For sometime now, I'd been waiting for the opportunity to visit Terroir and spend some quality time with George Howel and Peter Lynagh. George needs no introductions, he has been at the cutting edge of the quality in coffee movement for many years, and has set new standards in storing and roasting coffee to deliver the highest quality in the cup. Peter Lynagh is well known in professional coffee circles as a master roaster, a coffee aficionado and a frequent judge in cup of excellence and other coffee competitions.

Jim Schulman got the invite and asked me if I wanted to tag along. Without a second thought I filled up the tank and headed north to Acton MA. Acton is a small town about 30 minutes from Boston and 10 minutes from Walden Pond, where Henry David Thoreau spent two years in the woods writing: "simplify, simplify, simplify". The New York Times book critic responded by saying "One simplify should have sufficed".

We arrived on Wednesday afternoon and took a tour of the facility.


They use a Probat roaster


It is a hand-on-the-throttle roasting technique, no pre-programmed roasting profile.


Beans would take about 3-4 minutes to cool down after the roast

To preserve green coffee freshness, a growing number of green coffee farms ship their greens vac packed. Vac packing at origin has been George Howell's crusade for sometime now and it is a very slow starter. Daterra initiated that practice but this year, for the first time, the cup of excellence winner from Brazil was also shipped vac packed.


Vacum packed green coffee

As soon as the coffee arrives, it is stored in a freezer until the day it is roasted. The benefits of vac packing and freezing became very apparent on Day 2 when we cupped frozen coffee against non frozen. More about that later.

They have a variety of brewing machines in their tasting lab. I counted two Mistral espresso machines, one L/M Linea, a Clover, ten Technivorm drip brewer, press pots, and an Eva Solo pot. They also use two small sample roasters for cupping. After a few shots of espresso, we checked into our hotel and headed out to Harvard Square for wining and dining. The wine of the evening was a great piedmont whose name I forget.


A sample roaster

In the next two days we talked coffee and cupped a great sample of coffees from around the world. Here is a partial list from Jim's memory:

We started with a cupping of various Guats and El Salvadors, comparing existing samples from Terroir stock with newly arrived samples. George was trying to determine whether to buy some of those coffees or not. Then we did a Kenya cupping, again their stock Mamuto versus a newly shipped sample.

The highlight of the day was an experiment designed to determine the effect of damaged beans in the cup. Peter took what looked like a high quality shipped & sorted green sample and sorted it further by hand to separate flawed beans from the perfectly processed ones. He then roasted the samples and served them for cupping. All cupping was done blind. The difference in the cup between the healthy and flawed sample was noticeable. The healthy one was vivid, round, wonderfully acidic, sweet and aromatic. The damaged sample was flat, slightly astringent and weak in comparison. Bottom line, better sorting of beans at origin has the potential of drastically improving cup quality.


Separating flawed beans from the pack


Cupping. From left: Peter Lynagh; George Howel & Jim Schulman

On day two we started with around the world samples. Daterra (Brazil), Sumatra, Pedra Petra (Brazil COE), Santos Maria Las Sauces (Colombia), Addis Ketema (Ethiopia) and Mamuto (Kenya).


Around the world in 30 minutes.

We followed it by cupping a DP Yirg agains a WP Yirg. The fermentation of the DP was very obvious. The highlight of this day was the following two experiments. In the first one, we cupped three different samples of the same coffee (La Minita). Sample one was green coffee that was vac packed and frozen as it arrived. Then roasted, ground and immediately brewed. Sample two was the same coffee as sample one, only, it was ground ten days ago. Sample three was not vac packed or frozen but stored as green coffee in a standard burlap bag. The results were startling. Sample one was as expected the best of the three: vivid flavorful and sweet. Sample three was the worst. It was flat, faded, and tasted a bit like the bag it was stored in. The biggest surprise however was sample two. That coffee was ground 10 days ago, and it still tasted better than Sample three.

Vac packing the greens at origin and freezing them at arrival has a huge impact on cup quality.


Can you make my coffee taste like a burlap bag? From left: Jim Schulman, Peter Lynagh & Abe Carmeli

The second experiment of the day was a showdown between three brewing methods using the same coffee: Drip using the Technivorm; Soak pot using the Eva Solo, and Soak/vacuum using the Clover. This was very informal, and we did not spend the time to fine tune each brewing method, so the results should not be taken as more than casual play. The winner was the Eva Solo, followed by the Technivorm drip. The Clover came in last by a large margin. We were using the Mamuto as our sample coffee.

That experiment concluded the 2nd day after which I said my goodbyes and headed back to New York. Jim stayed after lunch for another round of the flawed versus healthy beans comparison. Many thanks to George Howell, Peter Lynagh and the team at Terroir for their great hospitality.


The humble winner - Eva Solo soak pot

Thanks for the early post, Abe. It covers the events very well.

However, as we did these cuppings and listened to George, both of us were struck by how all the things he was doing grew out of a very thought through view of what specialty coffee should be -- something he's been developing since his Coffee Connection days. In effect, "Cupping with George" is a post graduate coffee seminar, where one goes beyond the text books and gets into visions of how the field should develop from here.

In a nut shell, everything from cultivar selection and farming practice, to prep, sort, transport, storage, cupping, roasting and brewing is about conveying what is potentially in the bean without additions or degradations. Lots of people preach this same point, but George Howell has worked out all the details into a very comprehensive whole.

There are things George does that very few people in the business agree with; and I don't think anybody agrees with the whole package. However, because his stance is consistent; the details more or less follow from the premise. This means that arguing with George's detailed practice requires that one thinks through ones entire approach to coffee, so one can argue both the details and the principles from which they stem.

This is not something that can be done in a forum post; so I'm going to write a longer article and post it on a webpage. Once it's up and checked for accuracy, I'll crosspost links on the various coffee fora. I'm hoping this will stimulate one of the more interesting discussions we've ever had.

Abe, Jim - thanks for great writeup.
It makes me wonder though. Given the tasting results of vac-pac & frozen vs. burlap & room temp for the green coffee - how do other roasters handle this? If results are so incredible theoretically Terroir coffee will always be significantly better then from pretty much anywhere else. Is vac-packing at roaster too late? Does it have to happen at the farm to see the difference?

Vac packing has to be done immediately after processing at the farm. Daterra takes it a step further and keep their beans in lab controlled warehouses monitoring temperature humidity and other parameters. This is one reason why Daterra's coffees are so different from the traditional Brazilians. Higher acidity, more fruit & floral notes etc. Roasters often complain that between the time they cupped the coffee in the farm and the time they get it (a few months later) most of it died on the way. The first to go are the high, floral & fruity notes.

As to a comparison between Terroir's coffees and other roasters, I haven't done a study on it. But from the experiment we conducted it is pretty obvious: If you store your beans in a burlap bag they will fade in a matter of months, sometimes weeks. High grown coffees will fade slower.

Edited 5-2-07 to correct an error.

Abe Carmeli wrote:As to a comparison between Terroir's coffees and other roasters, I haven't done a study on it. But from the experiment we conducted it is pretty obvious: If you store your beans in a burlap bag they will fade in a matter of months, sometimes weeks. The more acidic high grown coffees will fade faster.

Generally speaking I agree with that statement. After all, I have been vac bag sealing my greens for 6 years and have also done many different greens storage tests. However I believe it's way too generalized a statement "in a burlap bag". Greens stored in breathable bags (be it burlap or cotton or whatever) and those breathable bags stored in temperature and humidity controlled environment would fair far differently than stored in uncontrolled humidity and temperature. Even a single day in extreme humidity and higher temps can kill beans flat. If you know, it would be good to know the storage environment of the burlap greens tested and how long stored. Too bad they didn't include vac'd and stored in same environment as the burlaps in the comparison.

OTOH I've only done one 6 month vac bagged home freezing greens compared to vac bagged at usual home temp. (air conditioned in summer) Results while leaning towards the freezer stored were inconclusive and the entire final tasting process flawed to heck & back for various reasons. (Done at the end of PNWG IV last year after a full day of playing with dozens of coffees and many different espresso machines etc. and even worse issues of cracking press pot carafe during brewing, getting distracted and over infusing all.) Was going to run another test but never got around to it. That said personally going to work on figuring out freezer space for around 100# already vac bagged greens myself.

Also, do they fully defrost the greens before roasting or adjust profile for maybe longer low temp early stage to include thawing during the roast?

miKe mcKoffee wrote: If you know, it would be good to know the storage environment of the burlap greens tested and how long stored. Too bad they didn't include vac'd and stored in same environment as the burlaps in the comparison.

The beans in the experiment were all one year old. The burlap bag was kept in the roastery with no temp control.

Also, do they fully defrost the greens before roasting or adjust profile for maybe longer low temp early stage to include thawing during the roast?

They defrost the beans before roasting.

Abe and Jim, this is really an incredible post!

I get really excited to find out that there are pros out there looking to squeeze quality out of every bean/roast/brew. I am sure Terrior is not alone but your reporting on this is stellar.

Thanks!

Abe Carmeli wrote:...The highlight of this day was the following two experiments. In the first one, we cupped three different samples of the same coffee (La Minita). Sample one was green coffee that was vac packed and frozen as it arrived. Then roasted, ground and immediately brewed. Sample two was the same coffee as sample one, only, it was ground ten days ago. Sample three was not vac packed or frozen but stored as green coffee in a standard burlap bag. The results were startling. Sample one was as expected the best of the three: vivid flavorful and sweet. Sample three was the worst. It was flat, faded, and tasted a bit like the bag it was stored in. The biggest surprise however was sample two. That coffee was ground 10 days ago, and it still tasted better than Sample three.

Vac packing the greens at origin and freezing them at arrival has a huge impact on cup quality.

The second experiment of the day was a showdown between three brewing methods using the same coffee: Drip using the Technivorm; Soak pot using the Eva Solo, and Soak/vacuum using the Clover. This was very informal, and we did not spend the time to fine tune each brewing method, so the results should not be taken as more than a casual play. The winner was the Eva Solo, followed by the Technivorm drip. The Clover came in last by a large margin. We were using the Mamuto as our sample coffee.

Agreed, an outstanding trip report. And two big surprises: downfall of the mighty Clover, and validation of preground coffee. I just don't get the latter. Whenever I use preground, even just a few hours old, my espresso suffers. In particular, there's a noticeable reduction in crema. How was the preground sample stored after grinding? Vac pac I assume...

Abe Carmeli wrote:The beans in the experiment were all one year old. The burlap bag was kept in the roastery with no temp control.

Thanks for in the info'. Not surprising at all the one year old burlap stored were flat and baggy. Unfortunately that's exactly the easy and inexpensive type of greens storage seen too often even quality focused Third Wave espresso cafes & roasters like Stumptown. Rows of stacks of burlaps full of greens sitting just a few feet from a big Probat...

Abe Carmeli wrote:They defrost the beans before roasting.

That's what I did for that one greens freezing test I did too, didn't really know better just made sense.

RapidCoffee wrote:Agreed, an outstanding trip report. And two big surprises: downfall of the mighty Clover, and validation of preground coffee. I just don't get the latter. Whenever I use preground, even just a few hours old, my espresso suffers. In particular, there's a noticeable reduction in crema. How was the preground sample stored after grinding? Vac pac I assume...

They didn't say sample 2 pre-ground 10 days tasted good, just that it didn't taste as bad as sample 3 one year burlap stored in the roast room greens roasted and then fresh ground!