RaspberryPi Camera for Astrophotography

[martindtm]

Question for the group:
Does anyone have a list of the best settings for the camera for astrophotography? When I am trying to focus on a target, due to the latency and the automatic cam settings, it wavers so drastically it is very difficult to find the sweet spot.

[gordon77]

Here's some pictures of the latest housing in the Pi Display case, and Pi Noir camera on HDMI extension.

[ab1jx]

Enjoyed skimming this, was looking for examples of what the v 1.3 Pi camera could do with different lenses before I start trying to do it myself.

What would happen if you mounted 2 Pi cameras side by side in the telescope eyetube? Would they be far enough apart to give stereo?

[gordon77]

I think you need something like this...

http://www.rothervalleyoptics.co.uk/cel ... r-125.html

And 2 pi cameras

[ab1jx]

Oh, OK, I was just curious because the bare cameras are small enough that 2 would probably fit into a 1-1/4" tube. Somewhere I saw stuff about using 2 for stereo. And the old version 1.3 is going for $8 or less each in places like gearbest.com, dx.com, mininthebox.com. I don't know how you'd arrange them, maybe put 2 of the Sunny boards back to back with the cables coming off at a right angle so the cameras are side by side.

I wasn't impressed with the quality but I expect to have 3 of them in another week or so. I'm not sure how much to blame on the lens. This is completely stock with raspicam and defaults.

picam_2017-10-11_15-36_orig_1024.jpg (98.35 KiB) Viewed 5337 times

Compared to a Nikon

DSCN8253_lev_cur_shp_960.jpg (150.5 KiB) Viewed 5329 times

[gordon77]

You could try 2 of these..

https://thepihut.com/collections/raspbe ... ry-pi-zero

Or these...

https://thepihut.com/collections/raspbe ... spberry-pi

They should fit.

[ab1jx]

Just got my 3rd v 1.3 camera and I'm wondering if there are variations in quality between individual cameras. This one seems a lot better than the last. Haven't had it outdoors or devised any good tests yet. Like finding a rare sunny day in December and trying it outdoors with a Macbeth chart https://en.wikipedia.org/wiki/ColorChecker

[Fer]

Just got my 3rd v 1.3 camera and I'm wondering if there are variations in quality between individual cameras. This one seems a lot better than the last

Most probably, what you're seeing is variations in focus.
Raspberry Camera lens is screwed into mount and then semi-locked in place by Loctite (or something like that).
Since focal length is extremely short (3.6mm if I remember well), a tiny difference in how much the lens was screwed into the mount (during assembly) translates in sensible focus error.
Moreover, from tests it looks like the lens is "factory-focused" at about 2m , while you're trying it outdoors (100m +). Depth of field isn't enough to cover the difference.
When correctly focused, the tiny Raspberry Camera is actually quite capable.

[gordon77]

I've been experimenting with cooling the Pi camera to reduce the noise, so far looks promising...

Sadly it didn't prove sucessful, yes reduced the noise BUT made the pictures unusable, maybe condensation ?

[gordon77]

The cooled pi camera... REMOVED

[ab1jx]

How cool? Dry ice (-79C) and liquid nitrogen (-195C) https://en.wikipedia.org/wiki/Liquid_nitrogen are both widely used in chemistry, biochemistry, physics. And once you're equipped to handle them they can be surprisingly cheap. I left that job in 1990 but I seem to remember something on the order of $0.50/liter for nitrogen. It's mostly made by big compressors, in this case one college owned one and sold off nitrogen to neighboring colleges essentially at cost. Sort of like the origins of the internet.

Liquid helium (-269C) is expensive enough that we had a substantial amount of plumbing to recover the boiloff. If you're really trying to get close to 0 degrees K that's about the best you can do. Google "optical dewar" for ways to use it. We were mostly passing a light beam through a sample at those temperatures and measuring what came out in a straight line. A few instruments that looked at fluorescence also measured what came off at right angles. Mostly we had cuvettes that were 1 cm on side, not that different from the size of a bare Pi camera.

Dangerous? To a point. Both are very cold so they need gloves. Both are gases trying to regain their gaseous state, so don't confine them in all directions. I use to make liquid oxygen by bubbling oxygen from a tank through copper tubing submersed in liquid nitrogen, then it went into a transparent dewar used in lecture demonstrations to show that it was sky-blue. Most of us handled them every day, thought little of it. Some cameras in astronomy are cooled by these methods, Google it, also "dark current". I worked in chemistry, not astronomy.

Think of the bare camera at the bottom of a test tube pointed up. Lower that into liquid nitrogen, you can still angle it to point at the overhead sky. The obvious problem is going to be the focal length of the lens vs. the length/depth of the test tube. If you don't have a dewar a thermos bottle is close (just don't put the cover on) or maybe about 3 styrofoam cups in a nested stack. Less thermal isolation just means having to replace the nitrogen or dry ice more often.

[gordon77]

Nice idea but a peltier is a bit more practical for me for experimenting

[ab1jx]

Aw, shoot, I used to drop dry ice into my beer to make a fog and look cool. Tasted awful though.

OTOH, I seem to get about 1 clear night every few weeks, without that I can't do anything.

[gordon77]

M31, 12 x 3mins, processed in Paint Shop Pro. Sony A6000, Canon 400mm 5.6 guided on NEQ6 by PiAG.
(Looks noisy on here due to resizing)..

Also M42, equipment etc as above...

[gordon77]

I just downloaded your Autoguider software and got it up and running using a Celestron Digital Microscope. Next I'll fab an interface board and try it with my CGEM.

Q: How do I "train" the autoguider about the orientation and magnitude of movements? Other autoguider systems go through an orientation process since the guide camera could be at some arbitrary rotation angle relative to RA and Dec axes. Is this done manually with your software?

Thanks!

I finally got around to adding some calibration to the script, consider it beta at present until it's proved. Tried it on a target in the garage, hence the square star below, but I don't expect to see any stars this week due to the weather . Note you can't test RA with a fixed target.

Set the star to the centre of the screen , try zoom setting around 4 or 5. Click on the star and then click on Esc (will go yellow) and then click on Dec+/Dec- or RA+/RA-. It will go into CALIB DEC or RA mode, where it will move the telescope for 5 seconds (set by Calibration_time).
When complete it will set 'scale DEC+' and 'scale DEC-' or 'scale RA+' and 'scale RA-'. If there is an error in the aligment of the camera > 5 degs it will show you the amount it needs to rotate (just above R3 button).

[gordon77]

Still work in progress. I managed to test using a setup I built to simulate a star moving at sidereal rate. Very Heath-Robinson but seems to do the job

Click Esc then telescope 'centre' to activate calibration.

Now does both DEC and RA as one operation. It will show 2 angles, DEC and RA . If similar then it will take the average, and you need to rotate the camera. If one is very different eg 179, -5 then click on DECi / RAi as appropriate and try again.

[gordon77]

Managed some testing last night.

M42, 17 x 2mins, processed in Deepskystacker and Paint Shop Pro. Sony A6000, Canon 400mm 5.6 guided on NEQ6 by PiAG.

calibration will now automatically switch Deci / RAi as required.

10e11 now has the option to change the camera angle manually or automatically after calibration. Set 'auto-rotate = 1" in DISPLAY section for automatic. Manual control appears when calibrating, set the angle shown after calibration. Only works with pi camera at the moment.

https://github.com/Gordon999/Pi-AutoGuider

[gordon77]

Calibration hopefully now working better. Click on Esc , then CAL...

You can set the CAL time if too long to stay within the window with your setup.

You can also set the CAM angle if you haven't aligned tha camera correctly (I recommend rotating the camera if too far out of alignment)
Run calibration, this will give you an angle error. If you want to use software rotation enter this error figure in the CAM Angle setting.

https://github.com/Gordon999/Pi-AutoGuider

[gordon77]

Now includes more control of a USB web camera, may depend on model of camera.

Note controls based on a Philips SPC900NC, Philips PCVC740K ToUcam Pro and Logitech C270 webcams, others may vary.
If you have a diffferent model then try 'v4l2-ctl -L' command in terminal as see what controls it lists.
If different upload the results to here and I will look at modifying the script to suit.


10f2 USB webcam Auto_Gain added
10f3 I found an issue with the webcam display and zoom, needs more work but OK with the 2 cameras above.
10f4 Issue with USB webcam CAL, hopefully fixed
10f5 Hist (stretch histogram) added for noisy images, main use webcams. Note still only supports webcams named above.
10f6 one more option to Hist added, some bug fixes
10f7 Control of Logitech C270 added. Set USE_PI_CAM = 0, WEBCAM = 1, USB_MAX_RES = 4.
10f8 minor bug fix to user config
10f9 control of Sainsmart USB 4 relay card added (you need to install crelay http://ondrej1024.github.io/crelay/ )
See here for help installing crelay if required. https://www.raspberrypi.org/forums/view ... 0#p1336648
11a1 minor fixes, Pi Camera added as webcam, set use_Pi_Cam = 0, Webcam = 2. Note NO zoom
(You'll need to install imutils, sudo pip3 install imutils)
12e minor fixes, Pi Camera added as webcam, set use_Pi_Cam = 0, Webcam = 3. Note NO zoom
(You'll need to install opencv). ConfigX.txt changed format.

https://github.com/Gordon999/Pi-AutoGuider

[PaulR2019]

I have ben reading this thread and am finding it very interesting. I have been mounting my iphone to my telescope 10MM lens to take some pictures and that was going well. The thought about using the raspberry pi camera modules I have. Currently I ma having little success. I can't seem to find the right combination. I have tried the PI camera with lens through the 10MM eye piece like I do with my iphone but no luck. I then tried removing the lens and using an adapter and the prime focus method I don't seem to have the range to focus on the moon. I can see it but I can't get a good image. I have ordered an eyepiece extender with the hopes that will give me the range I need.

I have seen some of the pictures of fully build or mounted Pi camera but nothing disassembled, could someone show this? are these builds all prime focus or are they using the eye piece lens.

Thanks for the input.

oh and here are some pictures of the prime focus mount I am using.

Paul

[gordon77]

Hi,


I use this type of adapter, simply screws in the 12mm mount you have on the camera. Simply insert it in telescope in place of the eyepiece.
You may need some extensions to get the required focus point.

https://www.rothervalleyoptics.co.uk/rv ... r-125.html