As a scientific imager or astrophotography enthusiast, we hope you'll enjoy this brief newsletter. We occasionally send out announcements and information that we think you'll appreciate, but if you prefer not to receive these messages, you can unsubscribe by following the link at the bottom of the page.
With this newsletter, we're beginning a series of technical articles we've titled, "Understanding CCDs". The first article in the series is about Dark Current versus noise. Read the full article below.
In this Issue...
1. Take Delivery of an Award-Winning QSI 583 in just days!
2. Compare the Total Cost!
3. Don't Take Our Word For It
4. Understanding CCDs ??Dark Current versus Noise
5. Astronomy Magazine Picture of the Day
6. New QSI Dealers
7. QSI in the News
8. Upcoming Events
Take Delivery of an Award-Winning QSI 583 in just days!
The 8.3mp Kodak KAF-8300 has revolutionized the market for dedicated cooled CCD cameras with a simple equation - More pixels for less money. But that has also created a problem, the KAF-8300 is now in short supply. QSI anticipated the demand and placed large orders for sensors months ago so we have plenty of sensors in inventory and a steady supply of more KAF-8300's scheduled to arrive over the coming months.
The QSI 583 has become one of the top selling dedicated astronomy cameras with a combination of exceptional imaging performance, innovative features and great value. Order a QSI 583 today and image with it all summer long.
QSI 583 http://www.qsimaging.com/583-overview.html
Lead time direct from QSI is just 2 to 3 weeks. Some QSI dealers may have camera in stock for immediate delivery. Check with your preferred QSI dealer today!
Here are some of the reasons why a QSI 583 should be your next camera:
a.. Most Awarded KAF-8300 camera!
b.. Lowest noise allows you to capture higher quality images in less time.
c.. Fast 2-second Focus Mode for quick, easy focusing.
d.. Built-in color filter wheel provides unvignetted images to f/5 with standard 1.25" filters or all the way to f/2.8 with low cost 31mm filters.
You don't need larger, costlier filters required with external filter wheels.
e.. Shortest back focus with a color filter wheel
f.. Use Canon or Nikon lenses for wide field imaging
g.. Full support for 32-bit and 64-bit Windows 7, Linux and Macintosh.
h.. QSI 583wsg adds the Integrated Guiding Port for guiding with your main scope while providing 100% of the available light for guiding.
No more low SNR guide stars while trying to guide through color filters.
i.. Unbeatable value. Compare the total price of the QSI 583 to any complete camera system including the color filter wheel, filters and possibly an off-axis guider. We're sure you'll find that the QSI 583 is the best combination of performance and value in a cooled, 16-bit scientific camera.
Learn more about the benefits of QSI cameras>>
Compare the Total Cost!
If you're also looking at cameras from other manufacturers be sure you're considering the total cost of configuring a complete camera system. The 583ws and 583wsg include an internal color filter wheel that can use low cost and widely available 1.25" filters to deliver unvignetted images with optical systems as fast as f/5.
We also offer the optional 31mm filter wheel that provides unvignetted images with the 583 as fast as f/2.8. Even with a Hyperstar at f/2, the small amount of vignetting is easily correctable with a flat field. Astrodon 31mm filters are the same price as 1.25" filters. This can save hundreds of dollars over the larger 36mm or 48mm filters required with other KAF-8300 cameras. Compare the total cost and we're sure you'll agree that the 583 is the best value in a cooled CCD camera.
31mm Astronomik filters Now Available!
A complete set of Astronomik 31mm LRGB filters are now available for just $299 with the purchase of a QSI 583 making the total cost of a QSI 583 with internal filter wheel and filters of just $3,489! Just select the Astronomik 31mm LRGB Filter option when configuring your camera on the QSI 583 Order page:
Don't Take Our Word For It
Sky & Telescope magazine selected the QSI 583 as a "Hot Product for 2010". This after the WSG cover was selected as a Hot Product for 2009. Now AstroPhoto Insight has awarded the QSI 583 as the "Stellar Performer for 2009."
Read more on the QSI 500 Series Reviews page, where you'll also find a link to another extremely positive review of the QSI 583 that was recently published in AstroPhoto Insight magazine.
Understanding CCDs - Dark Current versus Noise
This is the first in a series of technical articles discussing the operation of CCDs and CCD cameras in more detail.
With warm summer nights approaching, we've seen more questions about the benefits of cooling the sensor in a CCD camera, how much is necessary, and what effect another 5C or 10C drop will have on noise. Most of the information in this article was previously posted in the QSI discussion group, QSI-ccd, but it generated so much interest, we figured it would be a good subject for the newsletter.
First, thermal, or dark current builds up in a CCD at a predictable rate whether it is being exposed to light or not. CCD Imagers exploit this fact to remove the thermal current that builds up in our light frames by subtracting a dark frame.
Thermal current is measured by subtracting a bias frame from a dark frame. The difference between the dark frame and the bias frame is a result of the thermal current. Now, click on the link below to look at two slides below from a presentation given last year at MWAIC showing Mean Thermal Current and Max Thermal Current over a wide range of temperatures and exposure lengths. This particular data was acquired with a QSI 516, which has a Kodak KAF-1603ME sensor, but a similar effect will be seen with most other popular sensors.
Mean and Max Thermal Current:
And here's a reduced set of mean thermal data from a QSI 583:
The key thing to note is that the mean Thermal current (Dark - Bias) grows very nearly linearly over time and that the thermal current is cut in half approximately every 6C (6.3C for KAF-1603ME, 5.8C for a KAF-8300). So for the vast majority of "normal" pixels, the behavior is very predictable.
However, note in the "Thermal Max" graph (click the link above) that some of the bright, or "Hot" pixels display distinctly non-linear behavior. For these pixels, small changes in time and temperature may produce quite different results.
As a side note, this non-linear behavior of "hot" pixels is a key reason why we don't generally recommend scaling dark frames. The "normal" pixels will scale quite well, but many of the hot pixels will not.
Now a very key point. Dark current is NOT noise, it's unwanted signal. But, like any signal it does contain noise.
Here's a table of some of the data points in the "Kodak KAF-8300 Mean Thermal Current" graph above showing mean dark current in a 5-minute exposure at various temperatures -- and the noise, or uncertainty -- in that unwanted dark current signal:
Noise (ADUs RMS)
The noise, or standard deviation of the signal, is calculated as the square root of the signal, in this case the unwanted dark current signal.
The dark current (unwanted signal!) will be removed by normal dark subtraction, but the noise in the resulting image will go up. Subtracting darks taken at a lower temperature will add less noise, but as you can see decreasing the temperature beyond a certain point offers diminishing returns.
It's very important here to understand how noise combines when adding or subtracting two images with multiple sources of noise. Noise combines as the square root of the sum of the squares. That may sound a bit intimidating but it's easy to understand and calculate in practice.
As an example, let's assume you took a 5-minute light frame and a matching 5-minute dark frame. In this example light frame, signal levels vary from 900 ADUs for the sky background to 40,000 ADUs in the brightest parts of the image. Mean dark current is as shown above. To be consistent with the QSI 583 data shown in the table above, read noise is 8e- RMS with a gain of 0.5e-/ADU, or 16 ADUs RMS. Some details in the calculations below are slightly simplified, but that doesn't materially affect the result. "SQRT( X )&rdquo' means the square root of the expression inside the parenthesis.
Total noise in a dark frame is:
SQRT( dark noise 2 + read noise 2)
Read noise is consistent at 16 ADUs RMS. Dark current and therefore noise is reduced by cooling.
Total noise in the 0C dark frame is:
SQRT( 3.7 2 + 16 2 ) = 16.4 ADUs RMS
Total noise in the -10C dark frame is:
SQRT( 2.4 2 + 16 2 ) = 16.2 ADUs RMS
Total noise in the -20C dark frame is:
SQRT( 1.8 2 + 16 2 ) = 16.1 ADUs RMS
Note in each case how the largest source of noise dominates the combined result.
Shot noise from the light in the light frame ranges from 30 ADUs RMS in the sky background (SQRT(900)) to 200 ADUs RMS (SQRT(40,000)) in the brightest parts of the image. You can quickly see that the shot noise in this example will dominate the noise in the dark subtracted image. Adding in the read noise yields total noise in the sky background of SQRT (30 2 + 16 2) = 34 ADUs RMS. Even subtracting a single dark at 10C would yield this total noise in the sky background:
SQRT( 34 2 + 16.4 2) = 37.8 ADUs RMS
(compared to 30.0 ADUs RMS of shot noise in the sky background of a single subframe)
Dropping the temp to -10C yields this:
SQRT( 34 2 + 16.2 2) = 37.7 ADUs RMS
Dropping the temp further to -20C yields this:
SQRT( 34 2 + 16.1 2) = 37.6 ADUs RMS
So, dropping the temperature of the CCD from 0C to -20C results in a drop in the total noise in the sky background of this example dark subtracted image of just 0.2 ADUs (37.8 - 37.6).
With higher signal levels in the brighter parts of the image, the effect of all sources of noise other than shot noise has an even smaller effect. Building master calibration darks by combining multiple individual frames further reduces the noise contribution added by subtracting darks. Here's a graph showing noise in a master dark frame made by combining up to 100 darks.
You'll note that the noise in the combined image is continuing to drop even after combining 100 darks (one of the benefits of nearly perfect Gaussian noise distribution in QSI cameras), but that the increased benefit diminishes with more and more frames. For my own imaging I generally take 15 darks because, for me in most cases, the additional benefit of collecting more darks is not worth the additional time.
Cooling does offer other benefits such as reducing or eliminating some fixed patterns that are visible at higher temperature and reducing the number and intensity of hot pixels. Precisely regulated cooling offers the additional benefit of providing a stable amount of dark current for any given combination of time and temperature. If the temperature is changing by even a few tenths of a degree, dark current response in the lights and darks will vary from frame to frame, resulting in a "blur" of the data when multiple frames are combined and reducing the benefit of dark subtraction.
You can acquire this kind of data with your own camera. Each data point in these examples is the average of 5 individual frames. To determine thermal current, the average of 10 bias frames was subtracted from each dark frame.
The Mean is the average of all pixel values. The median value would be somewhat lower because the mean is skewed upward by the small population of hot pixels and any tracks caused by energetic particles. Hot pixels typically account for about 0.1% of all pixels on Kodak KAF sensors.
The Max was determined from a median combine of the 5 thermal frames. This serves to smooth out the noise and get closer to the "true" value. A median combine was used to eliminate the impact of energetic particle strikes. This value ultimately represents the single, brightest pixel. Not all "hot" pixels exhibit non-linear behavior like this example.
Astronomy Magazine Picture of the Day
Are you interested in having one of your images published by Astronomy magazine? The photo editor of Astronomy, Michael Bakich, gave a presentation at AIC 2008 discussing what he looks for when selecting images to publish in Astronomy magazine and on the Astronomy.com web site. Michael then summarized the information on this hidden web page just for imagers:
You can also view Michael's complete presentation on this page:
The single most important thing that Michael stressed about submitting images to Astronomy was picking a good subject for the email. The email address for submitting images is: email@example.com. He prefers receiving emails with just the name of the object in the Subject, e.g. "IC 1396". His least favorite Subject line? "New image for you".
New QSI Dealers
We'd like to welcome three new QSI astronomy dealers:
Starizona in Tuscon, Arizona
Woodland Hills Telescope near Los Angeles, California
Northstar Astronomy in Greece
A complete list of QSI dealers can be found here:
Some QSI dealers have QSI cameras in stock and may be able to deliver popular models almost immediately. If you're in a hurry to receive a new QSI camera, call around for the best deal and quickest delivery.
QSI in the News
Several recent magazine articles have featured QSI cameras or people.
a.. Astronomy, June 2010 CCD Buyer's Guide.
b.. Astronomy, April 2010 Telescope Insider article featuring Kevin Nelson
c.. Sky & Telescope, Feb. 2010 QSI 583 used in review of C14-Hyperstar
d.. AstroPhoto Insight, Dec. 2009 Article on the Kodak KAF-8300 and the Stellar Performer Award, plus the 2010 CCD Camera Buyer's guide
e.. Amateur Astronomy, Summer 2010 Photometry article by Richard Berry with his QSI 532ws
QSI will be exhibiting at the following events in the coming months. Please drop by the QSI booth if you plan to attend:
Midwest AstroImaging Conference (MWAIC)
July 23-24, 2010 near Chicago, IL
MWAIC offers many of the top speakers found at other more well known imaging events with better access to the presenters and exhibitors ??and great food. Kevin Nelson of QSI will be speaking at MWAIC again this year. If you live anywhere near Chicago, don't miss this exceptional event. Click the link below for a link to the MWAIC web site. Sign up today!
Pacific Astronomy and Telescope Show (PATS)
September 17-19, 2010 in Pasadena, CA.
Also plan to attend the 1-day Riverside Astroimaging Workshop (RAW) on Sept. 17th, the day before PATS begins.
Advanced Imaging Conference (AIC)
October 22-24, 2010 in San Jose, CA.
AIC is moving to a new larger, better venue this year in order to better handle what has become the top imaging event in the world.
Complete details on upcoming events can be found on the News & Events page on our web site:
We look forward to seeing you there.
The QSI Team