What a healthy output looks like
A working RO membrane removes most of what is dissolved in the feed. How much is stated as salt rejection. The molded membrane in the LUXGEN range is rated at 96–98%, which means that for every 100 ppm in the feed, between 2 and 4 ppm gets through. Feed water at 500 ppm therefore comes off the membrane at roughly 10 to 20 ppm; feed at 1500 ppm at roughly 30 to 60. Standard TFC membranes reject a little less when new and a little less again as they age, but the shape of the answer is the same: the output is a small fraction of the feed.
What you read at the tap is that fraction plus anything added afterwards. A TDS controller mixes some water that bypassed the membrane back into the output, which raises the reading by design. A copper or mineral stage adds a little more. So there is no single correct output number. There is a correct ratio between what goes in and what comes out, and that is what to measure.
How to check it properly
- Measure the feed first. Take a sample from the inlet to the purifier, or from the nearest tap on the same line. Write the number down.
- Run the purifier for a minute before sampling the output, so you are reading fresh membrane water rather than what has been sitting in the tank overnight picking up minerals from the tank walls.
- Measure the output from the purifier's own tap. Same meter, same day, same temperature as far as possible.
- Work out the ratio. Output divided by feed. On a healthy membrane with the controller shut it is a few percent. With the controller open it will be higher, and that is fine as long as it is stable.
- Repeat monthly and keep the numbers with the dates. One reading is a snapshot; a trend is a diagnosis.
If you have not used a TDS meter before, the measuring guide covers rinsing, waiting for the number to settle, and calibration. Get those right or the trend will be noise.
Why the output creeps upward
A slow rise over months, with the feed unchanged, is the membrane ageing. Every membrane lets a little more salt through as it accumulates scale and its surface wears. That is normal, and it is why you keep the log: a membrane that started at 3% of feed and is now at 8% is still working, but it is on its way out.
A faster rise usually has a cause you can fix. Low pressure is the common one. Rejection depends on pressure pushing water through the membrane faster than salt can follow; when the booster pump weakens or a pre-filter clogs, pressure drops and more salt gets through. A blocked sediment cartridge is the first thing to check. Scale is the second. On feed above 2000 ppm, or on hard water at any TDS, calcium precipitates on the membrane surface and both flow and rejection fall together. Summer is the third: if the feed rose from 500 to 800 while the ratio stayed at 4%, the output went from 20 to 32 and nothing is wrong.
That last case is why you always compare output with feed rather than watching the output alone. Borewell feeds move with the season, and a purifier can look like it is failing when the well is simply saltier in May. The TDS reference page explains the seasonal rise.
Controller, or membrane?
Before blaming the membrane, look at the controller. On purifiers that have one, it is a small valve that lets some post-carbon water skip the membrane and rejoin the output. Turned up, it raises output TDS; if someone nudged it, or if it has drifted, the reading changes with the membrane untouched. Close it fully, run the purifier, and read the output again. That number is the membrane on its own. If it is a few percent of feed, the membrane is fine and the controller was the change.
If the purifier has no controller and the output is not a small fraction of feed, one of the seals is letting feed water round the membrane rather than through it. That can be the O-ring on the membrane housing or a membrane pushed in without its end seal seating. It shows as a sudden jump, not a creep, and the fix is to reseat rather than replace.
When the membrane is failing
Three signs together mean the membrane is done. The output has risen to a noticeable share of the feed with the controller closed. The flow has dropped, so the tank takes much longer to fill. And the numbers do not recover after a new sediment cartridge and a check of the pump. Any one of those alone has other explanations; all three point at the membrane.
Replace with the size that fits, which for most domestic purifiers is the 1812 form: the 80 GPD for a standard pump, the 100 GPD where the pump can drive it. On feed above 2000 ppm, fit the molded high-TDS membrane, rated to 3000 ppm at 96–98% rejection, and make sure the pump behind it — the LUX-PUMP-100 is rated 140 PSI — can supply the pressure high-TDS water needs. The filter change schedule gives the intervals for the other cartridges, and changing those on time is what keeps the membrane's numbers flat for longest.
Questions
What should the TDS of RO water be?
A small fraction of the feed, not a fixed figure. At 96 to 98 percent rejection a 500 ppm feed leaves the membrane at about 10 to 20 ppm. If a TDS controller is fitted, the tap reading will be higher by design; set it where the water tastes right and keep it stable.
My RO output TDS is 150 ppm. Is the membrane bad?
Not necessarily. If the feed is 3000 ppm that is 5 percent and healthy. If the feed is 400 ppm and the controller is shut, the membrane is passing far too much and is either scaled, bypassed at a seal, or worn out. Measure the feed before deciding.
Why is my RO output TDS almost the same as the tap?
Feed water is getting round the membrane rather than through it: a controller left fully open, an unseated membrane, or a failed housing seal. Close the controller, reseat the membrane and its O-rings, and read again. A membrane that has genuinely failed usually shows as a gradual rise, not a jump to feed level.