This morning I decided to do a little experiment. I have a little apothecary balance a pharmacist gave me many years ago after it had become archaic. I often use it to weigh small samples. I could not find my standard weights this morning so I used old grain-scruple-dram weights. Suddenly I was back in the early 20th century!

I weighed out half a grain of a mixture of KCl and NaCl. This is about 30 mg in modern units, 15 mg of each salt. I put this into an espresso cup, and then I put a second espresso cup to its right on the dining table. I started my faithful 1999 La Pavoni Europiccola and made two shots from a coffee I really like into a pitcher, then I split this espresso between the two cups. I stirred them both, just to treat them the same, and then tasted them. My goal was to see just how much the addition of Na and K influenced the taste. I had roughly 30 mL of espresso in the two cups.

I then would take a sip from one cup, then sip some water and then sip the next.

I was surprised at just how little the presence of the salts influenced the taste. I had difficulty detecting any difference, and this espresso had roughly 500 mg/Litre of each salt. I once tried a similar experiment with MgSO4 with similar results. When I used a glutamic acid salt--something for which we have a hypersensitive detector the change in flavour was extreme, and the change was, to me at least, NOT for the better.

The espresso with the salts had much higher concentration of Na and K than almost all tap water. Some claim that ions aid in extraction of the coffee. That seems to make little chemical sense because the ions are present in low concentrations and they would have to form very strong complexes to do this.

If one look into the boilers of my espresso machines they are spotless. They are never descaled, and they never have any water put into them except distilled.

You did the experiment backwards. The main function of the minerals in coffee brewing is to aid in extraction. Some flavor elements will not extract properly without minerals in the brew water before it hits the grounds.

For the record I just did the "real life" experiment. I have two Europiccolas. I made espresso with Girard, Kansas, city water, which is so hard that simply boiling water in a sauce pan will leave a deposit of CaCO3 on it! At the same time I made it using distilled water. I used the identical Nicaragua coffee. There was lots of crema with both water samples. I even put the test samples in identical cups!

When I tasted them there seemed no difference at all. Others should try this for themselves.

The explanation that the ions in the water improve extraction make no sense at all to me. Although transition metals bind to many ligands very strongly, Na, K, Mg, and Ca tend not to form especially strong complexes. The doubly charged ions Mg2+ and Ca2+ do form them with a few special bidentate ligands like EDTA.

If anything the infamous "salting out" effect should reduce the solubility of coffee components when there are lots of ions in it. If one put potassium carbonate in strong distilled alcoholic beverages the solubility of alcohol in the water is reduced so much that the alcohol separates from the water!

I did a couple of less controlled experiments a while back. It will always be distilled water for me!

Again I think people owe it to themselves to try an experiment like this. Put a hidden label under the cups after making the espresso, and move them around so you forget which is which. See if this results in your finding one horrible and the other wonderful. Then look at the label. Do not trust me or anyone else! Check for yourself!