Bug in SharpCap or a ZWO 585MC driver glitch

[ChrisR Oz]

Hi Robin,

I was using a Hubble Optics synthetic star lamp (an LED torch with 5 pinholes of various sizes) to test out a new Samyang 85mm T1.5 lens I plan to use as an (electronic) Finder (guide) scope. I used a ZWO 585MC camera in ROI mode. I found that as I panned the ROI around to centre the pinhole images, the image would flash occasionally. So I started a capture to SER and moved the ROI around. Then I converted the SER to FITS files and went digging. The flashes are worse when moving in the up/down "tilt" direction. Most normal frames look like this ...

A normal frame

Normal.jpg (20.04 KiB) Viewed 717 times

The 'flash' frames look like this ...

A bad "flash" frame

Flash.jpg (24.94 KiB) Viewed 717 times

The histograms are also very different. The normal frames have a typical histogram with a few saturated pixels in the "stars". The 'flash' frames have heavily saturated/clipped histograms, as shown in the garish colours above.

I have put a normal and a 'flash'/bad frame FITS file on Google drive at ...
https://drive.google.com/drive/folders/ ... drive_link

The question is, what's causing this? Is it a glitch in the ZWO driver or 585MC FPGA code or a possible hiccup in SharpCap?

Thought you'd find it interesting ...
Cheers, Chris.

[admin]

Hi Chris,

this kind of glitch when you change some camera parameter is relatively common - some brands of camera can be observed doing it when the camera exposure or gain is changed, some may show image 'tearing' when the ROI is moved. Typically you get one bad frame between the 'old' settings frames and the new settings frames. My assumption is that since the camera is taking images continuously, sometimes the new settings will arrive at a point which means they put the camera into an odd state (inconsistent) for a frame, at other times they will arrive at a convenient point in time (between frames) and no glitch is seen.

I don't think there's any need to worry about this unless you are doing things like processing the images from the camera in real time, in which case you need to be prepared to ignore dodgy frames just after settings changes. SharpCap does this internally for things like the sensor analysis.

cheers,

Robin

[ChrisR Oz]

Thanks for the info Robin.

If SharpCap is driving the ROI for planet centring, could these frames pop up in a captured SER? Or do you time ROI offset changes to avoid it?
Cheers, Chris.

[admin]

Hi Chris,

there's no real way to time the changes from SharpCap's view of the camera - maybe if you are the software developers at ZWO working with the full documentation of the sensor you could work out how to do it, but I don't have that sort of detail available to me.

So, that means that there is definitely a chance of that sort of frame being in the captured SER - I would expect the stacking application to reject it, as if it is badly overexposed relative to all the rest, it will have minimal detail and therefore be rated as poor quality.

cheers,

Robin