auto top off

[coralcruze]

I'm in the process of setting up my custom reef tank and need to set up an auto top off system. I'm planning to order the neptune controler to handle this and other systems in the tank. Additionally I plan on purchasing the vertex water filter with pump like this http://www.aquariumspecialty.com/catalog/product_info.php?products_id=3497

so the question I have is can anyone please explain how I should design the system so that I can control the water level correctly in the sump itself ahich sits below the tank. Unfortunatly I do not have the room for any reservoir tank so the top off water will be pumped directly to the sump itself. Is this a bad idea? I heard something about ro/di filtering better in bulk. when they filter in small bursts the water is of lesser quality? is this true? anyway, I plan on having some sort of float switch control the water level and perhaps a second one as a failsafe. additionally I need to control the pump on the water filter to turn on and off as the float switch indicates a low water condition. Please advise... preferably link me to some recomended items to complete the top off system of your choice. Thanks

 

[67Duster]

I would be very hesitant about connecting your RO to your system. I understand about not having the room for a reservoir, but without fully understanding the dynamics of what your attempting you might want to reconsider you options.

 

[jpruitt]

+1 on reservoir. I had my ro unit dump 40 gallons of ro into my system before i caught it. Limit the amount of water that the top off can allow to enter your aquarium just in case your top off malfunctions.

 

[coralcruze]

what about installing several failsafes? like the level-loc system on a timer offers 3 failsafes. I think that would sufice, don't you??? besides filling a reservoir also allows for another leak point and you also need float switches in that so am I missing something here?

 

[vangvace]

The other thing about having a reservoir is that your RO unit doesn't cycle on and off as often.

Oh and failsafe's fail. Just ask Murphy.

 

[coralcruze]

The other thing about having a reservoir is that your RO unit doesn't cycle on and off as often.

Oh and failsafe's fail. Just ask Murphy.

percisely my point as well... yes they do fail but isn't it true that you need float switches in your reservoir as well as your sump in the system you refer to? so that would mean that if you want to build in failsafes into the reservoir as well as the sump you will need twice of everything. that means twice the float swiches.

I also can't see if you set up say three failsafe devices how it could still fail. unless you never even look at the tank and do basic maintenence. For example what I plan on doing is having one float switch handle the water level. than set up another float switch as a backup to shut the system off via a solinoid/float combo. Than hook up the Ro/Di on a timer and if thats not enough I will also be using a controler to shut the system off in case of a detected leak. so thats 1 main water top off backed by 3 failsafes. If all 4 devices should still fail and Muphy gets his way than, heck it was meant to be and there is nothing futher I could have done,certainly adding a reservoir would not have done anything to prevent the disaster, right?

Again, I just want to be sure... isn't having float switches required to keep the level of a reservoir stable as well or am I missing something? is there an alternative failsafe device that I'm not aware of that others have used? Please advise.

 

[TeebD]

I wouldn't feel comfortable with 10 fail safes. I know it's not the most attractive, but you can always keep it outside the stand. You could build/buy a cabinet to match your stand. There's an acrylic builder on ebay that make ATO reservoirs that are as small as 4"x18"x18".

 

[OwenInAZ]

Why have a float switch in the reservoir? Fill it manually. Less room for automated failure. Maybe I'm just a KISS guy at heart, but I don't trust automation. When you don't think about something, you don't think about maintaining it. Then your failsafes fail. Precisely why I would never ever ever ever hook up an RODI unit straight into a tank. That's a recipe for a freshwater tank and living room full of dilute saltwater.

 

[collie man]

check out autotopoff.com they have a number of different options i am sure you can find one to work with. plus they offer a system with a solenoid that can be hooked up to an RO unit that has a 2 year warranty.

 

[collie man]

also i have heard that if you directly connect the RO unit to the ATO then it can extinguish the filters much more rapidly. those systems are meant to produce a large amount at a time. you would see more waste water than fresh RO going into your system.

 

[gjk]

Some RO/DI systems may operate better on long usage periods than short, but not ALL. My system is tested by two seperate and two different makes of TDS meters and it has run only short periods of time for two years. Four twenty minute periods per day to fulfill the requirements of a 400 gallon total system.

 

[coralcruze]

most posts have been great. some I think over the top... like having 10 failsafes and still not sleeping good at night, seriously... why even fill up a tank with water than? silicone seams have been known to spring a leak too Some posts have been more opinionated... still appreciated none-the-less.

for example many here have suggested doing a seperate reservoir tank. Thanks collie man and GJK... I have heard this as well but I think GJK is right that this is product specific and should be tested for. However, this is the ONLY reason I see for doing a separate reservoir tank... NOT for making a leak proof system. Again this is the point that I'm trying to drive home and understand for myself that no one seems to be responding to. People have suggested that setting up float switches in a sump can lead to overflow and failures... but then you set up failsafes to your system to guard against this, right? but here is the POINT>>> even if you have a reservoir you STILL need float switches, solenoids and pumps to shut off the system and pump water from one place (reservoir) to another (sump). so unless someone can explain to me how a reservoir system can eliminate ANY/ALL possability of an overflow, i just dont see the need for a reservoir system unless you absolutly need it to save the life of your filters.

The floor is open to anyone wanting to help a guy understand better. thanks

 

[collie man]

Honestly at this point the only factor is Murphy's Law. In my case i have a 5 gallon jug topping off kalkwasser through a aqualifter pump the the autotopoff.com ATO. in no way is my set up fail safe or the best but it depends on your situation. in your case having another body of water can allow for a greater risk but you must accept the risk of not having it. i say you should just do what fits your options. if you dont want to stress your ro unit just use a large tank to hold the water.

 

[67Duster]

How much is total evaporation per day?

 

[ibnozn]

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

 

[coralcruze]

How much is total evaporation per day?

The main tank will contain about 95 gallons and the sump about an additional 30 gallons...

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

Great article and I thank you for sharing it with me... I took the time to read it and my suspitions are confirmed by the author as he suggests to do a MANUAL refil of TO water only and not rely on ANY automatic system. The author states that an RO/DI hooked up directly to a sump could fail the same as an ro/di hooked up to a reservoir and then the sump. Its no different than plumbing your RO to your sump with the exception that you effectively added a faulsafe by implementing a float in the reservoir to maitain water level there and than another float to maintain levels in your sump. ANY automated water T.O system can fail and we've all agreed on that and I suppose even with failsafes they can still fail but at a lesser chance with more failsafes implemented.

The article states "If someone wanted to hook up his RO/DI unit directly to a reservoir like this, he could do so easily with a float valve. However, what I recommend is that once the reservoir is full, he should turn off the RO unit until it is time to refill the reservoir." again this statement confirms the authors idea of never automating any type of water top off. But what if you want automation? is it possible and at the same time lessen risk substantially? makes no differance wether you have a reservoir or not, if you hook up a float valve and feed the water from the reservoir automatically you run the same risks of failures as feeding water directly into your sump. so the question still remains... full automation of water top off or not???

I think the answer really comes down to personal gut and how well you can sleep at night trusting your systems failsafes are in place and will do what they are supposed to do. So here is what I think will make me comfortable...

I will guess that my system will evaporate between 1-2 gallons/day so lets say about 10-15 gallons a week. I will hook up the ro/di directly to my sump via a plastic ball style float valve to maintain main water level in the sump.

FAILSAFE#1: 1" higher than the shut off point I will implement an electrical float SWITCH and reverse the float on that switch so that if the contact in the bottom position breaks so too the power to the ro/di unit.

FAILSAFE#2: Hook up a timer to the ro/di so the unit will only allow to recieve power for the length of time needed to fill up the 1-2 gallons lost that day. this means since I will be getting 100 GPD of production water the power to the unit will be set to deliver only a total of 1/2 hour in any given 24 hour period. In other words since the RO unit produces 100 GPD or 4.66 gph and I only need 2 total gallons I would get 2.33 gallons in any given 1/2 hour period of time.

FAILSAFE #3 hook up a leak detector to the Neptune apex controller which will automatically sence any moisture between two contact points and will shut off power to the RO unit.

FAILSAFE #4: hook up a PH probe to the main sump area (which is a good idea when running a Co2 reactor which I will be doing) as well as a second PH probe to the Co2 reactor. this will ensure that any ph drop to a set level will also cut power supply to thr RO unit.

FAILSAFE #5: set the Neptune Apex controler to alert me of any alarm set via cell text and/or e-mail.

Beyond this point Murphy will have his way.

Please feel free to give me any further suggestions or advice on other failsafes to implement and if at all possible I will do it. Also give me your opinions and suggestions to the above... let me know what you think.

I will add just one other interesting point. My last tank had no automatic TO system. Terefore I was really bad about adding TO water as I would add it all at once about 5-7 gallons once a week. This practice most definatly stressed out my corals. secondly, looking back the only times that I ever flooded the floor (thank G-d) was two times and both times was when I was filling up (guess what?) a water reservoir. which at the time was a 35 gallon brute garbage container. I knew that once every 3 weeks or so I needed to fill up the container with fresh RO and it took about 5-6 hours. even setting up an alarm clock I still managed to flood the floor. Not bad but it was always a reminder that the worst failsafe in the world was ME

 

[coralcruze]

I don't want to beat this topic to death but I forgot to mention...

FAILSAFE #6: The Neptune controller will allert me if the salinity changes beyond a set point. So if the salinity drops due to fresh water being dumped into the system it will shut the RO/DI off and allert me of the event. The controller also has a battery backup in case of a power outage.

I thought to mention the above as I was talking to a reef buddy and he mentioned something no one did here. That is if something should leak OUTSIDE the sump, like my Co2 chamber for example than the RO/DI unit will continually deliver RO to the sump. This is assuming all other failsafes have failed at the same time so Murphy will need to play his hand in that one too.