DIY astromoding a Nikon D5300 Camera

[Blackrose]

Dear astrofriends,

 

I have recently modified my Nikon D5300 camera with tutorial from lifepixel. Everything went smoothly and now I have a modded camera. The only problem is now, that I can only focus via live-view and when doing by the eye-piece, autofocus or manual focus both fail. So generally there are two main concerns.

1) How do I make the focus(autofocus) work for the eye-piece again?

2) How can I make sure that the sensor is still parallel to the opening? The sensor board is spring loaded and you can control the height of sensor in comparison to your opening, via adjusting three screws. So for an extra turn of a screw, you'll lose the parallelism. How do I precisely achieve a parallel sensor with respect to the opening? Are there any tutorials that I can't find about this issue? 

 

Thanks in advance,

Clear skies

[beeboman82]

I previously modified my D5300 by removing the IR filter (no replacement filter), also following the Life Pixel tutorial (and some general experience from following Gary Honis' Canon tutorials).  Just today I went back in to try to regain autofocus.  Achieving infinity focus on Nikon lenses was actually my main goal, but I am hoping infinity focus will follow if I have autofocus.

Ideally when you modified you used one or more methods to mark the original position of the spring-loaded sensor screws.  This can include marking the height of the posts near the screws and marking a line that extends from one part on the screw onto the metal behind the screw so you know just where to line it up.  I also found the depth to be about 5 full turns from the point that the screws threads first catch when installing.

I've been assuming that as long as the screws are returned to the original location or are each rotated the same amount from that position then the sensor plane should be parallel to the focal plane.  In the case of my modification with IR filter removal and no replacement, I found that an extra 3/4 of a turn clockwise/inward was enough to push the sensor far enough forward to get pretty close on autofocus.  I perhaps needed just a little bit more, but 2 of the 3 screws seemed to max out around this point anyway, so I think I will leave it alone.  If you installed some sort of replacement filter, you shouldn't needed to adjust so far and probably have more flexibility to dial things in to your desire.

I have not ventured into any software to analyze sensor tilt, but it sounds like CCD Inspector or MaxSelector might be some options for that.  If you didn't carefully mark original screw positions, perhaps software might help with recovering.

Hope that helps,

Brian

[endless-sky]

I modified my D5300 myself, as well. Before removing the screws, I marked them and I also counted the turns I needed to unscrew each one of them. I, too, counted 5 turns. I did replace the stock filter with a UV/IR cut filter, as I didn't want a full spectrum camera and I read pretty much everywhere that without UV and IR cut, you get bloated stars, with purple halos, if using lenses - and I am using lenses.

The biggest problem I encountered was finding a suitable UV/IR cut filter. I couldn't find pre-cut, rectangular filters that would even come close to the correct dimensions. So, at the end, I had to buy a 2", round, UV/IR cut filter and cut it myself with a glass cutter - hardest thing ever. I completely broke the first one, even if I had 4 tries (4 complete rectangular filters would fit inside the 2"). I bought another one and finally manage to get 2 good samples. I mounted one and kept the other one as a spare.

I didn't care much about autofocus, but I tried my best to achieve it anyway. Main problem was that the new filter was a lot thicker than the original one and I tried some formula to determine by how much to correct the original distance. By the math, it seemed I had to do 4 turns and 3/4, instead of 5. But I had to stop at 4.5, because the screws were becoming really hard to turn (I suspect the extra thickness of the new filter made the sensor/filter/enclosure reach some mechanical stopping point before the 4 and 3/4 turns), so I didn't force it and stopped at 4.5. Screwing all 3 screws at 4.5 seems enough to keep the sensor planar (I have bad tilting on the lens I use, but the stars have exactly the same shape in the framing before and after the modification). When I get my imaging refractor next month I will do eventual adjustments, if needed.

I couldn't, however, retain autofocus. It's not a big deal, as I don't use the camera for everyday use and I use either live-view or the laptop screen for focusing, with a Bahtinov mask - didn't have any problems with that.

Edited August 2, 2020 by endlessky

[beeboman82]

Yes, good point on the UV/IR filter.  I was too scared to try glass cutting and balked at pre-made ones costing 1/3 the amount I paid for the camera.  That means that, at the moment, regaining autofocus accuracy and infinity focus is not useful for astro work with camera lenses (and was never needed for use with telescopes).  At some point I may look into a clip-in filter, but the options are not as plentiful as for Canons.  I do occasionally tinker with the camera with IR long pass filters for daytime work, so maybe sensor positioning was useful for that(?).  For now, arguably gaining autofocus was more for fun than necessity, but I'm hoping also an investment in future flexibility.

Brian