QHY5+PHD First Light (mixed success)

[MikeWilson]

Dear fellow stargazers,

Tonight (whilst my family watched Kung Fu Panda for the fifth time) I decided to give my QHY5 camera it's first light. A strange choice of night I expect you may think, given the almost 100% cloud cover that we currently have (you can't even see the moon at the moment) but I've had the QHY5 - new - for almost two months and I haven't had a chance to use it yet.

Long story short, I spent the best part of the last four hours trying to get a star to show up on PHD Guiding or GQVideo. In my indoor testing I noticed an orion-like pattern of hot white spots coming through and this was replicated outside. I tried using the PHD "take darks" function but this didn't remove the spots. Am I doing something wrong? I presume my CCD isn't defective but that I'm doing something wrong. I will try retaking the darks later if the clouds clear tonight.

PHD screenshot and 200% zoom magnification on hot pixels.

For several hours I struggled to try to see anything other than noise or the hot spots. A quick Google revealed (an SGL thread) that showed as I had attached the QHY5 to the ST80 via the t-threads, that there wasn't enough back focus on see anything anyway! I came back inside to find the 1.25" eyepiece adaptor and ST80 diagonal (I really wanted to avoid using these for stability reasons) and I found something that I'd picked up from Astroboot back in the summer - a 50mm long T-thread extension tube. Perfect! I was just able to see a star and move towards getting it in focus before the 100% cloud cover moved in. Thankfully Arcturus being the baddass star that it is was able to shine through the light cloud just in time for my eureka moment.

So in summary, a question;

1. Have any other QHY5 users found hot pixels on their CCD and if so, does the PHD "Take Darks" function adequately remove them? (If it does, any suggestions as to why it didn't work for me would be welcome)

and a point for future Googlers;

2. If you're using an ST80 and QHY5, don't forget the T-thread extension tube!

[Whippy]

I think my QHY5 has one, but it doesn't look anything like a star anyway. Plus shouldn't the calibration fail if you tried to use it as a guide star?

Tony..

[MikeWilson]

Upping the exposure to 2.5 seconds shows the hot pixels up better. There are quite a few of them, at least ten (brand new QHY5 from Rother Valley Optics). Should I be worried?

(50% gain, pointing towards Arcturus but clouds have rolled in produced this light pollution/cloud image with the hot pixels clearly visible)

[Whippy]

It does seem there's a few, are there any marks on the chip itself?

Tony..

[MikeWilson]

Hi Tony,

Thanks for your replies. I can't see any obvious marks on the chip and connecting the camera part separately confirms the hot pixels aren't part of the imaging train past the QHY5.

I took a couple of macro shots of the CCD to look for damage but can't see anything obvious. It all looks fine.

[martin_h]

Ive never used darks but I do use noise reduction.

[MikeWilson]

Thanks Martin,

I have to report that using 2x2 mean and 3x3median PHD noise reduction actually makes the problem worse and I see more 'hot pixels' than I would otherwise. (I suppose these stuck dots are more than just single pixels).

Have I got a dud QHY5?

(PHD noise reduction applied: 2x2 mean)

(with the main hot pixels circled in purple)

[mark7331]

Hi Mike, I've just been using my QHY5 tonight and finally getting pretty good guiding (10-15min subs) after playing with the settings again and again. So I thought I would take a shot of the CCD output with the lens cap on so I could see the hot pixels. I thought I had a couple but as you can see from the picture I have lots and lots of them. It doesn't cause a problem for me and I think, from the guiding I'm getting, that PHD is clever enough to ignore them.

I hope ths is useful and maybe will make you feel better about your QHY5.

All the best

Mark

EDIT: taken with 2x2 mean noise reduction. Second image added to show hot pixels with most of the dark noise scaled out.

[martin_h]

I assume I have hot pixels on mine as well but they have never been a problem (for me) I always choose a good round star to guide on - fairly bright but never the brightest. I dont think I've ever clicked on a pixel.

Just a thought but what is your camera gain set at? mine is at 95%

[mark7331]

My gain, if it helps, is set at 50% in the properties when I connect the ASCOM camera but it shows 95% in PHD.

[MikeWilson]

Thanks chaps,

My gain is set at 50% in the ASCOM properties and I didn't take note of what PHD reported the camera gain at. It's clearly high and shows the 'rolling shutter' banding effect.

I'll have to try guiding with it to see if my QHY5 isn't fit in practice. I'm concerned that I may not be able to tell the difference between a star or a hot pixel (given how many of them there are).

I've dropped an email to Rother Valley Optics just in case they know of similar issues with any other QHY5's they may have sold around the same time (mine might be in a bad batch?).

Just waiting for a clear night to try it out...

[martin_h]

A star will be bigger than a pixel.

[MikeWilson]

I had a call from Bern at Modern Astronomy today. A very helpful and knowledgeable chap who explained that all these CCD's (especially the cheaper uncooled ones) will have hot pixels. Hopefully the dark frame subtraction will reduce them (I noticed that PHD 'stores' dark frames and you need to create new dark frames for each exposure duration just prior to guiding) and as you say Martin, the guide stars should look totally different to the hot pixels.

I expect I'm worrying about nothing but will let everyone know how I get on.

[martin_h]

Not a fair comparison I know but this is my QHY5 through a ST80 on the double cluster.....spot the stars!

Taken about an hour ago.

[MikeWilson]

That's actually rather helpful, thanks Martin!

(by the way - where did you get your clear skies from?)

[ddtwelve12]

hi ive only used phd a couple of time as well as the qhy5 cam but when phd calibrates itsself to your mount it actually moves the mount in both the axis for calibration so if it accidently picks a hot pixel it wont calibrate itself as they wont move

[martin_h]

That's actually rather helpful, thanks Martin!

(by the way - where did you get your clear skies from?)

They have been on back order since August 8th last year - finaly deliverd today. :)

[Nexus 6]

Hi Mike glad to see you have managed some imaging, I can confirm that Martin has a pretty awfull place from which to image, I went to see him to pick up an item I had bought, I was astounded he can attain such great data from his location

[ddtwelve12]

well ive given up hope so all my gears up for sale

[jgs001]

I do get the hot pixels show up with my 5v, but... having said that, There's something going on with PHD, that seems to show more noise and hot pixels when there aren't sharp stars on screen (almost like some form of auto screen stretch) that makes them appear worse. When I've got focused stars on the screen, the noise and hot pixels practically disappear. I do use the PHD darks feature. I get this effect if the FOV is devoid of stars or I'm adjusting the position whilst looping. I've found that 1 second loops is normally sufficient. I can also tell how much dew is forming on the objectives by looking at the PHD display... the more noise the more dew.

[JayStar]

Seen a lot of the same problems - with a screen full of noise (literally thousands of pixels). I finally managed to get something sensible by turning the gain and gamma right down to 15-20 in the camera setting and the exposure in the camera settings down to about 250ms.  ( I am not sure how the exposure setting in the camera setting options actually affects guiding?  The main window is still set to 2 seconds...)

This was done before taking the dark frames.  Then using 2*2 noise reduction I managed to get a noise free screen when exposing with the lens cap on. I am only trying this in the day time at the moment, but when I then take the lens cap off I do get a normal image coming through, so hoping now I have it tuned to pick up stars when out imaging.

I am using phd2 and an ASI120MC for guiding.

[JayStar]

Correction, I am using 3*3 noise reduction to get a completely noise free image and Gamma is set to 40 with gain between 15-20.  I find that if the gamma is too low the image can flick between white and black.  This latter observation also seems to linked to exposure times - the lower the exposure, the higher the gamma needs to be set to avoid the black frame/ white frame issue.