How to plumb an automatic water change system!

[Bemmer]

Anthony,
I read in your propagation book about setting up an automatic water changing system. I am finalizing the design of my sump and want to include the water changing drain. The sump will be in two sections. The skimmer section 18"L x 30"W, then three baffles to the return section (24"L x 30W). I am planning on having the baffles at 9" high so the water level in the skimmer will be 9" high, then pour over the baffles to the return. The wate level in the return section will be about 7-8" high using a top-off device. When doing water changes I am planning on changing approx 45 gallons (about 10% of my total water volume). Can you give me some insight into where I should drill the water changing drain (skimmer or return section) and how high up in the tank? The height of the tank is 17". I am guessing there is a math formula or something like that to calculate the distance up the tank to determine the amount of gallons that will be removed...but I was not very good in math. :rolleyes:

Thanks for your help.

[Randall_James]

L * W * H /231 will yield your numbers :D (all in inches and 231 is the number of cu inches in a gallon of liquid)

Each inch of depth in the skimmer section will be 2.3 gallons
45 gallons will require 19" of water depth

Each inch of depth in return section will be 3.1 gallons
45 gallons will require 14" of water depth

So neither will provide adequate volume for a single operation.

I would look at how the detritus settles in the current sump. I would then put my fitting in bottom or as close to as I could the "dirtiest" section. (this is so I could stir up the loose detritus and let it flow out with the effluent.

With this method you will need to do multiple water changes to achieve your goal of 45 gallons.
This would also have only minimal effect on the actual percentage of 10% of your total volume. Otherwise you will have to add drains to the entire sump to drain it.

I have in fact put 1/2 holes low in my sump baffles to allow my single drain to empty the sump with very little effect on how the baffles operate.

[Anthony Calfo]

its rather easy...

you have your evap top off device set wherever you want/need the water level in the sump proper (last and lowest sump compartment).

Set the evap float swithc so that it is very finely sensitive (frequently doses small amounts) but is fed by a VERY slow pump (extreme here... 1/4" airline or icemaker line style stream... mere gallons per day rate of speed)

Then you have your auto water change overflow hole drilled either a) slightly higher than the evap-top off level by say... not more than 1/2", or b) drilled at the same level as the evap top off level but with an adjustable elbow that allows you to change the overflow level (my recommendation)

Thus... with the overflow hole/height being less than half inch over the evap-top off level... you can auto (timer) pump new and slightly cooler seawater (no more than 2F cooler) into the display tank which will displace warmer aged water to the sump and out/over the overflow hole! Automatic water change.

And from a waiting/reserve vat (barrel/buckets... whatever) of new seawater... your auto-timer (digital from Radio Shack, for example) can be programmed to do daily, weekly, etc water changes as you wish! Sit down once and calculate how many gallons per day/week you want to change... and how many gallons per minute the pump you use for make-up will push. Simple math.

Then... after just a little bit of evaporation (after each auto water change that bullrushes the sump and overflow hole) the sump proper resumes normal evap top off operations in wait for next auto-water change, which will again slightly overun the evap top off level by just 1/2" or less

Now you may be wondering... isn't that slight overun (<1/2" over float switch level) adding a little bit of salt to the tank and increasing my salinity slowly over time? Well... theoretically yes. But if you will worry about that, then are you also calculating the similarly lost salt due to salt creep? Heehe... point is, either way its a small amount of salt. And either way, with or without salt added or exported (creep)... you still need to make slight corrections over time to the tank by checking salinity. If its a big deal (large sump)... you can compensate then by making your new seawater used for auto-water changes just slightly less saline.

And voila... automatic water changes!

[Randall_James]

so this could be a constant water change as much as an automatic.. sure wish I had a floor drain..............

[Anthony Calfo]

Yes... and the floor drain is kinda crucial here ;)

[Bemmer]

Now...that's watta'm talk'n 'bout. All the calculations were making me a little crazy. :p

Sounds like a plan. However, i do not have a floor drain. The water change drain (WCD) will flow into a wall drain (sink drain on other side of wall), I can plumb into that drain pretty easily. It will be lower than the WCD hole. I can make it as low as I need to for the old saltwater to be gravity feed into the wall drain. I was thinking about plumbing in at 2" below the WCD hole. The water will be traveling a distance of about 36" from the WCD hole to the wall drain. How does that sound.

Hoping to order the sump today!!!!!

[Bemmer]

Anthony,
I forgot to ask you, how big should the bulkhead hole be for the WCD? 1 1/2" or 2"

[Anthony Calfo]

your wall drain is fine, my friend... of course, because it is lower than the overflow on the sump.

As for the overflow bulkhead size... only you can determine that. I have no idea what size pump you are using to do the auto-fill... how long the timer will be set for running it... how much water will run through, etc.

That said, bigger is better (2" bulkhead perhaps).

And moreover... chances are the percent of the water changed and speed at which you will be pumping remote new seawater will be slow enough that the single bulkhead of any reasonable size will be enough. But you must above all do your own math here to be sure.

[Bemmer]

Thanks for all your advice. :)

[Bemmer]

Anthony,
I would like to show you my proposed sump drawings, with the auto water change bulkhead hole and the other return pump holes, as well as an overflow hole. I want to be sure I am not missing anything before I order my acrylic sump. Thanks for your comments.

http://reefcentral.com/gallery/data/500/37817Sump_Drawing_Front_View.JPG


The left side view of the sump shows the overflow drain in the top back corner.

http://reefcentral.com/gallery/data/500/37817Sump_Drawing_Left_Side_View.JPG

[Anthony Calfo]

your diagram is fine by the measurements, love.

For the purpose of mentioning to others reading this/archiving... I should point out that the overflow hole as drawn is too high as it appears to be above the skimmer partition wall (which would flood the sump level over it and bypass the baffles).

but indeed as you have stated in the measurements... the skimmer wall is 1" higher than the overflow hole. And it will work fine. Quite fine actually... you are getting almost maximum sump-proper volume in this case.

And you will need to install the evap top off (float) device at or slightly below the bottom of the overflow bulkhead drain.

The extra plugged hole in the sump is a great idea too. Be sure to plug from both sides (threaded) so that you can add plumbing as needed in the future without having to drain the sump.

Latsly... if the bottom image is showing the display tank with a single 2" bulkhead drilled for the overflow... I am worried that will not handle enough water through the sump. In fact... I am sure of it. I don't see you even getting a decent (quiet) 1000pgh through it from the sump loop

[Bemmer]

Originally posted by Anthony Calfo

For the purpose of mentioning to others reading this/archiving... I should point out that the overflow hole as drawn is too high as it appears to be above the skimmer partition wall (which would flood the sump level over it and bypass the baffles).

but indeed as you have stated in the measurements... the skimmer wall is 1" higher than the overflow hole. And it will work fine. Quite fine actually... you are getting almost maximum sump-proper volume in this case.
[QUOTE]

Anthony,
Sorry, I read this part three times but I was a little confused by the two statements. :rolleyes: Your first statement says that the water changing drain (WCD) overflow is too high in comparison to the skimmer partition wall. Is the skimmer partition wall the baffles? That is how I think you mean? Then are you correcting yourself in the second statement, when stating the skimmer wall is higher than the WCD overflow? Maybe it is too late or too early for me to understand. ;)

[QUOTE]Originally posted by Anthony Calfo

Latsly... if the bottom image is showing the display tank with a single 2" bulkhead drilled for the overflow... I am worried that will not handle enough water through the sump. In fact... I am sure of it. I don't see you even getting a decent (quiet) 1000pgh through it from the sump loop

Anthony,
I should have been clearer on the labeling for the bottom image. That is actually still a drawing of the sump, not the main tank. I realized after I posted the drawing that I labeled it "tank". The 2" drain at the top of the sump tank on the left side is...well, literally an "overflow" drain specifically to avoid flooding the sump. If for some reason, the check valve from the return pipe were to get stuck open or one of the drilled hole below the 1" seaswirl outlet (my return to tank is through two 1" seaswirls) got blocked, I have a third safety feature. This "overflow" drain will be plumbed directly into the same wall drain as the WCD overflow hole. Hopefully, it will never get used. :D

The main tank has two 1 1/2" bulkheads at the top back corner of the tank. There is no overflow setup other than the two bulkheads. These bulkheads will drain directly into the skimmer section. One will drain into the ER CS12-2RC skimmer and the other into a Marc Levenson (Melev) bubble tower.


Now, to drain a specific amount of water ie 45 gallons or 10% of my total water volume, do I just add the 45 gallons of NSW at the lower temperature, while the sump drains equal amount out of the WCD overflow hole? Is that how the new versus old saltwater amounts are distributed. Cause, I was getting some confusing info on another post about putting the WCD overflow hole under water about half way between the bottom of the sump and the water level. The only way that makes sense is if I have a shut off/on valve coming off of the WCD overflow drain. Right? Sorry, if that sounds confusing. :confused:

[Randall_James]

I think he was just clarifying that the measurments although correct, the image itself is not. Even though correctly labled, it is just "drawn" in the wrong spot.

The hole for your wcd is going to be at or just above your topoff switch if I read his meaning correctly. As you start the water change it will in fact "overfill" the sump slightly. This "overfill" will go out your drain hole. As this occurs, the new slightly cooler water will sink to the bottom of the sump to be returned to display.

[Anthony Calfo]

Randall James has it almost exactly correct and quite clear/concise in answer to your post, Rebecca. Much thanks James!

The only clarification here is that new and slightly cooler seawater is sent to the main display, not the sump. And with mere partial water changes (10, 20% etc) it will sink in the greater volume of the main display and displace aged older water to the sump which will, as Randall states, "overfill". Indeed... overflow the drain as intended. And that drain is to be at or only slightly higher than the float switch that maintains the sump level with evaporation top off as described above.

As far as the chap that said putting the overflow hole halfway between the sump floor and running water level/surface... there is some kinf of misunderstanding or mistake. It's incorrect or not practical if even workable.

[Bemmer]

Excellent!!!! yeah!!!

I will make the drawing change to the WCD hole and submit the sump drawings to the tank manufacturer...Now, if only i could decide on a the sump tank manufacturer. :rolleyes:

[Randall_James]

http://www.melevsreef.com/

[Bemmer]

Randall,
Thanks for your input on this thread. I really appreciate you helping me understand what Anthony was trying to convey. It was 4 a.m. when I read his thread earlier this morning. :p

If only I could get my sump made by Marc Levenson. Marc has helped me out a bunch on the boards as I am putting together my big in-wall tank. Unfortunately, my sump is too big and he lives too far away. It is cost prohibitive for me and shipping issues on his end prevent us from doing business together. He is a terrific guy and I would like nothing more than to purchase my sump from him. :) Thanks for the suggestion.

[Gundo5000]

Thats awesome!

[robsmith32]

Ok, since mine is similar, and he's finished, i am doing a sump, unfortunately not acrylic. And was going to do a ~5% change twice a week, by putting a bulkhead with valve @ 2" higher than the pump bulkhead, and have a float valve for top off at 12" higher than bottom of that bulkhead making it exactly 10 gallons.
Then in rubbermaid mixing container next to it, refill it once it drains.
Is what you're saying above saying this way is not workable?

[fishguy0226]

maybe I missed something in the design, what is to stop this from overflowing with power outages and then on restart running a pump dry? Sorry if this has already been adressed somewher.

[Anthony Calfo]

good point... and without making the setup exceedingly complex, the really short answer is to have a solenoid on the overflow drain that fails closed.

Thus... when power is on, the WCD hardware (timers, pumps, etc) can operate as intended and the overflow drain will be open (uninterupted power keeps the solenoid on/open).

But if the power goes out... the solenoid kicks off (shuts/closes) and the sump will not lose excess water that was not part of a scheduled WC overflow.

Clear or no?

[fishguy0226]

crystal, thanks, I am just in the process of trying to figure out a new basement sump syst. and am getting ideas. I'm still having trouble with the head pressure issue though and the calc. here I don't believe will work.
thanks

[Kathy55g]

How's this for an idea:
so you have a running sump that keeps the level of water, perhaps, about 1/3 the height of the sump. When the return pump is off, the water level is 2/3 the height of the sump.

How about if you plumb a drain bulkhead just above the 2/3 height of the sump. If you always turn the return pump off before you do a water change, then the water you add to the display drains to the sump and "dirty" surface water drains to the floor drain. You just have to remember to turn the pump back on when the water change is done. Kind of semi-automatic. I am thinking of doing this, so I would like to know what you think of it.

Thanks,
kathy

[Randall_James]

Originally posted by kmleah
How's this for an idea:
so you have a running sump that keeps the level of water, perhaps, about 1/3 the height of the sump. When the return pump is off, the water level is 2/3 the height of the sump.

How about if you plumb a drain bulkhead just above the 2/3 height of the sump. If you always turn the return pump off before you do a water change, then the water you add to the display drains to the sump and "dirty" surface water drains to the floor drain. You just have to remember to turn the pump back on when the water change is done. Kind of semi-automatic. I am thinking of doing this, so I would like to know what you think of it.

Thanks,
kathy You will have something of a volume issue. As if designed correctly, the sump should only pickup a minor amount of water when everything is shut down. Perhaps a few gallons at best?

Now I suppose if you set this up to kill the pumps every day that the repeated smaller changes would work out well. Only problem there is that your topoff water and your water change water may run into balance issues. 1 gallon of topoff and 2 to 5 gallons water change water a day.

Now you need a timer to kill the topoff and bring on water change water during the same time interval. Starts to sound like a lot of "possible" failure paths..

[Kathy55g]

But top-off float valve is at the 1/3 sump level in the example above. As you kill the return pump the water volume in the sump goes up past the 1/3 sump level, closing off the top-off valve. The water goes up to the 2/3 sump level. Then you start adding water 2 degrees colder to the display tank. The display tank overflows to the sump which raises the water level higher than the 2/3 sump level and down the draining bulkhead which is just above the 2/3 level.

I am making these levels up to give an easy number picture. The real levels will be different, as you say only a couple of gallons difference.

Probably I would not do this every day, but rather weekly.

[Randall_James]

Then all you did was move the drain hole up from the original posistion, I do not see any advantage to the change. The nice part about the lower drain hole is that you technically have a full time water change going on (small amounts a lot of times) and from what I have read, this is more desireable than larger infrequent changes.
I change 34g of my 100G system weekly and 20 per week on my 24g nano. My 125 gets about 15 (fo), I wish I had a floor drain :D

[Anthony Calfo]

kmleah... FWIW, its sounds to me like the long way around the barn, at best, and frankly... a host of problems beyond the challenge to plumb more likely.

I think you might be missing our point here... or we are missing yours.

[Biotoper]

Hi Anthony,

Thanks for coming out to Boston last week - it was a great talk, although I had to run out before the end. :(

Anyway, this sounds like a good system, but I don't see how it can be run truly constantly. Not sure how frequently you could have the pump go, but too frequently or in too small volumes each time and the auto-top-off would not function very well, resulting in increasing salinity. Correct?

I was throwing around an idea of potentially doing something similar to pH monitor-based CO2 injection system, by hooking up a dosing pump to a digital salinity meter (I'm no electronics expert, so have no real idea how to do this). To maintain salinity at the target leve, the pump would dose water, but it would be hooked up to brackish water rather than RO/DI, so it would need to overcompensate the evaporation to maintain salinity, resulting in excess water out the overflow drain.

Ryan

[Anthony Calfo]

thanks kindly Ryan... Boston was a truly wonderful time/club. :)

As per your salinity concerns... 'twas answered/addressed earlier in the thread:

"Now you may be wondering... isn't that slight overun (<1/2" over float switch level) adding a little bit of salt to the tank and increasing my salinity slowly over time? Well... theoretically yes. But if you will worry about that, then are you also calculating the similarly lost salt due to salt creep? Heehe... point is, either way its a small amount of salt. And either way, with or without salt added or exported (creep)... you still need to make slight corrections over time to the tank by checking salinity. If its a big deal (large sump)... you can compensate then by making your new seawater used for auto-water changes just slightly less saline."

Indeed... without a WCD, your salinity can/will still change and stray over time through various habits (kalk dosing, salt creep, misadjustments on make up water, etc). We are always and often using our hydrometers/refractometers for various things. So for the huge time savings and benefits of a WCD... the occasional check and tweak of salinity is a small matter.

[Biotoper]

I caught your point that continual evaporation of that extra 1/2" that's not topped off will drift salinity up (not accounting for other salt export from creep, skimmer, spray, etc).

My point was just that the surge from the wc has to be infrequent enough so that the 1/2"+ evaporates and you get some FW top-off. In my 20g nano, daily evap is ~1/4" - so with daily SW inputs to a 1/2" high drain, the auto-top-off would never trigger! In other words, this system probably has some minimum tank size limit to work.

In fact, I guess I could get rid of FW top-off and just use your WCD but with brackish water instead of SW. The equation is something like:
salinity of WCD water = target salinity * (1 - e/(e+x)); where e equals estimated %evaporation per day and x equals the target %water overflow per day.

Having a salinity meter-controlled system would be much better, but I don't know if it's possible (and much more expensive for equipment)

Ryan

[Anthony Calfo]

More often... folks with WCD systems simply make their WC water slightly less saline to (over)compensate for the diff... unlike a meter (hobby grade salinity meters have been historically unreliable/inaccurate), the adjusted SG water is at least reliable (cannot fail) and is simple enough.

[Kathy55g]

However, if your WC water is less saline, it will be less dense than the tank water, causing it to rise in the water column, so the water you skim off down the drain will partially be new water, defeating the purpose of the automatic water change system.

It would be better to manuallly adjust the salinity back where you want it by removing some saltwater and replacing it with fresh on occasion.

[Kathy55g]

Originally posted by Anthony Calfo
kmleah... FWIW, its sounds to me like the long way around the barn, at best, and frankly... a host of problems beyond the challenge to plumb more likely.

I think you might be missing our point here... or we are missing yours.

Perhaps I did miss the point. Your point was that this is for Automatic water changes, not simply easy ones (my point). Conceivably, one could put both the return pump and the water change pump on a DAY/Night timer to turn off the return pump when the water change pump came on.

[Kathy55g]

Originally posted by Randall_James
Then all you did was move the drain hole up from the original posistion, I do not see any advantage to the change. The nice part about the lower drain hole is that you technically have a full time water change going on (small amounts a lot of times) and from what I have read, this is more desireable than larger infrequent changes.
I change 34g of my 100G system weekly and 20 per week on my 24g nano. My 125 gets about 15 (fo), I wish I had a floor drain :D

Advantages: If power outage occurs, pipes drain to sump and water is contained there because the drain hole is just above this level. When power comes back on, typically when I am not around, the water will still be there, not down the drain. Return pump will resume its job, I hope.

Disadvantage: I must be there to conduct water changes, unless I put both the return pump and the WC pump on day/night timers to turn one off while the other is running.

I'm going to look into the solenoid thing, though.

[Anthony Calfo]

"Disadvantage: I must be there to conduct water changes"

... you see, that's the not-so-automatic part of your automatic water change system ;)

[Kathy55g]

Gotcha! :)

[Randall_James]

Originally posted by kmleah
Gotcha! :)

Not sure I understand what you mean "gotcha"?,

this is an automatic water change system, not a "Gotta be there" water change. Also the "solenoid" program creates another point of failure and then 2 more timers? If you are going that route, there is an even easier way to do this. But again defeats the purpose of this job

If you have power failure, you lose 2 to 4 gallons, the system can make that up pretty easily when it has a 35G reservoir behind the fill system.

The idea is keep it simple, reliable and fail safe. Otherwise you just may as well break out your python and drain the thing by hand.

Thread title "How to plumb an automatic water change system!"

[Biotoper]

Originally posted by kmleah
However, if your WC water is less saline, it will be less dense than the tank water, causing it to rise in the water column, so the water you skim off down the drain will partially be new water, defeating the purpose of the automatic water change system.

It would be better to manuallly adjust the salinity back where you want it by removing some saltwater and replacing it with fresh on occasion.

If the less-saline new water was dripping into the main tank at a relatively slow rate (dependent on the difference in salinity of the new water compared to the tank, and the estimated evaporation rate for the system), and the overflow drain was in the sump below, the mixing would be sufficient to still make the water change system pretty efficient. I'm talking a slow drip system, like those used for FW flow-thru racks, rather than the on/off pump system you and Anthony are discussing.

One idea would be to incorporate a rotifer tank in this system. Something like:
a. At the top, a 10g tank with new water at ~1.019 and frozen phyto added manually each day, dripping into -
b. 2 2g rotifer bottles (for redundancy in case one crashes), overflowing into -
c. the main tank at ~1.024, overflowing into -
d. sump, with overflow drain to sink, or enough extra height so you can manually remove water every few days when you add new SW to the 10g on top.
You've got auto-water change, -top off and -rotifer feeding all in one. You'd need to check salinity regularly, and keep a careful eye on the roti pops as you would with any live culture.

Ryan

[Randall_James]

Originally posted by Biotoper
If the less-saline new water was dripping into the main tank at a relatively slow rate (dependent on the difference in salinity of the new water compared to the tank, and the estimated evaporation rate for the system), and the overflow drain was in the sump below, the mixing would be sufficient to still make the water change system pretty efficient. As I recall, Anthony did a chart about this. Am not quoting the chart but it essentially stated
100 Gallon tank
100 Gallon water change = 100% water change
100 one gallon water changes = 90% water change

I think, I can not find the chart now but hey that is plenty effcient if I am even close to his chart.

ON the roti feeder, couldnt you just keep this system seperate from the auto top off? What I mean is that the auto topoff is pretty important (even more so when on vacation, and the only reason I had to make one btw) I like the idea but again the KISS principal is sort of the goal here? Neat concept however.. As long as it would not contribute to the failure rate of the system (key word was "careful eye on roti pops")

[Biotoper]

Originally posted by Randall_James
As I recall, Anthony did a chart about this. Am not quoting the chart but it essentially stated
100 Gallon tank
100 Gallon water change = 100% water change
100 one gallon water changes = 90% water change

I think, I can not find the chart now but hey that is plenty effcient if I am even close to his chart.

ON the roti feeder, couldnt you just keep this system seperate from the auto top off? What I mean is that the auto topoff is pretty important (even more so when on vacation, and the only reason I had to make one btw) I like the idea but again the KISS principal is sort of the goal here? Neat concept however.. As long as it would not contribute to the failure rate of the system (key word was "careful eye on roti pops")

The problem is that the continuous drip from the top would result in the sump level slowly rising, so there's no where to set-up a water-level-controlled auto-top off. That's why I thought of using less saline water in the drip to (try to) keep salinity constant given an estimated level of evaporation (salt creep/spray is very low in my current setup, but that would be another more-difficult-to-measure issue).

In terms of failures, there's no electricity involved at all in the water change/top-off system (just the sump pump). As long as the sump can hold the main tank overflow w/ pump off + ~14g, if I drop dead it still won't overflow.

The roti culture will be tricky - I haven't read about anyone having it auto-drip to the main tank, so that might be a problem. If it crashed, at worse I'd be dripping frozen phyto water straight to the main, which isn't bad (I'm staying away from growing live phyto in the top 10g - I think frozen concentrate will be easier and better).

On %water change, I think if you keep total % the same (e.g. 1/24% every hour, 1% a day, or 7% a week) and assume removal or excretion of X molecule is constant (e.g. not concentration-dependent), the concentration of X molecule in the water asymptotes at the same concentration. There's an article in Advanced Aquarist that shows graphs for this.

Ryan

[Randall_James]

kmleah
The salinity of the water for the change system is offset by the temperature. The cooler water is going to drop to the bottom of the tank. By the time it warms it will have diluted enough with tank water to offset the salinity issue. ( I had to try this first btw in the kitchen, my better half thinks I have stepped of the deep end)

Biotoper
The problem is that the continuous drip from the top would result in the sump level slowly rising
That was the first thing addressed by the "drain" at or just above the auto top off level in the schematic.

[Kathy55g]

Originally posted by Randall_James
Not sure I understand what you mean "gotcha"?,

sorry, I meant that now I understand the point.



this is an automatic water change system, not a "Gotta be there" water change. Also the "solenoid" program creates another point of failure and then 2 more timers? If you are going that route, there is an even easier way to do this. But again defeats the purpose of this job

If you have power failure, you lose 2 to 4 gallons, the system can make that up pretty easily when it has a 35G reservoir behind the fill system.


Unfortunately unless the power failure happened to coincide with a water change time on the automatic system, the sump, lacking a solenoid on the drain 1/2 inch above top off level would refill with RO/DI topoff water.:(
I am beginning to think that keeping it simple, means at least turning a couple of valves by hand while I watch.

[Kathy55g]

Originally posted by Randall_James
kmleah
The salinity of the water for the change system is offset by the temperature. The cooler water is going to drop to the bottom of the tank. By the time it warms it will have diluted enough with tank water to offset the salinity issue. ( I had to try this first btw in the kitchen, my better half thinks I have stepped of the deep end)

That was the first thing addressed by the "drain" at or just above the auto top off level in the schematic.

The reason that warm water rises above cooler water is that warm water is less dense than cool water.

The reason that more concentrated salt solutions sink below less concentrated salt solutions is that salt adds density to the water. More salt, more dense. Less salt, less dense.

When you tried this for yourself in the kitchen (my husband would sympathize with your wife) did you try only the temperature test, only the salinity test, or did you test both salinity and temperature differences to see how they offset each other? It would be interesting to see how both lowering the salinity and the temperature would affect water levels. How much temperature offsets how much salinity? How did you distinguish one "kind" of water from another to determine the result?

Thanks,
Kathy

[Kathy55g]

Originally posted by Biotoper

One idea would be to incorporate a rotifer tank in this system. Something like:
a. At the top, a 10g tank with new water at ~1.019 and frozen phyto added manually each day, dripping into -
b. 2 2g rotifer bottles (for redundancy in case one crashes), overflowing into -
c. the main tank at ~1.024, overflowing into -
d. sump, with overflow drain to sink, or enough extra height so you can manually remove water every few days when you add new SW to the 10g on top.
You've got auto-water change, -top off and -rotifer feeding all in one. You'd need to check salinity regularly, and keep a careful eye on the roti pops as you would with any live culture.

Ryan

Interesting idea Ryan; how did you know I am starting to raise clownfishes and the system I am noodling over is a grow-out for them?

The problem I see is the rotifer tanks overflowing into the main tanks. Rotifers pollute their water beyond belief, and were I to simply let them overflow with their water into my larval tank, the larvae would die very quickly from the incredible pollution. I prefer to spend the time to filter and enrich my rots before I feed them to my babies. I would not overflow them with their nasty water into my display tank either. Yours is not a bad idea. It is an idea that we have all noodled over at one time or another, but there is no getting around the funky water issue, so I'll just keep doing it the old fashioned way.

I keep my rots in water jugs by the basement sink so I can filter off some rots, do water changes on the jugs, feed and treat for ammonia, and wash up the equipment very conveniently.

Cheers,
Kathy

[Randall_James]

Kathy
I tested as it would be in the tank. I set the 81 degree water at tank salinity 1.026 and the 78 degree water at 1.024 (water had some dye in it so I could see it) I dripped the water in from 1" above water level and it sank to the bottom It did appear to loose some of the content on the way down but the vast majority of the new cooler, lower salinity water did make the bottom of the tank.

I dripped 1 gallon of the cooler less saline water into a 10 gallon tank full of water fresh out of my 75G sps tank. (water change day) so it only cost me 1 gallon of good water :D, I am as cheap as I am curious :lol:

I have read your thread and your endeavors with the fish fry in the breeders forum I think it was.

I am beginning to think that keeping it simple, means at least turning a couple of valves by hand while I watch
Well making it "Fail safe" means watching it. I personally like the "fail safe" method. I would open my valve, start my pump on a 1 hour timer or what ever it needed and walk away. Leave a flag on the fishes food bin that the valve is open so that I do close it sometime in the next 24 hours. I am a tough sell on anything being fully automated. I have tried them all, AC controller 2 was a nightmare and I am greatful I did not loose my tank to it.

[Kathy55g]

Wow, great experiment, and thank you for telling us the details. It is good to know that that combination works.

I am also as cheap as I am curious.

I hope you enjoyed the third attempt thread. The fish are nice and healthy and starting to be ready to leave home. I don't know whether to be happy or sad.

The fail safe method you describe is probably my best bet: semi automated for the absent minded.

:)
thanks,
Kathy

[Biotoper]

Originally posted by kmleah
The problem I see is the rotifer tanks overflowing into the main tanks. Rotifers pollute their water beyond belief, and were I to simply let them overflow with their water into my larval tank, the larvae would die very quickly from the incredible pollution. I prefer to spend the time to filter and enrich my rots before I feed them to my babies. I would not overflow them with their nasty water into my display tank either. Yours is not a bad idea. It is an idea that we have all noodled over at one time or another, but there is no getting around the funky water issue, so I'll just keep doing it the old fashioned way.

I keep my rots in water jugs by the basement sink so I can filter off some rots, do water changes on the jugs, feed and treat for ammonia, and wash up the equipment very conveniently.

Cheers,
Kathy

Thanks for the advice. I didn't know you were breeding ;) - I have some filter-feeders in my tank and planning on getting an Ocellaris pair, so I was thinking roti culture.

One question I have is whether the roti jug water would get that funky if ~0.5g of new SW was trickling through it per day? Of course, a population of X density would create the same amount of waste w/ or w/o flow-thru, but it would be less concentrated in waste than if you had added water straight from your isolated roti jug. But I would really like to find someone who's tried a flow-thru roti system (success or failure).

Cheers, Ryan

[Randall_James]

Originally posted by kmleah
Wow, great experiment, and thank you for telling us the details. It is good to know that that combination works.

I am also as cheap as I am curious.

I hope you enjoyed the third attempt thread. The fish are nice and healthy and starting to be ready to leave home. I don't know whether to be happy or sad.

The fail safe method you describe is probably my best bet: semi automated for the absent minded.

:)
thanks,
Kathy Watching your thread very carefully actually, I have mated yellow strip maroons that I want to breed but I do not know where to start, you are thankfully helping me lay out some idea of how to go about this :D thx

[Kathy55g]

You are most welcome Randall and Ryan. Both of you may want to check out Joyce Wilkerson's Clownfishes book. And if you are going to culture phyto (wait, that's a different thread) , and/or rots, the Plankton Culture Manual by F. Hoff is good info also. Clownfishes is everywhere, but Florida Aqua Farms has the culture manual as well as other great things like live cultures, and mesh filters. Were I starting out, I would save money and buy the kit. I chose to be cheap, bought a little at a time, and ended up paying big time in shipping costs etc.

Hope this helps, and sorry to hijack. The topic of this thread is :
AUTOMATIC WATER CHANGES.


Sorry Anthony.
Kathy

[Casperson]

Anthony, just got your book!
Great stuff! Makes Sprungs books look like a beginner. :-) Hehehe, just kidding Julian.

Could be off the thread here, but it's my first post!
I've read every thread in regard to CLM's but could not find anything on what tool to use to
cut holes into acrylic. I have an existing 55 gal Tenacor acrylic tank with only one floor hole in it, the old corner overflow. It's filled with water and coral. I don't want to empty it. I would like to cut some holes where they should be near the top, to add additional water flow. Along with that I would like to add a CLM for better water circulation. Right now I'm using the old "Power heads in the display technique":-)

So let me have it! I guess after 15 years I'm a beginner again. :-)

Peace

Tom C

[Anthony Calfo]

Thanks kindly, Tom... but yes, your question is off topic. The title of the thread is "Automatic water change systems." Please start a new thread for such topic if you cannot find the answer in the archives.

I'm not even sure what exactly your question is here, my friend. Do you want to drill a tank that is full of water? You'll need to drain the tank indeed. The stress (resonance) of the water and its weight/force are a risk to that tank integrety otherwise.

As for the hole saw... just notice the mfg details on the saws available at the local hardware store (for metal, for wood, for plastic, etc)

[Casperson]

Thanks Anthony,
Sorry about the thread issue.

[Kathy55g]

In a previous post we were talking about how to prevent the emptying of the sump in the event of a powerfailure. solenoids are complicated and expensive as a turning off mechanism. I recently thought of those battery operated garden watering timers that you hook your sprinkler to. They are not very expensive, and can use standard hose fittings, and will open and close the sump drain on a timed schedule. These fail closed, so if the batteries wear out, you are OK, if the power goes out during water change time, you are out of luck, but what are the chances of that? You can schedule the water change for a time when you are likely to be home, so that you can manually shut down if needed.

Just a thought.

[Anthony Calfo]

they operate witht he same principal, my friend... both fail off.

yet the garden hose option is lower quality... not something you want to run your expesive reef and live animals on.

And worse... the batteries require a level of ongoing maintenance that not everyone will remember to do.

I do feel strongly that you get what you pay for with an industrial solenoid valve: quality and reliability.

[enigma32]

[first post]

What solenoids do you folks typically use?
i.e., what manufacturer? Is the solenoid plastic?

I did a quick google search and found mostly solenoids intended for CO2 injection.

How do the solenoids stand up to the water?
I've worked with brass solenoids in salty water environments (they typically lasted 4-5 years, sometimes longer), but in a very different situation than a reef tank.

[Avast Marine]

I would love to see some pics of this system!

[skeeter_ca]

Wow, Complicated thread was hard up with but i think i understand. I have a semi-automatic water changer and top-off on my tank. Since i don't have a drain lower than my sump i use a pump to pump the water out of the sump to my laundry drain and then a top-off switch to turn on a return pump from my SW reservoir. If i leave my main pump running i can remove 4 gals then replace with 4 gals of new SW from a reservoir in the garage without distrubing anything. Could be made automatic easily with timers. If i shut off my main pump it can change 14 gals + another 3 if i pull my durso and let the overflow drain. I can also turn on both pumps at the same time and exchange all of the water in my reservior and the new SW inlet is at the pump outlet of the sump, so the new water has to go through the entire tank before it comes to the old SW out pump.At first I designed it to be automatic but now i like being a part of it as it helps me stay close to the tank and see whats going on.

[porthios]

i'm really sorry to dredge up such an old thread but since it's referenced in the stickies i feel i have to point something out. the solenoid on the wcd is not going to save you from draining extra water in the sump due to a power outage.

best case scenario, power is cut to the return pump and solenoid at the same time. power is then restored to both at the same time and they race to empty the extra water in the sump. the return pump at whatever rate it flows and the wcd at the max rate it'll flow (1" = 600gph, etc..). you'll suffer less dilution than w/o the solenoid for sure but you don't eliminate it..

worse case, power is not cut/restored to the return pump and the solenoid at the same time (wired on different circuits, breaker trips on return pump).

worst case, chuck norris doesn't like you but that's another thread..

i should point out also, that power is often lost and restored a few times during a home's power outage, so a week of storms while you're away on vacation might result in quite a bit of dilution with ro/di..

in the end, i think kathy55g gave up too easy ;). there's no reason her method can't be automated with a proper controller

1. cutting power to the return pump
2. allowing time for extra water to settle in sump
3. pumping new water into the system and overflowing excess through wcd
4. restoring power to return pump

i think i've talked myself out of automating water changes now ;) i'll just strive for 'easy as possible to do manually'. i don't think the automation ends up buying you that much in the end..

[reverendmaynard]

What about the 2 channel dosing pump method? One channel is pumping water from the reservoir into the tank, the other channel is pumping water out of the tank and into the drain.

[Brad Black]

bump for a nice thread. i am in the planning stages of changing from an under-cabinet sump/return to one that is in the basement (1 floor below display). i intend to use several (much larger) containers for various functions and i would really like to see these ideas (auto WCR) in the form of a drawing or pics to assist in my planning.

i get the idea but maybe someone has a link or can help out with this? thanks

[Paradiddle7]

Nice thread, but nobody mentions that the fresh saltwater mix will also increase in density over time? How do you compensate for that? It would seem that the larger the reservoir of fresh saltwater you have (meaning the longer amount of time between manual intervention by making a fresh salt mix), the density in your system will increase (Adding 1.030 water to a 1.026 tank etc...).

I have a controller with a Salinity probe, so I could pump out small quantities of saltwater, allowing my ATO RODI water to maintain the proper salinity - OR - I could use another density probe to maintain the fresh saltwater at the proper density. Obviously the first method is cheaper and easier and I just need to make sure my probe is always clean and calibrated, with an extra on hand just in case.