×

INDI Library v2.0.7 is Released (01 Apr 2024)

Bi-monthly release with minor bug fixes and improvements

Raspberry Pi HQ Autoguiding?

  • Posts: 11
  • Thank you received: 1
I’m thinking of acquiring a Raspberry Pi HQ Camera for autoguiding using PHD2; it has (from specs) decent sensitivity, capacity for binning and raw output, 12mp resolution, possibility to remove the IR filter as well, better at long exposures than Pi Cams v1 and v2. It does gave a rolling shutter however.

With an aftermarket 25mm c lens, it’d be equivalent to a 350mm lens on my crop sensor Nikon, and the 25mm lens is pretty fast at f/1.6. Whole thing would cost about 75 USD.

Any thought about this setup? Beats acquiring a dedicated guide scope or guide scope+cam, although a c-to-t mount exists if it is felt a guide scope would be necessary, and it’s be pretty easy to mount on the scope (straight-up 1/4 screw mounted on the ring) or on my astro dSLR body (hot shoe to 1/4 adapter). It seems to be supported by PHD2, but any optical or indi-related reason not to go for this setup?
3 years 10 months ago #56802

Please Log in or Create an account to join the conversation.

  • Posts: 163
  • Thank you received: 26

Replied by Bart on topic Raspberry Pi HQ Autoguiding?

You shouldn't thing in terms of '35 mm equivalent'. It only describes the field of view equivalent, not the viewing angle per pixel.

Do the calculation: What is the angle per pixel?

But it still could be an interesting cheap guide cam. Just need a old Thorlabs 160 mm 2" doublet for example. 3D print the housing and off you go.

I have designed a 50 mm guide scope with a Thorlabs 200 mm doublet using my 3D printer, some lathe-work and a piece of carbon fiber from e-bay. It works superb!
Super light weight at only 200 grams -including the Touptek IMX290 camera-, yet is as good as a-thermal and can be well focused.

Cheers! Bart
3 years 10 months ago #56804

Please Log in or Create an account to join the conversation.

  • Posts: 11
  • Thank you received: 1
Thanks for the response!
How would you go about calculating this? I googled rapidly and didn’t find a satisfying answer. 25mm lens states a FoV of 19.6 deg, do I just go ahead and divide by the horizontal (or vertical?) resolution or is it more difficult? And what is an appropriate deg per pixel for a guide camera?
3 years 10 months ago #56805

Please Log in or Create an account to join the conversation.

  • Posts: 1957
  • Thank you received: 420
I usually use this

astronomy.tools/calculators/ccd

or other calculators on that website.
3 years 10 months ago #56813

Please Log in or Create an account to join the conversation.

  • Posts: 1229
  • Thank you received: 568
For the record, the equations are pretty simple (and are used in the internal guider) and only a function of pixel size and lens focal length:

arcseconds_per_pixel = 3600 * (180 / PI) * pixel_width_in_mm / focal_length_in_mm

So, if you have a 600mm focal-length refractor and 3-micron = .003mm camera pixels, you'd have about 1 arcsecond/pixel = 3600 * (180 / 3.14) * .003 / 600
or the RPi High Quality Camera, 1.55 micron pixels, with a 200mm lens: 3600 * (180 / 3.14) * .00155 / 200 --> 1.6 arcseconds / pixel
3 years 10 months ago #56820

Please Log in or Create an account to join the conversation.

  • Posts: 1187
  • Thank you received: 370
To be honest: I would not recommend it for guiding. It’s a wonderful camera and I use it as an allsky camera, but it has small pixels and each image is 12MP big. Both are not good parameters for guiding cameras, which needs fast image download and good star detection with short exposure times.

HTH
Wolfgang
3 years 10 months ago #56950

Please Log in or Create an account to join the conversation.

  • Posts: 3
  • Thank you received: 1
I've been trying to get the Pi HQ to work as a guide camera for a few weeks now but have not succeeded yet.

Here is my story:

I have a C-mount to 1.25 inch adapter and then use the HQ camera in a Svbony 50mm guidescope (190mm focal length).

With either raspistill or RPi_Cam_Web_Interface, I'm able to get a star field. It is faint with never more than a few starts visible yet astrometry.net is able to resolve the images so it is adequat. See for example: nova.astrometry.net/user_images/3821744#annotated

With my 50mm Svbony guidescope and its focal length of 190mm, the field of view is as follows according to astrometry.net image:
Size: 1.91 x 1.43 deg
Pixel scale: 1.7 arcsec/pixel

My main scope is a 650mm focal length with a Nikon APS-C crop sensors that gives me 2.0 deg wide field of view. The guidecope would thus have a near 1:1 field of view and nearly matching arcsec/pixel.

Here are my 2 issues: First, I'm trying to get it to work with PHD2 via the indi_v4l2_ccd driver running on the pi. I'm on version 1.8.4 and I'm not able to get good images at night. During the day, I can see in PHD2 daylight images of what ever the guide camera is pointing too. At night however, I get a lot of noise (white dots all over that change position on every shot) and can't get stars. I obviously have an exposure, white balancing or gain issue.

Second problem is that indi_v4l2_ccd (or V4L2 by itself) does not support the camera's high resolution out-of-the-box. Someone has identified that if you manually set the camera resolution before starting indi, then it works but at that resolution, PH2 complains and I get no images (daylight or night). BTW: I use the following cmd:
v4l2-ctl -v width=4056,height=3040,pixelformat=BGR3

If anyone has figured this out yet, let us know. I'm going to upgrade to indilib 1.8.5 and continue hunting from there. I'l also try to see if I can get exposure/gain to work at when imaging using commandline v4l2 tools instead of indilib.
The following user(s) said Thank You: Eric
3 years 10 months ago #56989

Please Log in or Create an account to join the conversation.

  • Posts: 1029
  • Thank you received: 301
Would you share some log excerpts Marc?

-Eric
3 years 10 months ago #57015

Please Log in or Create an account to join the conversation.

  • Posts: 11
  • Thank you received: 1
Just finally got my HQ cam in the mail. Still waiting for my c-to-t adapter before I can properly test.
12mp is excessive and, as someone mentioned, light sensitivity is not the best, so I was thinking of doing the following:
1. Mate it with a svbony guidescope (30mm aperture, 120mm focal; not the widest but better than the official rp hq cam lenses, and a solid fixed focuser).
2. Do pixel binning - dual advantage of reducing file size and increasing sensitivity. Still will have sufficient angle per pixel since I’m planning on using a aps-c sensor dslr with a 200-600mm focal as main imager

Will try and work out the software kinks waiting for the c-to-t adapter, and the guiding thereafter. Will let you know how that goes!
3 years 9 months ago #57783

Please Log in or Create an account to join the conversation.

  • Posts: 8
  • Thank you received: 0

Replied by Damien on topic Raspberry Pi HQ Autoguiding?

Did you have any luck using the HQ camera?
3 years 2 months ago #67998

Please Log in or Create an account to join the conversation.

  • Posts: 3
  • Thank you received: 1
No I did not succeed and I moved on to use a TC7 (ASI120 clone) and it works. With the PI HQ, I could not get it to take images in the dark that looked anything like a star field with INDI. When I used some of the Raspbian web-based video tools. I was able to see bright star. This tells me the HQ camera can image in the dark but that it was INDIlib that could not configure it correctly. Note that back then, I was using the V4L2 driver. Am I mistaking or is there a driver for the the HQ camera now?
3 years 2 months ago #68031

Please Log in or Create an account to join the conversation.

  • Posts: 3
  • Thank you received: 1
So I used the V4L2 driver and it never occurred to me to check for a native driver but there is a rpicam driver here: github.com/indilib/indi-3rdparty/tree/master/indi-rpicam

The comments in cameracontrol.h states:
"Raspberry Pi High Quality Camera CCD Driver for Indi."
Last edit: 3 years 2 months ago by Marc Carrier.
3 years 2 months ago #68033

Please Log in or Create an account to join the conversation.

Time to create page: 0.802 seconds