Jump to content


Photo

Which camera?

  • Please log in to reply
70 replies to this topic

#1 norden

norden

    Mariner 2

  • -----
  • Posts: 232
  • Joined: 08 Sep 2008
  • Loc: Norden

Posted 14 February 2013 - 10:36 AM

Hello friends,
I am an astronomy interested man who lives in Sweden, far up in cold northern Europe, I am a very happy owner of a Celestron CPC100 SCT. I love it much and I use every clear night. I devote myself exclusively to visual observations but would like to reach a little further out into the universe and catch a little more detail than what I currently are able to do with my equipment. I have long thought about a MallinCam which I think is an excellent addition to my visual observations. I have read many posts on Cloudy Nights and studied Rocks' website thoroughly and I am sure that MallinCam is an amazing nice and useful camera suitable for my purposes. I would appreciate your opinion and I would like to get some advice from you.

My SCT telescope is mounted on the original ALT / AZ mount so I think exposures of 60 seconds will be enough for my purposes due to sky rotation. Do you think HYPER PLUS is a suitable camera for me? If not, what camera do you suggest?

I'm very interested in the distant galaxy groups. Should I choose a B/W :confused:? Of course I would like a color but can I still reach faint galaxies, like Hickson 50, around magnitude 15 within 50 seconds? My observing site is rather dark, thank god.

My telescope has F10, but think about F4 is appropriate for my purposes . I would like to keep my 2 "diagonal when using the MallinCam. What special equipment do I need to buy?

Is MallinCam HYPER PLUS in Pal format and what type of screen do I need?

Grateful for any advice and suggestions.

#2 mclewis1

mclewis1

    Thread Killer

  • *****
  • Posts: 10808
  • Joined: 25 Feb 2006
  • Loc: New Brunswick, Canada

Posted 14 February 2013 - 12:21 PM

Mikael,

What is your light pollution like?

I think you will be happy with 30-60s exposures with your alt az setup. You will see a little bit of field rotation in some areas of the sky, and if you stay away from them or keep your exposures down to 30-45s when in those areas then even that won't be an issue for you.

The difference between BW and color with the Mallincams is so minimal that I wouldn't suggest the BW camera, color has many advantages and really doesn't give up anything to the BW in sensitivity.

The MCHP camera has very manual exposure and cooler settings (7/14/28/56s) and on/off for the TEC. This means that you need to be accessing the camera directly on a regular basis. This is not a big problem but it means that you won't be able to automate things (like on those really cold winter nights) in the future. You can however get the MCHP upgraded at any time. Rock Mallin offers a upgrade service which takes a MCHP and turns it into an Xtreme model for what I think is a reasonable price.

F4 on a C11 is a good target for viewing small objects like galaxies. The Meade f3.3 SCT focal reducer would be a good choice. I often use one with my C11 and it's setup like this ... C11 > f3.3 reducer > Meade T adapter with 10mm spacer (these comes with the reducer) > metal T to C mount adapter > Mallincam. In this configuration I am at about f4 (the spacing is a little shorter than what Meade specifies). Another option would be the Mallincam MFR-5 reducer which won't require any adapters (it fits into any place where a 1.25" eyepiece is used) ... it's more flexible than the Meade reducer but also more expensive.

You can get any Mallincam in PAL video format, it's just an ordering option. I would also order the ExView HAD sensor upgrade at the same time. The extra sensitivity of this sensor would help to keep your exposure as short as possible which would help with tracking and field rotation issues with your alt az mount.

You have two display options -
1) Direct to a video monitor. This is the simplest setup but you will need to find a good quality PAL video monitor with a composite (or better an S-Video) input. Ideally a good quality CRT type of monitor will give the best images but good LCD/LED monitors can be very effective as well. If it's just you viewing the images (not a group of people on a regular basis) avoid going too big, an 8-12" model would be fine and will work better than a larger display (assuming both are of the same quality).

2) Using a PC/laptop. This requires a USB frame grabber in between. These units usually handle both PAL and NTSC video automatically (so you don't have to go looking specifically for a PAL version) and start off in the $20-40 range. Mallincam also offers a couple of very nice units, the basic one is $80 USD and the more interesting + model is $150 USD. This + model offers the ability to adjust the black levels which makes viewing fine faint details (like faint galactic arms and dust lanes) much easier.

#3 rmollise

rmollise

    Hubble

  • *****
  • Posts: 15469
  • Joined: 06 Jul 2007

Posted 14 February 2013 - 01:33 PM

Everyone of us has differnt tastes, and for some objects color doesn't make too much difference. For me? I used a black and white Stellacam for years, and after a color Xtreme, I will never go back. Just being able to pick out little galaxies due to the golden color is incredible!

#4 norden

norden

    Mariner 2

  • -----
  • Posts: 232
  • Joined: 08 Sep 2008
  • Loc: Norden

Posted 14 February 2013 - 03:08 PM

Thank you for your encouraging response! I live about 200 miles east of Copenhagen in a light blue to blue area where I have a relatively dark sky. The visual limiting magnitude is approximately 6. Unfortunately, the western part of the sky is always destroyed by the light from Copenhagen. Despite the cold weather, I like being close to my telescope and sky and nature and all the sounds in the darkness. So observations from the inside of my house is not for me. So with a Mallincam MFR-5 reducer can I still use my diagonal? Which do you recommend video monitor or a laptop in terms of best and clearest images? I understand that a laptop is great because the camera then you can more easily controlled. Please explain to me why the difference between B / W and color is only marginal.

#5 GlennLeDrew

GlennLeDrew

    Voyager 1

  • *****
  • Posts: 10586
  • Joined: 17 Jun 2008
  • Loc: Ottawa, Ontario, Canada

Posted 14 February 2013 - 03:58 PM

Because field rotation is worst at the zenith, you will not be pointing your scope straight up and might be able to dispense with a diagonal. Especially if the mount's controller allows you to set a maximum altitude limit.

The advantage of leaving the diagonal out of the picture is that you will be working at a somewhat smaller focal length. (When moving the focus farther back on SCTs, the focal length and f/ratio increase somewhat.) and if you obtain a Meade f/3.3 focal reducer (recommended by me), you will not be able to use a diagonal anyway.

Regarding field rotation limits to exposure duration. This depends to a great degree on the detector's pixel count. Video cameras, having only about 330,000 pixels, are fairly 'forgiving' in this respect. And your northerly latitude helps, by allowing a full minute over most of your sky, and even up to two minutes over fair portion toward the east and west.

You may wish to consider a camera which offers longer integrations than 56 seconds. Particularly because your sky is dark, where longer exposure times will be required to bring out fainter nebulosities. And if down the road you obtain a polar aligned mount, field rotation will not be a limiter.

#6 norden

norden

    Mariner 2

  • -----
  • Posts: 232
  • Joined: 08 Sep 2008
  • Loc: Norden

Posted 14 February 2013 - 04:17 PM

Hi Glenn,

Maybe you are right. Maybe better leaving out the diagonal. Perhaps one VSS+? I have to think about it. What is the maximum angle I can use the camera without hitting the base (including Meade f/3.3 focal reducer) ?

#7 GlennLeDrew

GlennLeDrew

    Voyager 1

  • *****
  • Posts: 10586
  • Joined: 17 Jun 2008
  • Loc: Ottawa, Ontario, Canada

Posted 14 February 2013 - 06:00 PM

I would think you should be able to get to within about 20 degrees of the zenith, certainly 25. This is based on an estimate made from examining a photo on Celestron's web site. In the product specifications there is no indication of the clearance between the rear baffle opening and the base when the OTA is vertical (which there should be.)

Can you measure this distance? I'll compare it to the length of my VSS+ and Meade f/3.3 reducer combo.

#8 nytecam

nytecam

    Voyager 1

  • *****
  • Posts: 11434
  • Joined: 20 Aug 2005
  • Loc: London UK

Posted 15 February 2013 - 05:46 AM

Hello friends..... I devote myself exclusively to visual observations but would like to reach a little further out into the universe and catch a little more detail than what I currently are able to do with my equipment. I have long thought about a MallinCam.... I'm very interested in the distant galaxy groups....but can I still reach faint galaxies, like Hickson 50, around magnitude 15 within 50 seconds? My observing site is rather dark, thank god. My telescope has F10, but think about F4....Grateful for any advice and suggestions.

You're on the slippery slope once you use some form of electronic imaging and much like aperture fever there's no way back :o On the rare occasions I attend a visual ob club meeting I groan under my breath at what is barely detectable whilst from my severely LP backyard at home remote gxys pop into view in colour if I wish :grin:

Your reference to Hickson 50 had me searching my images like below in 4m stack so could be difficult in only 60s exp which just recorded gxys A and B. My DSO gallery here but as you'll see, I use a regular CCD cam and not video cam - stellar penetration is similar and I get to see the images downloaded in realtime in brief exposures, but I have more control of finished results :grin:

Attached Files



#9 mpgxsvcd

mpgxsvcd

    Surveyor 1

  • -----
  • Posts: 1951
  • Joined: 21 Dec 2011
  • Loc: Raleigh, North Carolina

Posted 15 February 2013 - 08:41 AM

Regarding field rotation limits to exposure duration. This depends to a great degree on the detector's pixel count. Video cameras, having only about 330,000 pixels, are fairly 'forgiving' in this respect. And your northerly latitude helps, by allowing a full minute over most of your sky, and even up to two minutes over fair portion toward the east and west.


Actually it depends more on the sensor size than pixel count. When displayed at 640x480 or less on a monitor a .3 megapixel 1/2" sensor will show the field rotation much more than an ASP-C sensor even with 24 megapixels. Field of view matters most here and the smaller chip will yield the smallest field of view if they are both used on the exact same scope. That will make the field rotation very apparent with the smaller chip sensor.

The nice thing is that the smaller chip can use a focal reducer to get the focal ratio down drastically. That will decrease exposure times and widen the field of view. If you already have a fast inexpensive scope like a Newtonian then the larger chip is much more forgiving. If you have a long focal length scope like an SCT then the smaller chip is definitely a better option if you also get the focal reducer.

Basically the focal reducer is an absolute must with a small chip camera and a longer focal length scope like the SCTs regardless of the resolution of the camera.

#10 norden

norden

    Mariner 2

  • -----
  • Posts: 232
  • Joined: 08 Sep 2008
  • Loc: Norden

Posted 15 February 2013 - 12:23 PM


Regarding field rotation limits to exposure duration. This depends to a great degree on the detector's pixel count. Video cameras, having only about 330,000 pixels, are fairly 'forgiving' in this respect. And your northerly latitude helps, by allowing a full minute over most of your sky, and even up to two minutes over fair portion toward the east and west.


Actually it depends more on the sensor size than pixel count. When displayed at 640x480 or less on a monitor a .3 megapixel 1/2" sensor will show the field rotation much more than an ASP-C sensor even with 24 megapixels. Field of view matters most here and the smaller chip will yield the smallest field of view if they are both used on the exact same scope. That will make the field rotation very apparent with the smaller chip sensor.

The nice thing is that the smaller chip can use a focal reducer to get the focal ratio down drastically. That will decrease expsure times and widen the field of view. If you already have a fast inexpensive scope like a Newtonian then the larger chip is much more forgiving. If you have a long focal length scope like an SCT then the smaller chip is definitely a better option if you also get the focal reducer.

Basically the focal reducer is an absolute must with a small chip camera and a longer focal length scope like the SCTs regardless of the resolution of the camera.


I guess I need a focal reducer ;)

#11 norden

norden

    Mariner 2

  • -----
  • Posts: 232
  • Joined: 08 Sep 2008
  • Loc: Norden

Posted 15 February 2013 - 12:25 PM

Hello friends..... I devote myself exclusively to visual observations but would like to reach a little further out into the universe and catch a little more detail than what I currently are able to do with my equipment. I have long thought about a MallinCam.... I'm very interested in the distant galaxy groups....but can I still reach faint galaxies, like Hickson 50, around magnitude 15 within 50 seconds? My observing site is rather dark, thank god. My telescope has F10, but think about F4....Grateful for any advice and suggestions.

You're on the slippery slope once you use some form of electronic imaging and much like aperture fever there's no way back :o On the rare occasions I attend a visual ob club meeting I groan under my breath at what is barely detectable whilst from my severely LP backyard at home remote gxys pop into view in colour if I wish :grin:

Your reference to Hickson 50 had me searching my images like below in 4m stack so could be difficult in only 60s exp which just recorded gxys A and B. My DSO gallery here but as you'll see, I use a regular CCD cam and not video cam - stellar penetration is similar and I get to see the images downloaded in realtime in brief exposures, but I have more control of finished results :grin:


I would be very happy if I could get a similar picture in 60 seconds with a MallinCam. Is that possible?

#12 rmollise

rmollise

    Hubble

  • *****
  • Posts: 15469
  • Joined: 06 Jul 2007

Posted 15 February 2013 - 01:24 PM

Yes. The Hicksons are easy with a Mallincam. And no computer frippery is required if you don't want it. ;)

#13 norden

norden

    Mariner 2

  • -----
  • Posts: 232
  • Joined: 08 Sep 2008
  • Loc: Norden

Posted 15 February 2013 - 01:43 PM

Yes. The Hicksons are easy with a Mallincam. And no computer frippery is required if you don't want it. ;)


Thats great news! Thanks Rod! Soon time to order I think :jump:

#14 jambi99

jambi99

    Mariner 2

  • -----
  • Posts: 212
  • Joined: 14 Aug 2010
  • Loc: Québec, Canada

Posted 15 February 2013 - 03:08 PM

Whats wrong with the netbook\laptop? A 7" netbook has the same size and the display quality is better than any portable lcd screen. As i said in an other post, you can add a privacy filter to overcome the laptop brightness. This will also improve the contrast:

Link

Personnaly, the ccd camera\netbook combo is a cheaper(much more) and a more versatile option. Any ccd camera with the 1/2" Exview color sensor would be as good\sensitive as the mallincan, but more than half the price.

The main advantage of using a laptop\netbook is that you can stack images(using DSS) to reduce the noise. So while, your are observing, the next exposure will automatically stack the image so at the end your keep getting a better view. Its like a two stages live view imaging process.

Plus,you can use the ccd as a regular ccd or guiding.

Following ccd model would work really well for live view:

Starlight Xpress LodeStar-c ~650$
Brightstar Mammut ~650$
Meade DSI II ~250$ used(discontinued)
Orion Start Shoot II ~250$ used (discontinued)

#15 jambi99

jambi99

    Mariner 2

  • -----
  • Posts: 212
  • Joined: 14 Aug 2010
  • Loc: Québec, Canada

Posted 15 February 2013 - 03:15 PM

I forgot to add that for camera control(for remote observation), you can use VNC by connecting the camera to the netbook and controlling it with another lapotp/pc or even a smartphone\tablet(using vnc client).

You can not only control the camera, but the telescope also.

All that is free if you consider that most people already have a laptop or smartphone\tablet. No 250$ remote control + ~100$ video transmitter.

#16 norden

norden

    Mariner 2

  • -----
  • Posts: 232
  • Joined: 08 Sep 2008
  • Loc: Norden

Posted 15 February 2013 - 04:25 PM

Whats wrong with the netbook\laptop? A 7" netbook has the same size and the display quality is better than any portable lcd screen. As i said in an other post, you can add a privacy filter to overcome the laptop brightness. This will also improve the contrast:

Link

Personnaly, the ccd camera\netbook combo is a cheaper(much more) and a more versatile option. Any ccd camera with the 1/2" Exview color sensor would be as good\sensitive as the mallincan, but more than half the price.

The main advantage of using a laptop\netbook is that you can stack images(using DSS) to reduce the noise. So while, your are observing, the next exposure will automatically stack the image so at the end your keep getting a better view. Its like a two stages live view imaging process.

Plus,you can use the ccd as a regular ccd or guiding.

Following ccd model would work really well for live view:

Starlight Xpress LodeStar-c ~650$
Brightstar Mammut ~650$
Meade DSI II ~250$ used(discontinued)
Orion Start Shoot II ~250$ used (discontinued)


Some special 1/2" Exview color sensor camera you recommend?

#17 jambi99

jambi99

    Mariner 2

  • -----
  • Posts: 212
  • Joined: 14 Aug 2010
  • Loc: Québec, Canada

Posted 15 February 2013 - 04:44 PM

Thats the thing , there is nothing special about the Exview sensor. I mean, at the end it will always be the same sensor whatever the camera. It will be the same QE. There is no such thing as special sensor. All the camera i enumerated above will have the same sensor(ICX428/429) as the mallincam. Therefore, about the same sensitivity.

#18 GlennLeDrew

GlennLeDrew

    Voyager 1

  • *****
  • Posts: 10586
  • Joined: 17 Jun 2008
  • Loc: Ottawa, Ontario, Canada

Posted 15 February 2013 - 04:58 PM

Travis,
Working it out using the math, you'll see that the pixel array count alone is an excellent indicator of field rotation tolerance.

Consider two chips having the same pixel count, but one has larger pixels (and hence a larger size.) The field rotation tolerance is identical. Why? Because the ratio of corner pixel size to distance from center is identical.

How about difference in field of view for a given chip? There in difference in field rotation tolerance, for the rotation is an angular rate. Whether zoomed in to a few arcminutes or zoomed out to many degrees, at a corner pixel the same rate of rotation exists.

Fundamentally, the only factor is this. What is the angle subtended by a corner pixel as 'seen' from the center pixel? And how does this angle compare to the angular rate of field rotation?

(To be sure there is no confusion on this matter if pixel angular size, imagine you have been shrunk down the size of a pixel and are standing on the center pixel. A cube exactly as large as a pixel has been placed on top of a corner pixel. You wish to know the angle this cube subtends as seen from your location at the chip's center.)

A low pixel count chip by definition has a corner pixel which subtends a large angle as 'seen' from the center. As pixel count goes up, the angular extent of a corner pixel must only decrease.

Example 1. Consider two chips, each having 10 micron (0.01mm) pixels. Their arrays are 640X480 and 3000X2000. Their sizes (assuming no interstitial space between pixels, for simplicity) are 6.4X4.8mm and 30X20mm, respectively.

The smaller chip has a corner pixel which subtends an angle
ARCTAN(0.01 / (SQRT(3.2^2 + 2.4^2))) = 0.143 degree

The larger chip has a corner pixel which subtends an angle
ARCTAN(0.01 / (SQRT(15^2 + 10^2))) = 0.032 degree.

The high pixel count chip is 4.5 times more sensitive to field rotation. This makes sense, for the distance from the center to corner (18mm) is 4.5X the center-to-corner distance (4mm) on the lower pixel count chip.


Example 2. We now have two chips of the same size, they being 15X10mm. One has 10 micron (0.01mm) pixels and the other 5 micron (0.005mm) pixels. Their array sizes are thus 750X500 and 1500X1000 pixels, respectively.

The big-pixel chip has a corner pixel which subtends an angle
ARCTAN(0.01 / (SQRT(7.5^2 + 5^2))) = 0.064 degree.

The small-pixel chip has a corner pixel which subtends an angle
ARCTAN(0.005 / (SQRT(7.5^2 + 5^2))) = 0.032 degree.

The small-pixel chip is twice as sensitive to field rotation, which makes sense for its pixels are half as large.


To sum up. Field rotation tolerance depends only on the corner pixel size with respect to its distance from the chip's center. And this scales as the pixel array count.

#19 Dwight J

Dwight J

    Viking 1

  • *****
  • Posts: 978
  • Joined: 14 May 2009
  • Loc: Lethbridge, Alberta, Canada

Posted 15 February 2013 - 05:23 PM

Just one problem with that list of cameras, despite all having the same sensor, is that none of them are cooled and on a warm night (or day) excessive noise noise will be evident. Some of the heat from the camera can be removed by a fan blowing on it or placing a cold pack on the camera. Cooling the camera will cost extra. The new Orion G3 camera has TEC cooling and is now on sale for $399 in either mono or color and would be a steal. I would add this camera to your list of possiblilites.

#20 mattflastro

mattflastro

    Vendor - Astrovideo Systems

  • -----
  • Posts: 622
  • Joined: 31 Jul 2009
  • Loc: Brevard County , FL

Posted 15 February 2013 - 07:22 PM

Thats the thing , there is nothing special about the Exview sensor. I mean, at the end it will always be the same sensor whatever the camera. It will be the same QE. There is no such thing as special sensor. All the camera i enumerated above will have the same sensor(ICX428/429) as the mallincam. Therefore, about the same sensitivity.

IT's true that the Exview sensor is the same in all cams.
It is however not true that this means all cameras are the same if they use the same sensor.
After the sensor comes the analog front end, which among other things does a black level clamp , an optional amplifier stage, then another optional amplifier stage , a sample and hold and an A/D converter. The AFE chips used in cameras , either video cams , mallincams, samsung, lntech, stellacam, or digital usb/firewire/IP , vary wildly.
There are AFE chips with barely 18dB of gain and lots of noise while there are others with up to 46dB of gain and less noise .
A 20dB gain difference means a 10 fold difference in sensitivity . That's 10 times shorter exposure with the same exact ccd .
For some cameas, there's too little gain or too much noise in the AFE. For others, there's too much gain that can't be manually set to the optimal value and all the image highlights are blown , image is washed out with no color. Same ccd just different AGC .
Then there are various A/D converters with various resolutions ranging from 8 bits to 16 bits. The 16 bit is ihnerently 256 times LESS noisy than the 8 bit as far as quantization noise . That means you can strech that histogram and bring detail out of the dark levels while introducing 256 times less noise if you use a 16 bit AD vs a lowly 8 bit.
But quantization noise is not the only noise or error stemming from the A/D .
Some of these chips have useless specs , for example a 12 bit AD that has the lowest 3 bits buried in the noise floor . Its internal noise is larger than the 3 LSB , making it effectively a 9 bit AD (except if a large number of samples is taken and stacked, when autocorrelation takes care of the bits buried in the noise ).
Then there are various ways of driving the ccds for the long exposures and various ways to eliminate amp glow . Depending on the camera hardware, these different methods yield different amounts of noise and amp glow .
Some cameras exhibit digital noise that's radiated or conducted by its own digital or power circuits into the analog low level signal circuits. Power supply rejection and common mode rejection varies by tens of dB between various circuits. Then you find out one cam has much more digital fixed pattern or interference noise than another at high gain or long integration.
Then there are a variety of different DSP's used to process the image . Different speeds, different bit depths , different firmware algorithms all yield vastly different results especially when it comes to noise performance . Huge difference between some Chinese 12 bit old DSP , an Analog Devices Blackfin , or an ARM based 32 bit DSP .
For example performing an image calculation and truncating the result to 1 LSB vs. calculating it to the same 1LSB but without truncating might result in 40% more noise .
Then all this wildly different hardware generates very different power consumption levels, which all generate in turn various level of internal heating . A 6 deg of temperature difference doubles the noise . Some cameras will be more than 6 deg hotter than others (ccd temp).
Then there are various ways to cool the cameras and reduce thermal noise. Peltier , or just fans, or just a big chunk of heatsink . IF there's a Peltier, then it depends how deep can it cool the ccd, what's the thermal temporal profile , all influence noise and therefore the gain that could be applied. Is it set point or ...
So you get the idea, even though the ccd might be the same, the cameras end up vastly different in performance , and not by a small amount like a difference in QE of 20% but by orders of magnitude .

#21 rmollise

rmollise

    Hubble

  • *****
  • Posts: 15469
  • Joined: 06 Jul 2007

Posted 15 February 2013 - 08:47 PM

Whats wrong with the netbook\laptop?


Not everybody wants to bring a computer in the field, and the Xtreme is simply _more sensitive_. Have you used a Mallincam?

#22 jambi99

jambi99

    Mariner 2

  • -----
  • Posts: 212
  • Joined: 14 Aug 2010
  • Loc: Québec, Canada

Posted 15 February 2013 - 10:52 PM

I just bought a used StarShoot II for 250$. The camera is cooled(-20 C) and has the ICX428 color sensor. The mammut is also cooled(-20 C). The Lodestar-c and the DSI are not, however stacking would solve that problem. Judging by Nytecame images(he has the Lodestar-C, noise it doesn't to be problem.I think cooling is a plus, but for exposure around 60s its mostly trivial...

#23 jambi99

jambi99

    Mariner 2

  • -----
  • Posts: 212
  • Joined: 14 Aug 2010
  • Loc: Québec, Canada

Posted 15 February 2013 - 11:13 PM

Why the mallincam would be more sensitive ? What are the arguments about such statement? The mallincam use the same ccd. This mean that it has the same QE. QE is the bottle neck of any ccd camera. You can't produce photon that are not detected by the senor. Of course you can raise the gain, but that will never make your camera more sensitive. It will only make your camera look like it is more sensitive.You can also raise the gain up to saturation on these cameras.

I'm not bashing the Mallincam. I think they are really good camera. However, this is a forum and people come here for advices. I don't think that it is really fair give them the impression that there is only one option. There is other product that are more than capable of providing live view. This, at a much lower cost.

#24 mattflastro

mattflastro

    Vendor - Astrovideo Systems

  • -----
  • Posts: 622
  • Joined: 31 Jul 2009
  • Loc: Brevard County , FL

Posted 16 February 2013 - 12:41 AM

Why the mallincam would be more sensitive ? What are the arguments about such statement? The mallincam use the same ccd. This mean that it has the same QE. QE is the bottle neck of any ccd camera. You can't produce photon that are not detected by the senor. Of course you can raise the gain, but that will never make your camera more sensitive. It will only make your camera look like it is more sensitive.You can also raise the gain up to saturation on these cameras.

I'm not bashing the Mallincam. I think they are really good camera. However, this is a forum and people come here for advices. I don't think that it is really fair give them the impression that there is only one option. There is other product that are more than capable of providing live view. This, at a much lower cost.

That is correct, can't produce signal without photons .
However, different cams implement different design compromises .
Let's say you wanted the shortest integration time and were willing to sacrifice some othe aspects in order to get it.
Add a 10dB gain stage before the camera's normal AFE. Price you pay for this is lower dynamic range, lower singal to noise .
What you gain is 3 times higher "perceived sensitivity" . Can an imager live with the lower signal to noise ? PRobably not .
Can a video observer live with it ?
Yes, if the advantage is you only have to wait for 20 seconds for an image instead of 1 min (and the advantage gets multiplied by the fact that you may use a dob, that could track for 20 sec but not for a while whopping minute , or an altaz mount which would show too much field rotation in 1 min , or an eq mount with bad tracking or alignment ).
Add to this the horrendous pixel scale of 6-7 arcsec/pixel that most of these perceived very sensitive cams are using .
Can an imager live with such gross undersampling? Probably not . Imagers fret over 1.7FWHM seeing or mount tracking , use AO and guiding , etc. They don't want to use 7 arcsec/pixel.
Video astronomers like it to see stars the size of golfballs . This buys them shorter focal lengths and faster f ratios , which means a lot shorter exposures. Same camera, same gain, same S/N but double the pixel scale and you get 4 times the exposure .
Couple the 4 fold gained due to the pixel scale with the 3 times gained with the lower signal to noise due to higher gain and you have a camera that now can see in 30 seconds what it takes "less sensitive" imagers 6 minutes.
In theory all cameras with the same sensor model could have the same min/max gain adjustment (manual of course) and the same best signal to noise but due to many other issues, they don't. Up to this point it appears that very few have actually pushed for the shortest integration times even with degraded image quality . And the ones which are doing it use tricks such black clipping , look at the histograms of most of these great short exposure images . The black levels are clipped like with a chainsaw . Imagers don't do that unless they want to, but video gives no such choice . If you want to see it quick, you must live with the issues.

#25 mclewis1

mclewis1

    Thread Killer

  • *****
  • Posts: 10808
  • Joined: 25 Feb 2006
  • Loc: New Brunswick, Canada

Posted 16 February 2013 - 02:13 AM

I just bought a used StarShoot II for 250$. The camera is cooled(-20 C) and has the ICX428 color sensor. The mammut is also cooled(-20 C). The Lodestar-c and the DSI are not, however stacking would solve that problem. Judging by Nytecame images(he has the Lodestar-C, noise it doesn't to be problem.I think cooling is a plus, but for exposure around 60s its mostly trivial...

I would be very careful suggesting that noise isn't a problem from Nytecam's image. Take a close look at what he said about the Hickson 50 image ...

Your reference to Hickson 50 had me searching my images like below in 4m stack so could be difficult in only 60s exp which just recorded gxys A and B. My DSO gallery here but as you'll see, I use a regular CCD cam and not video cam - stellar penetration is similar and I get to see the images downloaded in realtime in brief exposures, but I have more control of finished results

It's a single stacked image from numerous subs, and stacking is a popular method of lower noise. The image is also somewhat processed ... the capture software is performing some video processing on the images (stretching and applying various filters). So to be clear that image is NOT representative of what you'll see as a "live view" from that camera.

Maurice gets some absolutely amazing images off of that camera/software combination (the best I've ever seen) but that little un cooled camera isn't quite in the same league as the cooled models or the video based Mallincams, particularly when using longer exposures (as he clearly states).

Cooling does start to benefit images at well under 60s exposures. It certainly helps those longer exposures but it's one of the main requirements for a continuous read video setup like in the Mallincams (who's sensor generates tons more heat at any exposure than the single exposure imaging cameras).






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics