All being well, the oscars new tanks will arrive on Tuesday. It's occurred to me that, since they will be on the other side of a fairly flimsy wall to the kitchen, I could plumb them into the cold water supply and have a continuous water change set up. I would need to filter the water through a HMA filter as I understand it. I was hoping that doing this would allow me to do away with the cannister filters. I would remove waste manually once a day which shouldn't take out more than 10L of water at the very most. The tanks are 450L and will have one 12"+ oscar in each one. Would I need to turnover the whole volume each day to stop ammonia build up? Would that be enough?
Continuous Water Change System
- Thread starter Blobfish
- Start date
April FOTM Photo Contest Starts Now!
🏆 Click to enter! 🏆
In principle it can work, after all, it's what nature does, however.....
When you add fresh water, it doesn't replace old water, it dilutes it, so a slow trickle of the tank volume over 24 hours won't replace all the water the way you want it to, you're looking at factors far higher to achieve that. It works better if you have a sump that drains and refills as that does actually do a % water change at once, but the plumbing is even more complex. Your 10l changes will also add to this, so you should be set from that point of view.
You do get a significant benefit in that it can essentially replace water changes, and may make the filter requirements less, but I can't see it practically replacing your canisters.
COI - I've got a thin wall behind my sumped tank and am planning to do the same, I do however have the sump set up as a filter and don't plan to change it.
When you add fresh water, it doesn't replace old water, it dilutes it, so a slow trickle of the tank volume over 24 hours won't replace all the water the way you want it to, you're looking at factors far higher to achieve that. It works better if you have a sump that drains and refills as that does actually do a % water change at once, but the plumbing is even more complex. Your 10l changes will also add to this, so you should be set from that point of view.
You do get a significant benefit in that it can essentially replace water changes, and may make the filter requirements less, but I can't see it practically replacing your canisters.
COI - I've got a thin wall behind my sumped tank and am planning to do the same, I do however have the sump set up as a filter and don't plan to change it.
If you're willing to expend that much water, it will work. I surmise, though, most of the ammonia will still likely be consumed by bacteria and plants and not discharged with the water change.
I also plan to connect my aquarium into my house's plumbing when I finish out the basement next year, so I've done some number crunching on this before. (My overflow output will be connected to a reservoir and lawn sprinkler system outside so as not to waste all that water.)
As DrRob pointed out, even if you're simultaneously taking out an equal amount of water (such as with an overflow into a drain), the amount of water added doesn't directly correlate 1-to-1 to the percentage of water change performed. However, it does correlate at an equivalent rate to the amount of control executed over added solutes (e.g, ammonia, nitrites & nitrates).
What I determined is that if at a steady rate over a given period of time a volume of water equal to your tanks' maximum capacity is added while an equivalent quantity of water is also removed at that same rate, you will have executed over that time period a 63% water change relative to its starting solutes and a 100% water change relative to any solutes steadily added during that same time period.
For example, let's start with 100 ppm medications and 10 ppm nitrates. We'll turn over the water 100% (450l) over the course of a week -- during which time we'll be gaining another +10 ppm nitrates. At the end of the week, we will have 37 ppm medications and 10 ppm nitrates. After another week, we'll have 14 ppm meds and still 10 ppm nitrates. After two more weeks, we'll have 2 ppm meds and 10 ppm nitrates.
To execute a 98% water change in that manner (relative to the start), at least four times the volume of your tank must be turned over. So, if anything spikes out of the ordinary, it will take an additional 4 turnover cycles before the spike is eliminated.
What this means in practical terms is that you need to know what parameter you're controlling, at what level you wish to limit it, and at what rate it increases to that level. The length of time it takes to get to that control level is the length of time during which you must turn over 100% of your tank's volume.
One final example: If you have a decoration that raises your water hardness by 1 dGH every 12 days, to counteract that decor's affect, 1/12th of the tank's volume must be turned over each day using a continuous water change system like this.
Since you're likely wondering from what nether region I came up with these numbers, I wrote a scripted iteration which loops a number of times equivalent to the number of drops (0.005 mL) in the water change.
aqSize := 450
period := aqSize / .005
ppmGain := 10
ppm := ppmGain
loop period
ppm := ppm * ( 1 - 0.005 / ( aqSize + 0.005 ) ) + ppmGain / period
clipboard := ppm
I also plan to connect my aquarium into my house's plumbing when I finish out the basement next year, so I've done some number crunching on this before. (My overflow output will be connected to a reservoir and lawn sprinkler system outside so as not to waste all that water.)
As DrRob pointed out, even if you're simultaneously taking out an equal amount of water (such as with an overflow into a drain), the amount of water added doesn't directly correlate 1-to-1 to the percentage of water change performed. However, it does correlate at an equivalent rate to the amount of control executed over added solutes (e.g, ammonia, nitrites & nitrates).
What I determined is that if at a steady rate over a given period of time a volume of water equal to your tanks' maximum capacity is added while an equivalent quantity of water is also removed at that same rate, you will have executed over that time period a 63% water change relative to its starting solutes and a 100% water change relative to any solutes steadily added during that same time period.
For example, let's start with 100 ppm medications and 10 ppm nitrates. We'll turn over the water 100% (450l) over the course of a week -- during which time we'll be gaining another +10 ppm nitrates. At the end of the week, we will have 37 ppm medications and 10 ppm nitrates. After another week, we'll have 14 ppm meds and still 10 ppm nitrates. After two more weeks, we'll have 2 ppm meds and 10 ppm nitrates.
To execute a 98% water change in that manner (relative to the start), at least four times the volume of your tank must be turned over. So, if anything spikes out of the ordinary, it will take an additional 4 turnover cycles before the spike is eliminated.
What this means in practical terms is that you need to know what parameter you're controlling, at what level you wish to limit it, and at what rate it increases to that level. The length of time it takes to get to that control level is the length of time during which you must turn over 100% of your tank's volume.
One final example: If you have a decoration that raises your water hardness by 1 dGH every 12 days, to counteract that decor's affect, 1/12th of the tank's volume must be turned over each day using a continuous water change system like this.
Since you're likely wondering from what nether region I came up with these numbers, I wrote a scripted iteration which loops a number of times equivalent to the number of drops (0.005 mL) in the water change.
aqSize := 450
period := aqSize / .005
ppmGain := 10
ppm := ppmGain
loop period
ppm := ppm * ( 1 - 0.005 / ( aqSize + 0.005 ) ) + ppmGain / period
clipboard := ppm
Thank you for your replies Dr Rob and Maehlice, much appreciated. I have decided against the idea as I don't have the room for the canister filters and the HMA filter, it would have to be one or the other. The continuous change system would be a problem if I needed to medicate the tank. It's a shame that I can't run a sump - would love to but I need to make two indoor dog kennels under the tanks so they are going to take up most of the room under there. Fortunately, I actively enjoy doing water changes so that's OK.
On the subject of dilution, I had wondered about this but I thought that this might be negated to some extent as the cold water coming into the tank would sink while the warmer water would rise and therefore be the first to be drawn out of the tank by way of the overflow pipe. Different temperature bodies of water don't instantly mix apparently.
On the subject of dilution, I had wondered about this but I thought that this might be negated to some extent as the cold water coming into the tank would sink while the warmer water would rise and therefore be the first to be drawn out of the tank by way of the overflow pipe. Different temperature bodies of water don't instantly mix apparently.
Blobfish said:
As long as you install a valve on the inline from the house (I imagine you'd have to), you could turn off the water and switch to manual changes during times of medication.
I'm not sure how your setup and display is, so I'm just throwing it out there ... your sump can go anywhere:
- Beside the tank. If the tank is using a HOB overflow, the only necessity is that the sump's water line is below the overflow's output. In the case of a DIY PVC overflow, that could feasibly be just a few mm lower than aquarium's water line.
- Above the tank. If a water pump is used to draw water out of the aquarium instead of back into it from the sump, the sump can literally be anywhere. In that setup, the overflow is built into the sump instead of the display.
Until I can find a compelling reason not to place my refugium and sump at a higher elevation than my display tank, this is how I will be setting it up when it reaches it's final home next year. Any tank less than 55g, I'm comfortable building and/or drilling, so I can eliminate the use of a might-be-faulty-one-day siphon type overflow and use a drilled sump instead. If my pump fails, the worst that might happen is water not making it to the sump; no check valves and siphon-breaks necessary. If my wife lets me, my shrimp tank/refugium will be in the living room above the basement aquarium.
If there's any room around the aquarium (in the house, really), and you really want a sump, it can be done.
Blobfish said:
I think unless you're pumping in serious quantities of water, the slower dilution rate should be negligible. At the flow rate where I think it would have an appreciable effect would be long after the fish died of temperature shock.
Unfortunately the sump really isn't an option, there is 3" clearance at one end before the door and 2" at the other end before the window starts. The top of the tanks will be 5.5ft high as it is. There is no attic above as the ceiling just slopes with the line of the roof and as it's a bungalow, no basement (but at least we have a solid floor which is a big bonus). The small kitchen the other side of the wall already has two tanks in it and as a result, barely space make a coffee and fry an egg. Underneath the tanks is largely dog space - necessary so I can fit another four tanks in elsewhere in the room - but there will be a gap of 12" which will take the two filters and electrics. Never mind, it will be great to have the new tanks, I'll carry on doing the water changes as I do now which is to run a syphon directly outside to drain and then use a 120L water butt which I fill from the kitchen sink, add conditioners and pump from the kitchen directly into the tank using a rain water butt pump, having a small place has some advantages! Thank you very much for you thoughts though.
Most reactions
-
326 -
203 -
181 -
149 -
140 -
138 -
125 -
125 -
121 -
119 -
92 -
73 -
70 -
69 -
66
trending
-
-
-
-
-
F-1 Angel... a bit nervous... it'll be my 1st F-1 fish...
- Started by Magnum Man
- Replies: 16
Staff online
-
FishmanicTFF Contest Czar