Hooking an RO/DI unit directly to your ATO is a bad idea IF you are using float switches. They do fail over time and just like most disasters in this hobby, it seems to always happen when you're not at home or vacation. Even with redundant float switches may not help.

I would say have an ATO reservoir no bigger than an actual failure where all the water got dumped in. It wouldn't lower your salinity too far below dangerous levels. Most things can survive it. On the bright side, it might solve your marine ich problems

As for the water changing board, for $40 bucks... I am extremely tempted... I am in the market for an AC3... wonder if this could integrate with it or if the AC3 can perform that function.

I wouldn't say it's always a bad idea. It's definitely a bad idea if you don't have a backup switch with an alarm, which is why one was incorporated into the design.

If you can program a output pulse for a few seconds from the AC3, you could use it in place of the start button and trigger a water change. As for programming the functionality into the AC3...I don't know about that.

Backup switch (another float switch?) and alarm won't help if no one is around to hear it. It works if you're home.. or someone is home. The only sure-fire way to ensure your water doesn't turn into fresh water is to physically limit the size of the top off source. Well, one of the sure-fire ways. If someone can think of something better...

I use to hook my RO/DI straight into the auto top off as well. It's not good for your membrane either. The first few seconds of water coming out has quite a bit of TDS. It clears up after a few seconds of running.

http://www.melevsreef.com/rodi_recipe_for_disaster.html

I think you can program an AC3 to pulse. When I get one, I'll see what all if it can.

The backup switch also disables the pump, in addition to the alarm. In order to have the ATO flood the tank, you'd need to have two concurrent failures. Lets run the reliability calculations.

Two assumptions:

Given that I've run a floatswitch ATO for two years without a single failure, I think a highly conservative estimate would be for a maximum failure rate of one per year for any given floatswitch.

We'll use a mean time to repair (MTTR) of about eight hours on the first floatswitch failure, based on most people's typical activities - unless you travel for work on a regular basis, you're going to be home almost every day, at least once.

So, we have an expected unavailability on the main floatswitch of 8 hrs per year. The backup floatswitch is tested after each water change, or once a week. Based on that testing cycle, we can anticipate a maximum unavailability of 3.5 day per year.

Those two numbers are converted to 0.009589 years/year and 0.000913242 years/year. The failure rate for a concurrent failure of the floatswitches is approximately 0.000008757 years/year, which corresponds to one failure every 114,194 years.

Lets assume now that you're not a normal person, and you're only home once every week, at which time you change water and test the backup switch. The MTTR for your primary floatswitch is 3.5 days, as is your backup. The unavailability of each is 0.009589 years/year, and so the concurrent failure rate is 0.000091949 years/year, or approximately one failure every 10875 years.


That's reliable enough for critical safety equipment at a nuclear power plant. Ask me how I know.

Did you read the link I sent? Melev's friend had two concurrent failures. He should have played the lottery.

I believe your numbers. It's highly unlikely. ATO float switches can fail. Mine failed after only 8 months. Maybe mine was just cheap. If you are gone for work for 8 hours of the day and the ATO fails in the morning, your RO/DI unit will have 8 hours to add RO/DI water into your tank. If you have a big tank, it is probably not too bad... with a 60gpd membrane, that's only like 20 gallons (actually, that's probably a lot even for a 200g system). But that's a lot for a nano. It's not a great idea but certainly not horrible either. It is possible and has happened even if the numbers say it is unlikely. It's better to have a non-working ATO for a few days as evaporation would be gradual and it would not kill anything in your tank. You'll still have clean carpets and no floods. AC3 has a water-on-the-floor monitor which can SMS you. You can add something like that as a third means. I think home security systems has water-on-the-floor monitoring services as well. I think insurance does not cover aquarium floodings either.

One other small draw back is the RO membrane does not produce pure water at first. If you have a dual TDS meter hooked up, you will see the RO membrane produces a higher TDS at first (not a ton, maybe 4 or 5 TDS higher than normal). This means your DI has to absorb it. If it is hooked up to an ATO, this means it will stop/start numerous times a day. It might make a significant dent on your DI. I am not anal about it but some guys are and have it rigged up to flush the first few cups of water prior to filling their ATO reservoir or salt mixing reservoir.

If you feel comfortable with it, go for it. For those reading this who do it without a second float switch, you're dying for trouble.

(Sorry for the off topic)

Back to the Auto-water-changer! Supposedly, there is someone who did it with an AC3 with the I/O breakout box. But it is iffy. I am definitely interested. I hate doing water changes. If you can add a conductivity monitor, you can make it mix your salt for you too. Like the dialyseas.

This needed emphasis for anyone skimming the thread.

I think we're in total agreement that running without the second float switch is just asking for trouble. It's also not surprising that there are occasionally double failures - for every 100,000 tanks with a similar setup, I'd expect one failure per year. Melev's friend sure should have played the lottery, assuming there's a law of conservation of luck. One of my favorite "laws", personally.

The TDS is a small issue, one I was not previously aware of. I've put a little thought into how to mitigate the issue. Given a reservoir with at least a week's worth of RO/DI, a ball valve inline with a float valve might be a better option. Once a week, open the ball valve and let the filter run until the reservoir is full. Once full, overnight, or whatever, just shut the ball valve off to prevent the cycling of the system. I'm sure an electrically actuated valve could be found as well, to put it on a timer. The expense of doing so...I don't think it counteracts the slight increase in DI resin costs. An extra 4-5TDS for a second or two is not a lot of solids for the DI to filter out....

The next step beyond this is setting up an automatic SW mixer like you've described, but I haven't looked that far ahead. Maybe once I've got a little more room, maybe a fish room....inwall setup.....ah....dreams.

I currently use an ATO reservoir that holds about a weeks worth of water. It has both a float valve and an in-line ball valve on the input, which is then connected to the RO/DI. Every week when it is getting low, I open the ball valve and let it fill up. It takes about 2 hours, so I usually set a kitchen timer to just over 2 hours to remind me the ball valve is open. The float valve eventually stops the RO/DI input, and soon after that I turn off the ball valve.

I also have a timer on my ATO pump/float switch system. Let's say it takes about 35 minutes/day of pump operation to make up for evaporation. I set the timer to be on for two 20-minute blocks of time during each day. This way if the valve gets stuck there is both a limited supply of water available, and a timer with a hard cutoff to stop the pump regardless of the float switch position. I have not had any incidents of over-filling the tank, except when I stupidly left the switch on the timer to "always on".

My two cents...
http://www.instructables.com/id/Creating-P...-a-INKJET-Prin/
This is probably the cheapest way to get a pro-look. My problem with soldering a board like this is the all the time you are soldering the board you are exposing the components to HIGH heat and that not a good thing. but given the fact that the components used in this project are pretty basic I wouldn't go overboard with inkjet unless you wanted to sell some...
but out side of that well done. I have done projects like this and soldering a board with a design like this is a pain. lol so Well done!
ps. I have no EE degree ether, just a novice electronics hacker

Yeah, I've seen a few boards etched this way by some fellow students back in school, and just wasn't pleased with the results. I SO WISH I still had access to my university IEEE club's PCB mill. I think I might just build myself one. Looking at our final year projects, you could tell who etched and who milled a board. It was pretty bad. I trust the solder traces more (and it's easier than sourcing/disposing the chemicals.). Certainly doable, but I haven't yet been made a fan of the DIY chemical etch. Haven't tried it myself, but given the results I've seen, I'm not sure I'm willing to put in that effort. I'd rather build a mill.


For an update, after moving, I ran for a few months without the AWC. That was hell. However, I was now able to do some things that swore I wouldn't do while integrating it into the stand. My ATO reservoir now refills automatically when empty, using a latching relay. Due to the risk, I went a little crazy. In order to have a flood, I'd need one of the following scenarios - failure of three separate float switches. failure of two solenoid valves. failure of three separate relays. short circuit of 120V hot to the solenoid outputs.

Just don't see it happening. Now, I don't even have to top up the tank - ever. It's SO AWESOME. I cannot express the difference between manual topups and never topping up. It's ridiculous. I now love having a SW tank again.

I also set up a 55 gallon barrel in the basement for mixing bulk saltwater. Now, a water change needs only to fill a bucket, plug in a powerhead, move a few hoses, push a button, and dump the wastewater. If only I weren't renting, oh man, I'd so have a pump right from the barrel and a sewer drain in the living room wall.

I'll post pics...later...

You could always use a service like this one for cheap PCB manufacturing. BatchPCB by Sparkfun.com They batch up requests to a fab house in China that they use for all their own boards. They have good rates with them if you are willing to wait a little while to get your board back.

Me personally, I'm working on a system the uses a PIC microchip to run ATO, AWC, Heater, and Fan. That covers a large portion of the automation I see needed and can be had MUCH cheaper than any commercial controllers.

Has their been any advances on this project? Great job by the way!

I have had good experiences with this place for relatively inexpensive pcb's
http://www.goldphoenixpcb.biz/special_price.php

One thing that I think I would consider adding as an additional safety factor would be a timer relay to limit the time that anything can fill (ATO for example). It does add to the cost a bit but they can be found for fairly cheap and have an incredibly low failure rate from my experience (I used them in the design of an auto-flush for nano-filtration systems).

Great project by the way, I always love to see this kind of thing in this community.

As a side note to anyone considering/mentioning using a pic or other micro controller to perform similar functions be sure to set up high level overriding service routines such as a watchdog timer to make sure you don't get stuck in a loop that could result in an easily avoidable failure. Redundancy is your friend

-Ian

Thanks Ian,

The project, for me at least, is complete. The only modifications I've made have been cosmetic, building the unit into the stand. I have a terminal strip to secure but that's it. Need to find the right screws...

Control panel is on the left.



The board is inside the box at the top, with the unsecured terminal strip. The unit switches a couple of receptacles in the panel below it.



I'll take a look at that link and see if I can put together a PCB group buy.

Major update!

Easy-to-use single-sided PCBs are now available through BatchPCB for about $15!!! Thanks to taylorjonl for pointing out that batchPCB can be used for distribution, even for us non-americans!

Order your V5.0 PCB here. Note - V5.1 has been released with a cheaper relay.

This release is updated to V5.0, which now includes an optional fifth relay for control of a RO/DI unit solenoid valve. Before connecting a RO/DI unit directly to your tank, be aware of the consequences of a failure and take whatever steps you deem necessary to mitigate that risk. If the ATO reservoir is integrated into the sump, then the same emergency shutoff circuit for the ATO will also shut down the solenoid valve, with redundancy on both the 12V and 120V circuits. You could also double-up on the solenoid valve for added safety.

An appropriate relay for operating small maxijet pumps can be found at digikey. The price isn't great, but it's available without a minimum order. Omron G5V-2-DC12. I'm basing it off the datasheet, but it looks good to me. The sanyou DSY2Y-S-212L is known to work, if you can find a source.

If you find a lower-cost similar relay that you'd like to use, the PCB design can be modified to suit - let me know!

Well, I just couldn't leave well enough alone. I was disappointed in the choice of relays available for the V5.0 PCB, and I realized that they only reason I had used that relay footprint was because they were the relays that I already had on hand. Well forget that. I went shopping for a new relay and the V5.1 PCB was born.

The new relay is slightly larger, which naturally necessitated a larger board. In addition, the pinout of the relays is different enough that I was not able to make it both single sided and compatible with a perforated board. So, this design is PCB-only, and cannot be duplicated on a perf board. This was late in the design, so I will probably release a V5.1.1 shortly that completely abandons the perf board in favor of a smaller, cheaper PCB.

So, the links:
V5.1 PCB
Relay
Diode

Total cost for PCB and components - $32.25

The following relays should also be compatible - just ensure that the coil is not polarized, or if it is, that it matches the circuit:
Relay 2
Relay 3
Relay 4

You can also use any IN4004 diode in standard axial (DO-41?) packaging