Reef-Pi for PhD project

[butterflyeq]

Hi all,

I'm a marine biologist, now in my 1st year of a PhD studying the impact of environmental stressors (such as climate change, nutrient run off,etc) on the impact on intertidal seaweeds in the UK. (Side note- seaweeds are very important!)

I stumbled across the Reef-Pi and it would be such a perfect addition to my experiments. Basically my experiments consists of 48 tanks, in groups of 12. So thats 4 treatments. Using the reef pi to monitor things like pH and temperature in each treatment. But also to control peristaltic pumps- to add liquid stressors, ato, and other bits of equipment. The experiments will be running over the next 3/4 years in a temperature controlled room so i would like to run the Reef-pi for this time (might even make one for my reef tanks at home!). I'm also in the process of building my own LEDs, so if i could control these too then that would be great (currently controlled with an Ardiuno, but better if i could have it all in 1 unit!).

I wanted to start this thread for several reasons:
1) to say a huge thank you to who made this! (i bow to you!)
2) To show you guys what i am doing and hopefully get some advice, help and ideas along the way.
3) For help with kit in the UK.

I am following to Reef-pi guidelines for building it from Adafruits. have just finished making the power controller, away to test it today. Only issue is that i do not have the American power strip, what are people in the UK using? I do have a old IKS Aquarstar 4 plug bar which i was kinda hoping would work-i found it in the lab! Other than that, would there be a way to just an 8 board relay? Im building/ 3D printing the peristaltic pumps myself which i would be plugging into the power strip, so i don't think i necessarily need plugs? Whats everyones thoughts on this?

I do however need to make sure everything is super safe, and will pass the universities tests.

I am very new to Rpi, and fairly new to Ardiuno/code/github/electronics so would appreciate the help. Thanks

 

[Michael Lane]

Very cool! I love the idea of reef-pi being able to contribute to research!

I can't help much with sourcing parts in the UK, but I'm certainly interested in how this goes. reef-pi supports wireless Kasa smart plugs, so that might be an alternative you can use.

The power strips definitely aren't required, and you can use relay boards just fine. I'm not sure what safety precautions you might need, especially since you will be working with a higher voltage than in the US. On the other hand, most peristaltic pumps I've seen are low voltage.

 

[butterflyeq]

Thanks for replying!

I love the idea of blending my hobby with research too!

I was under the impression that the power strip was required in order to build the other parts? If this isn't the case then i could just skip this step, even for now as i'm not totally sure if i need it now. I was going to power some 12V water pumps to make a counter-balance for my tidal system, but i found another way to power this so i might not need them.

The peristaltic pumps i built myself, with a 3D printer and 12V water pumps.

Counter balance= The species i'm studying are intertidal species, so i need a tidal system in the lab. To have water draining and refilling would mean i would need 2 tanks for every seaweed (tank and a sump), as i have 48 seaweed tanks, thats alot of space taken up. So i decided to build a pulley system which uses a PVC pipe frame as a pulley, and seaweeds in baskets which rise up and down. I was going to do this by having 2 buckets with water pumped between them to move the pulley up and down, but i have came across a motorised roller blind build. So i have 3D printed the parts and ran the code to make the "roller blind" (pulley) rise up and down twice a day = tidal system.

The main things i think i want the Reef-pi to do monitor the water temperature (as this is an important potential stressor with climate change), monitor the pH (also another potential stressor), and have an ATO so that salinity doesn't become a stressor unless i want it to.

 

[Peace River]

Let's see if we can get some additional input...

#reefpi @Ranjib

 

[butterflyeq]

t

Let's see if we can get some additional input...

#reefpi @Ranjib

Thanks. I'm new to the forum/ forums in general so i didn't know how to do that!

 

[Urtoo]

Lurking on this, but thought, what about a surge tank. YouTube search “borneman” Surge or tank
You could plumb a few tanks to each device and surge them all day and night, instead of only moving the seaweed/kelp.
Seems the oxygenation and detritus would not be similar in baskets moving around vs a true surge.

 

[butterflyeq]

Lurking on this, but thought, what about a surge tank. YouTube search “borneman” Surge or tank
You could plumb a few tanks to each device and surge them all day and night, instead of only moving the seaweed/kelp.
Seems the oxygenation and detritus would not be similar in baskets moving around vs a true surge.

Thanks for your reply. I had a look at this but unfortunately it wouldn't work for my set up. The problem is i would then need two tanks for every piece of seaweed. The water in each tank must be isolated in order to prevent pseudoreplication. I.E mixed water = 1 true result, seperate water = each tank as a true result (n=48 in this case). Each tank has a airline tub to it so oxygenated and their is no detritus.

Also no Kelp here, just intertidal species.

 

[Urtoo]

A bell siphon made from 4” PVC? Just looking at less mechanical to worry about for 3-4 years AND grant movement to the tanks in study.
Would be small enough to stand above each tank and then it’s pretty much set and forget.
Unless you are using large vats

 

[Michael Lane]

Does this mean that you intend to monitor pH and temperature in 48 different tanks, with 48 different probes? That's a lot of probes!

I'd probably start by listing my requirements and modelling how a single tank would be set up, then sort of brute force it to scale out to 48 tanks. After that I would begin to optimize the crazy stuff (like 48 raspberry pis).

Here's what I hear as requirements so far:
1. Monitor pH
2. Monitor Temperature
3. Control pumps
4. Control lights

I'm sure you can find a lot of advice and opinions here. It's a great group, and full of very knowledgeable people. A diagram may also help convey your ideas and then focus on problems. Each of the 4 requirements I've listed above are the tip of the iceberg. They can each have many more details, so it will be imperative to focus on one at a time, and then move to the next goal.

Good Luck!

 

[Ranjib]

We super excited to see reef-pi usage in research/academics. Let us know how we can help you.

Start small and see what works and extend it from there. I am not aware of many ready to use UK powerstrips. Kasa/TP-Link is a good option if you have stable wifi (and ability to fix the ip , by dhcp reservation or static assignment).
The opensource ph circuit is lot cheaper compared to atlas, if you have budget constrains. I am not aware of anyone who has run that many units with a single pi, I doubt it will work straight away (due to circuit challenges i think). I have tested the ph modules with 2 to 3 circuit at a time at max.
Light control should be straight forward, use pca9685 -> meanwell power supply -> LED setup.
I think its safer/easier to build multiple independent unit for large setup like yours. This will simplify the circuit, provide some failure isolation and probably you can interchange a few things.
Since this will be a multi year setup, you should also setup some type of backup process. We can walk you through all that as we get to know more about your setup

Godspeed,

 

[butterflyeq]

A bell siphon made from 4” PVC? Just looking at less mechanical to worry about for 3-4 years AND grant movement to the tanks in study.
Would be small enough to stand above each tank and then it’s pretty much set and forget.
Unless you are using large vats

Thanks, but it still wouldn't work for what i need to do, still would be twice the number of tanks. My set up is fine, the mechanical side of the tidal system is all automated so its not an issue.

 

[butterflyeq]

Does this mean that you intend to monitor pH and temperature in 48 different tanks, with 48 different probes? That's a lot of probes!

I'd probably start by listing my requirements and modelling how a single tank would be set up, then sort of brute force it to scale out to 48 tanks. After that I would begin to optimize the crazy stuff (like 48 raspberry pis).

Here's what I hear as requirements so far:
1. Monitor pH
2. Monitor Temperature
3. Control pumps
4. Control lights

I'm sure you can find a lot of advice and opinions here. It's a great group, and full of very knowledgeable people. A diagram may also help convey your ideas and then focus on problems. Each of the 4 requirements I've listed above are the tip of the iceberg. They can each have many more details, so it will be imperative to focus on one at a time, and then move to the next goal.

Good Luck!

No i don't plan to use 48 reef-pi's haha! I plan to use 1 Reef-pi per treatment, this would be over the 3 years, a max of 7. But more likely to be 4.

I'm undecided about controlling lights atm, i have no real need for any sunrise/sunset/moon feature as my seaweeds don't photosynthesis during these periods. I just need lights on/off. Still thinking about this 1. I would like to do the sunrise/sunset etc, just to maintain the natural environment effect, but my supervisors don't think it's needed (and they are the bosses! haha!)

pH and temp, and ato are my main focus atm.

I will try get some plans drawn up or pictures to help.

 

[Ranjib]

No i don't plan to use 48 reef-pi's haha! I plan to use 1 Reef-pi per treatment, this would be over the 3 years, a max of 7. But more likely to be 4.

I'm undecided about controlling lights atm, i have no real need for any sunrise/sunset/moon feature as my seaweeds don't photosynthesis during these periods. I just need lights on/off. Still thinking about this 1. I would like to do the sunrise/sunset etc, just to maintain the natural environment effect, but my supervisors don't think it's needed (and they are the bosses! haha!)

pH and temp, and ato are my main focus atm.

I will try get some plans drawn up or pictures to help.

Even with 7 pi, each pi will have around 7 ph probe attached to them. Individual ph circuit has i2c address. Cross check if its possible to configure 7 of those (Each needs to have a distinct i2c address) in one i2c bus/pi.

 

[SDchris]

Even with 7 pi, each pi will have around 7 ph probe attached to them. Individual ph circuit has i2c address. Cross check if its possible to configure 7 of those (Each needs to have a distinct i2c address) in one i2c bus/pi.

edit
I read it as 4 treatments. Each treatment contains 12 tanks(48 total). Running time is 3-7 years
So it's 4 pi's. 12 pH probe per pi.

 

[SDchris]

Also keep in mind, if not already, to have some electrical isolation for the pH circuit. And consideration to length of probe wire.
If you wanted to tidy things up, maybe build a little PCB carrier board for all the sensor circuits. Have a quick look at EasyEDA.

 

[butterflyeq]

Hi All,

Only going to have 4 RPi with 4 temp and 4 ph sensors, 1 on each RPi. they are going in different rooms so that is why they each have their own RPi.

The 12 tanks for each treatment are in 1 water bath, this controls the temperature of the water, the temp probe will be in the water bath. The pH probe will be in 1 tank, which will be a representative of all 12 tanks in that treatment.

The max length of time of these various experiments, using the 1 set up will be 3 years.