h2o: pt. 1

I've been long thinking I need to write something about water. If not for my own need to organize all the data floating around in my head then to help make clear what, in my opinion, can be an exceedingly complicated variable in regards to the brewing of coffee. Please don't judge me...this is not an academic/researched article and I'm not a chemist. It is somewhat scientific but a bit more anecdotal and meant to hopefully clarify a bit something that we all kind of go fuzzy on after a few acronyms have been thrown out.

It is ironic though because water is THE most affecting variable when it comes to the extraction of your coffee. And isn't it a bit odd that we spend literally thousands of dollars on temperature and pressure stable (controllable) water delivery systems,100's of hours perfecting pours so that turbidity doesn't over-extract coffee and yet water is sometimes barely a passing thought? We should start with water, then go from there.

My real education in water began when I moved here in Feb. of 2005. I moved from a city that arguably has some of the best drinking water in the nation. Fed mostly by the Cedar and Tolt Rivers, Seattle, WA has water that is relatively clean and soft compared to Madison water which is very hard...and depending on the area of the city, pretty heavy in particulates. Softness and hardness in water is a reference to the amount of dissolved minerals...mostly calcium and magnesium, but iron, aluminium and manganese as well. The softer the water the lower the presence of these minerals...the harder the water, the higher amounts of mineral content. The more stuff that is dissolved in the water, the higher the Total Dissolved Solids (TDS), measured in parts per million (ppm). When water changes forms (liquid to solid or liquid to gas) these dissolved solids can solidify...thus the presence of white dust in your tea pot or white flakes from your melting ice. Since we like to heat water in the coffee industry, this presents a bit of a problem for more than a couple of reasons.

Seattle/Madison Water Specs:

Calcium: 17.0 ppm/69.0 ppm
Sulfate: 2.0 ppm/16.0 ppm
Magnesium: 1.0 ppm/41.0 ppm
Chloride: 4.0 ppm/13.0 ppm
Sodium: 4.0 ppm/8.1 ppm
Bicarbonate: 18.0 ppm/110ppm
PH: 7.8/7.4

The above numbers will hopefully mean something if you labor through the rest of this article...but suffice to say, Madison has dramatically different water than Seattle. What I started drinking (and brewing with) when I moved here was radically different than what I came from. I also had to start feeding salt to the monster in the basement called a softener. I didn't really get how adding salt to water helped with anything, but after a few months in the coffee industry I started to understand very quickly what happens to our water when it gets heated and boiled to make coffee...and what we do to it before it even gets to the boiler. Today, I spend most of my time servicing and repairing coffee brewing equipment. I have become quite a pro at figuring out ways to remove mineral build up or blockages from tiny orifices so that coffee tastes good, but I'm still confounded by water, the most abundant and life giving substance on earth.

Specific water chemistry is highly regional. We start with water, which in its pure form is a chemical substance made up of two hydrogen atoms and one oxygen atom. But water doesn't really naturally exist in its pure form. Due to its nature to be highly solvent (dissolves other things) water ends up containing dissolved mineral solids it picks up from the environment around it. We use TDS to describe how much has been dissolved into the water. TDS does not tell you what is dissolved, but it does tell you how much. If you know a thing or two about where your water comes from then TDS can be a very good way to make some judgments about your water. Here in Madison it can range from around 400TDS to as high as 600TDS in some neighborhoods. Distilled water is at 0TDS...

So what does this all have to do with coffee?

A couple of things really. Most basically, it affects what the coffee tastes like because the specific TDS and relative hardness affects waters' effectiveness as a solvent and therefore to extract or take on flavors of the roasted and ground coffee. Second, and maybe just as important, when we heat water, the dissolved minerals in the water turn into a solid. This substance is usually referred to as scale, limescale, calcium, rocks... When this happens inside your water boiler on your espresso machine, or the tube that carries water into that boiler, or the valve that opens up to dose the water or a myriad of other inconvenient places, you get a breakdown. (This usually happens at 8 o'clock on a Friday morning when you have a line of customers out the door). So, to avoid this, we apply a number of chemical principals in the form of filtration to either keep the dissolved minerals from solidifying or remove them all together. Depending on what method is chosen the water chemistry is affected and therefore coffee flavor will also be affected. I'll really get into this, but first I want to talk about the regionalism of water again...and beer too.

In my move to Wisconsin I quickly adopted the proud past-time of drinking lots of beer (not super into the Packers yet though). I like beer. I'm a seasonal drinker like most, preferring Ales in the Summer and going darker as the weather gets colder. I don't brew beer but have hung around enough people who do to know that specific water chemistry has a huge affect on beer flavor/body (probably more so than on coffee). As I understand it this is one of the reasons why there are international differences in beer...and why one can't really brew a proper Belgium beer in Wisconsin. The locality of water has helped shaped the cultural identity of many noteworthy beer-brewing regions. I'm curious if there is a validity in this reality as it applies to coffee brewing. My gut says no. I think there has been some discussion..and the SCAA has, to a vague degree, made an international stance by considering a TDS limit when defining the “Golden Cup” whatever that is... The specific water chemistry certainly affects the coffee and how it tastes but, in my opinion, the absence or presence and level of these substances in the water will not set a coffee's brew-locality apart from another. It will change the flavors/extractions, but not in a regionally important way. Furthermore, as roasters are assigning best extraction ratios and water temp./pressure, coffee in/liquid out proportions, it seems obvious that there should at least be a standard for water chemistry.

The real sad part here is that I know of only one roaster in Wisconsin that is actively testing (or playing with?) how changes in specific water chemistry effects the coffee's flavor. I think we owe it to the specialty coffee industry to start weighing in on the chemistry of the water we use to brew with and the effect it has on the extraction of our coffee.

So, with that said...what should we do...and what do we do to our water so we can brew great coffee?

h2o

In anticipation of a post on water and to try out imovie here are some bad water situations for your enjoyment...

Pressure Modifications w/video

Coming on the heals of my last post/link (4 months ago!) this post may seem somewhat out of place...but pressure profiling and messing about with pre-infusion has peeked my curiosity. I believe that there is definitely evidence to show that a lower pressure period of 'pre-infusion' yields a sweeter more balanced and even extraction.

Infusion begins as soon as a solid is soaked in a liquid so I suppose it stands to reason that pre-infusion is the process by which the solid becomes soaked. As it relates to coffee (not bacon and bourbon...a very good infusion) arguably the most gradual and even soaking method will yield an even and balanced extraction; in combination with a score of other variables including water temp, grind consistency, dose and tamp.

Lever machines are the inspiration for what we are trying to do now with our electric pump powered machines. As the lever is pulled forward water is allowed to pass from the boiler to the brew chamber as the group spring is compressed. The time that it takes for the coffee to fill the small gap between the bed of coffee and the top of the piston would be considered pre-infusion time. The water entering this gap is at a pressure of about 1 to 1.5 bars (depending on boiler set point) but since there is a measurable space or gap there is a period of time that it takes to ramp from 0 to 1.5bar. This gradual increase of pressure gives the coffee grounds time to absorb the water it is being exposed to before the water is forced through the mass of coffee at higher pressures. Once the coffee grounds are fully saturated the lever is released and the tuned spring ramps pressure up to approx. 9 bar before slowly ramping back down as the spring relaxes. What results is (potentially) a really tasty shot of espresso.

What has been on my mind for the past 2 years or so has been how to (simply) do this on some of our current espresso machines. My first concept was a 'pre-portafilter' essentially a spacer that would fit into the group head, but then have a similar bayonet mounted to the bottom so as to receive the traditional portafilter. Essentially this concept adds a small space between the diffuser screen and the mass of coffee. If the machine could 'pre-infuse' at line pressure then there would be a slow ramp up during the time that it took to fill the spacer...and if only pump pressure then maybe an adjustment on the size of the spacer could be made. There may still be something to this idea but I can't say I really have the time or cash to prototype and produce such an idea...maybe someday!

A Non-Technical Technical Layout of the Hydraulic and Electrical

(You can see my original thought was to have three lines...crazy!)

I'm not sure when or what caused my second idea to come together, but I don't really think it is anything new, necessarily. I haven't yet seen it specifically carried out anywhere, but it isn't rocket science so I imagine it probably has been implemented somewhere. It is akin to what Herkimer did here. But it is unclear to me how they are switching the pump bypass on and off...it may be on a timer which would be really cool and repeatable. What I ended up coming up with was installing a switched solenoid valve on a split water line. Mains pressure would go through the switchable (opened or close) valve and then using a pressure regulating valve (PRV) I'd reduce mains pressure down to 2.5bar or whatever one wanted. The two lines would meet at the pump. The pump would be calibrated to take mains pressure up to 9bar. My experiment, masked as a birthday present for Bradbury's (turned 3 earlier this year!), was designed for a Synesso Cyncra. The Synesso has a standard sized rocker switch on the front face panel to operate the hot water valve. I swapped the switch out for a Pavoni double rocker switch and had one side operate the hot water valve and the other one operate the solenoid valve that was installed on the mains pressure line.

Double Rocker Switch
(White for the hot water spout and Red to switch mains pressure)

Back-side of Switch
(I ran the wires through the hollow Synesso frame...a genius of design by them)

With the paddle at stage one (brew valve open/pre-infusion) and the new switch at the off position the pre-infusion pressure would be running through the PRV and would be at the set point, 2.5bar in this case. With the paddle still at stage one the switch would be turned to the on position and the valve holding back mains pressure (5 bar in this case) would be energized. At this point you could add a third stage of pre-infusion with the paddle at stage two (pump pressure) and the new switch off so that the pump was increasing just the pressure regulated line yielding 7bar of pressure. Then you could turn the new switch back on, opening the mains pressure and increasing brew pressure to 9bar. 3 stages of preinfusion, let alone 2, is tough to standardize and repeat...and quite arguably not even necessary...but fun to play with nonetheless.

Josh,Jill and Ruby were kind enough to let me mess around with this concept and hopefully it will yield some fantastic results...so far the only measured result it has yielded for me was a caffeine/espresso induced fitful night of sleep and weird dreams (in which Josh and I were security guard baristas preparing coffee for a female Scott Walker speaking to a crowd of Epic employees...yah weird) I'm sure Josh will have alot of feedback and thoughts on pressure profiling after some 'lab' time with it. Anecdotely the shots we pulled that evening were quite nice.

My next thought would be to install timers on the switches (brew switch and mains line rocker switch) to control pre-infusion times more accurately.

As a side note as well to the Marzocco w/volume control users who don't already know: I think you should at least disconnect the pump lead from the manual switches and use those as your pre-infuse switch then hit the continuous pour button to engage the pump. A really simple modification that in my opinion would impact the flavor of your espresso in a really good way. I did this over at True Coffee in Fitchburg with great results as well. I didn't at the time, but the manual rocker switch could be replaced with the same Pavoni switch I used at Bradbury's to separate control of the brew valve and the pump...this would be great for the semi-auto Marzoccos as well. Linea's by the way....GB5 users are out of luck until they make a programming parameter that can switch the pump at user determined values. Or you could lime-up your GB5 pre-heater....but that is another story!