It was around five years ago now that a major lightbulb moment happened for me with regards to espresso preparation. I was visiting the (now closed) La Marzocco factory in Seattle and was talking everything espresso under the sun with John Blackwell, one of La Marzocco and Espresso Specialists' resident tinkerers. As we rounded a corner made of industrial shelving, he said "want to see our next big project?" Of course, I said, and he showed me into a little side area where this big, dual hopper grinder "thing" was sitting - it was a late prototype for the La Marzocco Swift grinder.
I was intrigued, but also a bit dismayed to find out exactly what it was (dismayed only because tamping is such an integral part of the art of the barista for me). As most professionals know today, the La Marzocco Swift is a high speed automatic tamping grinder designed originally to try and keep Starbucks somewhat "honest" (as in sticking with traditional machines, but simplifying the task of building espresso shots). Starbucks was looking for ways to increase consistency across their entire chain and reduce training costs - in my opinion, not only was the Swift viable, but it would have been an ideal solution for them. Instead, the big green went super autos, and the rest of that trail is history.
| Swift Grinder |
Featured many innovations, and some interesting discoveries, when it first came out.
Back to that first day I saw the Swift, I learned something then that has since influenced the way I prepare espresso, and the way I preach espresso preparation to anyone who will hear it, from consumers just starting out to world barista champion candidates. Many took this advice to heart, but time to fess up - I learned this one thing from La Marzocco, Blackwell, and Joe Monaghan, who was then with La Marzocco.
What was it I learned that day? I still remember the conversation quite clearly. After Blackwell explained how the grinder worked, I said "so it tamps with, what, 30 pounds of pressure, or is it adjustable?" He started to say yes, then cut himself off, and said "No! Actually, it only tamps with about 8 pounds of pressure!"
Coffee compaction theories in espresso
Thirty pounds of tamping pressure was kind of religious in espresso circles back in 2000. David Schomer, a pioneer in espresso crafting was touting that number in his books and his speaking engagements. In the newsgroup alt.coffee, the general consensus was that 30 pounds was the magic pressure for tamping (not all agreed with this though). So I was pretty surprised to hear that the Swift only tamped with 8 pounds of pressure.
I asked why, and Blackwell went on to explain that during the development of the grinder, they initially set the continuous tamping device at 30 pounds of pressure, and found that they created solid bricks in the filter basket, no matter how coarse the grind was. They were a bit mystified by it all, but eventually figured out two elements were at work.
First, ground coffee is a wide range of particle sizes within it's scope of "espresso grind", and those grounds form a kind of interlocking puzzle when compressed. The more pressure exerted on the interlocking pieces, and the tighter the overall fit is.
Second, and most crucial, they believed that in manual hand tamping, no matter how much pressure a person exerts on the top of the bed of ground coffee, by the time one gets down to the bottom third of the bed, almost all that pressure is eaten up, and the bottom portion of the formed puck is barely compacted any more by the tamping pressure. Because the Swift grinder continuously tamps as the coffee is ground and added to the basket, all the pressure comes to bear on the forming puck, a millimeter in height at a time.
Simply put, when you tamp by hand, you're tightening up the pieces of the puzzle at the top of the formed puck quite a bit; less so by the time you get midway down the puck, and very little, if any tightening at all by the time you get to the bottom few millimeters. The swift interlocks the grind puzzle pieces firmly and tightly from the get go, right through the entire formation of the puck. Result? Water hits a brick wall, even if the coffee is ground coarser.
La Marzocco were not the only ones to talk about this. I remember discussing this with some folks from Illy back in 2003 at the Boston SCAA show, and they talked about similar theories.
Let's take a small break here, and look at a typical flat tamper, knocked preparation:
| || || || || |
| || || || || |
| Dose |
Partially leveled out.
| Leveled |
with index finger, groove is very visible.
| Tamped |
Tamped, but not yet knocked.
| Knocked |
One can make out slight seperation between the cake and the filter.
| Polished |
Finished off, the edge has harsh and may have seperation below the surface.
So all this stuff I learned, from La Marzocco, from other industry folks - all of this was a lot for me to wrap my head around, and even today, five years later, I still think through the process. It introduces a lot of interesting theories about extraction, water flow, edge sealing, compaction, expansion and swelling of the individual coffee grounds as water meets it, and so much more.
Over this time, I've come up with a lot of crazy ideas, and put many of them into practice. A lot of them were just plain wrong so I won't rehash them here (to save myself some embarrassment); but I will bullet point a few salient points that have stayed with me for five years now - regardless of how factual they may be.
- Almost no compaction occurs via tamping in the bottom 1/3 or 1/4 (or perhaps even 1/5) of the bed of coffee in a filter basket.
- because water always follows the path of least resistance, you want to create as even, level, and positive a seal and environment for the water to "hit" in espresso brewing, to ensure even extraction.
- the path of least resistance you want the brewing water to take is straight down through the puck - as water gets lower in the bed of coffee, it faces less resistance because the environment isn't as tightly interleaved and locked together as the top portion of the bed of coffee. So work with that - make water follow the path you want, and only the path you want through a portafilter.
- when you hand tamp, a compressed top layer is formed that will seal up to the edge wall of the filter basket, but the bottom portion of the puck is more fluid, and more susceptible to breaking away from the filter's walls, which can present a path of least resistance favourable to the pressurized water.
- the act of "knocking", long practiced by many of the world's top baristas, would seem to be a detriment to maintaining this even, level and sealed playing surface to the pressurized water - especially if the bottom third of the formed bed of ground coffee is loosely compacted, or not compacted at all.
| Three style |
Euro curve, US (or Australian) curve, and flat tampers.
The last point is one I want to build on a bit, but before I do, I should also talk about sealing up the side walls of a filter basket.
Within three months of learning this news about compaction in the filter basket, I switched to a convex-curved tamper piston as my primary choice - the type today referred to as the "american curve" because of the shallow convex shape to the bottom of the piston. Another piston style, the "euro curve" came about in the early 2000s because of one fellow - Alexander von der Lippe, of Temperato in Norway put a call into Reg Barber one day asking for a more predominant curve.
These curved, convex shapes came about because of a theory - the tamper's piston applies pressure to give an extra squeeze at the edges - the lower middle portion of the piston pushes down, and out (to the sides); or so the theory went. Great espresso minds like Schomer and others were saying this, and von der Lippe wanted to push the theory even further by having a more pronounced convex shape. And at the time, I agreed with all of them, because, well, who was I to disagree with Schomer's theories!
No knock tamping, in theory and practice
| Properly dosed |
After you get your ideal grind, it starts here - with a well balanced dose
| Properly leveled |
I find that using your pinky finger can do a very nice final levelling.
| First tamp |
Using a euro curve tamper, this is the result after the first tamp - the 4 corner staub tamp is next
| Final polish |
After the 4 corner staub tamp, then a polish. Edges seem better sealed, there's a deep bowl in the puck's shape. Note the wear on the filter basket walls from the staub tamp method.
Back some five years ago, all this tamper science, theory and just plain guessing was so much to think about and digest, and I took some time doing so. The real change in my own tamping technique came within six months of learning this whole compaction / no compaction thing. By that time, I had a commercial machine in my house (a La Marzocco Linea AV 1 group, with special manual controls), and I was spending literally weeks and dozens of pounds of coffee just experimenting with one thing - tamping. The change I settled on, and still do to this day is no knocking on the portafilter - at all.
The theory - one I strongly believe to this day - has already been fleshed out above in this article, but I'll repeat for emphasis. I firmly believe that knocking on the portafilter once you've settled out your coffee and tamped it will unseal portions of the bed from the side walls of the filter basket, thus introducing new paths of least resistance (and least extraction) to the pressurized water.
But people knock the portafilter while tamping for very good reasons. The primary one is that it settles any ground coffee that is loose, sitting above the compacted bed, so that one more tamping action can do a final seal and polish of almost every grain visible to the eye. This way, no loose grounds will be floating above the bed prior to full pressure being built up; there's many theories why this is a bad thing, but maybe I'll save those for the forum comments on this article.
So if I stopped knocking to settle those grounds, how would I deal with the invariable quarter-to-half gram of coffee that may be sitting up on the visible side walls of the filter basket?
I swear, I did not know of the "Staub tamp" method before I developed the following technique myself - but I heard about Staub's theories very quickly after I started talking about it to people. Turns out I came up with somewhat similar conclusions that Carl Staub was famous for discussing and writing about. Basically, the "staub tamp" is the act of tamping squarely, then tamping at for corners of the "compass" inside the filter basket to push down on the loose coffee, and further create a tighter seal at the filter basket's wall. A final polish, and wallah, tamping done. Problem is, Staub recommended going a full millimeter smaller or more on your tamper to do this - which means using a 57mm for a 58mm filter basket.
I developed my tamping technique using a 58mm tamper. It's hard to do the whole north-south-east-west thing with a tamper that's virtually identical in diameter to a filter basket because there's maybe 1/10th or 1/5th a millimeter to play with in most filter baskets, but it can be done, and I a half-decade's experience doing so.
Where does this all bring us? Well, about four years ago I was no longer knocking on a portafilter to settle ground coffee. I was using a shallow curve tamper (Schomer's Ergo tamper, 58mm size, aluminum construction). I was doing my own personal version of the famous Staub tamp.
Did it end there? Not by a long shot. Something I firmly believe in when it comes to all aspects of coffee and espresso - we're never done. We've never reached the summit. We should never, ever stop learning and evolving every aspect of coffee. In the words of George Howell, specialty coffee pioneer currently running Terroir Coffee, "we haven't even begun to grasp what coffee is capable of", and we've had, as the western civilization, some 400 or more years to figure it out. When it comes to espresso, it's even worse - we're barely out of the cradle - pure babes and neophytes, all of us, when it comes to espresso brewing, technology and techniques. People may not realise this, but modern day espresso, that is the crema laden 1 ounce drink we know today, is barely 50 years out of the gate.
My own espresso journey traces back to 1990, and my first genuine shots of espresso enjoyed in Italy. From the late 1990s to 2002, I was a tinkerer, and tried to be a sponge when it came to information, theories and practice in building shots. By 2003, I started to get really serious, and my curiosity and examination of espresso technology, culture, rituals and science remains strong to this day. One aspect I've always tried to pay a lot of attention to was the tamping part - the most hands-on part of the ritual of shot building the barista does. In my next article, I'll talk more about this progression and what the results are today... and where I theorize tamping's role in espresso brewing will be going in the future.
NB: there are two other parts to this article, which remain unpublished as I continue to work out some of the theories and practices. The development of this article has lead to me attempting to design my own tamper because I felt there was nothing on the market available to address some crucial shortfalls in espresso preparation. As that project progresses, I will amend and complete the remaining two parts of Espresso Tamping, Science, Theory and Practice.