Starsense Explorer Stand Alone Conversion

[DaveL59]

do bear in mind Mark that you'd need to buy a starsense explorer scope so you can get the unlock code, else the starsense app won't let you use the camera to platesolve. The LT70AZ is the cheapest one at £135 but stock is low like with most scopes at the moment. Performance wise it's ok but the mount is real shaky, if patient with it tho it does pretty well I found. Was able to give it a short try out w while back with the SV171 zoom but I expect it'll perform better with a proper star diagonal rather than the erecting one supplied.

Yet to have a chance to test this on an EQ mount but can't see any reason it won't work albeit the directions may get confusing being geared for alt-az but will report back when I get a chance to test it out.

Edited October 11, 2020 by DaveL59

[AdeKing]

Finally managed to finish off the mods to allow me fit the Starsense phone holder to my OOUK Dob mount.

Was hoping to get to try it tonight. Clear skies when I finished up and put the Dob outside.

Just been outside and it's horizon to horizon cloud, grrrrrrr.

 

[DaveL59]

always the way Ade, here it's not too bad tho cloud is drifting across now from the NW, but tired eyes and needing to be up and out early, need to go unload tools etc from the car before I head off to zzzz so won't be playing tonight. Would like the epoxy to have a chance to cure off properly too before I try using that home-made prism rig 🙂 

[markse68]

couple of questions- does it really need the funny looking funnel light shield at the front do you think?

And anyone know about prisms? I found this in our junk optical box- it’s from an old industrial laser printer iirc. I suspect it may be a beam splitter with partially reflective coatings but not sure, but if the camera looks through a short side it sees out the other short side but also partially (the curved part in image with my fingers in it) out the long side 🤔 That’s not normal is it? It’s a bit on the big side anyway

I’m just playing around thinking about it for possible future purchase

Mark

[johninderby]

Easiest thing is just to use a cheap plastic mirror star diagonal and cut the front and eyepiece barrel off.

[DaveL59]

3 hours ago, markse68 said:

couple of questions- does it really need the funny looking funnel light shield at the front do you think?

And anyone know about prisms? I found this in our junk optical box- it’s from an old industrial laser printer iirc. I suspect it may be a beam splitter with partially reflective coatings but not sure, but if the camera looks through a short side it sees out the other short side but also partially (the curved part in image with my fingers in it) out the long side 🤔 That’s not normal is it? It’s a bit on the big side anyway

I’m just playing around thinking about it for possible future purchase

Mark

 

 

 

If the prism will give an inverted view with the camera looking in one short side and therefore of what is in front of the other short side, assuming you have good internal reflection off the long face then you'd be good to go. An old 10x50 bino prism is what I'm using in the rig I made, but a plain regular mirror also works, doesn't need to be first surface from a test I did the other night. As John says a cheap plastic diagonal might be easier as that already has the box you can glue to a phone holder etc.

As for the big wings in front, they baffle the stray light from the mirror so I guess if there's total darkness you won't need them, but not hard to add as needed if you find it necessary. I thought it worthwhile as I've no control of neighbours turning bright upstairs lights on etc.

[markse68]

23 minutes ago, DaveL59 said:

If the prism will give an inverted view with the camera looking in one short side and therefore of what is in front of the other short side, assuming you have good internal reflection off the long face then you'd be good to go

this prism is doing this weird thing of partially seeing through the long face- this image is camera looking through short side and the curved bit of decking and fingers is seen through long side which it shouldn’t. I reckon it’s a semi silvered beam splitter or something- not the right thing. Shame as it doesn’t vignette otherwise. I liked the idea of a prism as it’d be easy to wipe condensation off in the field. 

Edited October 12, 2020 by markse68

[DaveL59]

4 minutes ago, markse68 said:

this prism is doing this weird thing of partially seeing through the long face- this image is camera looking through short side and the curved bit of decking and fingers is seen through long side which it shouldn’t. I reckon it’s a semi silvered because splitter or something- not the right thing. Shame as it doesn’t vignette otherwise. I liked the idea of a prism as it’d be easy to wipe condensation off in the field. 

hmmm I get that if the prism is in free air, but if you hold dark/flocked card against the long edge the the image correct to be only what is in front 🙂 When there's plenty of light you get the interference of light entering via both exposed surfaces unless they are silvered or blocked in some way.

For me I didn't see the point buying a diagonal even cheap when I have a couple spares binos just kicking around waiting for a good use to come along 😉 

Edited October 12, 2020 by DaveL59

[AlexK]

The best way to design that beam folder is to model it in 3D CAD software (there are plenty of free options). Myself, I've been tinkering with that idea for my Galaxy S8 and a 10x10cm first surface mirror I've got off eBay for like $5):

The yellow pyramid is the video frame FOV (quite wide 90 deg diagonally) of the main camera, the blue one is its reflection from the mirror (generated automatically as you move the mirror). Everything in the model is parameterized and snapped together, so you can change the angle of the mirror assembly and move it relatively to the camera chip location along two axes to see how the reflected FOV cone changes/behaves.

My primary conclusions:

1. The closer the mirror's edge to the camera the wider the FOV it is covering (kinda obvious for the Z axis, but it also important to move the edge by X axis (along the phone).
2. The best mirror angle (for my camera) providing maximal unobstructed FOV is 54 degrees from the sensor's plane.
3. A lot depends on the camera location in the phone body, as the primary obstruction you have to deal with is the back panel of the phone from the camera center to the top edge.
4. Even a 10cm mirror is too short to cover the entire camera FOV (depends on the phone model). So the wingetting is unavoidable. But looking at the above projects, it seems like the SSE software is quite tolerant to vignetting. 
5. The first surface mirror is still beneficial as in any optical application. Looking at the sharp angle it has to use at the edge of the camera FOV it is obvious that the regular mirror will introduce plenty of distortions as the glass thickness for the light to traverse is changing from its center to the edge. But again, SSE plate solving software could be tolerant enough to that as well. However, I would expect it to be "stressed" by all of the above, possibly enough to have the pointing less accurate or/and taking more time or/and losing the alignment more often than usual.

I'm not pulling the trigger on a SSE yet as I'm 100% satisfied with the accuracy of pointing with the QuInsight device at the moment. But already designing the minimalist phone cradle (pictured above the glove-tight-fit phone case and the temporary mirror holder for experimenting with the mirror position and angle) to 3D print for my field smartphone as soon as we finally get the standalone CSSE (or a competitor) product.

Edited October 12, 2020 by AlexK

[DaveL59]

Hi Alex

yes it does seem tolerant as in my testing so far the prism and plain mirror solved faster than the starsense OEM rig which surprised me. I was aiming the camera high to NW at the time and that way is the London light dome tho my back garden is reasonably dark. Others on CN have tried with plain mirrors, shaving and the like and reported it works just fine. The mirror I used was very thin glass and fell off a hairbrush so I figured why not see if it works. Losing some of the FoV doesn't seem the be a big issue so long as there's enough sky for it to solve, more a problem is cloud or brightness in the sky. It warns of a bright moon and suggests to look at other regions of the sky as accuracy can't be assured in the region of the devil's floodlight.

[AlexK]

That all makes sense indeed. I wonder what would be the minimal usable FOV? As a prism is what I had in mind for that design myself initially but then saw like 80% of FOV obstructed with it in front of the cam. That would allow to design a very compact cradle which is easier to install on the OTA conveniently and without making a snag hazard.

On a side note: I'm using the smartphone star chart extensively with the phone mounted close at the eyepiece so I would prefer the CSSE mounted at the same location too. Sadly, folks have already discovered that CSSE doesn't work at awkward angles, only with the phone X axis (left to right) horizontal. So a special shoe will be required to hold it near the EP like that (I have a design for that too, but want to refine it to be interchangeable with my QuInsight shoe). That's another reason I prefer to wait for a competitor app or for the Celestron improving upon its creation so folks with EQ and Ball mounts could benefit from it too.

And a little tip about the clouds and other "sky devils" : 
If you make your cradle a bit movable around the field, then you should be able to re-calibrate it pointing into slightly different direction as needed to catch more stars and less obstructions (it should work being pointed in an arbitrary direction as soon as you can still register the alignment star/terrestrial object within the camera FOV; e.g. closer to a certain corner of the screen). That needs experimenting though, especially with DIY beam folders, as optical defects of the camera and reflection along with vignetting might increase the image errors in a way Celestron programmers have never anticipated in their algorithms tolerances.

Edited October 12, 2020 by AlexK

[DaveL59]

The loss of FoV with a bino prism isn't too bad really, a squared prism like Mark showed would be better but I didn't have one of those laying around so tested with what was to hand. The rig I fabricated should work well enough given the first mockup was some acetate film and flocking and that located under the stars very well sat on a photo tripod. Not having a 3D printer, hacksaw and acetyl sheet was the easy route, not to mention something to do to pass the time!

I plan to use it on EQ mounts as most of mine are that way 🙂 Not had a chance to test it out like that tho as weather here has been typically cloudy/wet of late but hopefully it will be usable. Always a chance that Celestron will release it as a stand-alone one day with a finder shoe mount, but the base model scope isn't exactly expensive for a play and other than the mount it performs pretty ok as it happens.

[AlexK]

Wish you good luck with the sky, Dave! I'm eager to hear your findings on a GEM.

Tests I'm referring to were done on a Ball scope. In a nutshell, as soon as the phone screen looks too much to the side it's refusing to proceed with the pointing. I guess that's because their accelerometers data fusing has an angle restriction or has a check of the gravity vector to be in some indirect sync with what the camera shows for a given geolocation. Totally avoidable (in the algebra involved), but we cannot possibly know Celestron's rationale behind that (from a programmer's negligence all the way to preventing the widespread cradle hijacking 😎 ).

Re 3D printer: The tech is super-mainstream by now. You can have a decent tool in your man's cave for just $200 shipped. Given the well known UK/EU prices on trivial astro gear "issue" that cost will be recovered in no time.

Edited October 12, 2020 by AlexK

[DaveL59]

1 hour ago, AlexK said:

 

Re 3D printer: The tech is super-mainstream by now. You can have a decent tool in your man's cave for just $200 shipped. Given the well known UK/EU prices on trivial astro gear "issue" that cost will be recovered in no time.

thanks, hopefully will get a chance to test sometime in the next months 🙂 Interesting re the accelerometer  I guess that could become a problem, I guess we'll know soon enough.

As to the printer, yes I have thought about it but where I currently am, space and curious kitties are the issue. One day perhaps...

 

[markse68]

29 minutes ago, DaveL59 said:

Interesting re the accelerometer

I was wondering if there’s any reason you can’t just mount the phone vertically and use the camera without a mirror but i guess the code is written for mirrored image and non mirrored accelerometers if they’re using them as well. Interesting 🧐 

I think i found the prism issue- I found this smaller one in an old viewfinder and the long side is silvered and it doesn’t have the dark circular ingress when looking through it. It does suffer reflections though- possibly as it isn’t coated. A mirror probably would be better/simpler. This one is 26mm wide and doesn’t vignette but i don’t think you could go much smaller. The bigger one is encased at the back (long side) and not silvered and still shows the circular ingress.

Are prism diagonals silvered?

Mark

Edited October 12, 2020 by markse68

[DaveL59]

not 100% that the prism I'm using is coated but might be. Are you seeing reflections between camera lens and the prism then? Could be the light level, at night it may not cause any issue unless aiming at a bright target. As you can see I flocked the edge facing the camera as well as the long edge, cut a hole wider than the camera and flocked the area under the phone. Idea was to quash as much stray light as possible and to give the prism as optimal a housing as I could. Seemed to work quite well. A bit of vignetting doesn't seem to affect the ability to get a star fix so I've gone ahead with building it. 

I do see a section at the lower part of the image that has a darker area as per your pic but it didn't seem to affect it at night so could just be stray internal reflections in brighter light. Saw similar with a prism diagonal of older japanese vintage when looking through the prism while cleaning it and when back in the diagonal looking down the barrel. But in the scope day or night I didn't note the effect at all. I did black the ground edges on the prism I'm using just to try to dull off any potential effects, kinda mute really tho being it is now fully clad in a flocked housing 🙂 

[AlexK]

Mark. The prism effect you are having caused by the wide angle of the camera FOV. If you raytrace the edge of the incoming FOV cone you will see that it's forced to pass the prism at the "spectroscopy" angle 🌈. I.o.w. refracting it. So that's actually the vignetting. The curved edge you see is the "rooftop" of the prism. With the mirror you will see the back of the phone. In a telescope the FOV convergence is a rather small angle, thus no such situation possible.

Edited October 12, 2020 by AlexK

[markse68]

3 minutes ago, AlexK said:

Mark. The prism effect you are having caused by the wide angle of the camera FOV. If you raytrace the edge of the incoming FOV cone you will see that it's forced to pass the prism at the "spectroscopy" angle 🌈. I.o.w. refracting it. So that's actually the vignetting. The curved edge you see is the "rooftop" of the prism. With the mirror you will see the back of the phone.

But it doesn’t happen with the smaller silvered prism- ah you mean it’s exceeding the angle of internal reflection and leaking out? yes that must be it and the slivering prevents it. I guess with a prism diagonal there’d be no need for slivering as the light cone is much closer to parallel. Could also be the smaller prism is a lower dispersion glass I guess- it “looks” different

Thanks for that Alex- was very puzzling

Mark

[markse68]

5 hours ago, DaveL59 said:

Are you seeing reflections between camera lens and the prism then? Could be the light level, at night it may not cause any issue unless aiming at a bright target.

No i don’t think so Dave although it could be i guess- I think it’s partial reflection ghost images of brights like office lights that move away from the source as you tilt the prism/camera. Probably not an issue in real use. I get similar off the uncoated prisms in my adc. I think that’s the main advantage of coatings apart from higher throughput- they seem to work backwards too and prevent unwanted internal reflections 

Edited October 13, 2020 by markse68

[johninderby]

Mini ball head arrived for making a tilt adjuster for the Starsense unit.

As most finder shoes are at an angle and the Staesense unit should be level used this little ballhead to allow adjusting the angle. Straight forward job using the parts already there from making the mounting bracket. Needs a hex key to adjust but means it’ nice and solid. 👍🏻

[markse68]

11 minutes ago, johninderby said:

Mini ball head arrived for making a tilt adjuster for the Starsense unit.

As most finder shoes are at an angle and the Staesense unit should be level used this little ballhead to allow adjusting the angle. Straight forward job using the parts already there from making the mounting bracket. Needs a hex key to adjust but means it’ nice and solid. 👍🏻

What’s the rack and pinion for John? Is it part of the phone clamp? Seems very elaborate!

[johninderby]

3 minutes ago, markse68 said:

What’s the rack and pinion for John? Is it part of the phone clamp? Seems very elaborate!

 

Part of the phone asjustement. There is one for side to side and another for front to back.

Edited October 13, 2020 by johninderby

[Steve Clay]

John, what screws did you use to attach the acetal plate to the phone holder?

Cheers

Steve

[johninderby]

1 hour ago, Steve Clay said:

John, what screws did you use to attach the acetal plate to the phone holder?

Cheers

Steve

M3 self tapping screws. Think they were 18mm long.

Fortunately I had a few in stock. 😁

[Steve Clay]

Thanks John.