Loaches Online Forum

Another boring nightshift here so been researching more



Would the above work? Would still have normal external filters (2 x Fx5) as well for the tank. Would add an airstone under the biomedia part and possibly have plants in the sump too

I do have a few questions tho

1. Does the overflow have to be at a certain depth in the tank? ie below the surface of the tank the same amount of gallons that would be in the sump? not expained very well but if the sump had a 20gal capacity then the intake below 20gal of water?

2. if there was a constant flow of RO water into the tank would there be to pure water going in eventually and require minerals to be trickled into the RO water?

3. what stops the sump overflowing if there wasn't an overflow drain (just can't get my head completly round this)?

4. and following from above what stops the pump drying up?

Sorry for all the questions but we've got absolutly no work tonight!!!

adampetherick wrote:Another boring nightshift here so been researching more



Would the above work? Would still have normal external filters (2 x Fx5) as well for the tank. Would add an airstone under the biomedia part and possibly have plants in the sump too

I do have a few questions tho

1. Does the overflow have to be at a certain depth in the tank? ie below the surface of the tank the same amount of gallons that would be in the sump? not expained very well but if the sump had a 20gal capacity then the intake below 20gal of water?

2. if there was a constant flow of RO water into the tank would there be to pure water going in eventually and require minerals to be trickled into the RO water?

3. what stops the sump overflowing if there wasn't an overflow drain (just can't get my head completly round this)?

4. and following from above what stops the pump drying up?

Sorry for all the questions but we've got absolutly no work tonight!!!

I can see some things wrong...
The UV will eventually fill up with mulm caught in it, since the water is coming straight from the tank. The water should be filter first to drap all the detritus. The UV is typically the last unit the water runs through before it returns into the tank.

You can't really regulate how much RO water your mixing.



1. Not sure what you mean, but overflow boxes normally reach a certain height of the tank top rim.

2. Not sure if I understand.

3. Your sump doesn't require an overflow. If it does, then it would need to overflow somewhere, like another sump or refugium. According to your diagram, the water is returned back into the tank via a pump. If the pump stopped, the water would overflow from the sump, onto the floor.

4. From your diagram, water would stop the pump from drying up. The only way I could think of that the pump would run dry is if the tubing from the main tank, into the sump were to get separated. Of course, water from the tank would oveflow to the floor. The remaining water in the sump would be pumped back into the tank. As more water spills from the tank and more water from the sump is pushed back into the tank, the sump will eventually dry up.

crazie.eddie wrote: I can see some things wrong...
The UV will eventually fill up with mulm caught in it, since the water is coming straight from the tank. The water should be filter first to drap all the detritus. The UV is typically the last unit the water runs through before it returns into the tank.

You can't really regulate how much RO water your mixing.



1. Not sure what you mean, but overflow boxes normally reach a certain height of the tank top rim.

2. Not sure if I understand.

3. Your sump doesn't require an overflow. If it does, then it would need to overflow somewhere, like another sump or refugium. According to your diagram, the water is returned back into the tank via a pump. If the pump stopped, the water would overflow from the sump, onto the floor.

4. From your diagram, water would stop the pump from drying up. The only way I could think of that the pump would run dry is if the tubing from the main tank, into the sump were to get separated. Of course, water from the tank would oveflow to the floor. The remaining water in the sump would be pumped back into the tank. As more water spills from the tank and more water from the sump is pushed back into the tank, the sump will eventually dry up.

Ok, UV could be moved to the other side on the return pipe

1. So the intake (would be drilled rather than overflow box) has got a maximum depth

if the sump was a 1m x 1m x 1m for simplicity's sake and the tank was 1m x 1m x 2m tall then the max depth for drilling the hole for the pipe from the tank to the sump would be 1m

2. If the tank is 100% RO water then all the minerals that the fish need would be stripped out of the water, guessing minerals could be added to the sump on a daily basis to compensate

3. The overflow on the sump is to stop it spilling over onto the floor if the pump stops and be piped direct to the drain

4. Would the flow rate of the pump need to be matched to the flow that could flow from the tank to the sump

Ok. I think I understand a little better. Back to your original questions...
1. Same answer...Overflow boxes have specific heights already.

2. Unless you know how much RO water your adding, then it would be difficult to figure out how much minerals to add. Water evaporation could very during specific times of the year, so even if you measured it, you probably didn't calculate for a later part of the year.
It might be best to NOT automate this process. Becuase IF the intake hose were blocked or disconnected, then the RO water would continue to add water as the sump returns water back into the tank. The sump would probably not run dry, but your main tank would overflow.

3. Your overflow in the sump is a good concept only if you intend to over fill the sump. Typically a sump is filled partially so in event, like a power failure, the pump in the sump stops pumping water back into the tank, there would only be enough water returned from the tank to basically fill it up slightly, but not to overflow it.

4. As I mentioned in my original post, the only way I can think of for the sump to run dry is if the intake hose from the main tank were somehow blocked or disconnected.
If the sump were to be filled full of water using your overflow in the sump idea, then the pump would have to flow faster than the returning water. Otherwise, it would not matter, but best if it does pump faster.

Anyone else?

For clarity, you're pumping 45gpd of RO, then using the sump's overflow to drain excess, right? It's like a constant water change system with regular RO input?

1) Your overflow box will determine the water line, so you just want it at the lip of the main tank.

if the sump was a 1m x 1m x 1m for simplicity's sake and the tank was 1m x 1m x 2m tall then the max depth for drilling the hole for the pipe from the tank to the sump would be 1m

If you're talking about the siphon break on the return line (from sump to tank), drill it right below the water line of the main. This way if the pump loses power you break the reverse siphon quickly.

2) Yes, and this holds your design back. One option is to boost GH and KH and other minerals daily. Another is to just pump RO into a holding tank and valve the line to the sump. Then you can reconstitute RO in the holding tank and open the valve for a fast water change.

If you want to just use continous RO, you should do it as a top-off for evaporation. You could use a float switch and solenoid if you want to get badass.

3) Normally you just keep saftey room in the sump. For example, you could stick 120L into that ~160L tank, create your siphon breaks, and you're set. Now if you lose power to the pump the reverse siphon and tank drain should only drop a few liters into the sump.

If you're uising an overflow from the sump to the drain, you of course do not need to worry about the saftey room. Instead you should just make sure the sump's drain diameter is wide enough for saftey and redundancy, to ensure the sump never overflows.

4)

Would the flow rate of the pump need to be matched to the flow that could flow from the tank to the sump

No, you want the max flow (drain diameter) from the tank to the sump to be higher than the rate of return from the sump to the tank. This is for redundancy and best practice is to overbuild the drain by ~25%. For example, if you are moving 400GPH from the sump to the main, you'd want a 500GPH overflow, so if you get some blockage at the overflow you're still okay.

It doesn't matter if the drain rate is higher than the pump return rate: water will still drain at the pump's rate. It's not getting into the overflow box any faster. Get it?

The much bigger problem is losing the siphon in your overflow box. Focus here (or just drill the main tank) and you'll do fine.

Other points:
Your UV requires a dwell time and slow flow to be effective. Putting it into the flow pipes is pointless if you want to move a lot of water; instead, consider just sticking it in the sump with a second (small) dedicated pump. It'll work inline if you keep the UV's proper flow rate though.

You don't need really those baffles if you're just putting all the media into the biomedia/sponge tower. Baffles either serve as bubble traps (for SW) or direct flow. Why are you directing flow after the C component?

Good luck.

JoeKuhlii wrote:For clarity, you're pumping 45gpd of RO, then using the sump's overflow to drain excess, right? It's like a constant water change system with regular RO input?

1) Your overflow box will determine the water line, so you just want it at the lip of the main tank.

if the sump was a 1m x 1m x 1m for simplicity's sake and the tank was 1m x 1m x 2m tall then the max depth for drilling the hole for the pipe from the tank to the sump would be 1m

If you're talking about the siphon break on the return line (from sump to tank), drill it right below the water line of the main. This way if the pump loses power you break the reverse siphon quickly.

2) Yes, and this holds your design back. One option is to boost GH and KH and other minerals daily. Another is to just pump RO into a holding tank and valve the line to the sump. Then you can reconstitute RO in the holding tank and open the valve for a fast water change.

If you want to just use continous RO, you should do it as a top-off for evaporation. You could use a float switch and solenoid if you want to get badass.

3) Normally you just keep saftey room in the sump. For example, you could stick 120L into that ~160L tank, create your siphon breaks, and you're set. Now if you lose power to the pump the reverse siphon and tank drain should only drop a few liters into the sump.

If you're uising an overflow from the sump to the drain, you of course do not need to worry about the saftey room. Instead you should just make sure the sump's drain diameter is wide enough for saftey and redundancy, to ensure the sump never overflows.

4)

Would the flow rate of the pump need to be matched to the flow that could flow from the tank to the sump

No, you want the max flow (drain diameter) from the tank to the sump to be higher than the rate of return from the sump to the tank. This is for redundancy and best practice is to overbuild the drain by ~25%. For example, if you are moving 400GPH from the sump to the main, you'd want a 500GPH overflow, so if you get some blockage at the overflow you're still okay.

It doesn't matter if the drain rate is higher than the pump return rate: water will still drain at the pump's rate. It's not getting into the overflow box any faster. Get it?

The much bigger problem is losing the siphon in your overflow box. Focus here (or just drill the main tank) and you'll do fine.

Other points:
Your UV requires a dwell time and slow flow to be effective. Putting it into the flow pipes is pointless if you want to move a lot of water; instead, consider just sticking it in the sump with a second (small) dedicated pump. It'll work inline if you keep the UV's proper flow rate though.

You don't need really those baffles if you're just putting all the media into the biomedia/sponge tower. Baffles either serve as bubble traps (for SW) or direct flow. Why are you directing flow after the C component?

Good luck.

Thanks, that clears up a lot of my questions, the flow direction after the media basket was just to ensure that the RO water was mixed with the tank water thoroughly and to make sure it all flowed past the heaters