ASI071 Flats 'icing up' again?

[symmetal]

Had first proper use of the ASI071MC yesterday for a few hours without the moon since I bought it a few months ago and got 48 3 minute subs of M31 which looked very promising on quick stack and stretch.

Tonight as it's cloudy again, as usual, I went to take the darks and flats. Used my Huion display balanced on top of the scope with two sheets of ND filter, to get exposure of 0.4s, like I do with the ASI1600 and Atik One without issues. As soon as the cooler went below 0C the flats showed streaks like icing. The cooler it got the more severe the streaks. Here's a screen grab of SGP of a complete flat at -14C and one uncooled at 6C. The histograms however do look very similar though the 'icing' peaks are a little wider. I've stretched the brightest bump as that shows the effect most. It's a colour camera, hence the three peaks.

All the 'flats' were taken at 0.4s exposure. Ignore the 0.004s in the screen grabs. I just loaded the images into my desktop SGP to show the effects.

The anti-dew was enabled to heat the sensor window.

Flat at -15C

Flat at +6C

I'd already replaced the tablets in the camera. To remove the camera to look at the sensor I had to warm it back up and disconnect it as it's all screwed together on the scope. Hanging in the air I cooled it back down to -15C and the sensor looked fine with no visible icing. I let the camera look at the ground which was lit by the Huion panel and took another 0.4s exposure. It was very approximately a similar shape to the real  flat but showed no 'icing' streaks at -15C. Here it is with a similar stretch to the others.

Ground 'flat' at -15C

Put the camera back on the scope with the flats panel. As soon as the camera was cooled below 0C the streaks appeared again like in the first picture.

Can anyone explain this?  Is it icing or something else. The subs I took yesterday of Andromeda at -15C show no visual evidence of icing.

Alan

[Adam J]

45 minutes ago, symmetal said:

Had first proper use of the ASI071MC yesterday for a few hours without the moon since I bought it a few months ago and got 48 3 minute subs of M31 which looked very promising on quick stack and stretch.

Tonight as it's cloudy again, as usual, I went to take the darks and flats. Used my Huion display balanced on top of the scope with two sheets of ND filter, to get exposure of 0.4s, like I do with the ASI1600 and Atik One without issues. As soon as the cooler went below 0C the flats showed streaks like icing. The cooler it got the more severe the streaks. Here's a screen grab of SGP of a complete flat at -14C and one uncooled at 6C. The histograms however do look very similar though the 'icing' peaks are a little wider. I've stretched the brightest bump as that shows the effect most. It's a colour camera, hence the three peaks.

All the 'flats' were taken at 0.4s exposure. Ignore the 0.004s in the screen grabs. I just loaded the images into my desktop SGP to show the effects.

The anti-dew was enabled to heat the sensor window.

Flat at -15C

Flat at +6C

I'd already replaced the tablets in the camera. To remove the camera to look at the sensor I had to warm it back up and disconnect it as it's all screwed together on the scope. Hanging in the air I cooled it back down to -15C and the sensor looked fine with no visible icing. I let the camera look at the ground which was lit by the Huion panel and took another 0.4s exposure. It was very approximately a similar shape to the real  flat but showed no 'icing' streaks at -15C. Here it is with a similar stretch to the others.

Ground 'flat' at -15C

Put the camera back on the scope with the flats panel. As soon as the camera was cooled below 0C the streaks appeared again like in the first picture.

Can anyone explain this?  Is it icing or something else. The subs I took yesterday of Andromeda at -15C show no visual evidence of icing.

Alan

yes that is ice on the sensor.

[symmetal]

40 minutes ago, Adam J said:

yes that is ice on the sensor.

But why does it only appear when used with the flats panel. Icing like that would show up on all images wouldn't it?

Alan

[Allinthehead]

5 hours ago, symmetal said:

But why does it only appear when used with the flats panel. Icing like that would show up on all images wouldn't it?

Alan

The signal is probably hiding it. I think you'll find it's there in the background.

 This camera doesn't need to cool beyond -5 as the benefit in reduced dark current is very small beyond that. How are you storing the camera? I use one of those small disposable dehumidifiers in a small cupboard. Finally are you using the little adapter with the desiccant tablets that screws into the side of the camera? 

Unfortunately it's a known issue with this camera so you need to be sure you don't let any moisture into the chamber. I have no issues if i do the above.

[symmetal]

7 hours ago, Allinthehead said:

The signal is probably hiding it. I think you'll find it's there in the background.

 This camera doesn't need to cool beyond -5 as the benefit in reduced dark current is very small beyond that. How are you storing the camera? I use one of those small disposable dehumidifiers in a small cupboard. Finally are you using the little adapter with the desiccant tablets that screws into the side of the camera? 

Unfortunately it's a known issue with this camera so you need to be sure you don't let any moisture into the chamber. I have no issues if i do the above.

You must be right Allinthehead, although a very heavy stretch on the images doesn't show ice streaks like the flats. I'd have thought the ice would also act as microlenses/prisms so would also give random misshapen stars.

I did change the internal tablets a month or so ago and the camera is kept indoors attached to the scope when not in use. I did use the screw on adapter with the desiccant tablets before I changed the internal tablets. The 071 doesn't come with the adapter but my 1600 did. The 071 chamber  is meant to be airtight but it seems it's not.

I'll try keeping the camera in a tight lid food container with a desiccant bag or sachets when not in use. Probably better to leave the adapter plug off initially to enable quicker drying.

Alan

[symmetal]

Having stacked in Astroart and quickly processed with Startools the Andromeda image does show the effects of icing. The dithering tended to break up the streaks and they appear more as coarse blotches predominantly at the top. The full size crop shows it more. The icing is also probably to blame for the halos on the bright stars. ASI071MC + RedCat51, 31 3 min subs. There is some RedCat coma on the left but not enough to worry too much about. Used uncooled flats.

Camera now in food container in airing cupboard along with Lunt diagonal and stack of desiccant ready to try on the next clear night, whenever that may be.

Alan