RO Membrane in series

mattali · 2026-07-27T12:46:30-0400

I recently got my first first RO unit, it’s a 5 stage but wasn’t really happy with the TDS after the membrane being at 22 and does creep up and down, so it’s burning through my resin pretty fast. I believe there is water passing through the 2 o rings on the membrane, came with a pretty generic membrane.

The unit does come with an 80psi booster, so I’m having to add another membrane housing after the first one, not for the waste water but the 22tds water I’m getting, and running the waste water from the second back to go in through the first. Waste water from the the first down the drain.

Is this something anyone does? Will this work? The booster pump is located between the pre filters and membrane.

This is how the unit is currently configured, tap- 5micron sediment- 5micron carbon- 1micron sediment- 80psi booster-membrane-DI.

This is what I’m planning 1micron sediment (considering adding a 5 before it?) 2x 1micron carbon, 80psi booster- 2x membrane, 2 or 3 Di vessels.

I already have everything to do the new setup and received 2 new membranes today and carbon blocks today.

ICP on current setup shows everything 0 with 5ug/silicon, I get exact same results on ICP with just my 11l DI vessel that I use for water to wash my car although silicon is 6ug/l, everything else 0 including tds.

Oh, lastly mines a 400gpd unit.

dereksiz · 2026-07-27T13:56:36-0400

"is this something anyone does?" - yes, double pass RO is a real thing, but i don't think it'll work plumbed the way you've drawn it. permeate comes off the first membrane at basically no pressure, so it can't drive a second membrane in real time. commercial double pass setups run pass 1 permeate into a holding tank, then a second booster pump into pass 2. with your one pump sitting before membrane 1, membrane 2 would barely produce.

before you re-plumb it though, what's your tap TDS? 22 out doesn't mean much without it. if you're on 500ish tap that's about 96% rejection which is fairly normal, if you're on 250 that's more like 91% and something's actually wrong.

also worth checking that your flow restrictor matches the membrane. 400gpd wants an 800cc restrictor and generic units often ship with the wrong one. undersized restrictor puts more backpressure on the membrane and pushes more TDS through, which would look exactly like what you're seeing.

and yeah, a 5 micron ahead of the 1 micron is worth doing, it'll make the 1 micron last a lot longer.

Uncle99 · 2026-07-27T14:07:12-0400

Depends on your incoming source TDS before any filtration.
That will indicate, how efficient the system is in respect to your 22 TDS.
ICP, not all of them are accurate.

What’s source TDS?
If like 400 or well based, that needs some work. It contains silicates and some heavy metals like iron and copper.

If like 200-300, then something is wrong in your system as most RO filters, installed correctly, can pull easy, 95% of incoming TDS.

This should get you to 5 before DI, conserving your resin