UBUHANZI
UBWIZA N'UBUZIMA
UBUKORIKORI
UMUCO N'AMATEKA
IMYIDAGADURO
IBIDUKIKIJE
IBIRIBWA N'IBINYOBWA
UBWUBATSI BUSUBIRA INYUMA
UBUMENYI
SIPORO
IKORANABUHANGA
IBYAMBARWA
Reverse Osmosis: Pushing Water Uphill Through a Membrane
Mary

Yaremwe na

Mary

23. Nzeli 2026FI
20
0
0
0
0

Reverse Osmosis: Pushing Water Uphill Through a Membrane

The solar still and the alembic both separate salt from water by evaporating the water — and both pay the latent heat of vaporisation to do it, about 2,400 kilojoules for every kilogram. That energy bill is why distillation stayed expensive for three thousand years. Reverse osmosis does the same separation without a phase change. Osmosis drives fresh water INTO brine through a membrane that passes water and rejects salt; push back harder than that osmotic pressure and the flow reverses, leaving the salt behind. The idea was old, but it was useless until Sidney Loeb and Srinivasa Sourirajan at UCLA made a cellulose acetate film in 1959 with a dense skin a fraction of a micrometre thick on a porous backing — thin enough to pass useful water, sound enough to hold back the salt (US 3,133,132). Every desalination plant since descends from that film.
Hejuru
1 day

Amabwiriza

1

Understand what you are pushing against

Osmotic pressure is not friction and it is not the membrane resisting you. It is the pressure the salt solution would pull water in with, and you must exceed it before a single drop crosses the other way. For sea water at 35 grams per litre that is about 27 bar. For the brackish water most small systems treat, a few bar. This one number decides whether your build is a bicycle pump or an industrial plant, and the notebook later works it out for your own water.

Ibikoresho by'iyi ntambwe:

UmunyuUmunyu35 g

Ibikoresho bikenewe:

Iminzani ya DigitaleIminzani ya Digitale
Igikoresho gipima ibitembaIgikoresho gipima ibitemba
2

Get a membrane and treat it gently

Use a small domestic RO element — the kind sold for under-sink filters. Note which way round it runs: feed in, permeate out of the centre tube, concentrate out of the far end. Two things kill a membrane quickly. Chlorine destroys polyamide film in hours, so anything chlorinated must go through carbon first — and note the irony, because the previous rung in this batch was about deliberately adding chlorine. And letting it dry out permanently collapses the pore structure: a membrane that has dried never recovers its flux.

Ibikoresho by'iyi ntambwe:

Urwungano rw'osimoze ihindukiyeUrwungano rw'osimoze ihindukiye1 igice
3

Build the pressure side

Plumb a feed vessel to the membrane through a pump with a gauge in the line, and lead two outlets away: permeate to one jar, concentrate to another through a restrictor. You need BOTH outlets flowing. A membrane run with the concentrate closed off becomes a dead end where salt piles up against the film — concentration polarisation — and the local osmotic pressure climbs until flow stops and scale forms. Every RO system in the world throws away a stream of concentrate; that is not waste, it is what carries the rejected salt out.

Ibikoresho by'iyi ntambwe:

Umuyoboro wa SilikoneUmuyoboro wa Silikone3 metre

Ibikoresho bikenewe:

Pompe yo hasi y'igarePompe yo hasi y'igare
Igipima UmuvudukoIgipima Umuvuduko
4

Make brackish water and run it

Dissolve 2 grams of salt per litre — brackish, not sea — and run it through at increasing pressure while you watch the permeate outlet. Below the osmotic pressure, nothing comes out. Cross it and permeate starts, slowly at first and then rising roughly in step with how far past it you push. Seeing that threshold with your own gauge is the whole point of building this: it is the moment an abstract equation becomes a thing you did with a pump.

Ibikoresho by'iyi ntambwe:

UmunyuUmunyu20 g
Amazi YatunganyijweAmazi Yatunganyijwe10 litiro

Ibikoresho bikenewe:

Igipima UmuvudukoIgipima Umuvuduko
Igikoresho gipima ibitembaIgikoresho gipima ibitemba
Isaha yo GuhagarikaIsaha yo Guhagarika
5

Rejection, flux and the energy bill

Gupakira ikaye ya Jupyter…

Ibikoresho bikenewe:

Igikoresho gipima ibitembaIgikoresho gipima ibitemba
Isaha yo GuhagarikaIsaha yo Guhagarika
6

Taste, test, and notice what is missing

Compare feed, permeate and concentrate. The permeate should read near neutral and taste flat; the concentrate should be noticeably saltier than what you started with, because the salt has to go somewhere. RO strips almost everything — salts, metals, most organics, and microbes, which are enormous compared to the pores. That completeness is also the criticism: it removes the calcium and magnesium the body uses, which is why large plants remineralise afterwards rather than delivering the pure product.

Ibikoresho bikenewe:

Impapuro zo gupima pHImpapuro zo gupima pH
Icupa ry'IkirahureIcupa ry'Ikirahure
7

Flush it, and never let it dry

Flush with clean water at low pressure for a few minutes after every run, to wash the concentrated layer off the film before it can crystallise. For storage, keep the element wet and cool. If you will not use it for weeks, store it in the preservative solution it came in. A dried-out membrane looks identical to a working one, and behaves like a blank sheet of plastic — the most expensive way there is to learn this lesson.

Ibikoresho by'iyi ntambwe:

Amazi YatunganyijweAmazi Yatunganyijwe5 litiro

Ibikoresho bikenewe:

Pompe yo hasi y'igarePompe yo hasi y'igare

Ibikoresho

4

Ibikoresho bikenewe

7

CC0 Umurenge rusange

Iyi blueprint yasohowe munsi ya CC0. Ushobora gukoporora, guhindura, gukwirakwiza no gukoresha nta kwemererwa.

Shyigikira Umuremyi ugura ibicuruzwa binyuze muri Blueprint ye Komisiyo y'Umuremyi byashyizweho n'Abacuruzi, cyangwa kora verisiyo nshya y'iyi Blueprint ukayinjiza nk'isano muri Blueprint yawe kugira ngo musangire inyungu.

Ibiganiro

(0)

Injira kugira ngo ujye mu biganiro

Gutegura ibitekerezo...