochrome, 0 gain, 0,5 sec, 30.000 ADU (45-50% peak white histogram), brightness 10, white R and B at 50.

Here my master flat created by Sharpcap with bias enabled and monochrome: https://drive.google...YTNn4CmAmJtX433

 

Because 0 gain has the maximum level of dark current, is this the problem? Do i have to take flats for example at a minimum of 120 unity gain or upper?

 

Cheers

 

Andrea


Edited by ippiu, 27 March 2019 - 09:02 AM.


#9 ac4lt

ac4lt

    Messenger

  • -----
  • Posts: 433
  • Joined: 16 Aug 2017
  • Loc: Northern Virginia

Posted 27 March 2019 - 09:09 AM

Andrea, I’m no expert but my understanding is the gain & temperature for your flats should match your lights and darks.

Also, based on some analysis done by John Upton that you can find in another thread, it was recommended that exposures for this sensor be >= 3 seconds. So, instead of doing bias frames take flat darks.

I’m not familiar enough with SharpCap to say you that would happen there but I’ve been using Sequence Generator Pro and PixInsight. I still struggle with color mottling in the final integration but I think John’s recommendations do help.

The other thing I’ve seen recommended and have tried once but don’t have enough info to really say how well it works is to use a larger rejection setting when integrating master darks. Usually the defaults are 2.5 to 3 sigma. Try 7 to 8 sigma for your darks and see if that helps. It seemed like it did on the one case I tried.

#10 cuivienor

cuivienor

    Surveyor 1

  • -----
  • topic starter
  • Posts: 1929
  • Joined: 06 Oct 2010
  • Loc: Tokyo, Japan

Posted 28 March 2019 - 12:18 AM

Thanks for all the feedback!

 

@wesnikon: thank you for the flat! I see the same in your flat, so this is definitely an overall 294MC sensor thing :)

 

@ippiu: thanks for chipping in! Yes, that's exactly what I was seeing. I thought that I could take flats at gain 0 as well, so I could take longer flats (the sensor being so sensitive), which is what I do sometimes with the ASI1600MM. If you calibrate with flats taken at or above unity gain, the mottling is much, much less noticeable, and if you have a good SNR, it's basically invisible

 

@ac4lt: as flats represent the issues with the optical path, rather than doing what effectively is sensor modeling (as darks do), they can *in theory* be run at a different gain, offset, and temperature as the lights - care must be taken then though to make sure that if the flats are calibrated, they are calibrated with dark frames taken at the same gain, offset, and temperature as the flats themselves. In effect though because the 294MC sensor seems pretty prone to fits of anger and inconsistent behavior depending on temperature, etc., it seems best to take the flats at the same temp, gain, and offset as the lights.

 

Noted on the higher sigmas for the dark integration!

 

I'm kind of thinking I should have gone with the ASI071MC instead... oh well!

Cheers,

 

Yannick


  • wesnikon likes this

#11 ippiu

ippiu

    Vostok 1

  • -----
  • Posts: 182
  • Joined: 06 Dec 2010
  • Loc: Italy

Posted 28 March 2019 - 03:53 AM

 

@ippiu: thanks for chipping in! Yes, that's exactly what I was seeing. I thought that I could take flats at gain 0 as well, so I could take longer flats (the sensor being so sensitive), which is what I do sometimes with the ASI1600MM. If you calibrate with flats taken at or above unity gain, the mottling is much, much less noticeable, and if you have a good SNR, it's basically invisible

 

https://forums.sharp...31&p=8014#p8014

 

So, after above reply in Sharpcap forums, do you think my problem is caused only by 0 gain?? 



#12 cuivienor

cuivienor

    Surveyor 1

  • -----
  • topic starter
  • Posts: 1929
  • Joined: 06 Oct 2010
  • Loc: Tokyo, Japan

Posted 28 March 2019 - 04:14 AM

@ippiu : it sure seems like it! (unless I am misunderstanding your last post) It seems like 0 gain is simply not usable with this sensor (whether for flats or for lights). And even unity or more shows mottling (at least in my case - but the mottling is less pronounced) when I image from Tokyo center, e.g. when the SNR ratio is bad. I may be using the camera with a Triad Ultra filter to see whether images get better, but at this stage I'm thinking about selling it and getting the 071 instead...



#13 ippiu

ippiu

    Vostok 1

  • -----
  • Posts: 182
  • Joined: 06 Dec 2010
  • Loc: Italy

Posted 28 March 2019 - 04:27 AM

Thanks @cuivienor...

 

Why not 183 cooled?? Yes, it has huge amount of amp glow, but with simply a darks library taken at different gain, exposure and temp (for winter and summer) as i normally do with 294, you could have a valid cmos with less money...



#14 cuivienor

cuivienor

    Surveyor 1

  • -----
  • topic starter
  • Posts: 1929
  • Joined: 06 Oct 2010
  • Loc: Tokyo, Japan

Posted 28 March 2019 - 04:30 AM

That's a good point.... The well depth of the 183 worries me though....



#15 ippiu

ippiu

    Vostok 1

  • -----
  • Posts: 182
  • Joined: 06 Dec 2010
  • Loc: Italy

Posted 28 March 2019 - 04:44 AM

That's a good point.... The well depth of the 183 worries me though....

Anonimo1.jpg



#16 ac4lt

ac4lt

    Messenger

  • -----
  • Posts: 433
  • Joined: 16 Aug 2017
  • Loc: Northern Virginia

Posted 28 March 2019 - 05:39 AM

Yannick, one of the things that John Upton suggested in his thread was taking a dummy bias frame between each exposure. I keep forgetting to try this, but you may want to give it a try. You'll just throw those bias frames away but taking them seems to have a positive effect on the next exposure. I don't know whether it will help, but it's an easy enough test to try. He's suggesting a dummy bias between regardless of frame type so the idea is to do it for darks, flats and lights.



#17 bobzeq25

bobzeq25

    Hubble

  • *****
  • Posts: 16931
  • Joined: 27 Oct 2014

Posted 28 March 2019 - 09:40 AM

Thanks @cuivienor...

 

Why not 183 cooled?? Yes, it has huge amount of amp glow, but with simply a darks library taken at different gain, exposure and temp (for winter and summer) as i normally do with 294, you could have a valid cmos with less money...

I have an ASI183C.  The tiny pixels need a fair amount of light, will have too small an image scale at longer focal lengths.  Jon Rista has questioned whether the 183M works well at low and high gains, I've only used my color at 50 and 100 (unity, approx).  I bought it with use with an 8 inch F2 RASA in mind, the sweet spot for that camera.

 

The cooling is excellent, the dark noise low, I'm not sure many would need winter/summer darks.

 

Everything in this business is tradeoffs.  Some people like their 294s, some the 183, some the 071.  Personal choice, read the reviews/experience, make a decision.


Edited by bobzeq25, 28 March 2019 - 09:42 AM.


#18 cuivienor

cuivienor

    Surveyor 1

  • -----
  • topic starter
  • Posts: 1929
  • Joined: 06 Oct 2010
  • Loc: Tokyo, Japan

Posted 28 March 2019 - 06:37 PM

@ac4lt: thanks for the tips! Almost sounds like Voodoo magic, but as you say it's easy enough to try out, so I'll do so! Maybe I'll have a scheduled script that cleans up the Bias folder from time to time too.

@bobzeq25: thanks for the feedback on the 183! I'm guessing it would be perfect for my 135mm f2 lens, and decent for my 350mm f4 lens... I have heard of many people having issues with the 294, some going to the 071 instead in the end, some just loving the 294, and I have heard about the 183 having a specialized sensor, and... Well difficult to make a choice! For the moment I'll go with my 294 and get a triad ultra filter, if they ever get back in stock.

#19 f300v10

f300v10

    Ranger 4

  • -----
  • Posts: 302
  • Joined: 16 Sep 2018
  • Loc: Marietta, GA

Posted 28 March 2019 - 08:17 PM

I have been using my 294MC-Pro for a few months now, and have not had the issues with calibration some have had.  Here is a single flat, and the 15 sub master flat from last night.  8 second exposure, gain 120, temp -10C, with a t-shirt and LED panel:

 

Single:

https://drive.google...iew?usp=sharing

 

Master:

https://drive.google...iew?usp=sharing



#20 cuivienor

cuivienor

    Surveyor 1

  • -----
  • topic starter
  • Posts: 1929
  • Joined: 06 Oct 2010
  • Loc: Tokyo, Japan

Posted 28 March 2019 - 09:27 PM

@fv300v10: thank you! Yes, your flats look very clean. There is some mottling, but it is a bit better than wesnikons' and much much better than what ippiu and I got at gain zero.

 

Because you used longer exposures, it seems to be that longer exposures definitely seem to give a marked improvement over shorter exposures.

 

Cheers,

 

Yannick




CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics