Whilst trying to get some RGB stars for Mike for my recent NGC 3324 image I have encountered a problem with very saturated stars which I'm not sure I've seen before.
In the crop of the image below you can see running up from the bright star a series of dotted lines. There are three bright stars in the image and they all exhibit this behaviour.
It has shown up in some long Ha shots as well last night.
I don't think that I have ever seen this before with this CCD, I performed a driver update the other day. Is this something I can fix, or has it always been there I've just never seen it.
I occasionally see this with the QSI683 but only on one or two mono frames. Never repeatable to create a coloured trail. For the most part it comes out with calibration and data rejection.
I occasionally see this with the QSI683 but only on one or two mono frames. Never repeatable to create a coloured trail. For the most part it comes out with calibration and data rejection.
Hi Paul,
As you can see, it's repeatable, so much so that the LRGB subs all align and then it looks like real signal to the software so no data rejection will remove it. Calibration doesn't fix it because it's only in the light frames and moves with the stars.
I might try downgrading the drivers that I updated. I have also noticed the USB download speed is slower than it used to be.
I thought you were Sbig in Aus? I have sent David Morrow an email.
Cheers
Stuart
Yes, you could have PM'd me some .fits examples and, not having seen this before , I would have liased with SBIG...but that takes time. Quicker to ask David.
While that's not ideal, wouldn't dithering calibrate the data out?
No it doesn't as the lines move with the stars, hence they dither as well. That was my hope as well, but the lines appear to be real signal to software.
No it doesn't as the lines move with the stars, hence they dither as well. That was my hope as well, but the lines appear to be real signal to software.
Cheers
Stuart
If it started after you updated the driver then that is a logical place to start.
I have never seen this before on my ML8300 or any CCD for that matter.
No it doesn't as the lines move with the stars, hence they dither as well. That was my hope as well, but the lines appear to be real signal to software.
Cheers
Stuart
Just to be sure.....does this artifact appear when you use (SBIG's native) CCDops?
David Morrow has contacted me and I have sent some fits. I rolled back the driver (I think) by reinstalling a previous version, though I have to confess that with a Mac I'm not sure which driver is being used. David was confused by the lines being above the star, I think CCDStack automatically flips them from memory.
Next clear night I have to check to see if that worked, too smoky here last night.
Just to be sure.....does this artifact appear when you use (SBIG's native) CCDops?
Hi Peter,
Can't remember how to drive CCDops! From the further information I have found out I would need to take a 30 second guided image. Care to write out a simple protocol for me?
Can't remember how to drive CCDops! From the further information I have found out I would need to take a 30 second guided image. Care to write out a simple protocol for me?
Cheers
Stuart
Power-up camera
Start CCDops, close auto-launching differential guider dialog
Camera>setup> Set temp, guide & imaging CCD res.
Wait for CCD to cool & power setting to stabilize.
Camera>focus ....planet mode works best. Focus telescope as image updates.
Track>calibrate Set exposure time (~ 1 sec) motion vector time (~ 5 sec) & run.
Check for success, if no, check guide star present, or not too close to edge, or for similarly bright star(s) in same field, guide cable plugged in, dust cap removed, dome open, not cloudy.
Slew to desired imaging field, Track>Self Guide> set desired exposure time and correction interval.
Select start. Watch guide star/or Star Trek until exposure is complete.
After exposure is complete, save image.
Power-up camera
Start CCDops, close auto-launching differential guider dialog
Camera>setup> Set temp, guide & imaging CCD res.
Wait for CCD to cool & power setting to stabilize.
Camera>focus ....planet mode works best. Focus telescope as image updates.
Track>calibrate Set exposure time (~ 1 sec) motion vector time (~ 5 sec) & run.
Check for success, if no, check guide star present, or not too close to edge, or for similarly bright star(s) in same field, guide cable plugged in, dust cap removed, dome open, not cloudy.
Slew to desired imaging field, Track>Self Guide> set desired exposure time and correction interval.
Select start. Watch guide star/or Star Trek until exposure is complete.
After exposure is complete, save image.
If this doesn't make sense, RTFM!!!
Thanks Peter, it was easy once you know how!!
Anyway, just posted this over at the SB website...
OK, after much testing I believe that I have confirmed that this is a software problem.
On my Mac i can run Windows under VMware, or natively through bootcamp, of course I can run MacOS Mavericks. So for the sake of brevity, these will be abbreviated to W7V, W7N and OSX respectively.
Firstly to test the camera hardware I took both 10 and 30 second guided exposures with CCDops in W7N, neither had streaks. Immediately after this I use TSX (10.2.0 B7477) running under W7N to acquire the same images, again, no streaks. Rebooted computer to OSX, retook images with TSX under OSX, streaks in both guided images. Ran TSX under W7V, both images again had streaks.
Thinking that I may have different builds of TSX under W7N and W7V, I checked this and updated the TSX version (it is shared BTW) to build 7715. Again I had streaks in TSX W7V. When the computer was rebooted to W7N, the streaks were now present in the images. I reran CCDops, no streaks. I reran TSX, no streaks.
It seems that running CCDops corrects the issue, I can then use TSX running under W7N to acquire streak free images.
Re booted the computer to OSX, without starting TSX under OSX, I tried TSX under W7V, the streaks had returned. I started CCDops under W7V, no streaks in either guided image. Retried TSX under W7V (without shutting the program down, just disconnecting and reconnecting the camera), the streaks have returned.
I did not try running CCDops under W7V and TSX under OSX as I though that there may be guiding issues running across operating systems, but now I think of it the camera would guide using the relays, so I may be able to do this, though I do not see this as a solution.
I installed the latest build of TSX for OSX, this is totally unusable as the guiding is broken, I take a guide image, try to select a star (or have the software autoselect) and I get a dialogue box saying "Please take photo so the guide star can be automatically found", there are also three nested guider display windows.
I tried to take an image, then before the exposure was finished covered the scope up to test if this is a shutter problem, it did not change the appearance of the streaks, so I doubt that this is the problem.
I think that I'll just have to reinstall TSX up to two builds ago to get around this problem.
And, what must be frustrating is that you put all this up and so far the only "official" response is a totally irrelevant comment about CCDSoft (which you were not even using). I feel guilty for having put up my simple suggestion because it seems only the "short" posts get attention, and your massive post is not even read. FRUSTRATING!!!!! I went through this for months trying to prove my MX issue....It's a vexing support system.
Yeah, their customer service is, quite frankly, Shiesenhousen. Their method of requesting assistance is a model of how NOT to set it up. When they do help, they are usually quite helpful, it's just making enough noise to be heard that is frustrating. Maybe it's an American thing.
Anyone have all the daily builds for Mac archived so I can roll back the software??
Now they've removed one of my replies to Daniel's inane post about the latest Mac daily build being broken on another thread. They've also removed Daniel's reply about getting to the root cause. I was about to edit mine to say that instead of posting this they could have checked the &*(Y*^*%&^% thing before dissing one of their customers.