Hi all,
logged into the web interface for first boot up and was met with a message saying the camera was disconnected. I have checked to see that I have plugged the camera into the correct slot and I have tried a different cable to no avail.
Wondered if it was the board so played with the motor controls and they appear to be moving.
I cannot see an option for terminal on the webpage so I am assuming this can only be accessed by connecting a monitor directly? I have ordered a hdmi adapter to test that but if anyone has any other ideas and suggestions that would be welcome. Cheers
You can connect to terminal remotely through ssh, if the RPi is connected on the same local network. You will need the IP of your Raspberry Pi. You can use a program like MobaXterm to connect, as long as ssh is enabled. If it isn’t, you can take your SD card to your PC, go to the boot partition, and create an empty ssh file, which should enable it. See https://www.raspberrypi.com/documentation/computers/remote-access.html for more info on that. Then you can enable VNC through raspi-config, and setup a VNC connection to it, which would let you remotely access the raspberry and use your PC monitor to do whatever you want there. You can also take a still image that you save on the SD card, take it out, and check it on your PC. Alternatively, flash Raspbian for a short while, check and debug, then flash the OFM software. This will remove everything on the SD card and will require you to setup it again of course.
2 Likes
There are a couple of physical things to check as well.
Is it a Raspberry Pi Camera v2 (also called v2.1) ? This is the only camera that will work with the OpenFlexure software from the OpenFlexure web site. Pi camera v1 and Pi Camera Module 3 will not be recognised, nor will other Pi compatible cameras.
Are the cables type B, as shown on the part page Assembly Instructions ? There are occasionally type A cables, which will immediately damage the camera.
Are the cables the right way round in the connectors? The metal contacts should face away from the clip. This part of the instructions has recently been updated, in step 3 on this page: Assembly Instructions (note that links to the current development build, so people coming to this post later may find it is no-longer relevant)
I hope that you are able to find the issue and get your microscope working.
1 Like
Thanks for taking you time to reply, I appriciate it.
I also realised after I posted that I should have been more specific about my setup etc… fog of disappointment,
I am using 1Pi V4 with a pi camera V2.1. The OS is the full from this page.
Double checked the cable I bought and they are B type and the box says for Pi camera so hoping I am safe. Cables have also been double checked and are in the correct way.
I am hanging on for a microHDMI adapter so I can connect it to my monitor, my computer-fu is decent but I suspect it is adding in one more variable where I could make a mistake.
Might be a few days before I get around to troubleshooting the next part but wanted to say cheers to you both for taking your time to reply
1 Like
Sorry it isn’t working. Let us know how it goes when you find time to debug.
If you can get access to the terminal (via ssh or directly plugging in a monitor) it is worth checking if you can stop the server with
ofm stop
And then it is worth seeing if you can run the built in picamera commands such as raspistill (note that the OS is pretty old* so we are on raspistill not the newer tools.)
Good luck.
*Lots of work is going into getting one ready that is less outdated
1 Like
Seems I have camera now. I bought a new Pi Cam 2.1 just in case and tested it before taking the lens off, after lens removal etc etc and it works. Stage calibration seems to not like something but to be honest I feel a lot better just being able to see something for now.
I saw that there is a new update coming so I will keep an eye out for that!
Cheers people
1 Like
The camera stage mapping needs a slide with good contrast, detail over a reasonably wide area (more than one field of view in all directions beyond the starting frame), and needs to be focused. A fingerprint in drywipe marker on a blank slide works well.
If you have any dust or bits of plastic in the optics module, that can be a problem for the mapping. Any features in the camera frame that do not move when the sample moves can make the algorithm think that the sample is not moving. Our eyes are really good at ignoring problems in a small area. A high contrast small piece of dust appears more important to the tracking algorithm than the overall sample.