STL-11000M: Horizontal Features in Dark Images

Dear experts,
we have been observing some strange features in dark images for our STL-11000M (S.N. 100400160).

In particular, randomly distributed horizontal structures appear in several images.
A sample image is attached and a sample fit file is available here (https://www.dropbox.com/s/sfu05ljosq0hzhv/dark.fit?dl=0)

Their thickness seems to be affected by binning and readout speed (becoming thinner with smaller binning and slower readout).
In case this could be related to grounding or power stability problems, we also noticed that pin 6 of the voltage supply cable is floating.

Any advice?
Thanks a lot in advance!

Carlo

 

Hi Carlo,

What you are seeing in your image is "typical" - these are columns of pixels that turn on "more easily" than others on the chip - due to thermal noise. Most such columns should disappear when the STL cooler has run for half an hour or so - even more of them when set for even colder temperatures. Any remaining stubborn columns can easily be removed by dark frame post processing, or by erasing the artifact using nearest pixel averaging, by command in CCDSOFT, for example.

Perhaps your STL cooler is defective - or was not actually turned on. Remember that shutting down some software like CCDOPS of CCDSOFT, asks you if the cooler should be kept running after shutdown, since you may want to switch to other imaging software, or return to this program shortly - you have the option to keep the camera still cooling, so you don't waste more time re-cooling the chip again after a short interval.

Or, the power to the camera was slightly below one of the three "critical fallback voltage" levels, so the firmware saves you (for a while), by first stopping the TEM cooling, relying only on the lower power consumption fan, so you can at least save the long exposure image in progress. Then it shuts down completely, if the power level gets worse. It is a well thought out feature for a graceful STL emergency shutdown.

You are right, pin #6 is floating. The STL Operating Manual (2003) diagram for the connector is wrong, where it shows +12 vdc on pins (1,5,6). In fact, there is even a difference in cables between the cigarette lighter plug cable version, and the A/C Power Brick cable to the camera. That little white screw on the black connector body in the latter, actually touches an internal bracket inside on the plug body, which is soldered to the A/C "cable shield" to drain any possible current leak from the STL case ... to a safe EARTH ground (third pin on your A/C power plug). There is no such possibility for a two-wire system like the battery as source.

I opened my Power Brick DC and also Battery cable version (STL end) connectors, photographed and made a PDF file showing what each of the cable plugs look like internally - (see attachment).

Hope this helps answer your two questions.
Joe Z.

 

Thanks for the very clear answers, Joe!

The images were actually shot with no cooling (or in some cases with unstable cooling). So we expected to see the vertical streaks, but thanks for confirming they are "typical".
I was maybe not clear about the features that puzzled us, but I was instead referring to the HORIZONTAL banding.
Any further insight on that?

By the way, our suspicion was that these could be related to some grounding problem - so, maybe, in the meantime, we can try to acquire further images with the STL case grounded using your suggestion!

Cheers,

Carlo

 

Hi Carlo,

Don't understand your purpose of NOT cooling the camera - you might as well just use a DSLR consumer camera, which has no cooling.

I also have terrible HORIZONTAL bands & lines on my STL-11000, and on its tracking chip, and also RGH when plugged in (in fact that one is even worse). So far I just put up with it, and apply a smoothing filter to the image. Stacking images is the same, since that averages out the bands. Dark subtraction & Flats wouldn't help, since the patterns are random between frames. The banding looks far worse when taking Flats, where they show up very clearly. I always thought they are caused by the computer "NOT keeping up" with the image download speed. Perhaps the controlling PC interrupts the stream to handle other Windows tasks. I run TeamViewer between laptops, so that may interrupt and slow down the host laptop saving images at the scope. That is why I bought an Intel-i7 processor to handle the scope & CCD, but it didn't help. Maybe raising the task priority for the imaging program or driver might help - or more memory than just 8GB might - my next attempt. Of course, that may be wishful thinking and it really is a hardware problem requiring repairs.

If it was grounding issues, I would have expected to see a "herring bone pattern" rather than straight horizontal bands. Luckily, the STL-11000 is one of - perhaps the only - astro accessory with a "properly engineered" electronics system. Unlike my AP mount, the Kendrick Heater, DIGI Serial Port Hub, and RCOS TCC controller, which just share common Battery Return by tying their "case/chassis ground" to battery return. If ANY DC accessory loses its ground return, the positive charge will seek an eventual ground on some other accessory, by going through that circuit board - often using the LOGIC ground of a DB-serial port or other communications. I have reproduced that on my system. This current leak won't reach the STL - its case is TOTALLY isolated including its comms port pins, so any case charge or current, drains off to Earth (when using its power brick).

If you are running the STL from battery cable, I would suggest, as a test, first to run the CCD off its own supply. I'll try that next time as well. Otherwise, attach a heavy ground wire to the 1/4-20 "safety screw" (camera strap) hole, at the side of the camera, and ground the other end to copper plumbing, or a long rod in the earth (less sure), or the "Earth ground pin" of an AC outlet - easy way is to do that into a long extension cord. Just be careful not to plug into the HOT pin. Taking a Flat should show you if STL grounding was a possible cause.

However, I suspect it may more likely be a camera bug that SBIG has solved in the past. Hope you find the cause and report back your solution.

Joe Z.

 

Hi Carlo,

As I was reading some chapters in the "CCDSOFT USER GUIDE" (Ver 5 Rev 1.11) tonight, I ran into an explanation of the Horizontal Banding you mentioned. Download the PDF manual form SBIG and on page 205 there is a sample image (Ch #7 Image Reduction) showing the venetian blind type banding and the comment:

" Also prominent in figure 15 are horizontal bars. These are natural artifacts of some anti-blooming CCD detectors. They are completely removed by a good flat field.
SEE: Figure 15: An example of a flat field taken with an anti-blooming camera "​

However those thin horizontal bands are both "regularly spaced and perfectly symmetric" compared to my Flats, which vary in thickness and position and with every image taken. I wonder if your bands are as uniform as the sample image indicates. There may be some other cause to the bands besides just the result of the STL-11000 "Anti-Blooming Gate".

Since my image banding varies between frames, I don't think Flat Fielding will remove the artefact, but it may mask it by losing resolution while processing the image.

Joe Z.