Developing blends for different regions.

[John Oughtred]

Just curious if any Roasters out there are formulating their blends for specific water types? We are a B.C. (Canada) roaster and we sell coffee into many communities. Some hard water some soft etc... We always encourage filtration to be used by our customers however this is tough to monitor. Obviously our coffees change in different areas and if you look at our sales across the province the popular coffees are so area specific. I am about to start a program at our roasting plant of bringing in water from our largest market segments to test the difference the water has on the coffees. I see a lot of value in creating blends that suit a specific market and was wondering if anyone has suggestions or experience in developing a program such as this? This comes via complaints I was getting from a logging firm on Vancouver Island (north). I ended up going there yesterday to visit them to find solutions. I agreed with them on the coffee, it tasted bad when I brought it back home (Victoria), and brewed it this morning, it had a totally different taste profile and although not my "cup of tea" it tasted way better then it did up there. I have always known this to be the case but now that I have a complaint I am looking for solutions to this issue. Any help or insight would be most appreciated.

Kind regards,

[IanClark]

I was going to post this in an older thread but given that yours is new and relevant to what I wanted to share I'll post here! This is all probably a little more specific than the intended scope of this thread, but I think the example serves to emphasize the potential benefit of blending for water type!

Over the past few years we've had tremendous difficulty with seasonal changes in water chemistry affecting the way our espresso extracts. In particular, from the data we've gathered it seems highly likely that there is a threshold of around 36mg/l total carbonate hardness (alkalinity) below which there is an inverse relationship between required extraction temperature and alkalinity. The theory I've heard through word of mouth is that there is a certain level of alkalinity required to facilitate diffusion of solubles through the interstitial substances surrounding the coffee cells. Below this threshold you need to increase thermal energy to maintain the same extraction rate.

Interestingly this phenomenon seems to apply mostly to tastants rather than aromatics. When you crank up the temperature to prevent underextracted taste (sour) you still overextract the aromatic character and you also get thin body.

Anyway, here's the interesting update: according to our Municipal Water Quality Supervisor (who has been quite helpful in all of this!) last week our water's alkalinity peaked corresponding with the spring melt. As predicted, for about two weeks during this peak period we had ridiculously awesome espresso compared to the usual experience. Then a few days ago the alkalinity started to drop again and we had to start bumping up the temperatures again to prevent sours. I decided to take the opportunity to taste the components of our espresso blend as well as some other coffees of the same roast degree as single origin shots. Much to my interest most of them tasted pretty awesome at the same low temperatures where the espresso blend was brutally thin and sour. I then pulled shots of one particular component of the blend (at 34%) that was extremely thin and sour where the other coffees had been excellent.

So it would seem not all coffees have the same sensitivity to low water alkalinity. At the moment I'll decline to comment on which particular coffee this was, but I think this is quite interesting! By removing this one coffee from the blend we'll be able to ensure a more even extracted of every component in the blend. I should note that we can't be completely sure that alkalinity is the culprit here, although the correlation over the past couple years is extremely high. There may be something else going on correlating with peaks and troughs in alkalinity.

[nick]

Ian, nice post you got there.

Indeed, a little alkalinity is apparently good, and helps the coffee particles absorb the water. The "problem" with your post is that you're referring to relativities. It's hard to comment too decisively without knowing some numbers.

The brand new SCAA Water Quality handbook (a must-have for everyone reading this!) establishes an SCAA standard of "Total Alkalinity at or near 40 mg/L" and a "pH of 7 to 8." A higher alkalinity than that, and you'll start affecting extraction and neutralizing various acids. Lower, and you can amp-up the acids in the coffee, but the tongue detects at different rates each of which. If I remember correctly, phosphoric acid is one that is more "intense" than, say, acetic acid, molecule-for-molecule. A higher or lower water pH would effect both acids the same "on paper," but the greater taste-perception "swing" would occur with the phosphoric acid. My word-choice here sucks, but you know what I mean?

I think you are on to something though. Getting a water test-kit and exploring the variables would be something to try, and life is a lot easier when you have some metrics to wrap your brain around. For similar reasons, I'm personally a big fanboy of the water customization products from folks like Cirqua and Everpure. It's hard to fiddle with the other variables when you don't have a firm handle on what's going on with your water.

John, as far as formulating blends for water types, I realize that exercising control over water is a luxury that some of us will put at a higher priority than the average coffee shop out there, but all of that just reinforces the need for a handle on our water quality. That said, I've been thinking about that sort of thing too… just not about water types... I've been thinking about espresso blends on various grinder types.

[IanClark]

Hi Nick,

Thanks for the reply - since posting we've done some interesting work in profiling the water across locations and determining the influence of each locations' water on espresso extractions by controlling for the usual grinder and coffee variation (same coffee/grinder) and pulling shots with the same temperature, pressure and a range of normal brew ratios.

Overall we had confirmation of much of what the water quality handbook has to say, with some specific data on optimal pH and alkalinity.

We found the coffee to be most sweet, balanced, nuanced and syrupy with a pH below 7.9 and significantly improving as the pH dropped towards 7.6. Above 7.9 yielded very thin, dull flavoured shots with increasing sourness as pH creeped up towards 8.3 at the highest.

Alkalnity (KH) also had a tremendous impact on the extraction. The shops with higher KH around 35 mg/l demonstrated more balanced and pleasant acidity as well as an overall better extraction with respect to sweetness, flavour nuance and texture. With a KH lower than this the acidity became progressively more aggressive and the overall extraction suffered as well.

Of course, these two variables can either complement each other if they are both ideal (pH towards 7.5 and KH towards 40), mitigate each other if one is not ideal, or they can completely preclude the possibility of getting great coffee if they are both off-ideal.

For instance, comparing two shops with a pH of 8.1 where one shop had a KH of 35mg/l and one had a KH of 25 mg/l, we found the 25mg/l shots to be of greatly inferior quality (likely scoring a 1.5 to 2 on a WBC taste balance scale) while the 35mg/l shots were probably more like 3.5. Again, this is with the same coffee, grinder and with many other extraction variables controlled as well as we could. I should note that TDS was also constant at 90ppm here and we saw no significant differences in other mineral contents.

Another interesting comparison is the performance between our two best shops. With one at pH=7.9, KH=30mg/l and TDS=100ppm and the other at pH=7.7, KH=30mg/l, TDS=90ppm, we found a significant difference in overall quality. The first shop (pH =7.9) had slightly less aromatic nuance both in espresso and in a 12oz latte. I would say the shop with a pH of 7.7 would likely have scored a taste balance of 4 - 4.5 while at 7.9 it would be more like 3.5 - 4.

All this is to say that I also am becoming a great fan of water formulation technology and am viewing it is an absolute necessity for our sort of business (attemping awesome coffee across 11 locations and growing...) given the less-than-deal and varying nature of our water.

Also, I'm looking forward to the day when espresso machines come with a built in "equalizer" that allows the barista to twerak the calcium, pH, KH, etc shot-to-shot!!