Neon Rabbit's 1G Pico Jar

[NeonRabbit221B]

Reminds me of the MP40s when the magnets arent perfectly lined up. Are they just too weak? Maybe try a slower ramp up speed to see if it will spin?

Yeah, there is a lot to learn about stir bars and its not an easy learning curve. I think you are right about it not being lined up. The magnets are likely not 100% coupled so when it changes direction it throws it off balance. The disproportionate weight and buoyancy of my cross bar design probably isn't helping maters. Trial and error is my specialty

For the heck of it I stuck a wetside in there. It made a ugly noise so I pulled it out. Not my brightest idea and luckily it was a used spare.

 

[eggplantparrot]

i imagine a cage with a shaft would stabilize it and help. the propellors look way too big too.

 

[piranhaman00]

10% of your yearly salary is too much...oh no.

 

[NeonRabbit221B]

Brief Update. So below you can see the electronic mess that is the brains. A 12V input is fed to an arduino nano and to a LM2596 which steps the voltage down to 9V which will power the heating pad. The arduino has several IO
2 Pots to control the stir bar speed and temperature of the heating pad
1 DS18B20 waterproof temperature sensor
2 signal pins and 1 PWM to control the L293D for the dosing pumps speed and direction control
1 PWM pin to control a MOSFET that varies the voltage to the heating pad
I will add a 1" OLED to the front to display the current temp, speed, ect that will be an i2c based comm.

Please excuse the duct tape as this first print was the prototype... Basic coding will work like this
The potentiometer input voltage will be mapped to a PWM value from 75 to 225. The speed for each direction is chosen based off of a random function that varies that PWM value + or - 25 and the duration is randomly chosen from .5 to 3 seconds.

The plan is to have the heating pad and temperature sensor to do a PID feedback loop to control the temperature. The reason for this is that the heating is indirect and a simple on/off switch would not have sufficed. Right now the pot simply varies the voltage to the heating pad using a mosfet.

Before I start my aquascape I have a few more things to complete.

Electronics and water typically do not mix well. I need to figure out how to seal this incase the glass were to break
OLED coding and display
Find a safe way to hide the heating pad
Re-design and reprint some of the parts
The stir bar and impeller design

 

[NeonRabbit221B]

So life got busy but here is a quick update

I am not tossing out the idea of using my magnetic stirrer and heating pad idea but I have a few roadblocks
1. Coding for the heating pad is far from done but got it to make adjustments
2. I am terrified I will burn my office down with the heating pad without fail safes
3. Turns out the dosing pump shaft sits at the slightest angle which is why the stir bar wasn't stable. Need a new motor.

By the time I get this idea to a comfortable state I will have already found a new job. I will update on design changes but have gone a more traditional method and used a more robust and error proofed controller setup from reef-pi

I did a 45 minute build of a new reef-pi controller... I am just so comfortable with with the Pi and setup process it was an easy decision. #reef-pi
I got a IOT relay board for christmas but it can only really control 1 piece of equipment. The relay board also can power the raspberry pi itself which is pretty nice. I hooked up a 50W Preset from Aqueon and wired in a ds18b20 to control the temperature.
The new reef-pi build now supports wifi switches so instead of wiring up a relay board I snagged two HS103 so I can control my light. I think it will be nice to be able to move the jar without having the lamp teathered.
The flow will be coming from a cheap air pump and check valve. I fed the line under the rockwork and guard I made to secure the stir bar. I got the reef-pi build loaded up and set up inputs/outputs, filled her up and plugged her in.

Here is the aquascape I went with.

 

[NeonRabbit221B]

So a big update today but I wanted to give some people some insight into design changes. I started the jar using a simple reef-pi setup to control a 5W preset and my lamp. Just as I worried the bubbler was fairly loud, caused massive salt creep and insufficient flow for my "tank". The chords and heater/air line in the tank still bothered me so I went back to the drawing board and did the following.

I replaced the dosing pump with a TT motor to achieve a higher torque and lower RPM to reduce the vortex as well as the stalling. While this runs fine at 5V I kept the previous wiring for the 12V dosing setup and just reducing the output voltage to 6V so it runs on the faster end. I know this is tacky but I glued my magnets to the wheel which came with the motors. Here are some images of the electronics and guts

I got my 1" oled wired up (although it took me about a day to track down a damaged i2c pin on my nano board) and got it to display the pot outputs for the speed of the motor and pad voltage. After I got it all assembled I ran a test to check the flow with varying magnetic stirring rods (will be ongoing once corals are readded). I used the cross shaped bar found here Amazon product with the pot turned to about 90%. It was perfect and the stir bar no longer stalled or got through off the magnetic coupling. The flow seems greatly improved judging by the movement of the debris in the water.
The heating pad element was also very successful and allowed me to heat 1 gallon of water from room temperature to a perfect 79 degrees in 12 hours and plateaued. I will still be working on the PID loop to better control this. Here it is assembled without my rock or coral

Final steps before I call this one a success is
-build a translucent lid to replace the cork
- PID loop for temperature
-incorporate reef-pi as a redundancy
-oled cover/mount

My reef jar is now wireless

 

[SMSREEF]

Very Cool setup!