×

INDI Library v2.0.6 is Released (02 Feb 2024)

Bi-monthly release with minor bug fixes and improvements

Regress to QHY driver 1.5 on the Pi for the SSAG?

  • Posts: 4
  • Thank you received: 0
Hello,

I'm exploring whether I can use INDI server on my Pi and for the most part, everything seems great! The part that doesn't work as well is my SSAG. Judging from other research I've done on the web, the SSAG seems like the outcast QHY5 stepchild, and I can't quite tell whether it should even work at all with indi_qhy_ccd.

What's interesting is that the SSAG seems to work when I connect it directly to my MBP running EkosVM. I can connect, adjust the exposure length and take an image without an issue. While I haven't done any actual guiding off it, the exposure I take (of the ceiling) seems consistent with with what I'd get if I were to take an exposure with it using PHD on my Windows laptop.

When I run INDI server off the Pi, it connects and I can adjust the settings. The image I get back after taking an exposure that suggests something isn't quite right.

Other than the obvious differences between my MBP and the Pi, the QHY driver version on the Pi is 1.6 while the one on EkosVM is 1.5.

I've attached two screenshots, one showing the image taking from EkosVM when the camera is locally attached to the laptop. The other image is what I get back when KStars is talking to the Pi.

Is using the SSAG with INDI a crapshoot or have I come across a bug? Or Both? If the SSAG should be reasonably well supported using the QHY driver perhaps I could try regress to version 1.5 on the Pi?

Can anyone offer some insight as to what's going on or what I could try? (I've looked into regressing the QHY driver on the Pi back to 1.5, but I'm not having any success in finding the source for it.)

Thank you!
7 years 10 months ago #8531
Attachments:

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

We just "upgraded" to a newer QHY SDK which could be the cause of this issue. I believe Peter already informed QHY about the bug. I will also forward this to QHY. On the other hand, there is another preliminary SSAG driver in the repo, but I'm not sure if it is ready for consumption yet. Will check if QHY has any response on this.

Can you update the VM? (sudo apt-get update && sudo apt-get upgrade), it will upgrade to v1.6, let's see if this Pi specific or QHY is broken everywhere.
Last edit: 7 years 9 months ago by Jasem Mutlaq.
7 years 9 months ago #8533

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

  • Posts: 486
  • Thank you received: 87
That happened to me too last night, before the last update I was using it with the indi_qhy_ccd drivers and had no problems.
Some time ago I used the SSAG on WIndows with QHY drivers changing the VID:PID but had a hard time with this camera before getting it right.

I am getting the same pattern as attached. Using Driver version 1.6 interface 22, available in the Daily INDI Library for Raspberry PI.
My RPi B+ has Debian Jessie Raspbian installed so I can't upgrade to the latest version on the PPA as it is based on libstdc++ 5.2.
Last edit: 7 years 9 months ago by nMAC.
7 years 9 months ago #8536
Attachments:

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

I tried to build the latest SDK QHY on my machine and it fails to build *sigh*, so contacting QHY to inform them about the issue.
The following user(s) said Thank You: nMAC
7 years 9 months ago #8538

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

  • Posts: 712
  • Thank you received: 174
I let QHY know, sent them patches to make at lest QHY5/SSAG work, but have no response yet. Peter
The following user(s) said Thank You: nMAC
7 years 9 months ago #8539

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

  • Posts: 4
  • Thank you received: 0
Thanks for the speedy feedback, guys!

I updated EkosVM and the image is now the same as that taken from the Pi. Both are reporting version 1.6. I've attached a screenshot.

The connection also seems flakier now. After I've taken a shot, I'm usually unable to take another until I disconnect the device and shut down the local indi server and then fire it back up.

Several weeks ago, I started toying with the idea of INDI device development. I grabbed Eric Holmes' OpenSSAG code and was able to build it without issue. I was never able to actually use it successfully, though. When trying to connect I would always get the message from the library that it couldn't find the camera. The camera was showing up with no issue in the list of USB devices, but something was evidently going wrong (probably with my code). At that point, I realized I was quickly getting out of my depth for device driver development, so I let it go. :unsure:

I then decided to try the indi_qhy_ccd driver in EkosVM and it just worked! I always thought the SSAG's ID prevented it from showing up as a QHY device (but what do I know?).

Please let me know if there's anything I can do to help, even if it is just testing new releases of the QHY libs against the SSAG.

Thank you!
Garth
7 years 9 months ago #8544
Attachments:

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

  • Posts: 1
  • Thank you received: 0
Just wanted to give another example for the above mentioned issue. As far as I can tell there is no error thrown for me from the server, and the issue is present both when using the PI and when directly connecting to my laptop.
7 years 9 months ago #8548
Attachments:

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

  • Posts: 193
  • Thank you received: 46

We have both the SSAG and the original QHY5 here. I can say with a large level of certainty, they are identical hardware with a different vid/pid programmed into the eeprom on board. Some years ago, I confirmed this by writing a small program for the on board fx2 chip (long since lost) that re-wrote the vid/pid eeprom on both of them, so now the QHY reported as the SSAG and vice versa, saw absolutely no difference in operation after the swap.
7 years 9 months ago #8552

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

  • Posts: 712
  • Thank you received: 174
The have different VID/PID in uninitialised state, once firmware is uploaded (in case of linux by USB rules, not by INDI driver), SSAG and QHY5 are identical. Peter
7 years 9 months ago #8556

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

  • Posts: 193
  • Thank you received: 46

This was what got me going down the road of testing / checking years ago. Back then, we were still using windows for our telescope kits, and the QHY5 drivers were somewhat unstable, but we found the SSAG drivers to be relatively stable in comparison. They both reported different vid/pid even after firmware load on windows, so my goal was, turn them both into SSAG, which I accomplished by using sdcc and fx2lib to write new values into the eeprom for the uninitialized vid/pid of the device.

I had been toying with the idea of reverse engineering a few of the connections inside the gadget, and writing a completely new firmware for it at the time, but, it's a lot of work for very little return. Once we got our systems stable with both of them programmed to be an SSAG, I abandoned that idea.

Maybe some day when I have more free time that I have projects, I'll re-visit the idea. ofc, the concept of having more free time than I have projects to deal with, not likely to ever happen :)
7 years 9 months ago #8557

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

  • Posts: 486
  • Thank you received: 87
I had this issue in my RPi B+ and on a Ubuntu Mate PC. Have just updated the PC with the latest indi_qhy_ccd and the image is normal.
Disconnecting the device gives me a crash on the driver.
Added a Kstars log and a commented server output log.
Last edit: 7 years 9 months ago by nMAC.
7 years 9 months ago #8580
Attachments:

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

Please try tomorrow to see if you get the same issue after the update.
The following user(s) said Thank You: nMAC
7 years 9 months ago #8586

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

Time to create page: 0.487 seconds