Horizontal observation with ability to process multiple specimen in series

Request Help
1

Alright guys!

Up until now, I have not found anything like that. For my cause, I bought a Raspberry Pi 5 with 16GB RAM (still got it cheap), an AI-Hat with 26TOPS and the HQ-Camera to build the base for my project.

Here is a scheme:

                              MESOSCOPE SYSTEM
═══════════════════════════════════════════════════════════════════════════════
                    ┌─────────────────────────────────────┐
                    │         Raspberry Pi 5 (8GB)        │
                    │   ┌─────────────┐  ┌─────────────┐  │
                    │   │  OFM Server │  │ AI HAT+     │  │
                    │   │   (FastAPI) │  │ Hailo-8     │  │
                    │   └─────────────┘  │ 26 TOPS     │  │
                    │                    └─────────────┘  │
                    └──────────────┬──────────────────────┘
                                   │ CSI
                    ┌──────────────▼──────────────────────┐
                    │      HQ Camera (IMX477, 12MP)       │
                    │      C-Mount │ 150mm Achromat       │
                    │      10x/0.25 Objektiv              │
                    └──────────────┬──────────────────────┘
                                   │
                    ┌──────────────▼──────────────────────┐
                    │     OpenFlexure Microscope Body     │
                    │     Sangaboard v5 + 3× 28BYJ-48    │
                    │     XYZ-Flexure Stage               │
                    └──────────────┬──────────────────────┘
                                   │
                                   │  mounted on rail
                                   │
═══════════════════════════╤═══════▼══════╤══════════════════════════════════
  Linear rail              │              │
 ══════════════════════════╪══════════════╪═════════════════════════════════▶
                           │    ← →  │    │
                          Motor      │    
═══════════════════════════╧══════════════╧══════════════════════════════════
          │                    │                    │
          ▼                    ▼                    ▼
   ┌─────────────┐      ┌─────────────┐      ┌─────────────┐
   │             │      │             │      │             │
   │ {Bioreactor}│      │ {Bioreactor}│      │ {Bioreactor}│
   │             │      │             │      │             │
   └─────────────┘      └─────────────┘      └─────────────┘
         #1                   #2                   #n

At this moment, I am building the customized optics module, implementing a c-mount within the optics module to try, if removing it will be necessary.

Next step will be to position the optics module horizontally, so it can generate images of the vertically aligned specimen window of the bioreactors.

I have some ideas to fix the z-axis problem already, but I guess this will change a couple of times on the run.

So far, I have migrated some dependencies issues coming up on the Raspberry Pi 5 and made some sample pictures, mounting the HQ Camera on a oldschool microscope, which were satisfying.

Will keep this posted probably, this is something, I would like to have well document on many instances :smiley:

Best and happy tinkering :slight_smile:

2

Are you going to use (x,y,z) from the OpenFlexure microscope, or is the long stage providing one of the motion axes? For a bioreactor do you actually need all three axes, or are they pretty uniform inside so that it is enough to have the long rail taking you to each reactor (or a few fields in each reactor) and then the optics module just needs a focus axis?

If you only need the fine focus axis from the OpenFlexure system, then the separate_z_actuator.stl from the Upright version of the microscope might be what you need. With a relatively heavy horizontal optics module I would mount it with the optics hanging below the actuator. The optics above the actuator is probably also OK, but certainly not sideways. The design focal plane is 12mm from the mounting plane for that actuator.

If you do need all three axes of fine motion for the microscope, it looks as though you would need to mount the microscope from the stage top. Then your sample (in the bioreactors) stays still and the actuators move the whole body of the microscope. This I think would not work well as there would be a lot of sideways strain on the mechanism.

3

Thanks, I will look into the separate z actuator, as soon as I have a clearer picture of where my project is going and which adjustments, I need to take.

For the separate z actuator: do I need an extra stepper motor for that? The instructions had no pictures, I feel dumb about that.

4

If you just use the separate z actuator, it uses one stepper motor. I was only suggesting it if you do not need xy motion, so not an extra motor, just missing out the xy actoators and motors.