To baffle or not to baffle?

[Parker]

I have one baffle in my sump but only cause I'm too lazy to take it out. No micro bubble issues and my skimmer doesn't introduce bubbles. My auto top-off is manual at the moment but I'm planning on running a float valve in the sump backed up by a electric valve connected to a timer through my APEX.

[lastlight]

If I'm filling my 5g bucket with ro/di with my membranes etc as is right now the rest I can get out of my membrane is about 7 TDS before resin.

Now that I fill a 45g tank every couple weeks it runs for many hours and eventually goes as low as 2 TDS pre-resin so that helps.

I also see TDS creep after having the water turned off for even 10 min. Since I started bypassing water until the TDS goes down I have made my very first portion of resin last since I got back into the hobby.

[Delphinus]

At my old house I ran RO/DI to my tank top off valve full time but because the run from the laundry room to the tank was large, I had it fill a reservoir in behind my tank and the tank topoff was filled from that reservoir. The RO/DI was on all the time into this reservoir but because it was so far away (well 20' or so) from the filters, it buffered things just enough that I ran this way without problems for years.

Then I moved and now had a tank 5' away from my water source so I removed the inline reservoir and had the RO/DI direct to the sump. What a mistake that was. I ended up replacing my RO membrane 3 times that first year before I figured out that the problem now was the incessant cycling of the membrane. It would literally run for about 5 seconds every 5 minutes. Turns out this is bad for a membrane. Who knew!

So now what I do is run my RO/DI into a reservoir which then feeds my topoff downstream. I worry about filling up the reservoir maybe once every two weeks or so - I let it empty, or get near-to-empty, then I fill it to the top (letting the float valve shut the RO/DI off when it's full to prevent overflows). Since switching to this method I seem get about 2-3 years out of a membrane before I start noticing TDS creep that doesn't clear itself with an extra long flush. It's really the same thing as what a pressure tank does except instead of using a pressure based diaphragm to control the fill, I just open or close a valve to fill the reservoir in one big fill.

Originally Posted by christyf5

True dat, but filling a 5g reservoir is different than the half inch change in level in your sump IMO. Then again maybe it just keeps the 5g full and it doesn't completely empty. I dunno, I just heard someone on here talking about it, maybe they'll see this and chime in.

Originally Posted by PoonTang

Running small volumes through your unit is hard on the DI resin due to TDS creep. When the unit is shut off the solids will cross the membrane as time goes by and when the unit is turned on this large concentration gets taken out by the DI. I now with my online monitor it will read about 25ppm at the membrane outflow upon startup and then settle to about 2 after a minute or so. I don't know what small volumes does to the membranes tho.

Originally Posted by sphelps

Rapid shut off can cause cavitation in the membrane which can cause damage but a pressurized storage tank or accumulator eliminates this problem. So if you don't have a pressure tank or accumulator then you'll want to run your RO for longer periods of time less often to reduce the occurrence of possible damage.