• Images
  • First Light with new ASI585MC Pro.

w7ay I am going to try mine on my new Esprit when it arrives soon!

  • w7ay replied to this.

    Kevin_A

    Just tried unity gain -- oh boy , really noisy.

    Chen

    Kevin_A I am going to try mine on my new Esprit when it arrives soon!

    The nice thing about buying a Takahashi is you don't have to constantly "upgrade" the OTA :-). Saves money on the long run. I have yet to find any new 60-90mm scope that makes me want to open up my wallet. One of them needs to produce a chromatic spot diagram close to that from the Baby-Q (Askars are about twice the spot size, although the V series looks better).

    If I could take the weight, I would get the real FSQ-106; but as it is, I can barely carry that scope without camera and filters, and that is too much of a hassle. I have set everything up so I can just plop my scoe on the mount and not even do a polar alignment.

    Chen

    Kevin_A I tried at 25db and that was too much.

    One hour before we leave astronomical twilight, the ASI585MC-Pro at Unity gain, 180 second. 455mm-f/5.4. Already quite noisy, but workable. Just average more frames.

    Cooling of the small camera is no sweat at all. 8ºC ambient, -10ºC with just 25% cooler power consumption.

    Chen

    1/2 hour before leaving astronomical twilight; same settings as above. I guess the previous image included a lot of sky backgroud. This one looks more promising (M51 vs sky). The dark vignette is one of my antenna masts. Once it clears and I leave twilight, I can try some autorun in ASIAIR after refocusing.

    Chen

    Just left astronomical twilight (so not completey dark quite yet). Noise appears manageable with unity gain and 180 sec exposure (around Bortle 5 for that part of the sky with some high thin clouds).

    "Feels" a bit better than my cooled ASI183MM.

    Chen

      w7ay Idid my first light at unity gain and may drop down to gain 150 but I do not like dropping too much below Unity… here is just slightly less than 3 hours of 180s subs at Unity… better than my 183 for sure. This scope has a bit of lateral CA and Unity did oversaturate bright stars.

      • w7ay replied to this.

        Kevin_A Idid my first light at unity gain and may drop down to gain 150 but I do not like dropping too much below Unity… here is just slightly less than 3 hours of 180s subs at Unity… better than my 183 for sure.

        Clouds came in just a few minutes after I started an autorun. 1 hour later, my all-sky camera shows this :-)

        (Hey, night time sky is also blue. Just so dark that you think it is black :-). You can see the dipper on the bottom left. North is to the left.) Daytime is about -11.7 EV on this system (ASI294MC with a Samyang fisheye), the above image is at +8.5 EV. 20 f-stops difference.

        Chen

        Hoo boy, this camera is sure noisy at unity gain (180 sec at 457mm-f/4.3). (Doesn't help with the Antlia RGB robbing me of the photons.)

        Stable seeing tonight. This is what I got a little earlier in the ASIAIR autofocus routine:

        Notice you can actually see a pretty clean Airy disc in the 3 second exposure. ASIAIR Auto focus actually works with this kind of plate scale (and very fast too, with such a small sensor). Temperature is also not changing too much; 0.7ºC in 90 minutes. But I will soon need to refocus the FSQ.

        Autoguiding is pretty clean too.

        Chen

          w7ay I am going to test out gain 150 as I think it will be a bit better than Unity and I need to bring up the FWC to a point where stars are not saturating at 180s. My skies are bad again so no testing for now. Gain 150 has a good balance of noise and fwc and it is not so bad that photons start being dismissed due to rounding off in the electronics.

          • w7ay replied to this.

            Kevin_A My skies are bad again so no testing for now.

            I have not even created dark frames yet. That could take a while :-). Just trying to get some initial feel for this camera.

            I am going to test out gain 150 as I think it will be a bit better than Unity

            Perhaps a factor of 3 difference in full well vs unity gain, while losing not quite 50% of the quantization steps. Hate to have to make decisions like that.

            You said HCG (read noise of around 1e- from 4e-) made the total noise worse? I might take a look myself tonight if I get clear skies.

            A -20ºC (probably doable for the small camera) should be able to bring dark current noise by almost two. I should look at that too.

            Chen

              w7ay using HGC increased the hot pixels substantially only so I do not find that would be good overall. I did not find that dropping the temperature to -20c from -10c did much in reducing dark current either. I think gain 150 may be the best overall comprise.

              • w7ay replied to this.

                Kevin_A using HGC increased the hot pixels substantially only so I do not find that would be good overall.

                Back of the envelope... Dark current noise at -10ºC is about .002e-/sec, or 0.36e- with 180 sec exposure. At Unity gain, this (0.36e-) should be the actual (unity) contribution of random sensor noise, while the read noise is at a whopping 3.8e-.

                With HCG, the RMS random noise is increased by a factor of 2 (HCG kicks in at a gain of about 5.4 dB above Unity Gain), so the 0.36e- dark current noise becomes 0.72e-, while read noise falls to about 1e- : the total noise (RMS) is now about 1.25e-. This should be better than the 3.8e- at Unity gain.

                Perhaps I am missing something. If not, are we seeing mostly read noise at Unity Gain and 180 second exposure at -10ºC?

                Also, if dark current is the problem, then -20ºC should help. But if read noise is the problem, that extra 10ºC cooling will do nothing.

                Chen

                  w7ay I just tested the darks and from my visual observations which were very non scientific… I could not see much improvement in visual noise at the lower temperatures. My rationale would be also… how do hot pixels get treated in stacking? Would they have zero data and just get averaged out to a close regular pixel data? So is that just lost data that is fudged when replacing hot pixel data?
                  My goal with this camera is broadband targets like galaxies so I am ok with a 4-7k FWC but 2.5k may be too small to handle bright stars when adding all that Christmas tree light data.
                  Maybe you can test out 150, Unity and HCG and see what your thoughts are.
                  I may be wrong about dark current as I would have thought -20c would give a difference but I could not see much difference in darks but the hot pixels at HCG scared the crap out of me.

                  • w7ay replied to this.

                    Kevin_A Maybe you can test out 150, Unity and HCG and see what your thoughts are.

                    I just measured two sets of dark frames at -10ºC. One set is at Unity gain, and the second set is at HCG gain.

                    When I compare the HCG (at half the exposure time of the Unity gain, to equalize the exposure value) with Unity gain, the HCG definitely has lower standard deviation (a.k.a. RMS). Not by much, but definitely there.

                    The dark current noise is still very dominant -- i.e., you can see the noise rise with exposure time (read noise should be constant with exposure time -- one read per exposure).

                    This is just one noisy camera for long exposures.

                    From that result, I think I will try HCG tonight with 100 sec exposure. Also abandon the Antlia RGB and use the Baader Neodymium instead. Moon is going to be nasty -- with more than 60% illuminated (the neodymium might help a little).

                    The HCG gain also has slightly better dynamic range than gain 150.

                    I'm taking some -10ºC dark frames at gain 252 and exposure 100 seconds right now to prepare for tonight. I still don't have dark and bias for last night's Unity gain and 180 sec. So, can't process those yet.

                    Right now, I am not sure it is worth getting the smaller pixels vs using the ASI2600. I can always teleextend the FSQ-85 by 1.5x for the larger 2600 pixels -- slows it down to an f/8 though. I think the 585 is just a cheap cooled camera - better than nothing, but not competitive with the better sensors.

                    BTW, does it irritate you that the long edge of the sensor, relative to the USB connectors, is 90 degrees from the ASI2600 orientation. No adult supervision.

                    Chen

                      w7ay I use a rotator on my scopes so not a problem but why would they do such a stupid portrait orientation. WTF! Haha

                      • w7ay replied to this.

                        Kevin_A I use a rotator on my scopes so not a problem but why would they do such a stupid portrait orientation.

                        Built into the FSQ-85, so no problem either for me. Just that it is really poor engineering.

                        By the way, when you increased the gain from Unity to HCG, did you also reduce the exposure time by about a factor of two? Otherwise it is not a fair comparison.

                        43 dark frames collected so far. Even 32 frames reduces the noise by 3dB*5 (25 is = 32) or 15 dB. So dark frame noise is insignificant. Probably can stop at 8 (9 dB).

                        Chen

                          w7ay I did a few tests at 60, 120 and 180s on my camera. I found that 60s was ok and 120s was the max. At unity 180s looked good but I will test HGC at 90s but I think if I remember even 60s at HGC the hot pixels were plenty! I will recheck. I may try comparing 90s at HGC compared to my 180s Unity pics.

                          • w7ay replied to this.

                            Kevin_A I think if I remember even 60s at HGC the hot pixels were plenty!

                            Yuk. Too noisy (albeit there is no bias, darks or flats here, just simple stacking):

                            This is what I got in 2022 with the ASI2600MM with just 50 subframes (180s HCG):

                            I think I give up on the ASI585 and try the color ASI2600MC tonight :-).

                            Chen

                            Here was my preferences…. Unscientific noise profiles.

                            Less than 60s - HGC 252 @-10c
                            60s - Unity Gain 198/HGC 252 @-10c Tied
                            90s - Unity Gain 198 @-10c
                            120s - Unity Gain 198 @-10c
                            180s - Unity Gain 198/Gain 150 @-10c/-16c Tied

                            • w7ay replied to this.