Making water for Coffee Brewing | How to (kinda) fix soft water with Salts


 In my opinion. it’s very hard to get a good espresso or filter coffee in Glasgow. A big part of this problem lies with the coffee shops here generally not being as progressive or quality focused as their counterparts in other cities, yet. The other part is the water. Glasgow has some of the softest water in the world, around 29ppm on a typical day is all you will get from your tap water. I’m not going to cover why this is a problem, since water has very much been the hot topic in the industry the last two years, I’ll instead just cover what I’ve looked at to fix the issue.

(If you want to skip ahead I start talking about the solution below the dashed line after waffling on about how I found out about salts.)

  I was lucky enough to go to one of Maxwell Colonna-Dashwoods talks at the start of the year before the release of his book, Water for Coffee. After the talk I asked him about practical solutions to soft water. He didn’t have as much experience with it at the time since most of England suffers from hard-water. He did instead direct me to The Cupping Room in Hong Kong which has a similar issue. What they have done is half genius and half mad. They use what I would call reverse reverse-osmosis.

 Reverse-Osmosis is the process of splitting a hard water into purified water and super-hard water. Then some of the super-hard water is added back in to get the desired hardness. It’s what you’ll see in most coffee shops in London as the solution to their hard water, however it is costly to buy, run, and is very wasteful of water. From what I understand, the guys at The Cupping Room dump the purified water and continually filter more mineralized water until they have water hard enough to use. As I said this is really smart but also crazy, their electricity bill must be huge and they will be wasting far more water than they’ll be using. Both the cost and environmental impact of this solution meant it wasn’t suitable for us to use so we kept looking for alternatives.

Enter Rob Ashton. Rob, for those unaware, is a basket case who has turned his obsessive perfectionism and geekery to the coffee industry and is fast becoming a key player. He’s certainly a major force in improving Glasgow’s coffee scene. It was at Rob’s house, Cafe Ashton as it is better known, that I first got to cup coffee made with different waters vs our tap water.

We used Workshop’s Gachatha, a Kenyan, brewed with tap water (29ppm), Calcium rich water (120ppm), Magnesium rich water (120ppm) and a 50/50 split of Mg & Ca (120ppm). Trying the tap water first I could taste the usual signature I’ve come to expect of Workshop when I’ve brewed it in the past, very complex but lacking distinct flavours. The other three were wildly different, suddenly there was incredible sweetness and flavours of mango and papaya along with a fuller body, we were getting all the amazing tasting notes the bag promised but usually did not deliver. Going back to the tap water suddenly it tasted flat and acrid, devoid of all positive flavours. I cannot emphasize enough what a massive difference this was. The decision was made, I would never use tap water to brew again.

 Through his time spent on forums and blogs Rob was able to point me in the direction of a home barista who has done an impressive volume of work on making water for coffee. Spence, the man behind GrindScience, talks through three different methods of making water on his blog in great detail. 

 At this point I would urge you to go and read his post on making water then come back and I will tell you why I think method 2, using Salts, is the best.

Here’s the link to the first article in the series Making Water for Coffee


 All done? Good. I hope you message Spence to thank him for all the incredible work he has done, it certainly helped me a great deal.

 Now Spence obviously makes a clear case for the sodastream method, which is also what Rob used in his house for the cupping, however after using this for a few of our pop-ups we realized how impractical it was on a commercial scale. Not only is it costly but you also have to spend several hours mixing and filtering waters, you can’t be doing that in a coffee shop every day.

 Having given up on that we then opted to try his previous method, making salt solutions then adding this to the source water. I’m pleased to announce we’ve had great success so far and it’s dirt cheap, easy, and effective. Each of the ingredients can be purchased for around £5 per 500g on-line and this will last you a very long time.

What I first did was to make three solutions each at 10g/L concentrations.

  • For Calcium I use Calcium Sulphate Dihydrate AKA Gypsum Salts
  • For Magnesium I use Magnesium Sulphate AKA Epsom Salts
  • For Bicarbonate I use Sodium Bicarbonate AKA Baking Soda

 Through experimentation with our tap water we cupped different combinations to see what generally worked best. Maxwell in his book suggests a 2:1 ration of Calcium and Magnesium to Bicarbonate for the best balance, he also suggests 120ppm is the optimum hardness for taste and machinery health. Magnesium affects acidity and sweetness whilst calcium gives body (It also gives unhealthy deposits inside espresso machines so we tend to favour magnesium) Once we found a balance that suited our source water best we then made another solution with the three minerals present in the right ratios.

 In our specific case this is 1L of tap water with 7g Magnesium salts, 3g Calcium salts and 1g Bicarbonate. When we want to brew a filter coffee we take our kettle and fill it with 1000g of water then add 22g of the solution to bring us to 120ppm. Simple. Or maybe not…

 There are some big drawbacks to this method, most notably for espresso. You’ll need to run your machine from a container of water rather than the mains, meaning you’ll need space for a 25L+ container under your counter. This also has to be filled every day which becomes laborious. The other issue was covered by Spence and that is that the maths don’t scale up correctly. So 22g makes 1L of tap water 120ppm but 44g won’t make 2L of water 120ppm, you’ll need to adjust constantly, especially when making a big batch to last the day.

 Other issues that people have brought up with this method is that there is a “salty” or chemical taste and that there may be some health concerns. In terms of taste I did think it tasted a little weird at first when cupping when they were very hot. However I’ve also experienced this with the water Rob made with the Sodastream method. I haven’t though ever experienced it when the coffee has been brewed, only during cupping.  As for the health concerns the levels of all the added minerals (magnesium, calcium, sulphates and chlorides) are all far below the recommended safe level. Even at 120ppm they are still present in much lower quantities than you’ll find in most bottled waters.
 There you go, slightly long winded but that is our current solution to the water issue. It’s obviously a very specific one that fits our water and is by no means perfect. If you have a heavier water then you may still be able to improve it with salts, you’ll probably need to go over the recommended 120ppm but its the 2:1 ratio of Calcium + Magnesium to Bicarbonate that’s the important factor. Just be sure to thoroughly test the water out before you think about serving it, especially for safety concerns, and check with your engineer on its possible effect on your equipment.

Happy Salting


7 thoughts on “Making water for Coffee Brewing | How to (kinda) fix soft water with Salts

  1. Spence says:

    Hi James,

    thanks for the reference, glad you found the info useful. I completely agree that in large volumes that the SodaStream method is impractical at best. You do also have to watch the bicarb levels and due it it’s very nature the water is itself changes over time.

    Using salts gives you results that are better than soft tap water, it’s easy to achieve and is very consistent.

    Hopefully someone can come up with a decent re-mineralisation filter for coffee that will save us all time and effort 🙂




  2. Andy says:

    Thanks for the further research and pointers on this, James! One quick correction…you’ve transposed epsom and gypsum salts. Epsom is magnesium chloride, gypsum is calcium.


    1. Spence says:

      Hey Andy,

      Just to ensure accuracy Epsom Salts are actually Magnesium Sulphate Heptahydrate (MgSO4.7H2O) and Gypsum is Calcium Sulphate Dihydrate (CaSO4.2H2O).




      1. Andy says:

        D’oh. Right you are. That’ll teach me to type faster than I’m thinking.

        I guess that’s the chemical equivalent of Muphry’s/Skitt’s Law…


  3. James says:

    OK re-reading it again and I think it is making sense for me now. What initially threw me was that with Spence’s method, you make your initial 3 solutions (albeit not at 10g/L) and you then add them, in varying quantities to a litre of water to make your brew water. Your method has an extra step, in that you add your 22g of ‘combination solution’ to water to make your brew water. Or at least, that’s how I think I understand it. Working from Spence’s method I ended up with water at 190TDS (which is not right) and it tasted as if I had totally overextracted the coffee. There was also a very unpleasant ‘twang’ to the water. Clearly something went awry. I will follow your instructions and report back.


    1. James Wallace says:

      Well what I did orginally was use Spence ‘s method of a mixture of three solutions to test out what combination tasted best. I then picked the “mix” or combination that I preferred of these.

      To simplify I made this “mix” up in one solution. This way when I want to make 1L of brew water I can add 22g of this one mix rather than three different amounts of three solutions. It’s easier to make larger batches this way since the minerals are in the solution already at the ratio (to one another) I want them to be so, I just need to add the solution gradually whilst measuring to reach my target TDS.

      This is how I’ve been making all my water for pop-ups the last few months and it has been pretty reliable. So al the coffees I’ve given you have been made with this method, did you think any of them had a twang?


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s