Hy Murveit replied to the topic 'AstroPi3 Scripts revised' in the forum. 3 days ago

I must say that 98% of the issues I have with my rig boil down to WiFi/networking, so, no this isn't a silly question to me either.

First, re the hotspot password. I believe when I installed via the script, the hotspot password was something like raspberrypi_password., not 123456789. You can find out for sure by looking in the folder /etc/NetworkManager/system-connections/ and finding the file that configures your hotspot (should be obvious from the filename) and then looking at that file (e.g. by running 'sudo cat FILENAME') and there will be a line with the password in plain text.

Second, though the standard WiFi connection to my home wifi usually works, it often doesn't or drops out. As opposed to using the hotspot, my approach is to use the hardwired cable approach--that is, connect a cat5 or cat6 line between the ethernet connector on the RPi and your laptop (if you have a macbookpro like I do, you'll need a ethernet-to-usbC converter) and you should be able to see the RPi at the address you set on script installation (it's possible that that requires some configuration on your laptop). This is a much more reliable/fast connection than the hotspot IMHO, and also probably avoids whatever was messing with my other WiFi connection in the first place. Of course another approach is connecting a monitor and keyboard to the RPi (which is my ultimate work-around), but that takes more time to set up.

I wish I knew why the WiFi is occasionally flakey. I assume the reason is the wimpy WiFi antenna inside the RPi. I've bought this thing which I've seen referenced by Rob www.amazon.com/gp/product/B07695YGLZ/ref..._title?ie=UTF8&psc=1 but haven't tried to install it yet. Given the pain WiFi has caused, I should up the priority on that project.

Hy

Read More...

Hy Murveit replied to the topic 'Focus Module revisited' in the forum. 2 weeks ago

above should have said:
**one or more volunteer testers**

Read More...

Hy Murveit replied to the topic 'Focus Module revisited' in the forum. 2 weeks ago

I'd happily volunteer to code a unidirectional focuser, if I get one or volunteer testers!
If you're interested, respond on this thread or message me.
I've been thinking the same thing for a while.
I warn you, though, that I believe backlash is only part of the problem.
I believe the HFR calculations (and, of course, seeing) are noisy and contribute to the lack of good v-curves.
Hy

Read More...

Hy Murveit replied to the topic 'AstroPi3 Scripts revised' in the forum. 2 weeks ago

Rob,

I wanted to commend you on these scripts! I ran the setupAstroRaspbianPi.sh script and it work flawlessly. I had previously put together my Raspbian Astro setup manually, but decided to redo it, and wow--what a time saver!!

I hope you don't mind--when I like something I tend to want to make it even better, so here are a few small suggestions (nits really) for you to consider.

1) There are 4 questions that I needed to answer during the course of the script. The script takes a long time to run. Some of the questions were asked after quite a while. When there is a long running script, I have a tendency to go do something else, and the script stalls waiting for my response. Why not front-load all the questions? Ask them at the start and store the answers in variables and use them when the time comes.

2) A few times I really wasn't sure what to answer. E.g. is there a downside for a static IP address? (I wound up saying no static IP, but then realized I wanted it later, and manually executed your instructions). Perhaps add some more doc about the plusses and minuses of the answers, or better, a link.

3) This would take a little ongoing work for you, but why not have the Raspbian script compile the latest release, instead of the bleeding edge? In my case, after the script was done, I ran the following: cd ~/AstroRoot/kstars; git checkout 17dcb543a; cd ../kstars-build; sudo make -j 6 install
Yes, I had to figure out from "git log --oneline" that the commit for 3.3.6 was 17dcb543a. It would be great if there were tags for all the official release, but even in the absence of that, if you were willing, you could update a variable in your script pointing to the commit for each new release.

4) You say "Raspberry Pi 3" a bunch of times in the script. I'll bet pretty soon most people will be running Raspberry Pi 4's. s/3/4/ ?

5) First time I ran it, I reflexively typed "sh setupAstroRaspbianPi.sh". That fails (because it needs to run bash, not sh). Of course, that's my user error, but you probably could check to see if bash is running and if not let people know. (I know you have the #!/bin/bash at the top, but if you do what I did, you don't get bash).

Thanks again for the very nice scripts.
Hy

Read More...

Hy Murveit replied to the topic 'Bug report -- autofocus on filter change confuses capture' in the forum. 2 weeks ago

I re-ran the same job last night (that had failed the previous 3 nights) but this time, without autofocus-on-filter-change (I set autofocus every 30 minutes instead, and filter changes were about every 30 minutes). My job ran through the night, no problems, completing the meridian flip, etc.

I did try and see if I could find a bug in the filter-change-autofocus stuff, and admittedly, I didn't see anything, but am pretty unfamiliar with that stuff and it's complex.
I'll report back if I ever do get to the bottom of it.

My workaround is not to use the autofocus-on-filter-change. I do think that's a nice feature, so I'll make an effort to track things down.
Hy

Read More...

Hy Murveit replied to the topic 'Bug report -- autofocus on filter change confuses capture' in the forum. 2 weeks ago

Jasem,
Hope it's OK if I probe a little further. 
I understand what you're saying (that there's a potential issue with PHD2 at 02:48:38) but doesn't that happen after the original strangeness?
That is, looking at the 3 lines here:
[2019-09-25T02:47:46.643 PDT INFO ][   org.kde.kstars.ekos.capture] - "Capturing 240.000-second Green image..."
[2019-09-25T02:47:58.470 PDT INFO ][     org.kde.kstars.ekos.focus] - "Image received."
[2019-09-25T02:48:00.023 PDT INFO ][     org.kde.kstars.ekos.focus] - "Capturing image..."
It seems that Capture started a 240s exposure at 02:47:46.6,but then 12 seconds later focus receives an exposure at 2:47:58.4,and starts up another exposure at 2:48:00, and no word from capture in between, so there are overlapping requests take an image from both the capture and focus tabs on the same camera. 
So, perhaps, PHD2 might have not resumed because of the autofocus?
I've had this crash 3 nights in a row. I'm trying to re-run this tonight, but without autofocus-on-filter-change, and hoping that the crash will not recur and give us a clue.
Hy

Read More...

Hy Murveit replied to the topic 'Bug report -- autofocus on filter change confuses capture' in the forum. 2 weeks ago

Forgot to mention, on the log_01-45-25.txt log, I was running kstars v3.3.6 (really compiled from the 3.3.6 commit on a raspberry pi 4), and on the other log it was a bleeding-edge compile around v3.3.4 also on the RPi4. Used a ZWO ASI1600 mmpro with zwo filter wheel.
Hy

Read More...

Hy Murveit created a new topic ' Bug report -- autofocus on filter change confuses capture' in the forum. 2 weeks ago

I've come across what looks like an odd bug related to the focus on filter change feature.
The last few nights, my jobs shooting M33 have stopped (around 3am) soon after a successful meridian flip. From looking at the logs (and I was logging all modules, pretty much) it seems like after the flip, align, guiding re-start, the capture module restarted it's captures, but *at the same time* an auto-focus operation was ongoing. This concurrent capture and auto-focus brought down the job. 
I'm guessing that the reason for the auto-focus was because the post-flip align used a different filter than the capture tab, and that somehow triggered an auto-focus. 
Also, clearly, when one requests auto-focus-on-filter-change, one means only for the capture tab, not for the align tab. 
Anyway, perhaps my theory is wrong, but here are a couple of logs.

Note in the log named log_01-45-25.txt

[2019-09-25T02:47:46.643 PDT INFO ][   org.kde.kstars.ekos.capture] - "Capturing 240.000-second Green image..."
...[2019-09-25T02:47:58.470 PDT INFO ][     org.kde.kstars.ekos.focus] - "Image received."
[2019-09-25T02:48:00.023 PDT INFO ][     org.kde.kstars.ekos.focus] - "Capturing image..."
...

and in log_21-40-37.txt

[2019-09-24T02:47:11.823 PDT INFO ][   org.kde.kstars.ekos.capture] - "Capturing 240.000-second Red image..."
[2019-09-24T02:47:23.254 PDT INFO ][     org.kde.kstars.ekos.focus] - "Image received."

File Attachment:

File Name: log_01-45-25.txt
File Size: 4,016 KB

File Attachment:

File Name: log_21-40-37.txt
File Size: 7,450 KB

[2019-09-24T02:47:25.174 PDT INFO ][     org.kde.kstars.ekos.focus] - "Capturing image..."
...[2019-09-24T02:54:11.775 PDT INFO ][   org.kde.kstars.ekos.capture] - "Exposure timeout. Restarting exposure..."
[2019-09-24T03:01:11.775 PDT INFO ][   org.kde.kstars.ekos.capture] - "Exposure timeout. Restarting exposure..."
[2019-09-24T03:08:11.795 PDT INFO ][   org.kde.kstars.ekos.capture] - "CCD capture aborted"

The system seems to be doing two things at once in both these logs.
Hy

Read More...

Well, clearly 30 minutes won't cut it, but for me, using a laptop with my pi isn't a power issue since the laptop is closed for most of the session after things are set up and running on the pi. In fact, to just quickly check up on things mid session, I often vnc in with the vnc app on a phone or tablet.

Read More...

I have been using this monitor but really it is only a backup for me (in case the networking fails). Typically what I do is use my laptop (macbook pro) over VNC, which works nicely. I prefer the laptop, but the small monitor does work in a pinch.

Note: (a) I need to clamp down the monitor stand to my field table, which is not ideal, and (b) I needed to build a little cable to go from my anderson powerpole power connection to the 12V input for the monitor.

My emergency kit also includes a standard little usb mouse & keyboard with usb hub. I got this foldable keyboard with touchpad , hoping that I could use with with bluetooth, but it did not work, and now reading the fine print I see it does not say "Linux" as a supposed OS. Not sure why, and looking forward to other suggestions on this thread for a little foldable keyboard/touchpad combo.

Read More...

Hy Murveit replied to the topic 'Right Click--GOTO object' in the forum. 1 month ago

FYI, if I understand you right, I have a reasonable workaround that I have been using.

To set it up, one time only, go to the KStars settings menu, click 'Configure Shortcuts', search for slew, click on 'slew the telescope to the mouse pointer position', and under custom click input, and input some beyboard shortcut, and click OK. I use control-l (letter L) for the shortcut, which overrides something else, but of course you can choose what you want.

Then on the sky map all you have to do is click on your target and type your keyboard shortcut.

Hy

Read More...