water level in return pump section of sump not staying the same level

[mathieurochette]

I'm having a problem with the return section of my sump. I'm pretty sure i have everything setup properly with a ball valve on the In side of the sump and a ball valve on the Return side. the In ball valve is wide open and and the return throttled back 20-30% now the water level does stay at the same level for a while 8 hours or so but eventually the return side drops to the point where my Mag 9.5 is sucking air and water. I realize that their is evaporation and the my protein skimmer is taking a bit of water out of the system but even if i top off the system eventually the water level will drop in my return section of my sump?

Can anybody help me out with this?? Please.



[attachment deleted by admin]

[lost_at_sea]

Is that just one baffle (or divider) separating your middle section and the return section?

I think it's a fundamental problem with how your baffles are setup ie their individual and relative heights is causing the return compartment to become independent of the others

[mikerobart]

How big is the return section ? You're sure it's not evaporation?

Only thing I can think of other than evaporation and losses to the skimmer is the rate either of drainage to the sump or pumping back to the display is changing. How this is changing though eludes me. Air getting into the piping and displacing water .... clogging with algae or something else...

Are you running auto top off?

The water has to be going somewhere... either the level in the display is getting higher... ?

If the level in the display does not INCREASE when the return level DECREASES, it's gotta just be evaporation, skimmer, or somehow displacing air in the pipes and using that extra available space although I doubt this is the case.

If you auto top off the return area, does your display level go up and up and up then?

[mikerobart]

Having a hard time telling exact relative baffle heights from pictures but lost at sea might have something. I think it still depends on rates of drainage and return. If you get to a point where the level in the display has decreased slightly, the drainage could slow down and not keep up with the return. Have you tried restricting the flow even more on the return pump with your valve?

[mikerobart]

Or, could try adding more water to the system and keeping a higher level in the return area /sump in general.

[mathieurochette]

Is that just one baffle (or divider) separating your middle section and the return section?

I think it's a fundamental problem with how your baffles are setup ie their individual and relative heights is causing the return compartment to become independent of the others

the compartments are (from left to right) L10xW12xH9 baffles 1" space 1 1/2" bottom and 2" space fuge L8xW12X9 return L4XW12xH7 i'm thinking it's evaporation because the tank water level is dropping

[Funkmotor]

The only place in my system that the water level drops at all is in the return section of my sump.  I have my ATO (auto top-off) sensors set up there, and that automatically adds new fresh water to offset whatever evaporates.

If your water flows from tank drain to fuge to return then the water level in your drain area and fuge should always be kept the same...then if the water is dropping in the return section I'd say that it is completely normal.

Can you take a picture (or make a drawing) of your sump so we can see exactly what you're talking about?

[mikerobart]

I'm thinking evap as well given that your return section is fairly small and as mentioned just above, the return section is really where the levels should change if water is being added or removed. Water level in the display depends mostly on where your overflow is, assuming that overflow can keep up with what is being pumped etc.

What size is the tank itself or did I miss that?

I probably get about 1.5 gallons of evaporation a day on my 90 gallon with a 33g sump and 2x250w metal halides which only run a few hours a day right now, and a fan blowing over the tank when the lights are on. By the looks of it, your return area can't handle much more than one gallon of evap before your pump might start sucking air... your return section can hold total I think about  3.7 gallons not including displacement from pump. Might be time for an ATO ! Autotopoff.com is the one I use, cheapish and easy.. some sweat by tunze osmolator ($$) but apparently awesome... jbj makes one.. can also diy with float switch from hardware store which is basically what autotopoff.com has but with less work for you....

[Hookup]

Hey,
  Confirming some facts. 

From the Display tank to the Sump prespective.

1. The drain line has a valve, but it's 100% open, so we can ignore that fact...

2. The return line, on your Mag 9.5 has a valve, and it's closed allowing 20-30% of the flow up to the display

3. The system appears to run fine, but after a few hours (8 or so) the last chamber, where the return pump is, runs dry.


If I've got the above correct, could you try something for me?

Turn off all pumps to the sump...

Put a piece of TAPE on the water line in the sump... Lable this "starting water mark".

Turn on your system for 4 hours.

Turn off all pumps as above.

Check to see where the new water line is.


I want to see if we can eliminate some of the possiblties.  If the water line is very similar, then we should be able to eliminate evap, which in an 8hr period has me questioning.


My first guess is that the down-line is not able to keep up with the up line.  you're not flooding because your display has the capacity to take the 1-2 gallons that's separated in and around your return pump.

[lost_at_sea]

Good call.. are you using an Overflow box? take some notes on how the water level is with respect to that with all pumps off and then again 2hrs into and again 4hrs into the test.

My theory is that the water level in the first 2 chambers is increasing and the height of hte baffle's are preventing the flow through to the return section.

Also take not the water levels in each of the 3 chambers of your sump.

[Hookup]

The part that bothers me is... if the Mag is pushing more water up than can come down, once the mag is running "dry" it's effectively shuts down the upward flow until the down-line can return the water to the sump...at which point we start to oscciliate between dry cycle and wet cycle from the return pumps point of view.  The transition from DRY to WET cycle would be pretty fast... and the poster, I assume Mathieu, has not noted that behaviour.

So if that is NOT happening.. then my next best guess is water is being removed from the system over that 8hour period... but that makes little to no sence either because with a period that short, it would have to be a leak... even I do not lose that much water in 8hrs...


So doing the tape-test should help understand if we are dealing with a static volume of water and we have simple flow/return rate issues, or if infact the water volume is decreasing...


One last *thought*.  Is the skimmer connected direcly to the drain?  A change in water levels within the sump area, could cause the skimmer to run wet, thus emptying water from your system... which could cause this type of effect... but that's a complete stretch at this point.

[lost_at_sea]

I think I got it.

I actually think the fundamental problem is there is not enough water in teh system to begin with.

If you look at the original SUMP.pdf pic you'll notice hwo the 3 compartments all have differing water levels... this is causing the MAG pump to lose it's ability to 'pull' water from the other compartments to keep the flow going.

instead that section starts to run dry and the others only fill up only by small amounts of water flowing over the baffles.

I would look at adding a lot more water to the sump so that theere is a good 2-3 inches of water above the lower level baffles (ie the first baffle from the skimmer section)

I bet you that then causes the sump to run at a constant level in all 3 compartments and will stop your pump from running dry.

[groupie02]

The Mag 9.5 is a pretty powerful pump to be in that small return section. When it starts sucking air, how high above the pump is the water level?  My 120g is losing about 1g/12h but it's sitting right next to a register which is always pushing air around.  If you lose about 1/3g over 8 hours, that's probably enough to cause the suction to happen.

The solution is really to use an auto-top-off device.

[mathieurochette]

The Mag 9.5 is a pretty powerful pump to be in that small return section. When it starts sucking air, how high above the pump is the water level?  My 120g is losing about 1g/12h but it's sitting right next to a register which is always pushing air around.  If you lose about 1/3g over 8 hours, that's probably enough to cause the suction to happen.

The solution is really to use an auto-top-off device.

Ok so tonight i will do the tape test and mark the water level in the tank and mark the water level in the return section of the sump will have updates tomorrow.

the tank itself is a 75 gallon and the sump is a 20 gallon standard but i figure its holding about 10 gallons of water between the 3 compartments and the skimmer is not connected to the drain

[mdugly]

Looks like your return section is too small.
I would remove the baffle (shown in second pic) to make return section bigger, and less prone to evap issues.

[mathieurochette]

Looks like your return section is too small.
I would remove the baffle (shown in second pic) to make return section bigger, and less prone to evap issues.

so i did the tape test last night and the display tank had about 1/8" less water and the return section of the sump was down about 2" from the tape line. seems like i need to add more water to the system and i think mdugly is right and will have to scrap my plans for a fuge and make the return section bigger

[groupie02]

instead of removing it, you could move it.  The fuge does not have to be huge in order to work.