First light with Atik 314L+... and a bit lost!

[AndyUK]

Hi Dennis - Thanks for your comments - FLO have already sent on my frames to Atik and we're just waiting to hear back from them now. Regardless, I'll be sending this one back though.

The 314 is indeed a USB2 device - The cable was / is the one that came with it and is 3m long (which is quite generous as the manual said it was supplied with a 2m cable!). I do appreciate that the banding present in the darks would be "taken out" by the bias, but it was those bright patches that I was more concerned about - Even if they can also be subtracted, it's surely reducing the chip sensitivity in that area to "real" data...?

Also I do accept that, in reality, DSLR's aren't REALLY made for prime Astro imaging, but I guess this b*ggeration factor has really got to me... I could really do without it with everything else that's going on at the moment!

[FLO]

FLO have already sent on my frames to Atik and we're just waiting to hear back from them now. Regardless, I'll be sending this one back though.

A replacement is being organised

Normally we'd have this done and dusted in about two days but it will be a little longer in this instance as we recently sold out and our next delivery isn't due til early next week.

Thank-you for your patience.

[AndyUK]

Cheers Steve - Martin did indeed allude to it, so thanks again... Not having used a CCD camera before, I had no idea what to expect so didn't really know what was "the norm". To be honest though, I've learnt a lot from this exercise (and Dennis!) so I'll chalk this up to experience... and being "just one of those things"!

(I'd still be interested in hearing what Atik say the issue is... assuming there is one?)

[ollypenrice]

Just get back to Atik. They are helpful and respect your rights as a consumer.

I would certainly try a different USB cable though, because they have been the root of much, though not all, evil in my experience! I change them regularly.

BTW, if BrianB gets going on USBs put your tin hat on and get down!!!

He's quite right though...

Olly

[swag72]

I'm really glad it's being sorted for you Andy. It would be interesting to hear what the issue is with that camera.

[roundycat]

Andy,

quote;

I do appreciate that the banding present in the darks would be "taken out" by the bias, but it was those bright patches that I was more concerned about - Even if they can also be subtracted, it's surely reducing the chip sensitivity in that area to "real" data...?

Bias dealt with the banding, no problem. The difference between the band and the adjacent area was about 10ADU in 4000 but the bright area was reading over 13,000ADU.

Whatever the bright area, so long as it is not saturating, it can still pick up signal. Provided there is nothing reaching saturation at the end of the exposure the Bias will take care of the bright patch.

Dennis

[FLO]

I'd still be interested in hearing what Atik say the issue is... assuming there is one?

So will we, we haven't experienced a faulty Atik 314L plus before. We won't wait on their verdict though, we'll send you a replacement as soon as our next delivery arrives. When Atik report their findings we will be sure to let you know

We'll also be sending you something for the inconvenience.

Steve

[AndyUK]

Cheers Olly - I will try it with a different USB cable (just in case). Not knowing enough about USB (or RS232 or whatever), I think I'll be steering clear of discussing PC connectivity with anybody..!

Thanks again Dennis - My concern was that I was already potentially "losing" c. 20%... Again, maybe I'm making a mountain out of a molehill, but comparing them to those other frames provided by fellow SGL members (same temp/exposure and viewed in same software with same parameters), mine seemed quite poor in comparison.

And thanks again Steve - I really do greatly appreciate and applaud your customer service ! However, as mentioned above, I just hope that I'm not making a mountain out of a molehill () ... or maybe it might turn out to be related to a "hot" USB cable... (?) connection...

[AndyUK]

Okay... I finally got the courage to get some sample frames from my replacement 314L+ (thanks Steve!). Here's a straight 600s dark (-5C, no processing... nothing). Can I ask for any comments on this, please?

[roundycat]

Unless the camera outputs jpeg's you must have done something to it!

It looks very good, almost impossibly good, but with possible re-sizing the hot pixels could have been partly neutralised by interpolation.

Any chance you can send me a dark and a bias?

Dennis

[shaunster]

Well the 285 chip is so good a lot of people don't use darks anyway

[AndyUK]

Hi Dennis - Okay, yes, I had to convert it to jpg so that it was under 2Mb, but that was all... However (again), I'd forgotten about trying zip files (please find attached).

I have to say, compared with the comparable bias and dark from the last camera (same settings), this IS MUCH better, but should I be concerned about all those apparent rows of hot pixels (the white line) at the top of the image? I know I can crop it out, but for a new camera and having seen darks from 3 other 314L+'s, none of those seem to show it... However, again, maybe it's just my paranoia...!

BIAS_0.1s_0010.1_BIAS.zip

Dark 600s.zip

[roundycat]

Seems like an improvement. The camera obviously has a high level of Bias by design, don't know why that is the case. The light strip across the top is a mystery but no isolated blobs of light in this one. You can see from the bias subtracted dark (top right) that the bright strip is dealt with.

That suggests to me that whilst the light strip will be in the lights (as a camera characteristic?) it will be dealt with in routine calibration. People are always keen to say if the chip is quiet don't bother with darks. The reason you should ignore this advice is in the picture below. The bias will always be present and is taken care of by dark subtraction. If you think of getting rid of the light strip by bias subtraction instead you will still have some hot pixels to deal with.

With such a quiet chip it might be feasible to only take four or five darks to make the master. Either way, they are done on a cloudy night when you have a temperature controlled camera.

In the picture below the top left is the calibrated dark, it had bias subtracted - nothing else.

Dennis

[mark7331]

Andy, Dennis.

Firstly please tell me to be quiet if I'm talking nonsense

I've had a look at Andy's data and compared them to mine. I've found a much higher signal in Andy's fits files than in my camera. I've just taken a bias frame to match the one Andy attached above. This is at zero degrees (not sure if yours was at zero Andy - could this be the reason) and is 0.1s in length. The output data is below. I'm just wondering here, while I know the first CCD was faulty, is there an issue with either your PC or the version of Artemis that you're using?

A bias frame from my camera

Min ............ 156 @ 1168 , 60

Max ............ 442 @ 473 , 900

Background ..... 250

Average ........ 250.88736

Sum ............ 362594705

Std.deviation .. 19.332448

And from Andy's:-

Min ............ 4207 @ 268 , 1033

Max ............ 65535 @ 1390 , 0

Background ..... 4905

Average ........ 4964.6395

Sum ............ 7175140285

Std.deviation .. 271.7935

I also updated my 314L+ firmware when I first had it. Don't know if that helps.

Mark

[AndyUK]

Thanks for this Mark - Wow... That's quite a difference... and I certainly won't be telling you to be quiet as I'm now getting totally confused!

I've agreed with FLO that it's probably best if I liaise direct with Atik so that I can pass files direct to them as opposed to via FLO, but your figures appear to be quite... well, almost damning. My frames were actually both at -5C so in theory one might expect even lower values than the ones you've noted for yours.

Admittedly, I haven't checked the firmware, but this latest version was shipped direct from Atik last Friday... (I will check it though)

Dennis - Thanks for your input too. The fact that the light strip will also be in the lights (at that saturation)... Although I can see that subtracting the darks should remove it, it appears to be at full saturation - Wouldn't that mean I'd have no signal in that strip at all and therefore would have no option but to crop it out?

(I think I've gone through the despression stage on this now - It's now become a challenge...!)

Cheers both...

[AndyUK]

Ah.... The plot thickens... or maybe I do...

Steve from Atik came back and asked me to re-run the bias/dark, ensuring that the pre checkbox was unticked... I think this may explain the huge variance in the figures above and also what Dennis was seeing as a high bias signal as the resultant dark frame showed no white band at all (), and the figures I got out of Fits Liberator (Mark - Not sure where you accessed that info?) was as per attached.

(I think I'm happy now... )

[mark7331]

Andy that looks good to me! I can confirm that if I check "pre" I get a load of noise at the bottom of the image just like you did.

FYI: I pulled the figures out of AstroArt.

I'm glad you got it resolved mate.

Mark

[AndyUK]

Cheers Mark - Thanks for confirming that... I don't mind telling you, it's a HUGE relief!!! (Any idea what that pre button's for?) Now I can move on to the next issue - achieving focus with the MN190 (I think I have it sorted, but just need a clear night to test it)

[roundycat]

Being a MaxIm man I find some of this other software difficult to understand. What is this 'pre' check box all about? The really odd thing to me is the fact that Artemis capture only writes limited info into the FITS header. It doesn't even distinguish between darks and bias. Neither the dark nor the bias have an entry under the heading 'Image Type' in the FITS header, in fact the Image Type entry is not even there. How does processing software know what to do if the frames are not named by image type?

For the record, I measured these figures:-

Dark at pixel 1390, 10; 65,535ADU. (It doesn't get higher than that in 16bit).

Bias at pixel 1390, 10; 5799ADU.

Applying the calibration (Dark - Bias) in Maxim this pixel (and all its neighbours) came down to 0. Note, 65,535-5799 = 59,736. Not zero!

I have attached a screen grab showing what part of a MaxIm FITS header looks like with an SBIG camera. Useful info such as Image Type is clearly missing. Something to take up with Atik?

Don't give up, the answer is out there. Somewhere!

Dennis

[FLO]

I'm glad you got it resolved mate.

So are we! It is a relief and an embarrassment to discover the solution was as simple as a tick in a check-box

Steve

[mark7331]

Andy, Dennis. The "pre" check box (from the full manual not the little one with the camera):-

The "Pre" check box, activates the preview mode, where that image download will be about twice as fast. However, quality will suffer. This mode is specially useful for focusing the camera.

I can't say I've ever used this mode as I focus with subs of between 0.5s and 2s without any problems. It's meant to be clear tonight so I'll give it a go.

I think the issues that Dennis highlights lie with the capture program supplied with the 314 - Artemis - which isn't a really powerful tool. it's cheap and cheerful and so misses out on some of the features you'd expect from Maxim and even AstroArt (which has FITs header issues too).

@Andy have a look at this thread regarding focusing. I use a 'Y' mask cut from a bit of thick card and it works perfectly.

http://stargazerslounge.com/beginners-help-advice/140758-focussing-masks.html

Mark

[AndyUK]

Cheers Dennis - That SBIG FITS header is a lot more explanatory isn't it...?! When I first opened the Atik header I was expecting to see something like I get from my raw DSLR EXIF viewer results (3 pages of imformation), but the Atik output does look a little scant in comparison

From what I've read about Maxim (please correct me if I'm wrong!) you effectively "dump" all the files into one directory and Maxim then pulls out what image types there are from the FITS header - If the image type isn't in the header, then presumably this wouldn't work...?

I've asked for further info on the "pre" option - In summary Atik have advised that it selects the fast lower quality mode, but I'm not sure when (if ever?) you'd need to select it... Maybe framing / focusing perhaps?

EDIT - Sorry Mark - Crossed over...!

[ollypenrice]

Cheap and cheerful Artemis may be but an awful lot of people love it. I do. I have never touched this Pre thing though!

Olly

[FLO]

Andy, Atik will have received your first 314L today so we should soon have their report. We'll let you know what they find. We accept that a percentage of everything is faulty but we have not come across a faulty 314L before so are keen to hear what they say.

Steve

[AndyUK]

Hi Olly - Yes, I like things that are Idiot proof and simple to use... I've heard quite a few people say that the Artemis GUI is very easy to pick up and when I first saw it with its limited options, I thought how difficult can this be...? And yet I still found that pre option and for some reason decided to switch it on... I guess I must be more of an idiot than I thought...!

Still, FITS headers aside, as long as it enables me to focus, frame and then captures the files, I think I'll be very happy indeed. I'm sure that Maxim is VERY powerful, but aside from the cost, I'm not sure my brain has got the capacity for something that big at the moment...

Steve - Thanks for the update - I just hope it wasn't another option somewhere (or indeed the same one )