Just to give a full summary of the shared info :
- RO systems (0.0001 microns) might get the TDS or Total Dissolved Solids to too low a concentration for drinking. Some systems which does ion-exchange resins filtration (eg Zerowater) might also get it a wee bit too low. WHO and other organisations have put warned about the long term health risks regarding this. When such overly soft ultra low TDS water is used in cooking, it's been found that micronutrients loss from the food is quite high.
http://www.who.int/water_sanitation_health/dwq/nutrientschap12.pdf
It's probably not a big problem for most of us, but it's good to be mindful, as different people different health and different diets/intake of micronutrients.
- As far as i know, the only health risk is with lead leached from brass plumbing, even the "lead-free" versions that's sourced locally conforming to PUB regulations and standard code of practice are still found to be leaching lead somewhat. It exists in both insoluble form as particulates as well as soluble form as a heavy metal ion. Some people may want to filter chlorine and chloramines too, that's fine as it usually improves taste. A solid carbon block or even GAC (granular activated charcoal) filter improves the taste and can reduce chlorine/chloramines. As far as possible, keep the flowrate down as it helps performance as well as filter life. At our end users tap, we might have somewhat higher levels of Cu from the indoor copper pipes (if that's being used by your plumber, but not necessarily so). But that's usually ok as it only presents an issue at very high concentrations. As we are not an agricultural country here, i don't think we have herbicides risks, nor do we have risks like mercury, arsenic etc.
- for Iron oxide (the brown stuff) particulates, it's easily taken care by even a 1 micron PP sediment filter. Your 0.5 microns solid carbon block easily provides mechanical filtration for it.
You do not need a RO or even ultrafiltration (0.01 microns) for this, just totally unnecessary. UF might filter most of the viruses out there, but the thing is that our chloramine in the water does have much better persistence than chlorine in our pipes and it does keep the water safe enough to drink. And anyway it may be better to boil the water still, just in case you have secondary biological contamination due to bacteria biofilm growing on the post filter surfaces. (not necessary if you change your filter regularly as recommended by the manufacturers). All surfaces that are usually kept wet have such risks, not rocket science.
Ultrafiltration 0.01 microns does not filter soluble lead, the lead dissolved in water as an ion. You'd need RO 0.0001 microns for that as the stuff is just too small. That is also why RO systems reduce the TDS by a lot, by name it already says it, Total DISSOLVED Solids.
- It's best to have NSF/ANSI certification for the product that you are getting. Usually NSF/ANSI 53 is the standard that you are looking for (not NSF 41 which is only for material safety). NSF/ANSI 401 is not necessary, testing for emerging contaminants. A rare few are tested by the WQA, that's fine too. Look for lead reduction specifically, for NSF 53 standard it's very tight, it states a minimum of 99% reduction and at twice the rated volume spec i think (for buffer), so if for a say 100 gallon/378 litre catridge, you are very safe even when you are using till 350-400 litres range.
- NSF certified devices are not expensive. Eg the cheapest NSF 53 device that i could find is the Dupont premier water faucet filter on Amazon. It's usd 13 for the device which comes with 1 filter. Filter replacements are at usd 16.99 for a 3 pack of 100 gal rated cartridges.
If you are changing at a frequency of one cartridge every 3 months (378 litres every 3 months, you can do 3.78 litres of water daily, shd be enough even for a 4 pax family unless all are staying at home and not working/schooling), the cost is just 8 Singapore cents per day. Assuming that you are taking advantage of Amazon's usd 125 free shipping Free AmazonGlobal shipping.
- The size of the filter device or filter cartridge usually does not determine filtration performance. To a certain extent, it might determine the capacity of water filtered. I feel that this is important to point out, a small tap faucet filter could easily outperform the lower tier undersink 10-12 inch cartridges systems. This is due to the many different qualities of solid carbon blocks.
However, some people may have issues with faucet filters relatively low flowrates.