Alt-Az Astrophotography: Every Problem Solved?

Added: 2026-05-12

[00:00:00]How can we do long exposure astrophotography using an out as mount?

[00:00:06]So a mount that can only move or rotate horizontally and vertically. We know it's possible because a lot of professional observatories with huge telescopes have their their telescopes set up effectively in out as mode and there's some very expensive equipment from things like plane wave that uh also uses by default without a wedge out as uh for asphotography. And when I started as photography, I actually started with an alt as mount, but I was limited to pretty short exposures because afterwards I would get a phenomenon called field rotation. And yet and yet this year at NE uh it seems that primal luch or what is it? The primal luch uh has decided that out as astrophotography is back in fashion. uh because we don't like doing polar alignment. We don't like setting things up. We just want to roll the scope out and press a button and then it's imaging which you know makes sense as well for smart telescopes. Uh some smart telescopes don't even support equatorial mode yet like the Vespera uh smart telescopes and they have a lot of cool software engineering to make out as work. Anyway, uh and the sea star does that as well.

[00:01:32]The dwarf, uh does that as well, although they have the option to be used in equat mode. And so this has been I've been falling into this black hole of how can we make it work for normal amateur equipment. And so I started to look into it and I kept looking into it and I fell into the black hole of looking into it.

[00:01:51]I'm currently in the state of spaghettification uh at probably or I don't know when. I'm probably at the event horizon or something like that.

[00:01:58]So, I wanted to make this video to show you a bit what I found and some of the questions uh remaining for me. Uh especially when it comes to things like flats. That's the only thing that's kind of like unanswered for me. So, let's get back to the start. What caused this thinking? And what do the does the maker of that equipment actually tell us to do? Okay. So, what started all of this is this guy here for me. Uh this is Primaluche. They showed off this during NE and it's one of the things that I hadn't really, you know, thought about it or even like not noticed it. I didn't mention it in my video about NE because I'm like also do proper astrophotography like that you plebs.

[00:02:45]But obviously I'm completely wrong. I mean I'm exaggerating. I didn't react like that just to be clear. But whatever. We have this guy here which is a CBC uh 8 inch uh Schmidt Cassagrain telescope on AL out as mount by all accounts. It's awesome. I've used AL as stuff for a lot of things including visual um mostly visual astronomy. I love them. They're amazing overall. So what how does this work? Well, we mentioned that there is field rotation.

[00:03:15]Actually, I'm going to skip ahead and show you something. a paper with tons of math but that shows you an example of a picture across time of Orion of the Orion constellation if you're tracking the center star of Arian's belt and you can see that with time if you're tracking it in out as Orion will rotate so that if you had taken a single long exposure for that whole period of time what you would see is all of those red dots connected here as a line as a line as a line as a line etc. in terms of star trails and the only star that wouldn't have trailed is this guy here, right? That's what you can expect to see if you don't do anything. So, what can you do about it? You can add like this this this guy here, this red red thing here in the middle, a drot or a rotator that can drotate. Basically, we have this phenomenon of field rotation, right? we know where we are pointed at.

[00:04:18]So we know how to drotate that thing because we know the exact rotation uh speed, right? So let's just counteract it with another motor. There's an obvious and convenient fact about that is that to track objects you now need the to move the azimus axis, you need to move the uh altitude axis and you need to move a third motor, the drot, right?

[00:04:44]So you have three motors working instead of one technically when doing uh astrophotography on an equatorial mount.

[00:04:53]Although uh I mean we know when we guide we also guide in declination because not perfect polar alignment etc. So it's two axis versus three effectively. And the third one, the de rotator itself, well, it doesn't matter that much because as long as it has a proper somewhat decent encoder in there, it should be able to de- rotate pretty well. So overall, it doesn't sound like a bad idea. But once like if you if we ignore the fact that we're now using three motors instead of two, there's the elephant in the room here. We are having a Schmidt categorane telescope with a very long focal length compared to the telescopes that I typically use and show on this channel.

[00:05:37]And so guiding is very very important because we want to track our targets very very well. So how can we do guiding with that telescope here? Well, for that let's ask Primaluche and their manual directly. Right. So, I'm opening the manual with the guiding and what they tell us is something very interesting.

[00:06:03]You may have read that off-axis guiders are often recommended for long focal length telescope like the edge HD optics used in the CPC Deluxe HD8 in right an off-axis guider picks up a guide star from the same optical path as the main camera which eliminates any flexure between the guide telescope and the main telescope which is all true.

[00:06:25]I've already talked about that in previous videos. Uh it's very important actually to eliminate flexure if you're going to do narrow band imaging especially at such slow uh focal ratios meaning you'll probably take like 10 minutes exposures but let's keep going. However, with an ALAS M fork mount, an off-axis guider is not a practical option since a camera mounted laterally on the optical train would inevitably collide with the fork arms during telescope movement because uh we continuously rotate the image imaging cam to correct for field rotation. If I go back to this uh this picture, I'm thinking really really really like would it actually collide anywhere if we have like a small enough uh of access guider? I I I don't see it. I don't see this as a valid excuse.

[00:07:20]There's another reason and it has to do with PHD2 and I had to go and look at the code of PHD2 to convince myself of it.

[00:07:29]We'll get back to that later. I'm so sorry. This is going to be such a nice black hole of information I'm sharing with you. It's such a geeky video, but I'm sure I'm not going to get a lot of views because I don't get a lot of views on geeky videos. So, if you want to help me out, please watch this until the end.

[00:07:44]Please leave a like. Please leave a comment if you have experience, for instance, imaging in out as mode and how you made it work. And if you want to support me even more and you're planning on buying anything from AIA, High Point Scientific, Amazon, uh, ltstore.com, uh, if you do so after clicking the links that I have down in the description, it will help me out at no cost to you. And of course, you can become a Patreon member, a channel member link in the description or it's the join button next to the subscribe button to make all of these videos possible directly to become a sponsor of the channel. You guys make everything possible. Thank you so much. Anyway, let's keep going. So, it's probably might not be the main reason. Maybe it could collide with this this focuser here on the side, but then you have an option where you're using an in uh in the imaging train focuser as well uh from Primaruches. So, I don't see that as a valid reason, but okay, whatever.

[00:08:37]Let's let's keep going, right? So, they tell us effectively that we need to use a guide scope. Okay, so we use a guide scope and the advantage of using a guide scope is that we can basically uh deceive PHD2 into believing that it is doing uh equatoral mode kind of guiding because the axis will always be vertical and horizontal. They even tell you that uh you want to make the camera itself have the sensor be either like oriented like this little orange square uh rectangle here or that little or rectangle there so that we have like it's nicely aligned on the pixel so we can get the best guiding possible. Okay, good. But this there there's more stuff, right? Because there's something very important that they they make a good point mentioning is that the guide telescope must be perfectly parallel to the main telescope and the guide star must be selected as close as possible to the center of the field of view of the guide camera which is also very important because remember uh that diagram here with the the field rotation of the Orion constellation here. Well, that star at the center, it's actually the star that we're guiding on. So, imagine you're trying to take a picture of the um uh Great Orion Nebula and the Running Man, let's say around here, right? But you're guiding on the central star of the belt. Well, you'll get field rotation on your target because you're not guiding at the center of the field of view of your main camera. So you would need to uh actually make sure that you cumate align your guide scope and your main telescope so that you point at the same thing. So you'd probably spend a session going like pointing to Vega or to some bright star and make sure that the center of your uh guide camera is the same as the sensor as your main imaging camera. And then when you select a guide star from PhD2, you want to make sure that you select a guide the guide star that is as close as possible to the center of the field of view of your guide camera because otherwise you can get field rotation. So uh that is a very important point that you know it's nice that they mention it because it is indeed very important and gives us a first element of response about why an off-axis guider might not work well because by definition an off-axis guider will use a guide star that is away from the main imaging sensor. Uh how much does that impact us on a very long focal length?

[00:11:19]Actually not much. Uh because again going back to this guy here, uh this guy made a lot of computation. By the way, if you open I'll have all of the links down in the description. Don't forget to click on more pages to see all of the nice computations. By the way, don't be too scared. It's mostly simple trigonometry and matrices. uh and a lot of the uh stuff is just like boilerplate formulas about how to uh do a rotation transformation matrix around a specific vector of given coordinates. It's it's pretty basic stuff and there's some very interesting as well uh assumption for instance that the the time difference between your points is small that kind of stuff but it's a f fascinating uh little paper I'll have the link in the uh in the description and one of the things that it tells us uh is that probably for long focal lengths the the fact that we're guiding on an offaxis star shouldn't be such a huge issue. So off-axis guider is not completely disqualified based on on that. It also tells us by the way that imaging near the zenith is probably not great because then you have one of the axis that could actually reach the necessity of rotating at infinite speed which you know it's not the case in real life but uh it's one of the reasons why why the sea star and other smart telescopes in out as mode don't really like imaging near the uh zenith. Anyway, fascinating paper.

[00:12:52]I'll have the link down in the description if you're interested. And I I really love this this little diagram because it really shows us field rotation around the star that we are guiding on. Okay. Now going back to the uh secret sauce which is not secret because a lot of people have been doing that in the past already. But the documentation from Primaluch is uh PhD to G guiding right. So they tell us like once we set up PHD2 guiding is the new profile uh wizard and under the guide camera we we choose our guide camera etc. everything's gonna gonna work and uh then they tell us how to connect to the mount via ASCOM so that PHD2 has access to the mount and because we're using a guide scope rather than an off-axis guider we don't connect to the rotator or D rotator because the guide scope itself itself it's fixed in place it's always going to move vertically and horizontally that way we can deceive PHD2 without needing to give it rotation information. In fact, we we must not give it rotation information.

[00:14:03]So, it's actually critical to not connect the rotator to PHD2 uh in this case. Now they point out the very important settings to have in PHD2 then is that for both right ascension and declination. Instead of for declination what we usually have which is the resist switch algorithm, we need to use the same algorithm as is normally used for right ascension in equatorial mounts. Uh which is this guy here, hysterosis and we use it for both axes and this is because both axes basically perform the role of right ascension from the PhD2 perspective. Okay. And then they tell us make sure to uncheck multiple stars. So that's a huge drawback. We've already seen that to use this primal system, we have to give up on uh avoiding flexure, right? Which means that we could be affected even if we have perfect guiding by some weird uh elongated stars. If we go for very long exposures in narrow band for instance, I would love for someone to get that equipment and test that with long 15 or 10 minutes long exposures in narrow band. And then we have a second uh issue which is that we have to uh uncheck use multiple stars which has been a huge boon in PH2. It's made guiding much better because it allows us to average the seeing of a single exposure across multiple stars. Gives us better guiding figures and better guiding overall. So now we have to take that advantage of normal guiding with an equatorial mount away as well. And the reason is simple because your guide scope has a fairly wide field of view which means that your guide scope will be affected by the uh rotation. your guide scope doesn't have a de rotator right so uh it will be affected by the rotation of the stars that are outside your primary guide star within the guide scope will be seeing something like this right so if your guide scope is has this guy as the primary star and then it uses multiple stars to get uh information from all of the others it will try to average the changes between the stars because it doesn't know that it's in out as right PHD2 believes it's equal mode and then it will basically give spurious incorrect guide command to the mount. So no, we cannot use that. And then simple stuff, we don't use deck compensation.

[00:16:34]We don't reverse the declination output after the meridian flip because there is no meridian flip. We're in out, which is one of the big advantages that we have here. But you can see there's already in terms of guiding a lot of trades trade-offs. And then in their documentation they are very clear and this is actually very very good of the two critical requirements for successful guiding which is that we need to select a guide star as close to the center of the image as possible otherwise we'll still get field rotation right and they even tell us that if no suitable star is visible near this center slightly move the telescope position in respect to the target for not too weak galaxies like M81 you can even use the galaxy core to guide. Right? So this is another issue of that kind of setup is that now we actually have to change our framing of our target to basically agree with what we can guide from. Right? So we have another constraint on what we're able to do for the advantage of having a very simple out setup which is still pretty interesting. And then they tell us that the guide scope and the main telescope must be perfectly parallel. And they properly tell us that if the two telescopes are not parallel, the startup PHD2 is monitoring and the guide camera will not correspond to the same area of the sky that your main camera is imaging making guiding ineffective. Also uh the uh the rotation rate chosen by uh your the main imaging sensor will not agree with the star that we are actually tracking on which the de rotation should apply to. Right? This could be somewhat resol resolvable by software but because we're using standard software that is not made for this purpose it's not without some custom software there. So, I thought this was fascinating and uh there's the rabbit hole or the black hole goes deeper, right? Because we've already established that likely I don't know, I would need to to double check this, but the the the the clearance for the off-axis guider might not be the real reason why we cannot use an off-axis guider. We've already seen at least uh based on my understanding of this paper here. Uh please uh math people double check it for me uh and tell me if I'm wrong that with an off-axis guider even though we're using an off-axis uh star we're probably fine in terms of the the d rotation for longish exposures. So what could be the real reason that we can not use an off-axis guider for uh guiding in uh phd2 with an out as mount? Well, for that I went down the rabbit hole of the uh PHD2 source code. And don't worry, you don't need to understand how to code. Uh I'm going to make it very easy.

[00:19:19]Basically, in PHD2, there's a rotator, right? So, PH2 connects to your rotator.

[00:19:26]And one of the important things that does is is this thing here. Uh it takes the rotator uh position from your rotator. Even there's there's even like events that listen to rotation changes, right? So, PHD2 is always aware of your uh rotator angle, right? Or in our case, it would be the D- rotator angle, which sounds like it's good enough, right? It should be able then to keep track of the D- rotation. And then what what PHD2 does actually is that it has a function under its its mount actually implementation that it uh adjusts the calibration for the scope pointing which includes taking the rotator angle from the the hardware rotator. So this guy here, right? So we take the rotator angle and then it's it does a lot of stuff in in there with the rotator angle. What it does is it takes your current calibration and just transforms it to take into account your uh your rotation. Right? If you're using uh ZWCA with PHD2 and Nina or ASI air and you're rotating between a between targets, this is what PHD2 will do. It will uh transform the uh calibration data to adjust for the rotation. It doesn't do a recalibration. Right now the very important thing here is when does this uh transformation of the calibration data based on the rotator angle happen?

[00:20:56]Well for that there's something else.

[00:20:58]This is the guider CPP. Um, basically the the whole guider routine, the most important part, arguably of uh, PHD2.

[00:21:07]And in there there is this very important line. It's basically the only place where this adjust calibration for scope pointing is called for the mount in addition to this. Uh, but this uh, when PHD2 says secondary mount, they mean adaptive optics actually. So we can completely ignore that. Plus in adaptive optics uh PG2 ignores rotation because it assumes the glass of your adaptive adaptive optics rotates together with the camera sensor. So we're looking at only this. So only this in this case do we transform the calibration uh coordinates of the calibration data to take into account your rotator position.

[00:21:46]When does that happen? It happens in a longer function which is when your uh guider has a new state for instance guiding is stopped guiding is started uh etc that kind of stuff. your the guide state changes and it tells us that if the new state of guiding is guiding meaning that you manually started guiding or nenina or the SI air uh asked PHD2 because the Sier also uses PHD2 in the background by the way uh ask PHD2 to uh start guiding then with we hit this so PHD2 before it starts guiding it looks at your rotator information it transforms forms the coordinates of the the data from your calibration to adjust to the new angle of your rotator and then it starts guiding. But it's very important is that it does it only once when you start guiding. So if you're having like a schedule where you're imaging multiple targets and you change the rotation angle of your camera in between targets, your software, whether it's the SI air nina or something else will stop the guiding slooh to the new target, change the rotation of the new target and then start the guiding again. When it starts the guiding again, PHD2, bam, it adjusts, it transforms it calibration, its calibration data for the new angle without doing any recalibration. Right?

[00:23:17]Notice the problem. If you're using a drot, your angle changes constantly while you're guiding. And so, PHD2 is not able to support a drot with constant duration. So it would need to uh basically call this function this adjust calibration for scope pointing function after every single guide exposure to transform the calibration coordinates and uh or the calibration data so that the guiding can can happen uh properly with the rotation. If it did if it did that after every single uh exposure then it should work decently for uh the rotation but currently it's not possible and I suspect it's the real reason why primaluch decided that okay PHD2 it's not going to work with an offaxis guider with a de rotator because PHD2 simply does not have the structure in place to support that. That said, PHD2 is called open PhD uh guiding for a reason. It's because it's open source and that's why we have a version of it in the SI air.

[00:24:27]That's why we have I believe a version of it in the tube text vita that kind of stuff. So it means that technically anyone including Primaluche could go in there and you know make their own branch branch of PHD2 that supports uh changing the calibration data for your de rotation after every single guide exposure before starting the next guide exposure or even while starting the next guide exposure uh but before the next adjustment is done based on the data from the next guide exposure. Right? it can be done and shouldn't be too insanely difficult to do that. So, I thought that was very interesting. By the way, while I was digging into that black hole, I saw like some old threads like what 16 years ago almost with people using a me telescope with a deer rotator and getting super cool images of galaxies like that. Let me tell you in 2010 that wasn't easy. So, uh, mad respect, uh, to, uh, Marino, I guess, for for that. Amazing. And then even when I I I I searched all around and there's an onstep onstep. So, that's like a mount firmware and driver that's used by stuff like the cell 33 from MLS.

[00:25:43]onstep has uh some code that was added I I think in yeah in 20 24 for alt azimus drives like if you're using pulseguiding and your mount type is out as or or out at whatever that means then do whatever this does I have no idea exactly what's going on and I suspect that uh this is for that reason meaning that maybe onstep mounts which obviously the Celestron CBC isn't might work way better for out asimus guiding than other mounts. I don't know. Again, I'll have all of the links down in the description. So, if you can understand what's going on here or you're involved in onstep in anyway, please tell me exactly what's that go what's going on.

[00:26:33]But I I suspect that's actually the case. It's basically like PHD2 will always think like the the guide rate is whatever it was given but it's actually being changed dynamically by onstep which I think if that's the case it's pretty smart. Okay. So now we understand quite a lot. By the way this paper amazing again if you don't want to go through all of the calculations there's also people that have given us a table of rotations at different coordinates of the of the sky. So your altitude and your azimus on the left as rows and if your azimus is like 160 and your target altitude is uh 50.

[00:27:14]There we go. Uh the uh the rotation should be at this uh speed. I don't even know what the unit is. I don't care. Uh but you can see you have tables of rotation values. And I suspect because the calculation is actually quite complex that uh any drotation software will actually refer to such tables and has to adjust the de rotation rate like in real time to uh take into account your change in out asamus throughout your session. It's really impressive stuff and and quite a lot of fun. Right.

[00:27:47]By the way, this uh information from Primaluch doesn't come out of nowhere.

[00:27:52]There's like this quite old article about like uh guiding the panther out as mount. This was a whole deal. I I wanted one anyway. The panther out as mouse with PHD2 and we basically find exactly the same instructions here using the same like I'm I'm almost wondering if like what are wait no the screenshots are not the same. Uh yeah the screenshots are not the same but you can see it's very interesting. Now, one thing that they did not highlight on Primaluchi is this deck guide mode. And let's see what do they have in here.

[00:28:28]Auto auto. So, it's it's fine. Auto has the deck uh guide mode. Okay, that's that's good. Okay, we're we're we're good. Now there's another interesting article by the way that also has like those uh object position in the sky and then the uh the actual limitation in exposure time that you can take at some um focal lengths. So you basically get the uh the information from this table about the rotation angle or the rotation speed that you need to have and you can expand that into like how long could you expose in out as without a de rotator and uh without having star trailing due to rotation. So there's there's more to it right there's more to using uh to doing astrophotography without as if you don't have a d rotator. But now I had a final question. Another final question actually. I was wondering how do the uh the professional stuff like plane wave plane wave it's a maker that makes very expensive gear that is available to uh rich amateurs like not me unfortunately but like some of you presumably uh and so they have actually very large telescopes on out as mounts. They do have wedges available, but they have rotators, de- rotators available as well. How does it work for those guys? I wanted to see. So, I looked at their uh at their website.

[00:29:56]Very cheap, like $275,000 uh stuff. I mean, this is amazing. I mean, seriously, if you had the money, if you if you won the lottery, tell me you wouldn't buy this kind of stuff. I would. I definitely would if I won the lottery, but uh it's not gonna happen because I never buy lottery tickets.

[00:30:14]Gambling is bad. Don't gamble. Uh and yeah, Poly Market, that kind of stuff.

[00:30:19]Yeah, don't do it. Anyway, uh going to the manual. I was wondering what I would see in there about autog guiding. And we see that uh we operate in an openloop tracking mode. Basically, what they're saying is that they're using encoders, right? It's the it's the thing that makes sense. You have encoders on every axis. You have encoders in azimus. You have encoders in altitude. You have encoders in rotation. If you have very fine encoders, you know exactly what's going on. And then you also have sky modeling like we see with premium mounts like 10 micron etc. With sky modeling, you know what the atmosphere is doing.

[00:31:00]So you know what at atmospheric uh defraction, whatever refraction, defraction, I don't care, but what it does to your stars and so you know exactly how to track perfectly across the star based on the atmosphere and based on your coordinates for the the rotation speed that you need to get.

[00:31:18]And since you know all of that, it's all calculated and you have encoders to double check that all of the equipment is doing exactly what you're telling it to do based on the sky modeling, based on where you're pointing, based on everything, then there's no need for guiding, right? So it's this approaches enhances tracking accuracy by enabling the system to account for positional deviations while keeping the telescope aligned with the target. And they tell us closed loop tracking which incorporates real-time adjustments based on actual exposure data can be achieved through thirdparty software as drivers and autoguiders. But in this documentation they don't tell you how.

[00:31:59]In particular they have like some more information there like we use openloop tracking so we don't care about autog guiding. Uh we don't support ask a move access method etc etc. So it's it's not quite, you know, uh that friendly to to to use like that, but we don't need to because we have the encoders. As long as you have a good sky model and encoders, you don't need the autog guiding. But because the CPC it does not have all of that, then you do need the autog guiding and then you are restricted to the limitations that we went through for out as a as astrophotography. So, it seems we've we've closed the loop. And now I'm like I'm really wanting to see how well this will perform at such long focal length in the real world. So, uh please let me know if you ever get one of those telescopes down in the in the comments because they could I think they could be a lot of fun if they work well. My guess is that Primaluch being Primaluche Hello astronomers. Yeah. Uh are able to uh were able to test that it works well enough for long exposures. But I don't know about flexure uh when you you start getting to very long exposures like 10 15 minutes in narab bandan. So that will be very interesting to see. And then there's another tiny elephant in the room an armadillo let's say an armadillo in the room which is uh what about flat frames? I mean typically with if I rotate my camera I take new flat frames with my optics. Um it's not always necessary, right?

[00:33:38]Sometimes you can reuse the same flat frames even you have some rotation, but I don't I mean you're drotation you're drotation all the time. You you cannot have flat frames from for every single angle of your optics. Although I guess you could take flat frames in advance if you never change your imaging train and you take them for every one degree and then you have a new software that can map the flat frames to the right light frames at the right angle. But that does not exist as far as I know. So if you own a plane wave system and you're not using a wedge, please let us know down in the comments what you do about flat frames. Maybe the optics are so good that we don't need flat frames in the first place. That might very well be the case for the optics of the uh the primaluch thing with the CPC edge HD optics, right? But that's something that I was really wondering about and I wasn't able to get a good information uh from my Google searches about flat frames on out as setups with a d rotator. So again, if you have more information on that, let me know. My guess is just flat frames at beginning and or end of session and then you just hope that it all works out. Uh but yeah, let me know if I'm wrong. So there you go. This was my descent into madness. I went hiking in the mountains for two days, staying in the mountains one night yesterday and the day before yesterday.

[00:35:07]I have beautiful view of Mount Fuji in the in the morning and the rest of the Tanzawa mountains. Magnificent. And while I was hiking, walking, looking at beautiful landscapes, all the all the time I was thinking about how do you do out as as photography properly with autog guiding? It makes no sense. So, uh yeah, I went deep into the rabbit hole with that. Let me know what you think.

[00:35:32]Let me know if you have more information than I do. I'll have all of the links down in the description. Um I look forward to your comments. Thank you so much for watching till the end. It really helps the channel out. uh like if you want to help even more, comment and if you want to help yet more and you're planning on buying anything from AIA, High Point Scientific, etc. uh if you do so after clicking one of the links down that I have down in the description, it will help me out at no cost to you. And if you want to sponsor the channel directly, you can uh join my channel as a Patreon member link in the description or as a channel member. It's the join button next to the subscribe button.

[00:36:06]Thank you so much for your support with that. Thank you so much for watching.

[00:36:11]Don't forget whenever you can to look up at the stars and I'll see you next time.

[00:36:35]Heat. Heat. N.

[00:36:45]Heat. Heat.

[00:37:35]Hey.

[00:37:45]Hey.