Some of you have followed my other DIY threads that include skimmers, rockwalls, tanks, media baskets, etc. I've been keeping this project under the radar because of the complexity of it but wanted to go ahead and share it with everyone.

I've decided it is time for us to have an aquarium controller that everyone can afford. My goal was simple.. make a controller for under $100 that can do what available controllers on the market can do. Because of the manufacturing cost involved for a project like this, the only way to do it is by offering it in a kit form that everyone can build. Ultimately, this kit will have a very nice case that is CNC'd, button overlay with reefbuddy logo, and professional PCB ready for you to solder the kitted components to along with an instructional video to step you through the entire process of completing your build.

We have chosen the Arduino controller as the microcontroller for the project. The choice was simple. Arduino is Open Source and backed by a devoted following of people that are willing to help with projects free of charge. The community is rapidly expanding and there are many controller choices you can choose from. The controller already has the power system and usb designed into the product at a very low cost. All development software and tools needed for design are available free of charge so getting into this is relatively inexpensive.

Here is a list of available functions that will be available.

4 line LCD Screen
7 button interface
Screen and Buttons remotely mountable from the control box (Similar to Reefkeeper products)
4 switched inputs (can be used for ATO's, etc.)
Temp
PH 1
PH 2
Conductivity
ORP
O2
16 available timers
16 switched 110 outlets (available in groups of 4 or 8)
All modules will interface via a standard phone cable up to 10' in length

There has been a lot of progress made with Arduino by others in our community and was the reason I got excited about this initially. As I get further in the project I'll update here. TEMP, Switched Inputs, Timers and Power Outlets are all up and running. We are currently working on the Menu system. This is going to take the bulk of the programming and memory from the controller but will give complete flexibility to the end user. There are two of us completing the project. I'm working on all the electronics/PCB designs and some coding and my partner is working on coding. The PCB designs for the most part are complete and I'm hoping to etch samples this weekend.

So the big question is "What will $89 get me?" and "When can I get it?" Well right now I'm hoping it will get you the basic controller which includes the Button/LCD panel, control unit, 4 switched outlets, and temp guage. It should be compatible with available probes on the market and allow you to plug them into the control unit. Our list is long but we have set a goal to have the product in a final form by the end of the summer.

Anyway, I hope you enjoy this project as much as I enjoy working on it. I'll post more information as it is available. We might be asking help from our community in the form of letting us borrow probes as we move along (If you can help in this manner please IM me) any help in this manner will be appreciated.

Steve

Super Ugly Prototype

way cool when can we buy verry nice

Big fan of open source. I'll be watching this...

Yes Open Source has its place. The aquarium trade really needs open source because innovation is very slow and cost is very high. Patent on LED technology for Aquarium Lighting... Give me damn break!!

QUOTE

The best ideas are common property. ~Seneca

For those of you who are wondering what the final product might look like. Here is a sample of a controller that is designed and built by the company I work for. Sorry for the black out as this controller is for a specific project not yet released. This controller is not open source (sorry we have to make money on something). We'll be using the same case but the screen will be in a different position. The same company that builds our button overlays will also build the overlays for the reefbuddy project. Enjoy.

cant wait till the end of summer to see the final product

been looking at some controllers lately and this is def affordable

want want want PM me when ready

Put me in line....looks great.

So how many of the functions are currently working?

I've got to start working on the probes but we want to get the menu system straightened out first since it is the bulk of the programming.

Absolute awesome build.

I'm working on one with a friend and the probes are the issue at the moment. Especially conductivity since it needs to be AC or rectified DC.

Man, you beat me to it. This was in my list of projects to come. I am actually sitting in my microprocessor class right now. After I realized how simple microprocessors are, I wanted to do something like this. Out of curiosity, which processor are you using. I assume it is on that arduino board that you have. Where did you find that and how are you programming it. My problem in my project is that I cannot figure out which processor to use. I want to use a non-risc processor just because it would be more simple. The motorolla 68k series seemed to be ideal for this application, but I cannot find a premade board that lets me program it. I do not really want to build my own board due to my current lack of skill.

The controller that the company I work for builds is for fountain and pool systems and has a bunch of the same probe technology already designed. I'm planning to use a bunch of that for this system as well but modify for the arduino. We currently monitor conductivity, ph, and orp with our systems. Our microcontroller is made by Rabbit so the there will be some coding work to be done to move it over to arduino. However the backend circuitry is esentially the same for us.

nerbaneth. The processor is a ATmega328. Bootloader is available via opensource. It is possible we move to another ATmega if programming room becomes a problem as the 328 only offers 32k or memory. However our existing code is only at 6k. with essentially probe and menu coding left. Keep in mind menu coding is a big job but required for end users with different requirements. My partner has his Masters in CS and is a master at c programming and streamlining code to fit in our needs. We'll be using the built in eeprom to store the settings and the RTC has a battery for time. I can also recommend the ATmega644p. It has a bunch more IO's however it's biggest advancement is 64k memory. We are using I2C for almost everything. Essentially I'm only using 2 pins to run from the arduino to run 32 io's with our port expanders. There are a lot of processors available that would be very easy to use, however a big requirement for us is that the controller be easy to build so through hole construction is a must and DIP processors are required.

Looks like I'll be getting a PH probe really quickly thanks to the generousity of one of our members. I'll get it running while we are working on menus and post some updates on that soon.

Cool build hcsceo, I'll take one when they are ready. Let me know if you need any "testers"

We are definately going to need testers when we get closer. I'll need input from several people on the instructions and build process to see how easy or hard it is to build from our instructions. Then we'll need those people to use the product and give input on use. The idea I was working on last week is to have an application process and select 10 users to build and test the units on working systems of various sizes and complexity once we feel the product is ready. We'll select people based on tanks and equipment already in place with a special need for people that already have controllers to compare against. Anyway that time is a ways off. We are committing ourselves to Sept 1 to have the kit ready. Whether that means ready for beta testers or ready for public consumption I don't really know now. I do know that we are way ahead of schedule as our first goal was 1 year. We are only a month into the project and it has been easier than expected. Most all the hardware problems have been resolved with the exception of the probes. Since I have access to electrical engineers and a working product for another line that does most of this for commercial systems it should be relatively easy. We will see.

Steve

This is really a awesome build bro!!!

I will be interested for sure when the time comes!!!

I would LOVE to be a "tester" if you need me!!!

Kudos bro!!!

I would buy one.

Dont' buy it yet. Might flood your place

Just let me know, I am going to need another controller for the 90g eventually and I already have an ACJr on the 40 breeder. Would be interesting to try and report feedback, as long as it doesn't fry my tank, lol.

I'm sure your tank won't catch on fire but the house is another story LOL. I'll let everyone know when we start the application process.

Is it bad I still want one? I keep coming back to trying something like this.

I hope this isn't too off topic (we can PM if you prefer) but I have a couple more questions! (and I can have more organized thoughts than my last question since I am out of school now)

To get things started let me tell you a little bit about what I am planning and myself. I am an electrical engineering student at the end of my 4th semester(only about halfway through) I plan on having this project finished by the time I graduate. My goal is to make a temperature controller. I want to turn on a fan when the temp gets over 80 and to turn off the lights over 84 and possibly a chiller on at that point as well.

I am curious as to how to modify/create a safe power strip that would have two digital switches. (So I can turn the power of devices on and off with the controller)

I think I will use the parts from a digital thermometer and use a multimeter to find whatever signal on the thermometer corresponds to different temperatures. If there is an easier way I would be interested in finding out about it.



I have never heard of a port expander, but a demux would require 5 pins to run 32 i/o's (if I can recall correctly) I would be interested in knowing how the port expander works!

I noticed you are using what looks like a USB cable? Does that connect to your temp probe?


I've never worked directly with a device like the I2C before. To my understanding it is a modular integrated circuit type chip that is inside of the microcontroller. Could you tell me more about the I2C and how you are using it?

Being a college student, I would like to make it cheaper possibly in the $25 - $50 range. Do you think that is possible?

For Orp, Salinity and PH would you buy a probe from an aqua controller and plug it into your device and then let the device handle how to read it(and power it) or is there some intermediary device that sends your aqua controller the signal from the probe.

Sorry for bugging you with all these questions but the answers will help me out a lot!

Thanks,
Nerb

nerbaneth, everything you want to do can be don with the arduino controller. The best place for you to look is the playground at www.arduino.cc and check out sample code and forums.

The other item you want is regarding a relay to switch high power devices on and off. There are a lot of ways to do it but I think i would look into some ssr's in your case. Inside an ssr is simply a octocoupler and triac and a few resistors. For the reefbuddy controller we are building the ssr's completely. We need to be able to control what phase the relay turns on and off and take into consideration inductive devices that cause an ssr to perform poorly, not at all, or lock them in the high position permanantly.

Lastly, what you want is a thermister for your thermo readings. I can recommend the dallas one wire line for this. Your project should only take a few hours to complete with all the components ready to go.

The port expander we use is the MCP23017 and has 16 io's, is i2c compliant, can interrupt on each IO and has built in pullup resistors. On I2c you can do a google search and learn quite a bit. I2c is a protocal designed to allow communication between devices in a network. It requires only two pins and each device on the I2c network has an address that responds to it. The maximum distance for an i2c transmission is 9 or 12 feet I don't remember which. It lends itself well to robotics as multiple controllers can do jobs and report back to a master.

You should be able to do your project in the range you have selected. It seems very straight forward and all the code you need is already written and available at the playground.

Hope this helps.




Fosi,
I just now noticed your question/problem with conductivity. I think what you guys should focus on is PWM as the transport to regulate your power issue. PWM can create a squarewave which should break your electrical field without disrupting the ions in the solution. If you are using the arduino controller PWM is supported out of the box and shouldn't require any more addional hardware except possible an op amp. We have not got to conductivity at all so this is all a guess right now but it is my understanding that conductivity probes work with squarewaves. Hope this helps

Steve

For menu options having some sort of lighting with PWM for LED rig dimming use or possibly as a way to ramp up/down wavemakers might be nice. I know some of the powerheads and wavemakers give up when simply switched with relays but I don't know how they'd handle PWM or if there's some other method used on those.

-
S

!
Thanks so much
!!!!
You are awesome

I can't wait to go to arduino!
good luck with your controller!

Thanks again,
Nerb

we won't be including led control in the basic unit. We are thinking of doing it as a separate unit that is controlled by the controller. LED has limited use commercially as patent issues are consuming the advancement of technology in this area.

almost forgot about pwm and powerhead control. I don't think it will be viable as the pwm is DC currents and most pumps are AC. The trick is switching the relay on at the low side of the ac pulse so that the pump soft starts. We are working with a few octocouplers that do this.

Helps a little I think... What you suggest sounds better than building an extra board to give us rectified current.

I'll be glad to work with you on this. I've got a ph probe coming. I've got the circuits done for it so it should just be a matter of plugging it in. As soon as that is done I'm going to focus on Conductivity. Looks like it might pose the most challenge of the probes and could be why many of the controllers don't offer it on the low budget models. What probe are you guys using? I'm looking at a few. I'm really interested in the inductive probes as they are coated.

Dude, you know you can count me in on this.

We are planning to use a DIY titanium probe but we haven't built it yet b/c we haven't worked out how to do it.

I wouldn't have it any other way.

I think it is pretty simple. Two rods spaced exactly 1cm apart wire and potted then leads sent to gnd and arduino pwm. Might need a resister 1k or so on the IO side. PWM at 60 HZ and read the ohms and convert with formula and shift it for temp. I have the formulas somewhere but I think this is it basically. Now with all that said you need to be careful cause it is possible that you could separate the hydrogen from the water molecule and create brown's gas and have a fire hazard if you got a spark. I don't think this will happen cause we are only talking like 20ma.

In SW, a more serious concern would be the chlorine gas that is formed from the hydrolysis. More deadly in small amounts than a bit of hydrogen.

see this is why we need each other. I would have burned down the damn house. This is one of the reasons I'm really wanting an off the shelve probe. It'll be easier for our DIY people that will put this together and look professional when installed and mounting options are already taken care of.


If you polled the probe only once a minute for 5 seconds you should be fine.

I would definitely be interested in being a tester! let me know if you don't already have enough

Hey, once again thanks so much!

What thermister did you prefer? I found 2000+ different kinds ranging from 30 cents to 80 dollars! I don't think I want to build the probe myself because of waterproofing issues (even though its only approx 5v and .5mA with a 10k thermister I would prefer not to put that in the tank) Maybe heat shrinking tubing would make it water proof?

Is there some database you guys refer to when trying to build the titanium probe? I would have no idea where to find a drop of information about any of that. Even the library at my school with an extensive EE section wouldn't have something this specific(I don't think).

This is excellent for those electronics savey individuals who are good wiht an iron.....BUT
Dude you know you are going to get bombed with people who couldn't solder a single connection with absolutely nothing around it if their life depended on it. Those are the people that will be the majority of your 1st run I bet. They'll see the price and what it does then hop on it. Two days later destroy it with an iron and blame you.

I barely understood any of the posts in this thread, lol. Looks like a good project for us cheap @ss reefers. I'd love to be a guinea pig if at all possible, but would still probably buy one of these regardless.

/\
The defense rests.

Ouch, that hurt. I can solder a circuit, but didn't understand some of what they were saying because I'm not an electrical engineer. I'm a drafter/cnc machinist student, talk machining or drafting programs and I'll understand.

It seems like PWM might be the solution to the TDS problem.

The main reason we were thinking about a DIY probe was to keep the cost down. We are already going to have to buy a pH probe. The probes themselves will inflate the cost of the controller pretty quickly.

I've never had a chance to work with on of the higher-end controllers so I don't know what they recommend vis a vis their conductivity probes... It seems as though the longer it is in the tank, the more biofilm will grow on it and that should change the conductance. If so, the probe would have to be pulled out every couple days or so and cleaned.

I'd like to test one of these, BTW. You're a lot further along than we are.

I'm a be building one of these mama jamas soon !

CNC router

This is a very exciting project. I will start saving my money now. I can't wait until the end of summer.

Allen

Just saw this thread and wanted to say how exciting it is to see a new product get "developed" here on N-R! I I am a firm believer of contributing to the success of fellow nano-reefers. SapphireAquatics and Stevie T and evilc66 are some that come to mind. All have developed great products and offered them to us. This is one of the things that makes Nano-Reef such a great asset!

Good Luck with your project and I'll be looking forward to the updates!

I absolutely agree !

Thanks prop. We have already decided on how to handle this. The great thing about open source is that we are betting on a community of people that will be willing to help advance the project. One of the ideas we have going is to find some people that might be willing to assemble and test the products for individuals with less than savvy iron abilities. Of course this would come at a cost that would be established by the person building and the person hiring. Secondly, soldering work is very easy. Most of our assembly people are trained in 10 minutes and have never handled an iron before. We are thinking of including a test product. Perhaps a simple led 9v battery connection and test pcb to practice on and include a section on soldering in the training video. Finally, there will be the obligatory "try this at your own risk, if you screw it up we'll help but don't cry if you have to buy another".

That's great on PWM and the probe. At $89 we never expected to include a probe with the product of course one will be available for purchase on the site. You might look closely at an inductive sensor. There are none out that I know of for tanks but they are designed to work in corrosive and unclean situations fulltime. As long as the center hole does not get clogged it will keep working without loss of accuracy. The only thing I wonder about is the energy field it creates and whether it effects other probes around it. Anyway, I might get some titanium rods from ebay and try and build one. Like you said it would need to be pinned out like the other offerings available so that it interfaces with the controller correctly.

That seems more realistic in a real world scenario then a separate jobber for hire.
( well.. depending on the assembly fees I guess versus the cost of buying one assembled ? )
The test kit still wont do it for those who are truly turd burglars at soldering, and it would be fairly costly in bulk terms for you to purchase and include a test kit in each package.
I would just make the standard do it yourself warning loud and clear in bright ass orange writing and offer tech support if needed, but of course.. if you fudge the soldering up its on you because this is a DIY project.
All in all though It should turn out very well. I am POSITIVE you will have those with no skills trying it, but the average person I would think will have enough common sense to realize what this involves before they try it and recognize their own limitations. ( lets hope any ways LOL! )

Don't specifically know what specific projects others are working on... I'm starting to mess around with building an LED driver shield; basically to eliminate the LED buckpuck etc drivers.

Still in the preliminary stages - idea is basically to incorporate 4 LED drivers onto a shield format; probably needing to incorporate I2C, constant current output would be software configurable from 100mA to 1500mA, pwm dimming, realtime voltage and mA outputs to the arduino for monitoring (ie if a led burns - the arduino can adjust or kill power to save the string as needed).

Found a basic PIC based led driver circuit @ http://ww1.microchip.com/downloads/en/AppNotes/01138A.pdf - but it's going to take some considerable upgrading to make it handle the loads that would be useful for driving luxeons/cree's, along with adding in the digital pot's needed to handle the software configurable current output... Talking to some geek friends last night who know the EE side better - it's definitely possible - just working out the hardware side - then the real fun of figuring out software side begins...