DIY azoo refrigerated feeding system?

[halmus]

This is kind of a vague and silly question, but I'm not a stranger to stupid questions. I'm actually the subject matter expert. So, here goes....

I studying computer science and computer engineering. I have a senior project a ways in the future. Actually a year or so off, but I want to get a head start. I've got a background in electronics already and I'm a pretty good carpenter and getting better at acrylic work. I'm considering one or two projects.

1. I'm already building my own version of the Radion lighting system. Mine will have more LED's and a wider spectrum of LED's than the manufacturer's. It will be dimmable on all channels. If I go that route for the project, I'll be working to integrate a micro controller and a GUI to control it from my desktop.

2. I've considered building a small mini-fridge that would be accessible (and marketable?) to the reefing community. Specifically, those keeping coldwater azooxanthellate systems. There are other forums out there dedicated to just that community and they're always looking for a way to modify an existing minifridge for their various feeding substances with peristalic pumps moving the feeds. The system I would be building (once again) would include a microcontroller system maintaining the temp and making the pumps controllable through timing circuits or computer interface GUI's. It's just an idea, any feedback?

 

[phasezero]

For the sake of a senior project, I would go for option 2 as it seems to lend it self better towards a product design project. Option 1 is cool, but what will your project address that hasn't already been done by all the other controlled LED lighting systems available on the market?. Yes you'd be adding a broader spectrum by adding different LED's, but the software end has pretty much been done. Basically what i'm saying is, for your literature research portion of your paper where you find out what's currently available, would make your purpose sound not as important to your peers and faculty.

Option two is the opposite as I've never seen those readily sold in stores or online. I think this project lends itself well for researching and i'm sure there are a ton of different solutions that you can evaluate. This is important because it looks good when you can show that you've considered many different possible solutions and how you chose that particular design for implementation.

 

[fall0n]

x2 for option 2!

 

[halmus]

Thanks to both of you for the feedback

PhaseZero-

I have to agree with everything you wrote, and it's something I considered, but couldn't' articulate as well as you managed to. The only reason I was considering the LED thing was that it would give me the chance to work on a project with a wider scope. I can't help but feel a little limited in working with the chilled dosing system. If I was to tell everyone that I was going to design a new toaster that you could hook up to your computer, the logical question would be, "Why?" The dosing system really isn't all that complicated. I have a hard time justifying the software effort just to control a few peristalic pumps. I don't need software to regulate the temperature. It's a simple thermistor circuit and maybe some digital logic which I'd rather do with discrete components. Add a few buttons and a couple 7-segment displays to adjust the temperature set-point and you have a fancy looking unit . Although, I did remember when I first started brainstorming the chilled dosing system that it would include built-in magnetic stirrers. I already have the Erlenmeyer flasks and medical grade stir bars. It doesn't add much to the project in terms of engineering, but does add a big plus in terms of marketability.

I'm moving forward with the LED system for my own use. I'll start the chiller-dosser this summer. The first one will be a single unit proof of concept just to test the heat-exchange unit, regulation circuitry, and magnetic stirrer.

 

[phasezero]

I see what you mean about project 2's level of difficulty. In that respect, project 1 is better suited towards showing your expertise. It sounds like you might need to dive into more research on both subjects to identify the problem and what your project will address. I'm still slightly leaning towards project 2. As a reefer, project 1 has all the pimp factor. It's fun applying your engineering to your own hobbies. For my composites class a couple years ago, I designed and built a vinyl ester/glass composite housing for my LED fixture, it turned out like ass hahaha. I just recently redid my fixture and used rectangular cuts of gloss black arcrylic. It took 20 minutes to make and looks 1000x better than my 50 hours of work pos composite.

 

[halmus]

It's been so long since I started this thread, I'm basically starting the thread over. I went with the fridge idea for the project. I'm way ahead of where I should be for working on this project, but it will involve a lot of complicated physical design that's irrelevant to my degree. It will also require a lot of testing to ensure the components I'll be using will hold up long term.

I've had some parts on hand for a while to begin prototyping this, I just haven't had time to do anything with it.

This is the beginning of a project that will probably take me a year or so. My end goal is to build a mini fridge system that can hold fish foods suspended in water that can be fed into the aquarium. Some hobbyists feed phyto to their tropical tanks. Non-photosynthetic tanks especially need a reliable source of food for the corals and other critters. Many of the people that keep these systems buy mini-fridges like you'd see in a college dorm. Then, they drill out holes and run a hose through from their tank and the food gets pumped into that multiple times a day.

They big picture is that people who want to feed their tanks multiple times a day need to have a way to keep their food source chilled. You don't want the food to rot, and if you'd like to be "hands-off" for a few days while you travel, you have no way of providing a source of food.

My project starts with building and testing a magnetic stirrer. Here is a video some one else made of their end product:



This isn't my video. The idea of a magnetic stirrer is simple, and it's design is simple. Provide a way of stirring the vessel's contents without having to drill a hole in the vessel or have a stirring rod come down from above. A magnetic bar (sealed to prevent corrosion) is placed in the vessel. A motor spins some rare earth magnets below it, causing the stir bar to be remotely stirred. You can vary the speed by altering the values of the components used in building the voltage control circuit.

Here's my prototype for the stirrer only:

The stir bar sitting in the bottom of the flask:

The view looking down of the fan with the magnets glued on. There is an acrylic sheet separating the fan from the flask.

Action shot:

This is just a test that I'll run for a while. I want to make sure the fan doesn't burn out too quickly. My plan is to build a fridge that will hold 3 or 4 of these flasks inside. Each will have an independently controlled magnetic stirrer, an air source, and a line out to feed the tank. Once I verify that the fan will hold up for a decent amount of time, I can start building the fridge around it.

Until then, I need to order parts that will do the physical cooling of the enclosed fridge. The whole thing will be built from scratch with acrylic and have a foam shell. After that, I'll build an acrylic cover to hide all the foam and make the project look nice.

References:
variable voltage regulator: variable voltage regulator circuits, LM317
DIY magnetic stirrer: DIY Magnetic Stirrer - YouTube

 

[pandora32]

WAY above my abilities!!...lol. I get excited when i remember to do my waterchanges on time...lol. Looks cool though.

 

[NemCrazy]

I'll definitely be following along! So I know of a good build thread on this idea. It's on nano-reef in the biotope section. Aquaticengineer made one that might provide some insight. I talked to him about it and he said the biggest problem was how much food he went through. Making it not cost effective to run. I believe he ran a peristaltic pump that would kick on once an hour for 15 min. But since yours would be hooked up to a controller wouldn't you be able to dial it down more?

 

[DTECH07]

Great write up! I just order some more components and have on schedule to start building my own next month (I have a new tank to plumb and setup first). I'll let you know of the progress or pm you with Q&A. Thanks again "J"

 

[halmus]

WAY above my abilities!!...lol. I get excited when i remember to do my waterchanges on time...lol. Looks cool though.

Thanks for following along. What I've done so far isn't above anyone's abilities. You deserve more credit than you give yourself. Anyone in this hobby has to be a little handy and innovative just to keep their tanks alive and well. Just throwing together a few parts from Radio Shack is pretty easy and fun to do if someone's already done the "cerebral" part for you. That is to say, I've only copied and modified, at best, someone else's circuit diagram so far. There will be parts to the build that will get hairy once I start designing the controlling circuitry, but a stirrer and the cooling mechanism itself isn't bad. Besides, it's pretty rewarding to have a product you've built yourself that meets

I'll definitely be following along! So I know of a good build thread on this idea. It's on nano-reef in the biotope section. Aquaticengineer made one that might provide some insight. I talked to him about it and he said the biggest problem was how much food he went through. Making it not cost effective to run. I believe he ran a peristaltic pump that would kick on once an hour for 15 min. But since yours would be hooked up to a controller wouldn't you be able to dial it down more?

I'll have to check it out if I can find the thread. Did AE build a cooler or was looking for something like this for a nano-reef? Having too much food run through isn't really a stopping point for this sort of build, since the fridge is scale-able. I could see how food requirements could be demanding on certain tanks though. I'm a little lost on what problem he had? I have to research the peristalic pump thing more. I see two issues:

1. The food might be physically to big or "chunky" to fit through a 1/4" line. Although, smaller foods and phyto would be no problem.
2. A multi-pump system. I have to see how it works. I assume many of them already have built in timers. Ideally, I would be able to have four separate pumps controlled independently. I could design the controller to turn on a relay for 4 independent outlet plug-ins on the fridge. Those outlets could be programmed to come on once an hour for a minute or whatever the user wants.

This brings up the big issue I have to face on this project as it evolves. How complicated can I make it? How complicated should it be? What are the major functions I want it to serve and how easily can I make those functions controllable by the user? I'll answer these soon.

Great write up! I just order some more components and have on schedule to start building my own next month (I have a new tank to plumb and setup first). I'll let you know of the progress or pm you with Q&A. Thanks again "J"

Great "D"! I'll be glad to see what you come up with. I've been running the stirrer for a week now non-stop. It's holding up. That's not much of a test for longevity but it at least shows that the design isn't total junk. Are you building a fridge or are you going to drill into a typical college dorm mini-fridge? The latter is certainly simpler and in the end will be cheaper than what I end up with.

 

[halmus]

This is going to be a slow process. I've got a LOT of work I should be doing instead. However, I want to start testing all the major components of the system that don't depend heavily on electronics. The stir bar is spinning away. We'll see if it breaks down over time. Next, I'm going to throw something together to start testing the cooling system. It's based on Peltier thermoelectric technology. A simple simiconductor device that "magically" pulls heat away from objects and pushes that heat to the other side of the plate. (It's the small white square with two wires). The idea is that you sandwich that device between two heatsinks with the proper thermo paste. Mount a fan on each side to pull heat away from the hot side, blow the cool air away from the cold side. The cold side, in this case, is inside an insulated box (fridge) which will be a cheapo styrofoam cooler for now.

I hope to wire this up some time this week. It's a simple process, just 12V and some bolts to hold everything in place. I just have bigger priorities. I'll cut out a hole in the styrofoam the size of the heat sink. Cold side inside. I'll hook up a thermometer and see how cold I can maintain it. I'll post pictures when it's going.

By the way, this circuit pulls a lot of juice! This won't be a "green" project. Any time you deal with temperature control, you're going to run into a lot of watts consumed unless you're using a swamp cooler. Chillers, heaters, toasters, all of those things draw a lot of amps. I'm hoping that this will only draw that current for short bursts so that the amortized power consumption is reasonable. This particular little circuit (1.5" x 1.5" peltier) with fans will draw 1 amp. I am hoping that I will only need to run two of these at a time.

 

[halmus]

I appreciate everyone's feedback on this. Your comments and suggestions help me refine what it is I'm aiming for as a final product.

Currently, I am planning on building this with two separate compartments. Each compartment will be able to hold two Erlenmeyer flasks. Each will have its own independent magnetic stir control. Each will have LED grow lights mounted under the fans pointed up. The two bays will be able to be temperature controlled independent of each other. The idea is that while one is chilling at whatever temp (40 deg?) the other bay can be set at 80 - 90 deg with the lights on. That bay can grow phytoplankton while the other feeds the tank. The lights will have a built in timer for each of the four container positions.


I wanted to have a way to grow phytoplankton while in the fridge but phytoplankton doesn't grow we'll at the lower temp. At least that is my understanding. It grows better at higher temps and stores well at lower temps. I would like this to be a versatile system.


So here is my list of functions I'm now aiming to accomplish with the design (these goals may evolve over time as I figure out that they're too complicated, or too easy to accomplish):
- four independent magnetic stir stations each with a controllable speed.
- four independently dimmable and programmable grow lights (one for each station)
- two independent temperature zones with user set points from 35-90 deg
- a light that turns on when the door or doors is open.
- Four independently controllable outlets for feed pumps that are set by the user.
- An interface screen that allows the user to program everything located on the fridge.

*- a Graphical User Interface to allow the user to more easily set feed times, temperatures, lighting schedules.

My big concern for this project is that it be easy for the user to interface with. The computer GUI is a nice feature but may be above my skill level. I'll have to see as time goes on. The whole project may be a flop, however the basic goals are quite achievable. A cold box with spinning things in it isn't that hard. That's why I'll refine the design as I go on. Since this is a project for my courses at UWT, I'll be given guidance as to the complexity of the project. I'm far away from submitting a design for approval.

 

[DTECH07]

Looking good! i'm doing some research in the heat shrink to see if there is a differences in types and density that would make a large differences in overall output or cooling properties. I'll keep you posted on both and let me know on the Phytoplankton i have three other types other than the Nana.

 

[halmus]

Thanks D. I've got a few disks of phyto starter cultures sitting around. I won't be doing anything with them for a while. I'm just trying to get the mechanics down first. I'm placing an order for the LED's from Rapid. I'm planning on placing them on a 'sink below the fan that will be spinning the magnetic stirrer. I'll use clear acrylic for the plate the flasks will sit on. The LED's will shine up through that from the bottom. Should be a cool effect. They'll be dimmable, so I'm sure the user will be able to dial it in to the intensity needed.

The temperature ranges I mentioned above may be way out of the range of feasibility. I'm not sure how low I can reasonably get, but I'm sure heating up the inside of the container won't be hard at all. The tests will tell me quite a bit. Maybe it will take two in each section cooling? If the system were run at max cooling with four thermocoolers, that would easily approach 5-6 amps for the entire system with fans and all. That would pretty much make the whole system cost prohibitive to run full time.

 

[NemCrazy]

With thermo electrics it may be tough to keep it in the 30's but I'd be interested to see if they could. A feeder like this would allow for successful keeping of gorgonians This would be a great thing for Coldwater tanks and allow for healthy azoox tanks which I think would be gorgeous! I was thinking that having pods growing in with the phytoplankton would be a great one, two punch. Temp swings might be a problem with that idea.

 

[halmus]

With thermo electrics it may be tough to keep it in the 30's but I'd be interested to see if they could. A feeder like this would allow for successful keeping of gorgonians This would be a great thing for Coldwater tanks and allow for healthy azoox tanks which I think would be gorgeous! I was thinking that having pods growing in with the phytoplankton would be a great one, two punch. Temp swings might be a problem with that idea.

The ability to both heat and cool is just an added bonus I thought about throwing in to add functionality. It may be unrealistic to think anyone would use an incubator function like that. After all, in my experience, phtyo grows perfectly well at room temps, perhaps warmed a little by the grow light.

As I think I said before, I'm not sure if any of this is really feasible any way.

I was basing my plans off of some online videos I checked out a while ago. That might be the first problem. The videos I saw were basically just mini-coolers for a can of soda or something like that. I thought, why not scale it up a little?


I ended up randomly selecting a 95w peltier cooler unit. My thinking was, that it would draw only an amp at 120VAC. That's true, however, after the transformer stage and rectification, I'll be drawing ~8amps at 12VDC. I wasn't thinking far enough ahead. My power supplies for testing are high-quality but only rated for 3amps. I have a dual power supply; I hooked the two up in parallel. That should give me 12v, 6amps. I hooked up the circuit last night, turned it on this morning, and it dropped the temp in the little cooler 10deg F in a few minutes and then leveled off. That's not enough, but I was only drawing 4.75amps out of the power supply. I don't know why it didn't try to max out the current of the power supply?


I'm going to pick up a computer power supply. 300-400W. I'll hook that up and see what results I get. That's really pushing the limitations of what I can do. A huge power supply consuming a ton of power isn't really worth moving forward with. I'll see how it performs with sufficient power supplied to the circuit. It still may take up too much room and consume more power than traditional mini-fridges. (I guess that's why mini-fridges aren't run this way, huh?)


This is what the test phase is for. Proof of concept. I'll post some pictures of the current set up soon.

 

[NemCrazy]

Bump. Any progress? Is the stirrer still functioning Peltiers chilling yet?

 

[seedlessinseattle]

Wow, cool tread so much info my head still spining, nice taging along

 

[halmus]

Wow, cool tread so much info my head still spining, nice taging along

Thanks for checking out the thread, it's going to be a slow process.

Bump. Any progress? Is the stirrer still functioning Peltiers chilling yet?

I have to give a short update. I'm stupidly, crazily, insanely busy these days with everything other than hobby related fun. I did manage to do some testing on the Peltier I got in the mail. The short story is that even after jerry-rigging a computer power supply to ensure I was getting sufficient amperage DC, the Peltier wasn't pulling the amps (consuming the Power) it was advertised to do. It did, however, cool down the heat sink on the inside of the cheapo cooler.

My next step, possibly, is to go a little bit crazy. Cover more surface area of the inside of a cooler with the heatsink, (ceiling, walls, back) and apply more Peltiers. It's kinda the Mythbusters approach. The"myth" didn't produce the expected results (expected by me, maybe not by all you out there). So, just WHAT would it take to cool this thing? Is it possible even if it isn't realistic to implement? If I do this, I want to do it in a way that I'll be able to recover and re-use the aluminum heat sink material. That stuff's expensive.

That might be a while. I'm working on a research project with "wireless sensor networks" at UW and that's taking one more day away from me. I'm working on taking down my big SPS reef. That's taking time and an emotional toll. And a million other things.



....edit: The DIY mixer design is holding up great though! At least something's panning out.

 

[halmus]

Like I promised, this project is taking a long time. I haven't had much time to devote to it, but I did come across a fridge at Fry's the other day. It's a wine fridge. I did a little more digging, and it does actually use thermoelectric cooling technology. So, I know that the technology can be made to work. I'm not sure what they did to make it work, but the fridge advertises that it can drop 16 bottles of wine to 42deg. Here's a link to the fridge. I might pick it up and see if I can reverse engineer the technology to make it work.

Cuisinart® Private Reserve® 16-Bottle Wine Cellar - Bed Bath & Beyond

I just finished a course covering Verilog programming an Altera Field Programmable Gate Array (FPGA) chip. It's incredibly powerful. Similar to the Arduino's many people are working with on on their lighting systems. Only it doesn't use the "C-ish" code I hear they use. I'll be getting familiar with the Arduino over the summer. I've got two official research projects to tackle over the summer. One of them has to use Ardiuno. It's a testing system for local streams. (what is the water volume flowing through a stream? what's the water depth? more data...) The other one, I can use whatever I want, but it's going to come down to testing the systems and whatever is going to be most effective.

The water stream project is already bringing a few ideas to mind that are applicable to our hobby. The water depth sensor I'm working on is a great alternative to the float sensors we source online. They always fail over time. I've come up with a way to seal magnetic reed switches in a PVC pipe with an enclosed magnet on a float that's also enclosed in PVC to ensure no metalic corrosion over time. The benefit is that it's way more robust, although it's really just a larger version of the cheepo floats we use.

Anyway, I'll keep you all posted as I make some real progress.