DR Diaries: Summer 2012 Part II, Off to the Mill

Our coffee director, Byron Jackson Holcomb, is not just a buyer but a farmer himself. This is his latest dispatch from a trip to visit his own coffee farm, Finca La Paz, in Los Frios, Dominican Republic, earlier this June. The 2012 crop of Finca La Paz has just arrived at our roastery, and it’s better than ever.

I spent the better part of today trying to ask the right questions. One of the best lessons I learned in Peace Corps was that if you don’t ask the right question, you won’t get the right answer. A huge assumption that most of us make is that the other person understands our position. Usually people (especially in the DR) they just try and give the answer that will make the other person happy.

For example: Peace Corps Volunteer to a random Dominican in the street: “Where is the CODOCAFE building?”
Dominican: “Oh that is just that way, keep going straight.”  
Volunteer: “Thanks”.
Several blocks later. It plays out with different directions. Several blocks later, it plays out with different directions. Eventually someone actually knows where the building is and gives accurate directions. The first two good Samaritans just wanted to be “helpful”.

Today I had pretty simple plans: get to the export warehouse and dry mill cup some offerings and really dig into what happens at a mill.

The cupping was pretty straightforward. I’ve found that usually when I’m clear on specs or profiles and have previously cupped with a company, the cuppings get smaller and better. They don’t try and show me everything, just the stuff I asked for. So there were about 7 lots on the table. All were solid. The Lagulito was slamming. The acidity was pretty insane. Overall it had everything we wanted. I’m going to try a
few different blends in our lab and build the lot for this year.

I can’t tell you how many dry mills I’ve visited, enough to see all the major brands of equipment and know what each machine does before I’m told. My questions were all about the control that the workers had over the sorting. See, dry mills are all about sorting or “extracting” the good beans from the bad. In the dry mill all the machines are using physical attributes to indicate defective beans. Extracting may not be the best word but it seems appropriate for this analogy.

Dry mills work by a few major steps. All of them are different. But there are some standard machines. Here are some basics that need to be understood.

Pergamino: Coffee dried to about 11% moisture, with the papery parchment layer still on
Hulling: Removes the parchment
Size sorting: Separates the beans into about 8 different sizes
Density table: Separates beans by density (the denser the better)
Catador: Uses airflow to pull out smaller and broken beans
Optical sorter: Uses laser lenses to read the color of the beans and reject beans based on the color

Walking through most dry mills, usually I find a few workers babysitting a couple machines and most of the machines are “self-managed”. Watching the workers, uh, mill around, always made wonder what they were doing. It is like a watching a barista make a shot but not knowing the important controls.

I spent most of the time working with the optical sorter and the density sorter. The density sorter has 4 major controls: feed rate, air flow, vibration rate, and the paddles at the end of the table. Some people call them Oliver tables but that is a brand, just like Kleenex. The paddles at the end direct the coffee into different slots. The worst coffee is pushed out directly on the lowest side. The middle coffee is usually either run again or pulled as a slightly lower grade. The best is usually sent to the optical sorter.

Time is money. When the sales rep is booking coffee faster than they can prep it, that is a problem. When a buyer asks for X Grade of prep, how does the dry mill adjust the machines to fill that grade?

Dry mill equipment is sold on how many lbs or quintals (100lbs) of coffee the machine can process in an hour, and they are all usually connected so no one must move the coffee once the system starts. When
a box has to be prepped the clock is ticking to make certain the coffee is ready before the boat leaves. The natural tendency is to push as much coffee through the system as possible. Run the density table at its maximum “feed” rate, run the vibrations high, run the air medium to low (increases the first grade) and the paddles are left untouched. As the feed rate changes, everything else is affected, much like adjusting the grind for espresso. The airflow pushes the coffee harder towards the lower qualities. So more air means cleaner coffee but more time cleaning. Remember the dry mill only sorts by physical appearance, size and density. So the workers aren’t milling around: they are vigilantly monitoring the quality.

There are only two places in the dry mill that have good light: the density table and the laser sorter. This is because the workers need to be checking the quality as the coffee is flowing by looking at the

The optical sorter had some similar adjustment but it is a lot more complicated. I was working with an Xceltron brand from Costa Rica. First the machine must be calibrated to expect certain defects in bean
color: black, white, red, red/black, and yellow. Each color can be set on a scale of 0 to 94 individually. Setting everything to 94 would make the machine hyperactive and cause it to reject too many beans. Setting it to zero means every bean would pass.

Optical sorters take coffee from a hopper, run it straight down a channel of by two rotating cylinders and then at the last second the bean flies straight off the end in to the air where the coffee’s color is read by a laser and a blast of air pushes defective beans out of the path the good coffee into the path of the defective coffee. This all happens very fast.

Again, time is money. Running a coffee twice through the machine adds a couple cents to the dry mill’s cost. They can be run at different rates because they are built just like a weigh-and-fill machine. The lenses need to be cleaned often. Dust is a major issue in dry mills. When dust covers the optical lens, it can’t read the color and defects are allowed through. Once again, these machines are baby sat. If all the
sudden they stop rejecting any coffee. . . something changed. Either the coffee became perfectly clean or some dust covered some lenses or the machine fell out of calibration. Often times the machines just
keep running until someone steps in and fixes it. This is why sometimes people exporters and importers say, “oh you just found a bad bag of coffee.”

The optical reader and density table will keep running until told otherwise. And if out of calibration, they could be pushing the wrong quality into the export silo. The dry mill workers are the final custodians that bring the coffee from a seed to green coffee that can be roasted. Once the coffee goes into the burlap bags and is put on a sealed dry container, that is it. You can’t really re-run the product once it gets to the United States.

So if someone is taking a long lunch and comes back to find a machine pushing the lower quality coffee in the “first” quality slot and they just adjust the machine and pretend like bags of coffee weren’t just
compromised it is hard to say if anyone will catch it until it gets to the roaster. That is where one sample from the same lot should show brilliance and the other defects.

The dry mill processing cost is typically not a huge percent in the final green price but it is incredibly important. From what I hear, the dry mill costs range from 7-30 cents per pound for prep. It depends on the
defect percent going in, level of automation, and whether you are looking at the real cost or the charged cost.

Even though it was only one day, I feel like I’m starting to grasp where the quality control points in a dry mill are and how they are manipulated to produce brilliant coffee. Dry mills make the coffee dance in a million different directions to try and shake out the dirty from the clean beans. The attention of the workers and the calibration of the machines is paramount to quality coffee.

Again, it just confirms that any delicious cup of clean coffee is a miracle.