Water Questions & Reports

[rolex]

Hello all, I was directed to this forum for espresso machine water recipe questions

My local store has self serve options for

Spring Water (Source: Formosa, ON. pH: 8.33. TDS: ~260 PPM+)

or Reverse Osmosis or Distilled (pH: ~6.6-6.9. RO TDS: 11-20 PPM)

My understanding is that any of these 3 choices on their own is bad for the machine for their own reasons

My plan of attack was to fill some 18L big jugs with a blend of distilled+spring or RO+spring until the TDS reaches an appropriate level using their meter, I called them already and they are able to assist!

My questions are

1. What is my target water PPM? Specifically for espresso machines (Currently an old Lelit Mara but upgrading to a La Marzocco this year)

2. Is RO or Distilled plus a specific additive more desirable than bringing up the TDS of either of those with good quality spring water? If so, which of the two, and why?

3. If my idea will work, would distilled or reverse osmosis be best to make up the majority of the water blend? Or is there ultimately no difference as they're similar?

4. If anyone is able to click the two reports I linked to above, I think all relevant information is on the first or second page.. anything look bad?

Much appreciated looking forward to learning!

[homeburrero]

What is my target water PPM? Specifically for espresso machines (Currently an old Lelit Mara but upgrading to a La Marzocco this year)

TDS readings are not the best way to choose water for use in an espresso machine. You can have an 'ideal' TDS and a terrible water. In this case, where you are using a water with high carbonate hardness and very low undesirables (chloride, sulfate, silica) I think you could use TDS reading as a good enough guide. If you were to dilute that Formosa spring water to 20% (the rest being RO or de-ionized) you would come out at roughly:

• TDS: 51 ppm (might read about 40 ppm on an inexpensive NaCl calibrated conductivity meter)
• Alkalinity: 46 ppm as CaCO3
• Calcium hardness: 31 ppm as CaCO3
• Total hardness: 52 ppm as CaCO3

That gives you a near zero LSI at 120 ℃ (i.e., should not create much if any limescale), and has near ideal alkalinity for corrosion protection.

Is RO or Distilled plus a specific additive more desirable than bringing up the TDS of either of those with good quality spring water? If so, which of the two, and why?

The advantage of using a recipe like rpavlis (distilled plus a small amount of potassium or sodium bicarbonate) is that you would have no chance of limescale accumulation even if you run a hot steam boiler and fail to periodically flush it. The advantage of the Formosa spring water mix is that it contains hardness minerals (calcium and magnesium carbonates) that some believe to be an aid in flavor extraction.

If my idea will work, would distilled or reverse osmosis be best to make up the majority of the water blend? Or is there ultimately no difference as they're similar?

You can use either, the 10 - 20 ppm of mineral in the RO is not that significant. The de-ionized is comparable to distilled. If price is the same, go with de-ionized or distilled.

f anyone is able to click the two reports I linked to above, I think all relevant information is on the first or second page.. anything look bad?

That's an impressive analysis report, and that Formosa spring water looks like an excellent clean source of calcium and magnesium carbonates. If you were to use it straight of course it would create limescale problems.

[rolex (original poster)]

Amazing!! That's a great reply, and very informative and answers my questions! Thanks very very much. I tend to think using the Formosa Spring water blend would be ideal for me as I definitely prefer the taste of spring water when possible. I'm going to give it a shot and if I think of anything else I'll be sure to check back, thanks again have a wonderful Halloween!

[JamesB517]

Adding Third Wave to distilled or reverse osmosis is my go-to. You just put a packet into a gallon and shake. If they have only 5 gallon containers, I am sure you can just do 5 packets.

[rolex (original poster)]

Hmm, I am unsure if those are available here. I have seen them mentioned on Reddit or similar forums before occasionally though

I do have another head scratcher back related to my OP, though!

So I visited the water store. Picked up both some RO and some Spring Water. Their lab tested numbers are supposed to be under 20ppm for their RO which both of my meters report a match, around 10ppm or less. But it all goes haywire from here!

Multiple spring waters now with confirmed label PPM under 300 are testing as 500+ on my cheap meters. The Formosa. Nestle Pure Life. And Kirkland Costco. All of these are reporting around the 500 mark which totally contradicts their own labels.

Furthermore, if I add 400ml of RO to a 500ml vessel with the intention of adding 100ml of spring as discussed earlier, by the time a mere ~30ml of spring is added, my cheap meters show a new TDS reading of 100ppm. If I added the full 100ml to reach a 20% ratio as discussed above, the meter will read nearly 300ppm!

Is this simply a malfunction of the meters, and I should trust the actual labels and stick to a 400ml RO to 100ml Spring ratio, even if my meter is wigging out and reporting it higher?

Or is the meter somehow more trustworthy and it's more likely that the lab results on these 3 Spring Water varieties are wrong?

Thanks again for your help, hope this made sense

[DeuxInfuso]

The cheap meters are not temperature compensated. Take your readings at 25 C and the TDS estimate will be more accurate. You are really measuring EC, electrical conductivity, anr that is affected by the specific cation/anion composition of your water, plus temperature. Typical natural waters have an EC/TDS ratio (coefficient) between 0.5 to 0.7, and EC rises with temperature because of ion diffusion coefficients and the viscosity of water.

[DeuxInfuso]

By the way, EC goes up about 2% for each degree Celsius, at Temps near 25 C anyhow (it increases faster at higher temps), but your water samples would have to be hot, i think, to get such high TDS numbers for typical bottled water TDS levels, thus it may be as you suggest, that your meter is wigging out! Recommend a better meter! I'm happy with an HM Digital COM-100, good instrument, about $65, lasts for years if you don't drop it on a tile floor!

[homeburrero]

Or is the meter somehow more trustworthy and it's more likely that the lab results on these 3 Spring Water varieties are wrong?

It seems unlikely that two different TDS meters would be that far off the mark but of course possible. One thing that might cause that is if someone were to mis-calibrate both those meters. (Most have a calibration screw that's easily accessed.)


The Formosa spring water analysis report looks very reliable to me, but it's possible that the water you bought was not actually from that source. It's very common for bottled spring water to come from a variety of water sources. That's the case, I believe for the Pure Life Spring water - - If you look at the 2021 analysis for this water: https://www.drinkpurelifewater.ca/, the report indicates a TDS of 54 - 580 ppm.

Crystal Geyser is a good example of a natural spring water that comes from 7 different sources, but in their case you can identify the source by checking the fine print on the bottle, and then look on their FAQ page to find an analysis for any particular source. The Crystal Geyser reports are nice because they also tell you the conductivity. For example the Johnstown NY source lists a TDS of 160 ppm and a specific conductance 390 of umho/cm. So you can use it to sanity check your meter - - using the common factor of 0.5 for the inexpensive TDS meters you know that it should indicate about 195 ppm when read at 25 ℃. (As explained earlier these things are calibrated at 25 ℃, and even if the meter does temp correction it's best to try to read them with the sample at that temperature.)