Astrophotography - Tips and tricks please.

[theandies]

So I finally got a good consumer DLSR, the T6i.  My T mount adaptor comes in friday.  I didn't get a shutter release yet but with the T6i I think the remote control through my cell phone will keep the vibrations down.......I guess.  I've always wanted to get into astrophotography but never had a camera that could do it.  I've read quite a bit about it over the years but I'll be new to it just as I'm new to anything other than a point and shoot and cell phone cameras.

 I have an old school (no electronics but I know how to use it very well) Meade 8" Schmidt–Cassegrain that I've had for years.  I do have an off axis corrector and a very good tripod.   I'm just looking for some hints and tips from the astrophotography buffs here if there are any.  What I think I'll try first is planetary pictures and pictures of the moon. After I get the hang of the basics I'll get more ambitious.  Anything would be much appreciated. 

Thanks

John

[TCampbell · Accepted solution]

Most LX200 scopes are f/10, but there were some f/6.3 models and the new LX600 and LX850 models are f/8.  You'll want to know what focal ratio you have because when you use a camera with a telescope there's no adjustable f-stops like there is for a lens.  You'll be doing manual exposures and you'll want to know what your native focal ratio is.

When you connect the camera, connect it "straight through" (no 90º diagonal).   I have a 14" LX200 ACF -- which has longer forks and I have a LOT of clearance behind the optical tube.  My scope can point straight up through the fork arms with an electric focus and my Canon 60Da attached and still have plenty of room so that there's no danger of the camera hitting the base if the optical tube tries to point straight up.  

Check the clearance on your scope (with camera attached) to make sure you are comfortable knowing that you have enough clearance -- as this will avoid potential damage if you slew to a new target in the dark and the camera crashes into the base.

My LX200 has an electric focus with visual back (standard 2" tube opening for a 2" eyepiece or 2" diagonal) and so my camera uses a 2" diameter nosepiece with the EOS t-ring.  I don't use the SCT T-mount.  

When focusing the scope, always "finish" turning the focuser into focus in the counter-clockwise direction.  The LX200 has mirror-shift when you reverse the direction of focus.  This is because the focus knob either pushes or pulls on the primary mirror at the back of the scope and the mirror rides on a central baffle that has to allow at least a tiny bit of slop so that it can slide easily.  When you "push" the mirror forward, it gets the best support from the bottom and also forces the mirror to rest in a properly collimaged position.  When you "pull" the mirror back, it slightly de-collimates your optics and ALSO the mirror may eventually settle on you in mid-exposure (that would ruin the image being captured at the time).   By finishing focus while rotating the knob in the counter-clockwise direction, you give the primary mirror the best support, collimation, and it reduces the potential for the mirror to settle and move after you finish the focus.

The LX200 has a mirror-lock knob and when I do visual astronomy I don't use it.  But when I do imaging, I DO use it.  My scope has an electric focuser on the visual back.  That means I can adjust the main focus, lock the mirror, then use the electric focuser to fine-focus the telescope to my satisfaction.  

I prefer to point the scope to a pinpoint star to focus.  Focusing on the moon or a planet is is not quite as easy to see if you achieved perfect focus.  But when you've minimized the size of a pinpoint star, you'll have more accurate focus.  

There is an application called "Backyard EOS" which is built for astronomy astro-imaging using Canon EOS cameras.  It has focus aids as well as both planetary imaging modes and deep-sky imaging modes.  It controls the camera and performs image acquisition for you (it's basically "tethered" remote shooting control of the camera.)

I normally use a Bahtinov focusing mask on my scope to achieve fine focus for deep-sky images.  But a lot of imagers I know have electric focusers and use a program called FocusMax which is exceptionally accurate focus.  FocusMax deliberately de-focuses a star (you see the donut shape) and starts focusing and imaging and focusing and imaging.  It measure the size of the donut and does several runs creating plots.  It mathematicaly determines where the perfect focus point must be located (even if sky conditions are horrible.)

Both "Backyard EOS" and "FocusMax" require Windows (I use a Mac so I don't use them, but they are popular.)

The moon is a VERY easy target... the correct exposure for the moon follows the "Loony 11 Rule".  That rule says that at f/11 (and it only works at f/11) you can set the shutter speed to the inverse of the ISO sensitivity.  So at f/11 and ISO 100, it'd be 1/100th sec.  At ISO 400 it would be a 1/400th sec exposure, etc.  

If your scope is really an f/10, it's close enough... f/10 is approximately 1/3rd stop more light than f/11 (it's not a significant difference) but it means the moon will be slightly bright.  You can adjust the shutter speed 1/3rd stop faster to compensate.  E.g. instead of 1/100th sec you should shoot at 1/125th sec. and be bang-on the accurate exposure again.  If you have an f/6.3 scope then you are 1.6 stops faster (almost, but not quite 2 stops).  That means insead of ISO 100 and 1/100th, you'd want to shoot at ISO 100 and 1/320 sec.

Planets are slightly more difficult... they are dimmer than the moon so you'll use a higher ISO setting.  But they are typically imaged by shooting about a minute's worth of video frames.  Stacking software is then used to identify the best frames out of the video and those are combined to create a composite image.

If you happen to image Jupiter, note than Jupiter has a fairly fast rotational speed... the surface changes enough in MERELY 10 minutes that if you try to combine images shot more than 10 minutes apart you'll get blur.  All the data you capture for Jupiter needs to be captured within that 10 minute window of time.  There is a program called Registax that is particular popular (and free) for planetary image stacking.  

If you image Saturn, it's low in the sky this year (because it's an "outer" planet and our northern polar axis is pointed toward the sun.  That puts saturn low relative to the horizon for astronomers who live in northern latitudes.)  This means you'll get some atmosphereic dispersion when you view it.  Atmospheric dispersion is a form of chromatic aberration -- except it's caused by our atmosphere working like a lens.  The atmosphere splits "white" light into the rainbow spectra of light.  You'll see a "red fringe" on one edge of the planet and it's rings... and a "blue fringe" on the opposite edge.  Not to worry... one of the features of Registax is that it can separate the single color image into red, green, and blue color channels and it lets you "shift" them back on top of each other.  This greatly improves the focus quality of your image.

Deep sky objects are particularly difficult.  This may cause you to lose all your hair.  You have been warned.  🙂

To detail how to take deep-sky images would take a while... but I can sum up:

The scope needs to be mounted on a "wedge" (e.g. such as a Meade Superwedge).  The wedge is moutned to the tripod and the scope is then mounted to the wedge.  The wedge is tilted so that the tilt angle is adjusted for YOUR viewing latitude.  If the scope is merely on an alt-az mount then you'd get field rotation as you imagine and that would create blurred results.

You mentioned your scope does not have "electronics" but it would need to minimally have an RA drive that can track at sidereal speed.  Do you have this?

The mount needs a "precise polar alignment" (which takes a bit of effort.)  

You take numerous long exposure images (e.g. 5-10 minutes would be typical).  

You also need to capture "dark" frames, and it's also helpful to grab "flat" frames and "bias" frames.  I can explain what these are if you haven't heard of them.

The images are then stacked using image registration and integration stoftware (for deep sky objects there's a free program called "Deep Sky Stacker".  I use something called PixInsight do do my registration & integration (stacking) but PixInsight isn't free.

You mention this is an "old school" LX-200 with "no electronics".  Are the electronics fried?  I've seen a number of LX200 models and some have very old primitive electronics, but I've never seen one that doesn't have any electronics.    There are places that will either repair or refit the scope so that it does have working electronics.  Finding the very old original boards in working order is tough.  There were some bad capacitors used which would dry out over time and then blow.  The guys that service the scopes know which ones blow and they replace them with modern equivalent capacitors that wont blow BUT it's critical that they do it BEFORE the scope has a problem.  If the capacitors blow before being replaced they often take out other electornics on the board and now it's a more serious repair.  Those "more serious" repairs can involve trying to find replacement boards that aren't made anymore, haven't been made in years, and are becoming increasingly rare.

So... now there are services that simply pull the original boards and replace them with completely new boards, but the refit kits give you electronics that work like an LX90 (not an LX200).  LX90's dont' have PEC -- so they aren't as precise as LX200's, but it's still better than nothing.

Moon shots are easy (no elecronics needed.)

Planetary shots are fairly easy but it helps to at least have a working RA drive that can track at sidereal speed (clock-drive) even if there are no computerized go-to electronics.

Deep-sky objects, however... will really need working electronics and an autoguider.  This are the most complicated images by far --- because they require such long exposure times.  And during those very long exposures, you can tolerate any movement or tracking errors -- otherwise it ruins the image.

[ebiggs1]

Very nice!  

I agree it is probsly from where you shot and not how you shot.

[theandies]

Ok, It's August.  20 days to go until the eclipse.

I've got my telescopes solar filter on it's way.  I've never tried to take pictures of the sun so any tips, settings etc. to get the best shots would be greatly apprciated.

[TCampbell · Accepted solution]

Yes, check out Fred Espenak's page (aka "Mr. Eclipse").  Fred is a retired NASA physcicist who does all their eclipse predictions and a top expert on eclipse photography.

http://www.mreclipse.com/SEphoto/SEphoto.html

Assuming you will be in the path of totality...

Also, it's best to have the camera under computer control so that you can enjoy the eclipse instead of having to pay attention to your camera.

If you have a Mac, then you may want to download "Solar Eclipse Maestro".

If you have Windows, then you may want to download "Eclipse Orchestrator" or "SETnC".

I'm familiar with Eclipse Orchestrator and Solar Ecipse Maestro, but not SETnC.

Solar Eclipse Maestro and Eclipse Orchestrator both allow you to script the eclipse capture but the timings are based on the eclipse path prediction data and your precise location (either via GPS or manually entered). 

The shots before and after totality are shot with the solar filter on.  

Double check frame & focus a minute or two before totality (with filter still on camera).

At 20 seconds prior to totality (and no sooner than 50 seconds prior to totality) you can remove the filter but DO NOT LOOK THROUGH THE CAMERA once the filter is off (that's why I mentioned doing the final frame & focus before removing the filter).

At about 9 seconds prior to totality you may see the "Diamond Ring" effect.

At about 1.5 seconds before you may see the "Baily's Beads" effect.

The software script can be set to announce warnings (e.g. 5 minutes to totality, 2 minutes to totality, when to remove filters, etc.) so you know when to do each step.

Once totality begins, it is safe to look directly at the sun.  You'll see the solar corona.  It has tremendous dynamic range and requires about 10-12 stops of bracketed exposures to capture the entire corona (you can merge the shots with HDR processing).

Once totality ends you'll likely get another Baily's Beads, followed by another Diamond Ring.  After you capture that, it's time to put the filters back on the camera (which should happen about 20 seconds after totality ends.)

Solar Eclipse Maestro is free for non-commercial use (he charges if it's meant for commercial use).  He does appreciate donations.

Eclpse Orchestrator has a free mode which limits it's functionality, but it's a paid license to unlock all features.

Both Eclipse Orchestrator and Solar Eclipse Maestro use the same scripting language.

I have not used SETnC.  What I've learned about it is that it (a) runs on Windows, (b) only controls Canon cameras (no support for any other brand), and (c) it's free.

Clear skies & good luck!

[theandies]

Awesome info as usual Tim - Thanks.

Unfortunatly I will be between 80 and 90% total.  I could drive down to my Moms place in Emerald Isle, NC and get closer to 95% but I'd have to take a few days off from work and convinve the wife it's a once in a lifetime event.  If in NC we could drive a few hours to get in the totality band.  That may be the plan.

[theandies]

Received my filter material from Thousand Oaks Optical and made my solar filter for my telescope, spotting scope and made some glasses for my kids.  The film works great.

The only problem I can see is when I'm using my camera with my scope the FOV is small and I can't get the entire sun in the frame.  Would a focal reducer allow for full frame shots of the sun?

[TCampbell]

For any camera, the goal is to get a focal length that allows the Sun to occupy about 1/4 to 1/2 (with 1/3rd being nominal) of the height of the sensor (in the short direction).

The Sun is about 1/2º from edge to edge.  So that means you want a field of view between 1 and 2º in the narrow direction (which would be about 1.5 to 3º in the wide direction).

For a Canon APS-C sensor camera, that translates to a focal length in the range of 391mm (you could call it 400mm) to 782mm (but 800mm would be fine) with 521mm being nominal.

For a full-frame camera, it works out to 634mm to 1268mm with 846mm being nominal (round to nearest available values).

You CAN use a focal reducer but there are a few things to know.

Focal reducers are commonly available with a .62x or .63x reduction.    Meaning, if you had a 1000mm focal length, it would reduce it to about 620 (with a .62x) or 630mm (with a .63x).    If you had an 8" f/10 SCT (that would have approx a 2000mm focal length) then a .62x reducer would drop it to 1440mm -- which is still a bit much.

The other caveat is that the amount of reduction depends on you getting the back-focus distance correct.  Most focal reducers expect about 105mm of backfocus (meaning the distance from the last element in the focal reducer to the sensor inside your camera should be 105mm away...  they are optimized for that distance and while they may work if the distance is shorter or farther the reducer to both focal length as well as focal ratio will not be accurate.

What brand/model scope do you have and also which camera model do you have?

[theandies]

It's a Meade 2080 f/10 F=2000mm as posted on page one of this topic.  Camera = Canon T6i APS-C

Thanks for the tip about backfocus.  I'll keep that in mind.  Since I can only hope to acheive ~1440 I'll have to live with it.  I can always make a filter for my T6i since I have some left over and put it on the tripod for full frame shots.

Found this on Meade's web-site.  Appears I can get my f10 to f6.3 with this.

I may pick one up:

Meade f 6.3 Focal Reducer/Field Flattener #07545. For Meade Schmidt-Cassegrain models. Improves edge-of-field correction and reduces exposure times by close to 50%. Threads into rear cell of any Meade SCT. Threads into rear cell of any Meade SCT. May be used to increade field of view and reduce magnification for visual applications.

An important advancement in high-resolution focal reduction systems, the Meade 4-element, multi-coated f/6.3 Focal Reducer/Field Flattener threads on to the rear cell of any LX90, LX200 or other Schmidt-Cassegrain or Advanced Coma-Free model and is typically used with an Off-Axis Guider or T-Adapter. Reduces the telescope's focal ratio by a factor of 0.63: f/10 telescopes are converted to f/6.3. Simultaneously, the 41mm-diameter lens system helps flatten the field of Schmidt-Cassegrain models, significantly improving edge-of-field corrections.

This handy accessory enables a reduction of photographic exposure times by about 50%, while producing an actual field diameter of 1.5" at the film plane.

[TCampbell]

The Meade f/6.3 on a 2080mm scope will bring your effective focal length to 1310mm.

See if you can get accurate info on the back-focus distance (if you buy it from someone like Oceanside Photo & Telescope (OPTcorp) they're probably the largest telescope & astronomy equipment dealer isn't the country and they have pretty good technical support, they can probably find that info.  Meade tech support used to be fairly bad but they've been bought out by a new company and I'm told things are better.  But you might be able to get the info direct from Meade.  

Anyway, the original reducer had a 105mm back-focus dsistance.  Then they redesigned the the product and I'm told that "briefly" their website listed the back-focus distance at 45mm (which is really surprising), but then redid their website and no longer list the back-focus distance.   I have, however, seen the "Q&A" section list that it's 105mm even though others claim it's 45mm.   I don't own the current model so I don't know the real story.

Canon EOS cameras have 44mm of space from the lens mounting flange on the front of the body to the imaging plane inside. So if it really is 45mm you'd basically want to find a T-Ring that can adapt Canon EOS mount to the Meade SCT size thread (and most T-rings adapt from EOS mount to "T-thread" which is not the same diameter as SCT thread.    So I'm somewhat skeptical of the 45mm back-focus claim.  Part of me wonders if someone didn't measure the distance of the extension tube at 45mm (and by the time you add in the 44mm in the camera you're at about 90mm ... and add in the T-Ring which usually adds another 5mm and now you're basically at a 95mm back-focus distance (and there are reducers on the market with 95mm back-focus distance).

The distance doesn't have to be bang-on accurate.  If the back-focus distance is supposed to be 105mm... you'd still be ok if you were anywhere in the 100-110mm range.  Technically it'll fractionally change your focal length and focal ratio ... but not by enough to have any meaningful impact on the exposure.

BTW, Meade used to make an f/3.3 focal reducer but you don't want that.  The flat field that it can generate is very small (about the size of your pinky fingernail) and it's designed to work well for small-chip cameras (e.g. Web-Cam size chips) that are often used in planetary video capture.  But on a DSLR camera you'll find only the very center of the field can be focused and the rest will be strongly blurred (and this is normal - it's not designed to produce a large enough flat field for a DSLR camera.)

[bhattvd]

Tim:

I greatly enjoyed your article on Astrophotography Tips & Tricks.  Very informative.  Seems like you have lot of knowldedge and the patience to coach new comers like me.  I have a specific question.

I have a Canon EOS T3i (cropped sensor) camera with tripod, intervalometer and three lenses. 1) 18-55mm kit lens; 2) 70-300mm kit lens; 3) 35mm F2 Prime lens.

I apply usually recommended Astrophotography settings (for manual focusing) like:

Both Camera & Lens on Manual Mode,

35 mm Prime Lens

F2.2 aperture

30 sec. exposure

ISO 3200 (1600-6400 range)

Here is my problem with "Focusing in Live View":

First, I focused on Jupiter, also focused with 10x zoom to get best focus.

Then, while maintaining my focus, I change the scene to Big Dipper.

At that time, I do not see any stars in my live view. (I know big dipper has a few low apparent magnitude stars which are bright enough).

My LCD is grainy and I am not able to identify stars at all.  If I can see the star pattern then I can navigate to a partical area of the constellation to capture M101, Pinwheel Galaxy!

What am I doing wrong?

You think I may have some wrong settings in my camera?  I set high ISO noise reduction off.  But, did not work.

I really appreciate your advise and time you spend on helping other upcoming astrophotographers.

-Vinay Bhatt, Austin, Texas

[TCampbell]

Trying to capture M101 (Pinwheel galaxy) with a 35mm lens on your camera means the galaxy will be *really* tiny.  The galaxy is about 23 x 24 arc-minutes wide.  A 35mm lens on a T3i is a field of view roughly 24 x 36°.  This means the galaxy will be a tiny speck.

At around 2000 to 2500mm of focal length it will nicely fill the frame ... or something a little less and crop in.  

AstroBin.com is an excellent resource for exposure ideas because you'll find loads of examples of images of any object you want, shot with pretty much any camera you want.  I did a search for "M101 Canon T3i" and found a selection of them... here is just one example:  https://www.astrobin.com/33218/

The photographer who shot that used a 6" f/6 RC telescope (see:  https://en.wikipedia.org/wiki/Ritchey–Chrétien_telescope ) with a focal length of 1370mm ... and then cropped the image to galaxy size.

But buying a decent telescope and a decent equatorial mount ... and a guide-scope and guide-camera ... will set you back probably something like $3500-4500.

So back to your question on focus...

I use a Bahtinov focusing mask (see:  https://en.wikipedia.org/wiki/Bahtinov_mask ) but I use a mask called the SharpStar2 by LonelySpeck (that's the brand).  Canon doesn't allow links to commercial products so you'll have to do a web-search for it.  BUT... most Bahtinov masks are a solid (non-transparent) card with slots cut into it.  The slots create a diffraction spike pattern when you get close to focus (see the Wikipedia page ... it shows an animation of someone focusing in and out and you can see the 3 diffraction spikes show up and then converge).  When the spikes perfectly converge at a common center, you've nailed focus.  But the problem with a typical mask is that they cut out something like half of the light.  That means the stars (which are already hard to see) are now only half as bright (so now they're *really* hard to see) and you barely get any diffraction spikes.  This particular mask that I use is completely clear.  It's some sort of clear polymer (possibly acrylic) and it has grooves etched onto it.  This means all the light comes through, but the etched grooves create the diffraction spikes.  This way stars remain as bright as they would be without a mask and the diffraction spikes are larger and easier to see.

But the SharStar mask is a square "slide in" type filters (not round thread-on style).   They make them in 3 different sizes (the 100mm square (aka 4") size is the most common).   It doesn't include a filter holder.  I use a Lee Filters holder, but the Cokin holders are less expensive.  The advantage of slide-in filers is that you also need an adapter ring that fits the correct size threads (for your lens threads) to the holder bracket.  But those adapter rings are cheap.  That means as you collect filters, you just get the adapter rings for each lens size you need (rather than needing another set of filters in a different size).

This setup makes it *much* easier to obtain decent focus.

Regardless of whether you have a focusing mask... do the following:

1. point the camera toward the brightest star in the sky.  Currently that'll be Arcturus and/or Vega.  Arcturus will be moderately high and roughly toward the southwest.   Vega will be VERY high (near the zenith).  
2. Switch the lens to manual focus.
3. Manually adjust focus to the infinity mark on the lens (not all lenses have a focus distance indicator on the lens barrel.  Your 18-55mm probably does not.  I'm guessing your 35mm probably does.).  The point of this step is that if you aren't at least semi-close to focus then you wont see *any* stars in the frame.  
4. Set the camera to Manual exposure mode.
5. Turn on live-view.  The camera has "exposure simulation" in live-view mode.  This means that as you increase your exposure settings, the subjects on the live-view display get brighter.  SO... to take advantage of this, you're going to crank the exposure to the max.  Set the shutter speed to 30 seconds.  Set the ISO to max.  Set the aperture to whatever wide-open is for your lens (e.g. f/2).  
6. If using a zoom lens, set the focal length to whatever you plan to use for your imaging and make sure you don't change it (because zooming (changing the focal length) will change the focus.... meaning a previously tack-sharp focus that you'll try to attain... wont be in focus anymore if the focal length changes.  Very few lenses are par-focal (meaning the focus is the same regardless of focal length).
7. Assuming you are pointing at a bright enough star (such as Vega) you should see it (and also assuming your lens is close to focus).  
8. Center the star and increase the live-view zoom to 10x.
9. Very carefully... adjust focus on the lens to try to make the star the tiniest pin-point possible (or if using a focus mask, until the diffraction spikes all center at a common point.)
10. Once you've achieved focus, be *very* careful not to bump the focus ring.  
11. Return to the section of sky that you want to image.
12. Return the exposure settings to something reasonable (ISO 800, f/2 and ... however many seconds you can manage.)

If on a fixed tripod (no tracking head) then use the "500 rule" to determine exposure duration.  Some people use the "600 rule" which is a bit more liberal (often good-enough as long as you don't inspect the image too closely).

This rule is meant for 35mm film cameras (frame size is 24x36mm).  You have an APS-C camera with a 1.6x crop-factor.  That means if you divide 24mm ÷ 1.6 and also divide 36mm ÷ 1.6 you'll get your cameras *actual* sensor dimensions.  But that also means you have to divide the 500 or 600 value (whichever you choose) by 1.6.  This will get you values of something like 312.5 to 375 (we could liberally call it about 350 if we split the difference).  DIVIDE this number (e.g. the 350 that I came up with) by the actual focal length of your lens (e.g. 35mm).  In your example (35mm lens) you'll get about 10.  That means 10 seconds... is the actual amount of time you can get away with exposing the camera on a stationary tripod and get stars that seem to be reasonably pin-point (not elongated and growing tails) as long as you don't look too closely.

If you get a tracking head (e.g. a Sky Watcher "Star Adventurer" or an iOptron "Sky Guider Pro") and you align it to the pole (the axis of rotation for the tracking head is paralllel to the Earth's axis of rotation) then as the Earth spins in one direction, the tracking head spins in the opposite direction ... and at the very same rate.  This causes the camera to remain fixed on the same piece of sky.  Doing this... you could easily take very long exposures (I've done 5-6 minute exposures with no problem.)

A tracking head will set you back roughly $300-400 (and mounts on your existing photo tripod ... which is hopefully a nice beefy tripod.  I tell anyone interested in Astrophotography that you want a heavy tripod (this is not the time to try to save weight).  The more solid the tripod, the less likely to have vibration during very long exposures and AP exposures tend to be long (many minutes).

At f/2 and ISO 800, something like 1.5 minutes should be long enough for most stuff.

BTW, I mention ISO 800 because it's the ideal ISO for your particular camera.  That's the point where the firmware starts to switch from using analog gain and starts using digital gain.  When the camera uses digital gain, each stop of ISO results in a loss of a stop of dynamic range.  In astrophotography you are going to end up with image data where the subject you care about is very dim (down in the lower 1/4 of the histogram) and you'll "stretch" the image data to brighten up the stuff you care about.  This means you need as much range as possible to differentiate between all the subtle differences in luminosity.

I own a Canon 60Da ... same sensor in a different body.  I use ISO 800 for nearly all astrophotography targets.

One last comment... traditional cameras have a filter in them that blocks some of the light in order to mimic the sensitivity of the human eye.  Turns out the human eye ... while sensitive to reds ... isn't *that* sensitive to reds.  So the filter in your camera starts to slightly block the light at wavelengths above 500nm ... and more aggressive at 600nm ... and completely blocks everything by 700nm.  Hydrogen alpha (which is the 'red' light we see in most deep-space nebulae) is 656nm ... at this point a traditional camera's filter is blocking roughly 75-80% of the light.  An astrophotography camera is typically modified with a filter that still blocks the UV and IR (everything shorter than 400nm or greater than 700nm) but tries to avoid blocking anything in between.  This means these cameras are something like 4-5x more sensitive to the "red" of Hydrogen alpha light -- and all those nebulae.  That means photographers don't have to run such long exposures to pick up the reds.  Lots of people will have their traditional DSLR cameras modified (send them in to a service that pulls out the factory filter and replaces it with a new filter more suited to AP work.). The Canon 60Da was a special edition of the 60D where Canon modified the filter.