ASI533MC - Smart Histogram recommends very short exposures

[Borodog]

Just got this camera and SharpCap is recommending very short exposures, 6.8 - 12.7s. Bottle 5, f4.8. The resulting histogram is not off the left edge, either. Gain 101, brightness 5. Is this normal for this camera? This is very different from my uncooled 183MC.

[admin]

Hi,

sounds reasonable (the exposure anyway)...

Comparing back to the 183, the 533 has

* 3.76 micron pixels, not 2.4 - each pixel covers 2.4 times as much area
* A read noise of 1.5 at gain 101, compared to about 2.2 for the 183

These both count in the favour of shorter exposures - I estimate a sky background brightness of 1.75e/pix/s for your setup, so with a read noise of 1.5, an exposure of 12.8s comes out as the minimum sub length.

Can't comment so easily on the brightness - I expect that a careful look at the histogram (turn on the saving of a CSV histogram, then snapshot an image and look at the CSV) will reveal that the histogram is - just - separated from the left hand side, even if that is not clear to see in the on-screen graph.

Remember you can go *longer* sub length than recommendation if you want (until you get to the point where your mount guiding/tracking becomes an issue).

cheers,

Robin

[Borodog]

Thanks, Robin. I had done the same calculation and come up with similar numbers; it's just that I typically see people using much, much longer subs, so I wanted to double check.

My thinking had been that, if the histogram in Live Stacking is above the x axis on the left side that meant there were probably zero ADU pixels in the stack and that the something probably needed to be increased (exposure, gain, or brightness). Is that not the case?

Thanks for the help.

[admin]

Hi,

the trouble is with trying to work out separation from the graph is that it's hard to be sure...

1) The horizontal scale of the graph is such that it only displays a maximum of about 1000 brightness levels, but your camera has 16000 levels. That means the bottom 16 levels of your camera may be combined into the leftmost pixel of the graph.

2) The vertical scale of the graph is logarithmic - that means that even though you can see the line being above the x axis right at the left, the total number of pixels may be very small (<10), which isn't worth worrying about.

As I say, have a look at the histogram CSV from a captured dark in Excel or notepad - that will tell you exactly how many pixels at each brightness level.

cheers,

Robin

[Borodog]

Thank you very much.