Alt/Az + De-rotation = No Diffraction Spikes?

19 replies280 views
John Stone avatar
I was just thinking that since I really hate diffraction spikes it's been keeping me from getting the big/fast reflector I feel I just "have to have" :-) for years.

Just tonight I was watching a guy run a L350 + CDK14 from his driveway in Alt/Az and at the end he showed his processed picture and it didn't have diffraction spikes!

So my thinking is that while the de-rotator keeps the stars in place on your camera through out the night the diffraction spikes continue to rotate.

I guess what you'd get in long exposures is a smeared spike along the arc of rotation for that subs duration?  But when you stack all your subs together it seems like it all will get rejected during stacking??

Maybe a 1 sub's less SNR in the diffraction spike "sweep" area?

Does anyone have experience with long exposure astrophotography in Alt/Az with a de-rotator? 

Am I way off base here or will this really work?

Thanks in advance for any advice you care to give.
Respectful Engaging
Tony Gondola avatar

I’m gonna take a stab at this. I think you have the right idea in that with a alt/az mount the spider is in a fixed orientation relative to the sky while the sky and sensor rotate. That should give exactly the effect you’ve described. The best way to confirm the rest would be to ask for a set of subs and see what happens.

I suppose if you are already going alt/az this would make sense. If not there are easier ways to eliminate or diffuse the spikes in a reflector as I’m sure you know.

andrea tasselli avatar
Do the math…, simples!

P.S.: actually, not that much…
SonnyE avatar

Back when I was starting out, I used to do extremely long exposures. It was the only way I could get an image with my first camera (POS Orion G3 color)

One night, it and my POS AVX mount were having a good night and I was doing a long exposure, (Because neither I, nor 5 other good imagers could stack the dog-doo subs that thing put out.)

That particular night I ran a 5400 second sub. Non-stop. 1 hour and 30 minutes.

It was a picturd, but it worked. One good friend told me “Sonny, you just stack with time.”

Ahh, those crazy days of Picturds. (And chitty equipment)

Engaging
John Stone avatar
Tony Gondola:
I suppose if you are already going alt/az this would make sense. If not there are easier ways to eliminate or diffuse the spikes in a reflector as I’m sure you know.


I'm thinking with my idea the spikes are completely rejected instead of diffused/blurred.   

The data in the stacked image will be just as sharp under the spikes as other places in the frame (except for the small loss in SNR), right?
Tony Gondola avatar

John Stone · Jan 21, 2026, 02:23 AM

Tony Gondola:
I suppose if you are already going alt/az this would make sense. If not there are easier ways to eliminate or diffuse the spikes in a reflector as I’m sure you know.



I'm thinking with my idea the spikes are completely rejected instead of diffused/blurred.   

The data in the stacked image will be just as sharp under the spikes as other places in the frame (except for the small loss in SNR), right?

Have you asked your friend with the alt/az setup? Maybe he can turn you on to a set of subs that you can test the idea out on?

John Stone avatar
Tony Gondola:
Have you asked your friend with the alt/az setup? Maybe he can turn you on to a set of subs that you can test the idea out on


Can't really do that since I don't know the guy.  I just watched this old video:

https://www.youtube.com/watch?v=iUkOg0r5yhc
Tony Gondola avatar

Boy, can’t tell much from that single cropped image. With my newt I don’t get spikes on all the stars, just the very brightest ones so you can see why it’s hard to make a judgement.

📷 m32 final render-3.jpgm32 final render-3.jpgThere must be someone on the bin that’s using an alt/az mount with a rotator that could comment.

andrea tasselli avatar
I'm pretty sure these were done with such a system:

NGC6072 NB-RGB - AstroBin

AG Carinae NB-RGB and HOO Composition - AstroBin
andrea tasselli avatar
The main argument is as follows:

An Alt-az mount would track a given patch of the sky by keeping the center of the field at the same position while the field would slowly rotate around it's center. The speed of the rotation would depend on both the object's coordinates as well as the latitude of the observer's location. From the point of view of the sensor thus the whole field experiences a rotation as if the whole OTA would rotate around its axis as well. Adding a rotator would counteract this and keep the image centred and free of rotation about its pivot point, thus being in fact equivalent to an equatorial mount, thus the spikes would retain the same rotation about the field as if it were on an equatorial mount. 

Or otherwise, check this image from the ESO VLT (amongst others). That is an Alt-Az mount with an optical de-rotator if I'm not mistaken:

The Helix Nebula* | ESO
Helpful Insightful
Tony Gondola avatar

andrea tasselli · Jan 21, 2026, 06:09 PM

I'm pretty sure these were done with such a system:

NGC6072 NB-RGB - AstroBin

AG Carinae NB-RGB and HOO Composition - AstroBin

And these certainly have diffraction spikes.

Tony Gondola avatar

andrea tasselli · Jan 21, 2026, 06:23 PM

The main argument is as folIlows:

An Alt-az mount would track a given patch of the sky by keeping the center of the field at the same position while the field would slowly rotate around it's center. The speed of the rotation would depend on both the object's coordinates as well as the latitude of the observer's location. From the point of view of the sensor thus the whole field experiences a rotation as if the whole OTA would rotate around its axis as well. Adding a rotator would counteract this and keep the image centred and free of rotation about its pivot point, thus being in fact equivalent to an equatorial mount, thus the spikes would retain the same rotation about the field as if it were on an equatorial mount. 

Or otherwise, check this image from the ESO VLT (amongst others). That is an Alt-Az mount with an optical de-rotator if I'm not mistaken:

The Helix Nebula* | ESO

I’m still not sure. With an Eq mount the spider rotates with the rotation of the field. With an alt/az the spider doesn’t rotate with the field, the rotator does. If we accept that as true then I think that’s a fundamental difference.

andrea tasselli avatar
There is no rotation of field involved with an equatorial mount, that's the whole point of it, the scope image is static relative to the stars.
Well Written Concise
Tony Gondola avatar

Yes but with an alt/az mounted ota, there is.

Craig Towell avatar

Tony Gondola · Jan 21, 2026 at 06:31 PM

andrea tasselli · Jan 21, 2026, 06:23 PM

The main argument is as folIlows:

An Alt-az mount would track a given patch of the sky by keeping the center of the field at the same position while the field would slowly rotate around it's center. The speed of the rotation would depend on both the object's coordinates as well as the latitude of the observer's location. From the point of view of the sensor thus the whole field experiences a rotation as if the whole OTA would rotate around its axis as well. Adding a rotator would counteract this and keep the image centred and free of rotation about its pivot point, thus being in fact equivalent to an equatorial mount, thus the spikes would retain the same rotation about the field as if it were on an equatorial mount. 

Or otherwise, check this image from the ESO VLT (amongst others). That is an Alt-Az mount with an optical de-rotator if I'm not mistaken:

The Helix Nebula* | ESO

I’m still not sure. With an Eq mount the spider rotates with the rotation of the field. With an alt/az the spider doesn’t rotate with the field, the rotator does. If we accept that as true then I think that’s a fundamental difference.

Yeah the way I see it in my mind is that with an Alt Az scope, the angle of the spider vanes changes relative to the target as it tracks, and with no rotator the spikes will be in the image whilst the target will be blurred (by rotation).

However with a rotator the camera is turned to counter the rotation of the target, but the spikes in the image would stay aligned with the spider vanes, which are static with respect to the target, in effect the spikes would rotate within the image. Whether that makes them invisible or not is another matter i think and would be dependent on sub length, overall exposure time and field rotational speed (target location etc).

Edit: seems like it isn’t possible to sigma stack them out? See here https://pixinsight.com/forum/index.php?threads/diffraction-spikes-rotation-on-alt-az-telescope.14364/

Craig Towell avatar

andrea tasselli · Jan 21, 2026 at 06:23 PM

The main argument is as follows:

An Alt-az mount would track a given patch of the sky by keeping the center of the field at the same position while the field would slowly rotate around it's center. The speed of the rotation would depend on both the object's coordinates as well as the latitude of the observer's location. From the point of view of the sensor thus the whole field experiences a rotation as if the whole OTA would rotate around its axis as well. Adding a rotator would counteract this and keep the image centred and free of rotation about its pivot point, thus being in fact equivalent to an equatorial mount, thus the spikes would retain the same rotation about the field as if it were on an equatorial mount. 

Or otherwise, check this image from the ESO VLT (amongst others). That is an Alt-Az mount with an optical de-rotator if I'm not mistaken:

The Helix Nebula* | ESO

That Helix nebula image was taken by an EQ mounted scope… https://en.wikipedia.org/wiki/MPG/ESO_telescope

andrea tasselli avatar
Helpful Respectful
andrea tasselli avatar
Or this one, closer to home (but you can browse the others), all taken with an Alt-Az scope :

Chilescope
Craig Towell avatar

andrea tasselli · Jan 21, 2026 at 08:22 PM

Or this one, closer to home (but you can browse the others), all taken with an Alt-Az scope :

Chilescope

Thanks yes you can clearly see the smearing of the spikes due to the rotation… same as in the PI forum I linked to: it doesn’t appear there is a software solution to the issue

📷 IMG_0705.pngIMG_0705.png

andrea tasselli avatar
I can't see any smearing in the one I linked or mine, for that matter.