Over the last decades, there’s been this trend of telling people that they should try eating “alkaline foods” and drink alkaline water and alkaline teas. Here in Germany, there is a choice of teas labeled Basischer Tee, Basischer Kräutertee, Basentee, or Basen Kräutertee.

How about the coffee? I wanted to make a quick investigation on the matter, and here’s what I learned in the process.

The pH measuring device

The most popular pH measuring device is this kind of generic Chinese pen that can be purchased online for random prices in the €5-€15 bracket (VAT-inclusive).

I got one for less than €8, shipping included. So, regardless of the outcome, at least the cost was symbolic.

Such devices suck big time, but they can still be useful. But let’s not put the cart before the horse.

1. The calibration

The way they come, such devices cannot be used right away. They require calibration. This is not because of careless, cheap manufacturing, but for a little-known and even lesser understood reason that I will reveal later. For now, I’ll stick to practicalities.

Ideally, such a device is calibrated in 3 points, using buffer powder sachets that you dissolve in distilled water:

  • Potassium acid phthalate → pH 4.005 at 25°C, usually labeled “4.01” but sometimes “4.00”
  • Dipotassium phosphate + monopotassium phosphate → pH 6.86 at 25°C
  • Sodium tetraborate (borax) → pH 9.18 at 25°C

In practice, the cheapest devices come with only two calibration sachets, for 4.01 and 6.86.

Note that:

  • You must follow the exact calibration procedure specified in the accompanying leaflet. There’s no way to calibrate such a device by inserting it in a solution with a known pH and then telling it, “this is the correct value.” As there’s no way to enter any digit, the calibration procedure is hardcoded in the device: first for pH 6.86, then for pH 4.01, in this exact order!
  • The calibration should be performed with each solution having exactly at 25°C. However, at 20°C is also fine for such purposes, as the pH deviation will not exceed 0.02.
  • The pamphlet would tell you to use distilled water, but such water is typically slightly acidic. Some other instructions say deionized water, which is better, but its pH also drops in time as it absorbs CO₂. For our purpose, Brita-filtered water is a good practical choice, but it will often be slightly acidic, too. Typically, tap water filtered through a Brita filter in the middle of its use window (say, after 2 weeks of use) has a good chance to reach pH 7.0. A newly changed filter would be too effective in removing the calcium and magnesium ions, so the resulting water could be as acidic as 5.5-6.5. On the other hand, right after the filtration, CO₂ is absorbed from the air. Just let the filtered water sit for a while, and the excess CO₂ should escape back into the air.

Now, some people complained (on Amazon) that the device is crappy. It is so, but they complained from day one. Unless they didn’t calibrate it at all, here’s what happened:

  • Calibration is not as simple as it seems, and they didn’t double- and triple-check.
  • There is a kinky reason for that (I’ll come to it later).

The calibration involves these steps:

  • Dilute each sachet in water, and make sure you know which recipient holds which buffer solution.
  • Immerse the probe end in the pH 6.86 solution, wait until the reading stabilizes (it could take 5-10 seconds or more!), then press the “CAL” button for 5 seconds, then release it. The screen will flash “6.86” and eventually display “End” (and stop flashing).
  • Clean the probe end with distilled or filtered water, shake it, then wipe it with a paper towel or a cosmetic tissue. Make sure you don’t press any button in the process!
  • Immerse the probe end in the pH 4.01 solution and repeat the procedure.
  • In theory, the process should be completed.

Well, it is, but maybe it isn’t. There are high chances that you need to repeat the procedure one or two more times! Signs that you need to do it:

  • Subsequent measurements of the same buffer solutions don’t give the expected readings anymore.

You don’t need any other sign.

And here’s why you need to repeat the procedure (just trust me):

  • The electrode came dry from the factory.
  • It needs to stay in the pH 4.01 solution for more time.
  • As you can’t repeat the calibration for pH 4.01 only, you need to calibrate again using both solutions.

So don’t get rid of the two solutions too soon!

2. The working principle

What you can see at the measuring end of the device is highly confusing:

Gemini to the rescue:

It’s perfectly normal to be confused! A pH meter looks like a simple digital thermometer, but it’s actually a sophisticated electrochemical sensor.

To measure pH, the device needs to complete an electrical circuit through the water. Those two components you spotted are the two halves of that circuit: the Glass Electrode (the bulb) and the Reference Electrode (the felt tip).

1. The Glass Bulb (The “Sensor”)

The transparent element with the wire inside is the measuring electrode.

  • The Glass: It isn’t just regular glass; it is a special, ion-sensitive glass membrane. When you dip it in water, a very thin “hydrated layer” forms on the outside of the glass.
  • The Reaction: Hydrogen ions (H+) in your aquarium water interact with this glass layer. If the water is acidic, more ions “clump” onto the glass; if it’s alkaline, fewer do.
  • The Wire: Inside the bulb is a stable liquid (usually pH = 7 buffer) and a silver wire. The wire detects the electrical potential (voltage) created by the ions sitting on the outside of the glass.

2. The “Felt Tip” (The Reference)

The little white spot that looks like a felt tip or a ceramic wick is called the junction. This is the part most people overlook, but it’s where most pH meters fail.

  • The Purpose: For the meter to measure the voltage at the glass bulb, it needs a stable “ground” or baseline to compare it against.
  • The Mechanism: Inside the yellow body, there is a second chamber filled with a salt solution (usually Potassium Chloride, KCl). The “felt tip” is actually porous. It allows a tiny, slow leak of that salt solution into the aquarium water.
  • Closing the Circuit: This leak creates a liquid bridge, completing the electrical path from the bulb, through the water, into the felt tip, and back to the meter’s processor.

Why this matters for maintenance

Understanding these two parts explains why these pens are so finicky:

  1. Don’t Let it Dry Out: If that “felt tip” dries out, the salt crystals clog the pores. The circuit breaks, and your meter will show “fixed” or wildly jumping numbers. Always keep the cap moist with storage solution (KCI).
  2. The “Pin” is Fragile: The glass bulb is extremely thin to allow ion interaction. Never touch it with your fingers or a cloth; skin oils or scratches will ruin the accuracy.
  3. Calibration: Over time, the chemistry inside the “felt tip” and the bulb changes. This is why you use the “CAL” button with those powder buffers—you are telling the internal computer: “Whatever voltage you feel right now, call it pH 6.86.”

Unexpected, eh?

3. The tragedy

My adventures with this device went as follows: I happily used it to measure the pH of tap water, Brita-filtered water, and various coffee preparations.

Then, about 3 days later, the catastrophe happened!

  • Upon starting the device, after the initial “0.00” reading, if I waited enough, the device would display random values in quick succession!
  • Trying to measure the coffee gave wrong values (too low, meaning too acidic).
  • Trying to measure bottled water with known pH also gave too low values (at least 0.5-0.7 too low).

Shutting down and restarting the device didn’t help.

Gemini explained, in the continuation of the same thread:

Ludditus: “Always keep the cap moist with storage solution (KCI).” No such solution is delivered with such devices. They are dry. The included flyer never mentions KCl. After each measurement, you’re supposed to clean the device with water, but of course it will soon dry up!

Gemini: You’re absolutely right to call that out—and it is one of the most frustrating things about these “budget” yellow pH pens. The manual often leaves out the most critical part of the sensor’s “long-term survival” guide.

The reason they are delivered dry is simple: cost. Shipping a liquid-filled cap requires a leak-proof seal and more expensive packaging. Since these are mass-produced, they ship them dry and hope the glass membrane survives the shelf life.

Here is the “unspoken truth” about why your flyer might be misleading and how to actually keep the device from dying in a month:

1. The “Clean with Water” Trap

The flyer tells you to rinse with tap or distilled water to remove aquarium gunk. This is good for cleaning, but it is terrible for storage.

  • The Problem: If you store the probe in distilled/deionized water, it actually “leaches” the ions out of the glass bulb and the reference “felt tip,” making the sensor sluggish and eventually non-functional.
  • The Result: This is why many people find these pens work great for two weeks and then suddenly won’t calibrate or give wild readings.

2. Why the KCl (Potassium Chloride) Matters

The “felt tip” you saw is a porous bridge to an internal salt solution.

  • If it dries out completely, the salt inside crystallizes and blocks the “pores.”
  • Once those pores are blocked, the electrical circuit is broken, and the screen will just bounce around randomly.

3. How to Hack It (If you don’t have KCl)

Since your kit didn’t come with storage solution, here is how to handle it like a pro without buying extra chemicals:

  • The Sponge Trick: Look inside the clear/black cap. If there isn’t a tiny piece of sponge there, cut a tiny sliver of a clean kitchen sponge and stuff it into the bottom of the cap.
  • The Liquid: Dampen (don’t soak) that sponge with pH 4.0 calibration buffer (the red solution you make with the powder).
    • Why? It’s acidic and stable, which keeps the glass membrane “conditioned” better than plain tap water.
    • Note: Never use the pH 9.18 (blue) buffer for storage; it will degrade the glass over time.
  • The “Never Dry” Rule: Before you snap the cap back on, ensure the bulb and the felt tip will be in a humid environment. It doesn’t need to be submerged, just kept in a “high humidity” zone.

4. If it’s already “acting weird”

If your readings are jumping or taking forever to settle, your probe has likely dried out too much. You can often “revive” it by soaking the tip in 1-2 inches of your pH 4.0 buffer solution for 24 hours. This rehydrates the glass membrane and the reference junction.

So you know all that needs to be known.

Such devices are dirt cheap.

The Chinese manufacturers never bother to ship some KCl powder. Gemini exaggerates when it mentions the shipping of a liquid-filled cap. A brand-new, never-used device can be shipped dry, provided that some KCl powder is also shipped so that the user could prepare the necessary solution! Then, of course, the design of the device should accommodate the storing of the probe in the storage solution.

Remember what it said: “If it dries out completely, the salt inside crystallizes and blocks the pores.” But a never-used device didn’t touch any liquid, so there’s no salt to crystallize!

Once you start using the device, you actually start its very short useful life… 🪦

At the very least, those bastards could have added some more sachets with pH 4.0 and instructed the user to utilize them to prolong the device’s life.

But for €8, it was a useful experience, because I was able to make some measurements. The results of this “study” are shown below.

Water, coffee, and measurements

The pH of the resulting coffee depends on several factors, including the water, the coffee, and the preparation method.

But what pH should we expect? Pardon my French, but Wikipedia is eating shit:

  • Gastric acid: 1.0–1.5
  • Vinegar: 2–3
  • Orange juice: 3.3–4.2
  • Black coffee: 5–5.03 (Such precision is absurd! Just write ~5 and call it a day! But for what dilution is this value? Are we measuring the powder?!)
  • Milk: 6.5–6.8
  • Pure water at 25°C: 7
  • Sea water: 7.5–8.4
  • Ammonia: 11.0–11.5
  • Bleach: 12.5
  • Lye: 14 (False! Sodium hydroxide aka lye aka caustic soda has 13.9.)

Much more useful is this picture:

The water

Water poses a specific conundrum: which of the possible sources of water to use?

  • Tap water (where drinkable) leads to scale deposits on the coffee preparation device, which will require periodic descaling using citric acid.
  • Tap water filtered by Brita or a similar carbon-based filter has some particulates removed, including some of the salts of calcium and magnesium that are responsible for the limescale deposits, but it’s more acidic than non-filtered tap water. Reverse osmosis filters are to be avoided, as the water is “too distilled” and may become corrosive!
  • Bottled still water, technically the safest choice, can be very mineralized, possibly even more than the tap water. It can also be very low in minerals, especially those labeled as “Suitable for baby food preparation” (“für die Zubereitung von Säuglingsnahrung geeignet”). The pH can vary widely, but most still waters tend to be alkaline, while all carbonated waters are acidic.

Basically, filtered (Brita, not reverse osmosis!) tap water is usually recommended to those who want to reduce the frequency of the periodic descaling of their espresso machines, whereas bottled still water within specific ranges of pH and mineral salts content is often recommended by baristas.

For instance, the Specialty Coffee Association (SCA) recommends water with:

  • Total dissolved solids (TDS): 100–150 ppm
  • Calcium hardness: 50–175 ppm
  • Alkalinity: 40–75 ppm (<80 ppm CaCO₃)
  • pH: 6.5–7.5

On the practical side, the choice of a bottled still water depends on your country of residence (or Land, province, or state), which dictates the availability of one brand or another.

In Germany (more precisely, in the Stuttgart area), most water brands are easily available in carbonated form, but fewer of them can be found non-carbonated. What’s worse, most waters have a high content of SO₂, which is passable in a carbonated water but tastes unpleasant in a still one.

Given that most water brands in Germany do not specify the pH, the following quick list is affected by this lack of transparency:

  • My preferred water is Fonte Tavina naturale (imported from Italy): pH 7.8 (value taken from here)
  • A “luxury” choice is Acqua Panna (also imported from Italy): pH 8.0
  • Adelholzener Naturell seems great, but the pH is not from official sources: pH 8.45 (estimated)
  • Waters to avoid, because they are not alkaline enough:
  • To avoid completely, as they are acidic even in still form:
    • Gerolsteiner Naturell, pH 5.9
    • Teinacher Naturell, pH 6.8

In Romania, where I performed this test, the choice seems to be easier:

  • Aquavia: pH 9.4 (not my first choice, though)
  • AQUA Carpatica still: pH 8.2
  • Carpatina (Băile Herculane): pH 8.08
  • Dorna Izvorul Alb: pH 7.9
  • Aquatique (Bușteni): pH 7.9
  • Borsec still: pH 7.8
  • Not alkaline enough:
    • Bucovina still: pH 7.31
    • Artesia (the still form of Tușnad): pH 7.03

Cool, I listed 8 brands for each country! 🙂

The coffee

This is where things get more complicated, and I try to avoid the various propaganda and misinformation that can be found online. Let’s keep it simple.

  • 100% Arabica is to be preferred to Arabica/Robusta blends, if not because of the way Robusta smells, then because it contains roughly half the caffeine levels of 100% Robusta.
  • The dark (sic) side of this: 100% Arabica also has roughly twice as much citric acid as 100% Robusta! Fortunately, citric acid is only one of the many acids in coffee. One study claims that the overall acidity is slightly higher for Arabica than for Robusta, but the two cited sources are quite divergent (a lower value = more acidic):
    • Arabica: pH 4.85–5.15 vs. Robusta: pH 5.25–5.50
    • Arabica: pH 5.0–6.1 vs. Robusta: 5.40–5.50
  • However, I would avoid the so-called “specialty coffee roasters” (Kaffeemanufakturen, torréfacteurs artisanaux), as they usually offer only light roasting degrees, and they don’t use normal terms for roasting (light, medium-light, medium, medium-dark, dark), but hipster-speak (City Light, Omniroast, Full City, City+, Italian Roast, French Roast), and they definitely misunderstand what a typical French roast and a typical Italian roast actually are!
  • The lighter a coffee roast is, the higher the resulting acidity! This is why “blonde” roasts should be avoided. I prefer a “medium to medium-dark” roasting, which can easily be found in Italian coffees.
  • Unfortunately, “single-origin” coffees (which in most cases is a misnomer because it usually means “from a single country,” not “from a single farm”) are typically offered in light or medium-light roasts only. The pretext: “If we roast such origin coffees darker, they would lose their individuality.” It’s true that dark-roasted coffee tastes almost the same, regardless of the quality of the coffee. But the true Italian espresso roast (medium-dark or midway between medium-dark to dark) is not meant to hide the qualities of a good coffee! (On the other hand, Robusta is usually added to coffees meant for espresso, one of the reasons being that it increases the crema.) In too many cases, poor-quality “single-origin” coffees are roasted lightly in a desperate attempt to preserve the little aroma that an overvalued, overpriced coffee barely has.
  • Fun fact: Starbucks coffees that can be found in supermarkets, both whole bean and ground, and roasted in Spain or Portugal in the case of Europe, are completely catastrophic. The blonde roasts are light roasts indeed, and they’re aromatic but acidic. The medium roasts, inexplicably, are all burnt coffee and nothing less! There is no middle ground!
  • Beyond using a darker roast, other means of making a coffee less acidic include:
    • Using a coarser coffee, which limits you to French press and [AmE] drip coffee makers aka [BrE] filter coffee machines (Filterkaffeemaschinen), not counting pour-over (V60, Chemex) or other hipsterish methods.
    • Using a shorter brewing time, which, in my opinion, favors both the portafilter espresso machines (Siebträgermaschinen) and the moka pots (improperly called Espressokocher in Germany; Italy: moka, caffettiera moka, macchinetta [del caffè], macchinetta Bialetti; France: cafetière italienne).

Here’s something stolen from here:

Coffees I put to the test

The coffees that I had available were:

  • Three normal 100% Arabica coffees, of which two can be found in Germany, Romania, and other EU countries where DM and LIDL are present.
  • One decaffeinated coffee (Tchibo).

In detail:

❶ DM Bio Espresso, gemahlen, 250 g, 100% Arabica (Peru, Honduras, Mexico, Ethiopia, Papua New Guinea), strength 4.5/5 (not true), MHD 03/2027: €4.25 in Germany, 24.95 lei in Romania.

There is a thing with the strength rating of German coffees. When separate indications for body (Körper), aroma, roasting, and acidity are not given, the overall “strength” is a compound supposed to grade the overall perception. A higher value might be the result of a stronger body or the result of a darker roast, and aroma can also add to the value. But it’s not entirely relevant.

In the case of DM Bio coffees, the “strength” is bollocks, and it mostly reflects the roasting, not the body per se:

  • Kaffee Klassik, gemahlen (Filterkaffeemaschine, Kaffeefilter, French Press), 500 g, 100% Arabica (Peru, Honduras, Mexiko), Intensität 2.5/5.
  • Caffè Crema, ganze Bohne (Kaffeevollautomat), 1 kg, 100% Arabica (Peru, Honduras, Papua-Neuguinea), Intensität 3.5/5.
  • Caffè de Peru, gemahlen (Filterkaffeemaschine, Kaffeefilter, French Press), 500 g, 100% Arabica (Peru), Intensität 3.5/5.
  • Caffè Crema de Peru, ganze Bohne (Kaffeevollautomat), 500 g, 100% Arabica (Peru), Intensität 3.5/5.
  • Espresso, gemahlen (Siebträgermaschine, Espressokocher), 250 g, 100% Arabica (Peru, Honduras, Mexiko, Äthiopien, Papua-Neuguinea), Intensität 4.5/5.
  • Espresso, ganze Bohne (Kaffeevollautomat, Siebträgermaschine, Espressokocher), 1 kg, 100% Arabica (Peru, Honduras) + 25% Robusta (Uganda), Intensität 4.5/5.
  • Espresso de Peru, gemahlen (Siebträgermaschine, Espressokocher), 250 g, 100% Arabica (Peru), Intensität 4.5/5.
  • Espresso de Peru, ganze Bohne (Siebträgermaschine, Espressokocher), 250 g, 100% Arabica (Peru), Intensität 4.5/5.

I preferred the ground “espresso” coffee from 5 countries to the one exclusively from Peru because I was pretty sure that coffee from Peru is more acidic. But the declared strength of 4.5/5 (or 9/10) does not reflect the reality. I’d rate the body to 7.5/10, the acidity to 3/10, and the roast to 8/10.

❷ LIDL Bellarom Barista Caffè Crema “Columbia, Brazil & Central America,” 100% Arabica, whole beans, 500 g, strength 3/5.

Many LIDL coffees under their fake Bellarom brand display the same “strength 3/5, acidity 3/5, roasting 3/5,” but they definitely can’t have the same olfactory profile! In this case, I’d rate the body to 6/10, the acidity to 6/10 and the roast to 6/10.

The grind level was… empirical. Let’s say, finer than that “universal” grind claimed by some supermarket coffees.

❸ BEANZ Brazilian Gold, 100% Arabica, ground, 250 g, body 8/10, “Full City” roast 8/10, acidity 4/10.

The current online price at the Romanian manufacturer is 44.57 lei; better prices range from 41.29 lei at Auchan to 43.49 at Carrefour. I got my stash when it was discounted (Aug. and Oct. 2025) at 23.99 and 24.99 lei (1 euro = 5.09 lei).

From the BEANZ range of ground coffees, “Brazilian Gold” has the darker roasting and the lowest acidity. And yet, the batch I’m drinking now is rather like this: body 7/10, acidity 5/10, roast 7/10.

❹ Tchibo Exclusive Decaf, ground, 250 g, Arabica and Robusta, intensity 3/6, acidity 2/6, roast 2/6.

This is an “international” variant. I’m not sure that I saw it in Germany, but I needed a quick fix for evenings when I’m in Romania, and a quick fix it was.

The presence of Robusta can be felt, and it also makes me unable to properly assess the real acidity. Tentatively: body 5/10, acidity 5/10, roast 4/10.

Notes on the aroma

Aroma is, for some, the most important qualitative aspect of a good coffee. It’s always difficult for me to assess it, because when an Italian coffee claims a 9/10 aroma, I always disagree. But then, the Italians sometimes also give grades for dolcezza (which is not the same as the antonym for amarezza!), rotondità, and persistenza del retrogusto, and I’m not sure about the scientific definitions or the measurability of the first two!

Until not that long ago, I was chasing 100% Arabica coffees with 8/10 or 9/10 ratings for aroma and 4/10 or 5/10 for body. The roasting wasn’t that important as long as the acidity wasn’t overwhelming (less than 6/10). Meanwhile, I reconsidered, and the acidity (as low as possible!), the roasting (preferably medium-dark), and the body (5-7/10) became essential.

I have to admit that light roast, acidic coffee can sometimes be fabulous. I fondly remember a coffee that has been discontinued, Taylors of Harrogate Colombia San Sebastian beans, 227 g. It was blonde, it was acidic, but it was amazing! The best Colombian coffee I’ve ever drunk!

My palate lacks the proper calibration (my sinuses, too!), but I’ll try to add aroma assessments to the aforementioned coffees:

VarietyAromaBodyAcidityRoast
DM Bio Espresso (ground)4/107.5/103/108/10
LIDL Bellarom Barista Caffè Crema (beans)6/106/106/106/10
BEANZ Brazilian Gold (ground)7/107/105/107/10
Tchibo Exclusive Decaf (ground)5/105/105/104/10

Notes on the preparation methods

The filter coffee machine I use is a “microscopic” Melitta Aromaboy, designed for 2 cups of 125 ml and a maximum of 300 ml. The standard German coffee-to-water ratio is 1:18, but I weigh 13.0-14.0 g of coffee per 250 ml of water. I use the official Melitta size “100” paper filters, the smallest that exist.

I own several moka pots, including an “Italian” Bialetti “Made in Romania” (at least, not in China!) and used in Germany, but in Romania I use a yellow G.A.T. Bella (Made in Italy). The coffee-to-water ratio is defined by the capacity of the 3-cup funnel and the coarseness of the coffee. I don’t weigh anything.

Water pH measurements

WaterMeasured pHOfficial pH
Tap water7.9—
Brita-filtered tap water6.9—
Still Dorna Izvorul Alb7.97.9
Still Carpatina8 (7.96-7.98)8.08

When the device started to display random numbers while in the air, and the measurements of the bottled water showed severe deviations, I stopped the tests. I wished I could test more water brands!

Water pH measurements (finally!)

CoffeePreparationWater pHCoffee pH
DM Bio Espresso gemahlenmoka pot6.95.5
DM Bio Espresso gemahlenmoka pot7.96.2
DM Bio Espresso gemahlendrip machine7.96.4
LIDL Bellarom Barista Caffè Cremadrip machine6.95.5
LIDL Bellarom Barista Caffè Cremamoka pot7.96.0
BEANZ Brazilian Golddrip machine6.95.6
BEANZ Brazilian Golddrip machine7.96.0
Tchibo Exclusive Decafdrip machine6.95.3
Tchibo Exclusive Decafdrip machine7.95.4

The water with a pH of 7.9 was Dorna Izvorul Alb. The water with a pH of 6.9 was Brita-filtered tap water.

This is all I could measure before the pH measuring pen went amok and committed suicide.

Conclusions

Given the limited number of measurements, these conclusions should be taken with a pinch of salt.

● The last coffee is an outlier. The impact of Robusta on acidity isn’t quite clear, but the decaf Tchibo blend had the lightest roast of all, and despite the perceived acidity being average, the measured acidity was the highest among the tested varieties. This coffee was also unbelievably resistant to the impact of more alkaline water.

● Notwithstanding some discrepancies, the use of alkaline water helped to some extent (0.4-0.7 pH points), and the coffee with the darkest roast (DM Bio Espresso gemahlen) produced the most alkaline result (pH 6.4).

● The drip coffee makers (filter coffee machines) that use paper filters seem to further decrease the acidity by 0.2 pH points compared to moka pots. The reference is the same DM Bio Espresso gemahlen, but I’m confident that LIDL Bellarom Barista Caffè Crema would have obtained a slightly better pH with alkaline water in a drip machine.

● The BEANZ Brazilian Gold also proved some resistance to improvement by the use of alkaline water. I cannot explain this behavior; more measurements would have helped solve the mystery.

● DM Bio Espresso gemahlen seems to be a winner! Its darker roast, despite the decrease in aroma, ensures, through the use of a mix of coffees from five countries, not only a very pleasant olfactory sensation but also a rather decent pH of 6.4 when prepared using pH 7.9 water in a drip machine that uses paper filters.

● If you prefer lighter, blonder roasts, you should change your preference.

● Unless you can perform all the measurements you need in 2-3 days, don’t purchase such crappy pH measuring devices.