ReefBuddy DIY Aquarium Controller $89

[indigo]

Very nice project. I know this is continually evolving but it would be very useful to compare your product with the open source DIY project at

http://reefprojects.com/wiki/Main_Page

I was planning to build something along the lines of jener8tionx's but with my own spin, adding

1. charts on an graphic lcd (I have not found I2C but did find a serial-interface 128x64 for $21), and perhaps
   
2. either SD-card or direct-to-computer logging, and
   
3. web-charts with the direct connection, and ultimately also sms alerts.
   

 

Is the following characterization fair?

 

Reefprojects.com describes a dual hardware+software hack where you buy the parts individually from (many) different sources, use the open source software and hopefully contribute to the software and idea pool. Learn a lot but fry some components in the process and end up with something that's your own but with lots of dangling wires. A big problem/cost for this approach is in a powerstrip with individual solid-state relays - single relays are so expensive and the 25A ones on ebay seem overkill.

 

Your's seems more like a kit concept, which could have increased costs due to your own designed components (with a profit incentive to compensate your efforts) but also decreased costs due to bulk-buying and optimizing, and without paying for your own mistakes when you build from scratch. Here one will end with a slicker product but also less customizable? Also you may be able to provide a very cost-effective solution to the relay-managed power-strip.

 

My impression is that with the stand-alone kit concept considerable resources need to go into the keypad-interface while with the compiled hack version you just change parameters in the code and recompile. Then you can devote resources to the output rather than the input. Beyond the present sensor readings temp/PH/ORP etc. I really want a chart of the last 24 hours that reveal the swings overnight etc. I'll be able to see a problem at a glance. You also can also visually see when devices were switched on-off etc. Once you have such charts and a connected computer its fairly simple to make them available on the web for that stress-free vacation.

 

In trying to decide to proceed on one's own or wait for your product it would elaborate a bit on what your kit may include and to what extent it would be customizable. In particular

• if it would be easy to hack a graphic lcd and its supporting code into yours,
  
• what probes would be included, and also
  
• what your solution to a controlled power strip would be/cost.
  

Thanks

[Militant Jurist]

Yep totally possible. Would need to be programmed. BTW, I'm glad you responded. Since this is open source project we have to conserve resources as much as possible. We'll be looking for someone to help setup our trademark and disclaimer verbage. IM me if you are interested.

 

YGPM

[AndyL]

Personally, here's my .02 on where you should go...

 

Start with a basic system - the base arduino + a shield to handle the basic inputs & outputs. (can likely do this under 35$ - in a case with room for expansion).

 

As there's going to be people looking for other options - ie some will want the lcd - some will want wifi/web access instead; make those available as easily plug in-able modules / shields... Dosing pump control - definitely as a separate shield etc... Keep the base price low - then let the users pick/choose options from there...

 

I don't think the lcd should be an integral part - a basic case lid for those of us who don't so much want the LCD, as standard, then a replacement lid with LCD that plugs into an I2C shield (I think I2C would be best installed on that one basic shield)... Likely better option - then you can have the gamut of interface options from bluetooth control from your palm/crackberry/etc, to wifi to ethernet to various upgradable/scalable screens...

[hcsceo]

[*] if it would be easy to hack a graphic lcd and its supporting code into yours,

[*] what probes would be included, and also

[*] what your solution to a controlled power strip would be/cost.

Thanks

 

You are right that a bulk of our time is devoted to user input and customizability of the settings. Making things work is the easy part, creating a menu system that is both user definable and easy to operate is the challange. Effectively our system will do everything that the open source one will do except for the stray voltage check. As far as customizability, this project will be very customizable by the end user through i2c so you could add any features that are lacking for yourself and send that to us as a possible upgrade for all systems in the official software release. You will have access to the available ports on the arduino that are not being used currently via additional ports on the side of the case and there will be a modular jack that will allow you to add i2c compliant devices. The selling point of this project is simple.. A diy kit that will look store bought when complete and cost you very little to get into. In order to do this the basic unit has to be, well.. Basic. It needs to provide the most needed functions but expandability for additional modules and features down the road. If you want to make changes you are free to do so and we encourage this.

 

Graphical display will be difficult for two reasons. 1. The arduino processor has only 32k of ram and using that to store information is really not feasible. 2. The additional library required and coding for graphics creates the same problem with available space. We are thinking that a separate computer application that interfaces via the usb or a network interface would solve this problem on tracking. There is one final possibility.. It is possible to add a SD card chip to the project and store information there. However, cost will be a limiting factor for such an option.

 

Lastly, since this is an open source project all schematics and code will be available for you to hack and create your own projects from. All we will ask is that any changes or updates are made available to us so that can have the opportunity to include it in the official release. The name reefbuddy will be trademarked so any use of the name will be prohibited including "ReefBuddy compatible" unless it is tested and approved for use.

 

I hope this answers your questions. Keep them coming.

 

Steve

[hcsceo]

I don't think the lcd should be an integral part - a basic case lid for those of us who don't so much want the LCD, as standard, then a replacement lid with LCD that plugs into an I2C shield (I think I2C would be best installed on that one basic shield)... Likely better option - then you can have the gamut of interface options from bluetooth control from your palm/crackberry/etc, to wifi to ethernet to various upgradable/scalable screens...

 

I think we will include the lcd as cost is only about $13 and really needs to be part of a complete system if it is to be effective at all. We won't be using traditional Arduino shields as the control unit will have all the breakout for you to add your own modules. All this will be provided in a professional case with graphic button overlay that will rival any product on the market currently. Right now the basic unit is looking like this

 

Arduino Microprocessor

4x20 LCD

20 Key Keyboard

DC45F Polycase 6x4.5 with mounting flanges CNC routed

PCB and parts to build

PH

ORP

Salinity

Temp

I2C expansion jack - 6p4c RJ11 with 5v, gnd, SDA, SCL

4 digital inputs

up to about 6 additinoal IO's including PWM via RJ45 jack with ground and power available to you.

 

Hope this helps.

 

Steve

[indigo]

The memory requirements of a graphic lcd would depend a lot on how it is used. What I had envisioned was plotting the high and low of say temp and PH over 15 minute intervals which could be represented by 24x4x4 bytes and use 96 columns of the LCD and with the remainder of the 128 columns for the latest numerical values. So you need 0.5kb if you want to store the data and replot on queue but essentially none if you just write the latest update (overwriting that of 24 hours ago). One could also use a few rows to indicate when various pumps/lights/fans etc go on and off. I think this visual feedback adds tremendously.

 

The libraries/code for a graphics display would get larger for displaying the menu's etc related to the keyboard input in your implementation. Perhaps one could add the graphics display I envision as a 2nd display for the scale of $25.

 

The other question I had was about the power strip. For a commercial system the DC8 runs around $150. The reefprojects version with eight 10A solid state relays may still approach $75-100 and requires a lot of 115V re-wiring and beyond a heater and perhaps one big pump we typically do not need that much current. Are you exploring a solution? I think this is an area where there could be great demand if you can get around liability issues. Its frustrating that the power strip alone may cost more than the controller with display and probes!

 

--Jan

[disaster999]

WOW, first off, i thank you for doing this for us. will definitely be interested in a controller!

 

i know this is a long shot, but can the power supply and outlets be customized for other countries? if i was to purchase this, can i just switch the power supply for the adrino from 110v supply to 220v supply? can i change the USA style blade type plugs into the euro style plugs?

[hcsceo]

The memory requirements of a graphic lcd would depend a lot on how it is used. What I had envisioned was plotting the high and low of say temp and PH over 15 minute intervals which could be represented by 24x4x4 bytes and use 96 columns of the LCD and with the remainder of the 128 columns for the latest numerical values. So you need 0.5kb if you want to store the data and replot on queue but essentially none if you just write the latest update (overwriting that of 24 hours ago). One could also use a few rows to indicate when various pumps/lights/fans etc go on and off. I think this visual feedback adds tremendously.

 

The libraries/code for a graphics display would get larger for displaying the menu's etc related to the keyboard input in your implementation. Perhaps one could add the graphics display I envision as a 2nd display for the scale of $25.

 

The other question I had was about the power strip. For a commercial system the DC8 runs around $150. The reefprojects version with eight 10A solid state relays may still approach $75-100 and requires a lot of 115V re-wiring and beyond a heater and perhaps one big pump we typically do not need that much current. Are you exploring a solution? I think this is an area where there could be great demand if you can get around liability issues. Its frustrating that the power strip alone may cost more than the controller with display and probes!

 

--Jan

 

I'll have to give the Graphic LCD idea more thought. Backtracking code for that right now would push us a back a bit. However, if the space is there then I don't see why it couldn't be done.

 

On the power strips, you are exactly right. Power centers cost money and depending on how you do it will determine how much it will cost. Right now we are using an Octocoupler (less than a dollar) that switches at the zero point of the power (essentially soft start) and that controls a snubless triac (around a dollar). There are a few resistors involved but effectively that is it for each circuit. The snubless design gets around power issues normally associated with SSR's and the octocoupler isolates the circuit optically from the AC side. By the time you buy the AC jacks and box it and the PCB along with AC cable things start adding up. I am hoping to include one 4 port ac switch box with each unit sold. For reference here is a design sold commercially that controls 4 circuits for $18 Velleman Triac System This unit does not do zero point switching and is not a snubless design but it is essentially the exact circuit we use just with different components. Now our design is not fully tested yet. I've got to get it switching several thousand times a day for a week or so to see how it holds up and whether or not inductive devices are going to cause us issues.

 

Hope this helps.

 

Steve

 

 

 

WOW, first off, i thank you for doing this for us. will definitely be interested in a controller!

 

i know this is a long shot, but can the power supply and outlets be customized for other countries? if i was to purchase this, can i just switch the power supply for the adrino from 110v supply to 220v supply? can i change the USA style blade type plugs into the euro style plugs?

 

I'll have to check the ratings on the triacs but I believe they work up to 250v and 6 amps each. Don't quote me on that. It would them require only different jacks to be installed or a jack converter and you would be good to go.

[jener8tionx]

indigo: hcsceo's power idea is way more efficient than my 25A SSR's. I only went that route because they were readily available on fleabay and met my need. I hope that his project really kicks off. My stuff over at reefprojects.com was not meant as a kit or layman's project, but just me trying things out. hcsceo's idea takes it to the next level and really gets things going. I am excited to see where this all goes.

[hcsceo]

indigo: hcsceo's power idea is way more efficient than my 25A SSR's. I only went that route because they were readily available on fleabay and met my need. I hope that his project really kicks off. My stuff over at reefprojects.com was not meant as a kit or layman's project, but just me trying things out. hcsceo's idea takes it to the next level and really gets things going. I am excited to see where this all goes.

 

thanks for the nod Jener

 

OK quick update. I've got most of the circuits complete. The orp and PH probes should be here this week, however, I feel very confident the circuit I have laid out will work for both as both are just testing mv and reporting it to analog. The only circuit that has to be done before I can ship our design off to have a prototype pcb built is the conductivity (EC) circuit. There is very little information available on this circuit so it is going to require some testing. As it turns out the commercial unit my company sells for fountain systems does not include conductivity so this will have to be done from scratch. I need a probe to get this finished. If anyone can help with this please send me an IM. We are looking to use a probe with a mini din jack exactly like the aquacontroller probe so that our unit is compatible with accessories for other devices. I have a circuit diagram for the conductivity that just needs to be tested. Thanks for your help.

 

Steve

[indigo]

The Reefbuddy power solution does seem to reduce cost considerably and its great that 4 switched ac outlets would be available with all those probes and the control unit. I'd love one myself. I think Jener's power approach may be easier/simpler/costlier for a hobbyist -- but I'd like to understand building the Reefbuddy version from scratch to see if a novice might try it out.

 

I was tinkering with my graphics LCD idea and it may be better as a remote secondary display. Here is a *mockup* using 96 columns for 24h sampled at 15 min intervals. The top graph is the temperature with max/min values that trigger a heater and fan. The bottom graph is PH with a low boundary triggering a dosing pump. The 4 bottom bands indicate activity in the ATO, Kalk dosing after dark, heater and fan respectively. The vertical line is current time, which sweeps to the right with the latest 15-min sample added - requiring no storage. The catch is that there is not much room for the rest. The current temp and PH are 14 point Courier and the time is 9 point and won't fit anything after 10pm! (GraphicConverter does not have a smaller font.)

 

 

You would need to store 96 bytes per chart if you want to draw them at will and managing the menu's with a graphics LCD may be a pain. Also, while I'd love to know if my tank does anything like this over a day and I could live with this resolution (128x64 $27-84x50mm or $21-51x41mm at seeedstudio), it may be worthwhile to wait for higher resolution LCD and for cheap color. I'd be curious what other reefers think?

 

--Jan

[Militant Jurist]

I'm not sure if this would actually fix the time issue, but could you just switch to military time? That way, you won't need the 'p' to indicate PM, so you could use all 4 digits of the current time.

[The Propagator]

Yea you and I are on the same page. I expect the cost of having someone put it together will put it in the same range as the Reefkeeper Lite. You'd have to want this product more to actually do that I think. If I was a non DIY type I'd just buy another product that was ready to go with warranty. However, some people will try to do a project that is outside of thier abilities and be mad when it doesn't work. I suspect that by the end of the year China will be making these for $35 packaged and with warranty.

 

YOU KNOW IT !!

[hcsceo]

YOU KNOW IT !!

 

Man Prop can you imagine if that happened? How cool would it be to have an aquarium controller for $35 that does everything this will do? I can see the floods from failed controllers now

[scottyreef]

not to take away from you great DIY build but years back before reef controller were so cheap i build my own in college with these guys and got a 3" X 3" panel to montor it with and handheld micro controller the total build back then was $165 when everything was said and done sadly i sold it when i broke down my 180 4 years ago. I will hunt down some pic to show you how i build mine

 

http://web1.automationdirect.com/adc/Overv...le_Micro_Brick)

[hcsceo]

not to take away from you great DIY build but years back before reef controller were so cheap i build my own in college with these guys and got a 3" X 3" panel to montor it with and handheld micro controller the total build back then was $165 when everything was said and done sadly i sold it when i broke down my 180 4 years ago. I will hunt down some pic to show you how i build mine

 

http://web1.automationdirect.com/adc/Overv...le_Micro_Brick)

 

Wow that product is pretty cool. We use a lot of PLC's in panels for fountain systems. They are robust and packaged nicely. I might have been persuaded to go this direction from the get go had I seen the prices. There is an awesome build waiting for someone on that link.

 

Steve

[scottyreef]

thanks steve here is close to the user interface i used mine didnt have keys and i had to buy a small 6 key controller I got all the stuff used on ebay to how ever wants to know or if anyone needs any help. Sorry about the hijack

 

http://web1.automationdirect.com/adc/Overv...-Graphic_Panels

[basser1]

Its' been over a week since last post, any updates??

[hcsceo]

Its' been over a week since last post, any updates??

 

Not much to report here. I've been working on the PCB design and trying to get everything to fit in the case we have selected without any vias. At this point I don't think it will be possible which means home etching will be out of the question. I might try and do a design that has a few jumper wires. I've also sent for a few peristaltic pump test units. They should be arriving soon. My plan is to build an I2C dosing system that will be controlled by the controller. This probably won't be done in time for the first release of the controller but will be available at some point. Currently I'm waiting on parts to finish the ORP and PH circuits. I might go by Radio Shack today and see if they have some op amps that will work till those parts make it from Digikey next week.

 

The coding is really the difficult part of this project. Since I have changed the keyboard from a 7 button to a 20 button 4x5 matrix we are having to revisit the coding. The final design has been decided on and it will look exactly like this with a different graphic on the button overlay.

 

 

I've got pricing back on the engineering of the button panel and the final cost of each unit and I was happy though it could always be less. Once I get the PCB done I should have pricing on it and the case we have chosen and then I we should have a number for what we can sell these for in kit form. I'm constantly changing the design to keep that goal of $89 possible. Anyway, that is all I have right now. All that is left is everything

[davidr2340]

Man, this is going to be sweet me thinks!!!

 

Can't wait!

[Needreefunds]

+1

 

How sweet would it be to have something like this available for under a hundy?

 

An exciting project for sure Steve!

 

[The Propagator]

Man Prop can you imagine if that happened? How cool would it be to have an aquarium controller for $35 that does everything this will do? I can see the floods from failed controllers now

 

 

They actually already have one for about $75.00 on eBay from China but it only come sin 220v.

I bet it suck major arse though.

[Mr. Fosi]

Link?

[hcsceo]

oooo damn Prop that is pretty good pricing I wonder what features it has. I'd like to see the link.

 

OK guys PH and ORP are completed. I've got to finish the code to get it displayed on the screen but that shouldn't take any time. Conductivity is my last major hurdle hardware wise.

 

I've got sample Peristaltic pumps coming in and will be designing a Dosing system to be included with the controller software. It will work via I2c and will plug into the RJ11 lack like the switched ports. I might not get the dosing parts completed by Sept. but I will sure try.

 

Firewolf4 has suggested adding a Phototransister to the project to measure lumens. I think that would be an awesome feature and will try and include it in the controller. I believe we have a free analog port so this should work. Let me know what you guys think of that.

 

Anyway, we keep moving along. The Menu system is really the most difficult task and is giving me headaches but we'll get it worked out.

 

Thanks for everyones support and IM's with ideas.

 

Steve

[zjharva]

This is a cool project! I have an arduino, but never learned how to use it. I will definitely buy this if its made available! keep up the good work