The “park” position is with the cameras facing down and the RA axis horizontal. Note that the iOptron will not actually let you “park” in this position so we do the park command but then “manually” move the Dec axis to get the cameras facing down.
How does that seal work? Is it a tight fit around the lens hood? Is it made of a flexible material?
Those two images might actually be a bit confusing because they show the scope parked on the two different sides of the mount. The top photo shows it parked on the west side of a north facing mount (not used by PAN001 or PAN006) and the bottom one shows it parked on the east side of a north facing mount (correct).
I was quite impressed by Jonah’s weatherproofing of his scope, where it appears he has used some riveted together several bent aluminum sheets. However, given that you’ve been doing some 3D CAD and 3D printing, I wonder if you’d like to try an experiment…
The plastic covers for the electronics of the mount are held on with screws, and so can be taken off easily enough. I wondered about designing alternate covers, using the same holes, but specifically designed assist in weatherproofing the mount (e.g. slightly overlapping the bearings, etc.). Such 3D plastic might need to be sealed, and it might be necessary to use a rubber washer/gasket in the screw holes, but it is an interesting approach to weatherproofing the mount.
Yes, I should have pointed out that it doesn’t actually matter which side it is parked on and it should be set for whatever makes most sense at your location.
@james.synge Yes, it is just a very tight seal. It took 3 attempts to get the perfect fit.
Looks great! Nice progress.
I have received the new interface boards! Ill get in touch to get them over to you guys.
@zacharyt Hey all. I am redoing our main website, is it okay if I use some of these images?
@wtgee Yes, feel free to use them. I look forward to seeing the updates.
Yesterday we completed the new version of the head unit PCB and passed the last tests. We also used the UNO and documented the steps, so if anyone else would like to use those images to clarify the instructions we would love to provide our resources. We are moving on to the control box board, and have already made substantial progress(currently half way in terms of step numbers).
We are having some trouble with the NUC setup and could use some help. We have done all of the setup steps, however we are unable to run the pytest. When I enter the final command, “sudo pytest” it sends the error: “sudo: pytest: command not found”.
I tried uninstalling and reinstalling POCS, but this did not fix the issue. Upon closer review I saw that when the install dependencies command was run the install-helper-functions.sh script was giving the error, “Line 18: printf: ‘;’: invalid format character”. I tried editing that specific file to fix it, however I was unsuccessful. There were also errors in many other scripts, and I am hopeful that the install-helper-functions.sh script caused them.
I also have the terminal input and output saved as a text file if anyone is interested in using that to see where I went wrong.
It shouldn’t be necessary to use “sudo” before pytest. In fact, that likely will prevent pytest from being found, as you discovered.
Please confirm that you have installed miniconda (Anaconda python distribution, but without so much stuff).
And please send more details about the errors you encountered. I wrote the install scripts, so would like to get them fixed ASAP so that others don’t run into the same problem.
Because of the amount of errors, I would like to get you the text file before transferring all of that information onto the forum. Do you have a google account or an email I can use to send you the file since the forum currently doesn’t allow for text files to be uploaded?
We have solved the previous error but encountered a new one.
We found that because of multiple attempts at installing the POCS software copies of some libraries(including miniconda) and py scripts were created, leading to the install-helper-functions.sh script not knowing how to handle some circumstances. We solved this by wiping our drive and reinstalling our OS and then POCS.
Things were going well but when the install-dependencies.sh script was run, The following error occurred after the script successfully ran and avoided the error we had previously encountered:
Command “python setup.py egg_info” failed with error code 1 in /tmp/pip-install-e8n_jst_/astroscrappy/
The strange thing is that this file does not exist on our computer, so I wasn’t able to investigate that file any further. I don’t know what error code 1 means in this context, but I would assume that it occurs when a file that doesn’t exist is asked to do something. I looked into the setup.py script and found that two lines of code were blocked out into text, which if unblocked would treat everything in scripts as a script to be installed. I’m not sure if enabling that would fix the issue and create the file we need, or make things much worse.
Additionally, we are now using Ubuntu 18.04, which another team had issues with. If we can’t find a solution in a few weeks, we will probably move to a different version of Unbuntu to isolate any errors more effectively.
We have fixed our NUC problems and have identified an issue with the download process. All of our errors were caused by a bad download link to astrometry. There has been an update and(if I remember correctly) the link that was used appears to download the most recent astrometry version, not the specific one which the system expects.
Currently, we are very close to finishing our unit. We are waiting on a special part to make a cable to connect to our mount, but will then proceed with some final tests before deployment.