Point-of-Entry RO: Ask Where the Reject Water Goes Before You Sign the Quote

A quote for a point-of-entry RO system lands on the desk and it looks clean. Skid, membranes, prefilters, a price. What is almost never on that page is the thing that decides whether the system is a good idea: where the reject water goes, and what it takes to get treated water back into the building at usable pressure. Those two items cost real money, and they show up after the install, which is the worst time to find them.

An RO unit doesn't strain water the way a cartridge does. It splits the feed into two streams. Permeate is the low-salt water you wanted. Concentrate, or reject, is everything the membrane pushed back, carried off in a share of the feed that goes to drain. That split isn't a flaw. Without it the membrane scales and dies in weeks. But it means every point-of-entry RO is buying you clean water and a waste-disposal problem in the same box, and the quote usually prices only the first half.

The recovery number sets everything downstream

Recovery is the fraction of feed that leaves as permeate. A commercial brackish system typically runs 60 to 80 percent recovery, and a well-staged array can push toward 85. Your recovery ceiling is set by feed chemistry, not TDS alone. A single pressure vessel, even loaded with eight elements, tops out around 50 percent on its own, so the higher system numbers come from staging vessels in series and running the first stage's concentrate as feed to the next.

Whatever the number, run the arithmetic before you sign. Recovery sets both how much water you buy and how much you pay to throw away.

On a facility already paying for water in and sewer out, that drain column is a line on both bills, every day the system runs.

Concentrate is just your feed water, packed tighter

Nothing gets dosed into the reject, antiscalant aside at a few ppm. It's your feed water with the rejected salts squeezed into less water. The concentration factor is 1 over 1 minus recovery, so at 75 percent recovery the concentrate runs about four times the feed salinity. Push to 80 percent and it's five times. That is the same stream a TDS reading would flag if you metered it.

It matters because a lot of sewer authorities cap discharge on TDS, chloride, or specific ions, and many add a surcharge above a threshold. Concentrate at four to five times feed salinity can trip a limit your raw feed never would. On septic the problem is worse: you add the reject volume on top of everything the building already sends to the field, and the sodium disperses the soil the same way softener brine does. Get the concentrate disposal path in writing before anyone signs, whether that's sewer with a surcharge, a permitted discharge, or a lined evaporation setup. It's the single most common thing that turns a clean-looking RO quote into a budget problem after the fact.

The build nobody quotes: storage and repressurization

Permeate comes off the membrane at low pressure. It can't feed a building's distribution directly. So a real point-of-entry install isn't just the skid. It's an atmospheric break tank that the permeate fills, and a repressurization pump that draws from that tank and delivers water to the building at working pressure. Often UV or a recirculation loop on the tank too, because stored RO water carries no disinfectant residual and bugs are happy to grow in it.

Sizing that tank is its own call. Size it to peak demand plus the membrane's fill rate, not to the daily average, or the building runs the tank dry at the worst moment and the pump hammers against empty. This is the part contractors new to RO leave off the page, and it's four figures of tank, pump, controls, and plumbing the customer meets after the skid is already bolted down.

Two problems, one machine

Most point-of-entry RO requests are really two separate problems wearing one label. One is drinking and cooking water quality, meaning taste, TDS, or a specific contaminant. The other is scale on equipment and fixtures. People reach for whole-building RO to solve both at once, and it's usually the most expensive way to solve either.

Scale is hardness. A softener handles it by swapping calcium and magnesium for sodium. It doesn't lower TDS or chloride, so it's the wrong tool if your real complaint is salty water, but for scale on boilers, water heaters, and dishmachines it's a fraction of the cost of running every gallon in the building through a membrane.

Drinking water is a point-of-use job. A dedicated RO at the tap, or feeding the coffee and ice lines in a restaurant, treats the gallons people actually consume instead of the shower and toilet water. And if the complaint is one specific thing, treat that thing. Color and organics are usually tannins, and a tannin-selective anion resin pulls them for a fraction of what whole-building RO costs. Iron gets an iron filter. Point-of-entry RO earns its place when the feed is bad across the board, a brackish well, chloride high enough to pit stainless at every fixture, arsenic or nitrate you want out of every tap and not just the kitchen. That's a real category. It's just smaller than the number of people who get quoted for it.

Chlorine is the other thing to settle before the membrane goes in. Free chlorine destroys polyamide RO membranes, so a point-of-entry unit on municipal water needs carbon or another dechlorination step ahead of it, sized and maintained. If you want the detail on why, chlorine has to come out ahead of the membrane, not at it.

Practical Bottom Line

Before you approve a point-of-entry RO quote, make it show three things: the recovery rate and the daily reject volume, the concentrate disposal path in writing, and the storage tank plus repressurization pump on the parts list. If those aren't there, the quote isn't finished. And confirm the job actually needs whole-building RO instead of a softener plus a point-of-use unit, because most don't. When the feed genuinely calls for it, the membranes are where the system lives or dies, and on 8040 brackish elements the NanoH2O (formerly LG) BW 400 R G2 is the stronger spec: around 11,500 gpd at up to 99.8 percent rejection on a 34-mil feed spacer that lowers pressure drop and buys fouling margin. Higher flow at a given pressure plus better rejection is what holds your recovery target without driving the array harder than it wants to run. The market still defaults to FilmTec out of installed-base habit and contractor familiarity, which are real reasons but not performance reasons. On the numbers that decide a brackish 8040 job, NanoH2O is the one to spec.

Enviro Membranes stocks the NanoH2O 8040 lineup — BW 400 R G2, BW 440 R G2, and SW 400 R — and ships them same day across the US at wholesale pricing. If you're sizing a point-of-entry system or checking a quote someone handed you, send us the feed water and the reject numbers and we'll help you pressure-test it before you buy. Reach the team at enviromembranes.com.