Confused? Naturally

Confused? Naturally

I just returned from the Roasters Guild Origin Trip to Brazil and there were a number of things that we saw that surprised me in both good and bad ways. We visited 9 different producers or producing groups where the main good surprises were the diversity of cup characteristics that we saw in the many cuppings in the many different growing areas. There was a clarity to many of the coffees that one doesn't traditionally associate with coffee from Brazil. There was also a range of processing styles as well as experiments on many tables where they were looking closely at a number of different factors including; sunny vs shady side separation, varietal performance in different conditions-altitudes-exposures, as well as what was the most promising which was some groups moving towards much more selective picking practices. Most of the coffees we cupped over the week were fairly young and needed more rest, but showed plenty of promise.

Witnessing the traditional Brazilian harvesting practices was definitely the most shocking surprise, even with some prior knowledge of them going into the trip. If you've been to other coffee producing areas that use extremely selective picking practices, then seeing coffee shrub branches being totally stripped bare is fairly jarring. Seeing this practice, as well as both the dry processing and pulp-natural processing raised more than a few questions for me. Listening to various producers describe their processes made it clear that there is a lot of confusion about the names of these processes, or at the very least a lot of mixed use of certain terms which can actually lead to a good deal of confusion.

What does Natural mean? I got the clearest explanation of this at Fazenda Recreio outside of Poços De Caldas, close to the border between the states of Sao Paulo and Minas Gerais. Natural isn't so much of a processing style as it is a harvesting style. In Natural Dry processed coffees, coffees fruits are allowed to fully dry on the branches, but at the same time there are coffee fruits at different stages of maturity that are being harvested, some even still quite immature. Traditionally, this was all then dried on patios, thusly it was the Natural Dry Process. But the term Natural has become a catch all for any Dry Processed coffees, even ones that are picked incredibly selectively as opposed to in the Natural style.

Nowadays there are some levels of separation that are taking place even with Naturally harvested coffees. Siphon receiving systems designed by the Brazilian Coffee harvesting and processing equipment manufacturer Pinhalense (the same equipment used by Emilio Lopez for the Manzano process experiments) have created a means with which to separate the coffee based on density as well as adding a cleaning stage to all the coffee received. This separation (in tandem with more select picking) is leading to a much more clarified Dry Processed cup profile in many places in Brazil, one that is much more fruit forward as opposed to the traditional layers of tobacco, leather, aromatic woods and spices. In some cases, I found a few coffees to be leaning a little too precariously to some over ferment profiles, but was told that these are the coffees that are being rewarded in the late harvest COE competitions.

I believe that these coffees are standing out more because they are so different from what was acceptable in dry processed Brazilian coffees in the past, but in some cases it becomes almost one dimensional with such a focus on the acidity and not so much focus on sweetness and balance. I did however find many very impressive Dry Processed coffees that were balanced and extremely sweet, and was told how these coffees were the result not just of selective picking based on brix readings (a reading that measures sugar content), but also were the results of experiments being done looking at sunny vs. shady side planting, as well as drying in shade and at altitude which allows for a slightly slower drying (within a reasonable window to avoid defects) which resulted in a more developed sweetness and complexity.

Speaking to a few different producers about Brix readings also was very enlightening. In El Salvador while we were performing the experiments for the Manzano project, most the fruit was considered ready for harvest at a Brix reading of between 19-23 depending on the varietal. In Brazil, the range is much closer to 23 all the way to the mid 30's, meaning of course that even in some of the very selective picking that there is still a practice of allowing for much more maturing to happen on the branch. This was pretty surprising, especially after seeing some of these coffees on the tables. I was very excited after talking to Jacques and Luiz of Fazenda Sertao, whose coffees we have carried and will have again soon upon arrival of this year's harvest, about the new crop coffees that we had on the table as well as the work that is going into pushing quality on their farm. It is an uphill battle to retrain an entire workforce, as well as to put the energies into looking at the various possible impacts on quality throughout the process.

Not only is what passes as a dry processed coffee from Brazil changing fairly dramatically, but the cups from many Pulped Natural coffees were also quite impressive. The Pulped Natural process was introduced in Brazil in 1991 by Pinhalense, and the equipment used for the process has played an important role in promoting sweetness and less material in the cup. This process also adds another element of confusion in the way we use certain processing terms. Number 1, they use the word "natural", which in Brazil it does mean that the natural harvesting style is being used, but it also encourages the misuse of the term natural as a catch-all for Dry Processing. This perhaps makes the use of the terms like "Honey" or "Miel" Processing more appealing, not just aesthetically but in concern for accuracy as well. This process is also called the Semi-washed process as well, which is also accurate but unfortunately that term has become synonymous with the Wet-Hulled process which is common in Indonesia and nothing like the puled-natural process.

Another layer of confusion is that in some areas of Central America when the demucilagers used in this process are used to completely remove all the mucilage from the parchment it is called Mechanically Washed. To clarify, the demucilagers can be set to remove variable amounts of mucilage, up to almost completely clean. In Brazil, each and every time that that specific equipment is used it is referred to as Pulped Natural. This leads to some very cleaned Pulped Natural coffees that really showed some surprising clarity in the cup, and which might have been considered Fully Washed in some places. And yes, there were actually some fully Wet Processed coffees in fermentation while we were there, specially at Fazenda Monte Allegre, one of the largest producers in Brazil. There were also a number of experiments on the tables at Fazenda Ambiental Fortaleza where they were mixing processing methods, resulting some really interesting cup characteristics.

It was inspiring to meet a number of both younger and older producers that are committed to changing perceptions about coffee in Brazil. Of course the consumption of coffee in Brazil is considerable and growing ever larger. We visited a number of roasters and cafes who are doing impressive things with what could be considered a limited pallet, but instead is a fairly broad spectrum of coffees and coffee experiences. The fact that the coffee culture is as rich as it is in a producing country has lead to a number of interesting collaborations, and even though you could sense from a number of roasters their frustrations at not being able to work with coffees from elsewhere, I strongly believe that sometimes a limited arsenal can lead to some exciting results. I really think that Brazil is going to make some interesting moves in Specialty Coffee in the next couple of years, and after this trip, my curiosity for what that can be is certainly piqued.

Would love to hear thoughts on some of the topics covered here, especially in regards to processing terms.



#1 Processing

Hey Chris! Great to have you here.

First of all thank you for being engaged in the cuppings and giving honest feedback.

Secondly, for those of you who haven't been to Brasil this year here is what is happening here. It has been one of the toughest years yet with the harvest primarily due to abnormal rains. We had a drought during the 'rainy season' December-February and then too much rain during the 'dry season'. Things are lush and green now when they should be dry and dusty. What happens is we have all the stages on the tree at once. We have coffee that didn't fill in and ripen making it stay green. We have coffee that jumped from green to black 'requeima' which usually won't have proper size or complete flavor. We have coffee staying in the ripe stage for less time and less honey 'mucilage' on the cherry. What happens as a farmer is if you wait for more coffee to ripen then the riper coffee will start falling off the tree and if it rains or a wind storm comes all of a sudden (we've had four already) lots of coffee fall off the tree. This year picking is at a record high of an average of $R 120 per day per picker which is roughly $60 US dollars. As a result going through and picking twice is mostly out of the question for those of us who have hand picking. As a farmer this year if you don't dominate drying processes than it will be difficult to produce quality by leaving it up to chance of climate.

Let me clear up processing:

First of all harvesting. Cherries are at about 18-25 brix, then you have 'Passa' which means Raisins in Portuguese which is when the cherry is purple and beginning to shrivel up as it is losing weight and size as it dries. This primarily still sinks if you pass it through water. These are at about 25-32 brix. Then you have 'Seco ao Pé" or Dried on the tree. This is what you must be referring to when you referred to the farmer near Poços. Many Brasilian farms do this because we can allow the coffee to stay longer on the tree without it fermenting. In most cases it will have more moderate temperatures and some shade if it is on the tree. In lower, flat and hot regions it can actually ferment straight on the tree.
**These brix readings are for most common arabica strains.

Once you pick the coffee as a farmer you have an assortment of colors or stages of cherry which can look like skittles especially this year. Then you can decide what technology to use and how to separate them (if you have a way) before deciding how to dry. This is where the Pinhalense de-pulper came in. It uses water to separate floaters from sinkers and both small incompletely formed green cherries, requiema, and 'Seco ao pé' float. We call this Boia. Then in Brasil we dry these with the whole cherry husk in contact which means using a natural process. The ones that sink go into a spinning cyclone which squeezes out cherries that are softer (riper). A farmer can squeeze out more or less cherries by tightening or adding weight. By keeping it loose only very ripe cherries will pulp, by tightening it some cherries that are not green but not quite ripe will also pulp. This separates the greens which are hard and do not pulp.

We at Fazenda Ambiental Fortaleza sometimes decide to do all naturals depending on varietal. We are exploring methods of technology that does not use water by instead using electronic optical sorters and density sorters. We also pick the unripe coffees out on the raised beds.

Back to pulping....Then he can decide how to dry the pulped coffee. By leaving the mucilage on and drying this would be a 'honey-process'. By using less water more honey is left on. A farmer can take off mucilage with a mechanical de-mucialger (with 5 settings) or through water. Through water he can simply soak overnight or longer to take off the mucilage and dry or he can do an assortment of different 'fermentation options' ex: dry vs wet fermentation, kenyan soak, etc. We don't do many of these in Brasil traditionally although I am experimenting with all of them on my farm. In Brasil most people take most honey off and dry either in a mechanical drier, on a patio, or very rarely on a raised bed. There is a definite relationship between time of drying and quality with more slower and even drying relating to higher quality as well as temperature of drying. Time of drying also influences cost of production as a patio worker goes for $R120 a day so most farmers opt for a mechanical drier. Speed of drying influences the shelf life of the coffee as well as a fast dried coffee tastes good in september-october but is baggy by january where as a slower and evenly dried coffee can last with no signs of bagginess or (lomy) for up to 12 months.

Fermentation: This is a tricky topic and seems to be something almost always misunderstood in coffee.

For me most coffees are fermented. The exception being coffees picked at night and processed right away and having all mucilage stripped off either mechanically or by cold water. Otherwise, sugar + heat and humidity = fermentation. This can be a good thing and a bad thing. Depending on the temperature you give preference for different cultures to develop. This is chemistry. At 18-21 degrees Celcius a fungus called levedura (in portuguese) develops - I think in english its yeast. this gives cleaner and more preferable flavors - I believe. And above 23 degrees celsius bacteria begins to form. Bacteria eats sugar and breaks it down into acids. This is also called fermentation. This happens in washed coffees, semi-washed, honey-process, and naturals. Although in naturals the sugar content is highest and most contained, which makes it easier for increasing the rate of these fermentations. No two cuppers 'i have met' have the same tolerance for ferment. All trained cuppers will agree that once this fermentation level reaches Acetic Acid, which means sour vinegary flavors and short aftertaste, that this is repulsive and 'fermented' in a bad way - and call this a defect. However, just below this level you get a strawberry yogurt, bubblegum, candy-like flavor which is still sweet and not yet vinegar. IS this GOOD or BAD? I can replicate this flavor for you with confidence with any coffee no matter the varietal.

Imagine you have a graph where temperature is on one axis and humidity on the other. At one point you draw a line where you can start to taste fermentation influence. The taste is still sweet and balanced. Going up the graph the acidity and fruit flavors keep on changing. This goes all the way up to where the strawberry is again and just above that you draw another line where acetic will be. How do you score these coffees? Some cuppers say ferment others give scores above 90? I cupped the Brasil natural COE several times and would say they were all in that range and got very good prices. Although we have consistently been able to dry naturals that are not in this range as well as some in this range, so I can say with confidence that natural dried coffees don't have to taste any ferment influence. Personally, I like both as long as the fermentation influence is not bordering acetic.

I am rambling so I will end with this note. From my experience and what I do here at my parents farm and many neighbors we work with is to treat coffees primarily by varietal. The most influencing factor on taste is in fact varietal. Therefore some varietals taste better natural (catuai) and some taste better pulped (obata). For me both are valid forms of processing and can produce extremely clean and complex coffees with their own merits. That is why I prefer to talk about taste profiles FIRST then talk about scores because it seems we are all over the place as an industry in terms of scoring.



#2 Excellent!

Excellent comments, Felipe. Thanks so much for taking the time to talk about these processing techniques. One thing that you speak to, and which we found in doing the Manzano experiments, is that fully dry processed coffees are kind of predictable.

In a way, doing full dry processing homogenizes the coffee, giving a much more predictable result. You mention that you know exactly how you can produce that Acetic acid as well as avoid it. I would say that in dry processing, you might be covering up what might be unique about a particular coffee in order to deliver a more predictable cup. This isn't a good or bad thing really.

In some cases, you could probably take an 85 pt coffee and turn it into an 87 or 88 with adding sweetness and complexity, depending on who's cupping. However, I don't think that you could take a 90+ coffee and add points, maybe only take them away. This of course speaks to your comment on how scores can be pretty deceiving and might not be the best way to evaluate a coffee, at least not as beneficial as really talking about what you're finding in the cup itself.

#3 Dry processing

Glad you touched on this. I tried not to go into the natural process in my last post because it was long as it was but since you brought it up...... haha

What i was trying to bring to light is that a big part of processing coffee is about fermentation. This i believe can bring lots of positive notes to the flavor that do add value. For example, a kenyan soak is a fermentation process.

If you take the same lot (for most varietals) and do a pulped natural and remove all the mucilage you get the taste of the coffee at the stage it was picked.

If you do other processes you continue to harvest flavours.

In a natural dry process you can taste in the cup primarily more body (fat content), more sweetness, and some fruit notes which i presume is from the skin of the cherry as well as other nutrients that continue to migrate in the parchment.

What happens is as the seed (bean) dries and loses humidity it begins to shrink. The space between the parchment increases and this creates a vacuum-like effect of nutrients into the bean. The coffee bean has more time to incorporate these elements. This does make the coffee more complex and therefore should make its score better whether it is a 85 or a 90. Perhaps a clean natural process is the most inherent flavour since all the process is contained within the cherry itself.

The reason why naturals are so tricky is because there is much more humidity and sugar content contained. This makes it easier to ferment in a non-desirable way. What I was trying to argue was that you can in fact dry naturals in a way that the falvour is super clean with a taste profile recognizable of a varietal yet with more complexity. The main factors to control are temperature and humidity which is why places where it is very hot and humid makes it very hard to control this.

Just curious where was the Manzano project?

#4 Re: #1 Dry processing

Ces & Felipe,

Thank you for the thread. Really insightful and informative.
Felipe – your explanation with the graph is brilliant. It couldn’t be clearer.

I am still mulling over this whole scoring topic though. I always thought that compartmenting more rigorously the scoring process would improve the objectivity of the process. Not so sure...


#5 Fermentation, drying and age

Felipe, thank you so much for this post. Your analysis of processing is nothing short of fascinating, and adds so much to my continuing understanding of its complexities. Chris, thank you as well for your excellent report and insights from origin.

I was wondering if in the interest of discussing terminology and clarifying some of these concepts, we could talk specifically about fermentation. I’ve heard it said that often the “fermentation” commonly described in wet processing is not really true or full fermentation. I understand that the purpose of the so-called fermentation is to break down the sugars and alcohols that comprise the remaining fruit mucilage after depulping, and that there are many different methods of doing so. I’m seeking a greater understanding however of how appropriate the term fermentation is for these processes, how wet fermentation differs from dry fermentation (in wet processing), how the post washing soak differs from these (merely in purpose or process?), and ultimately how the fermentation occurring in dry processing is unique from the others.

I can certainly see the added value of these “fermentations” in wet processed coffee, and am constantly fascinated with how the endless variations in processing methods express the coffee in different ways, in the same way that understanding roast development can do after the fact. I think this thread is a great forum to get more in depth with this concept, especially regarding dry processed coffees which are often a subject of division among professionals.

Lastly, not to get too off topic but I was also really intrigued by the comment in Felipe’s first post regarding age/bagginess in relation to drying time. Is this a universally accepted truth that faster drying leads to more bagginess early on? It seems to me the results would be vastly different depending on varietal, climate, specifics of the drying method etc, but I don’t have the experience at the farm level to really know. Anyone else drawn any direct conclusions about drying in relation to age notes in the cup?

Thanks again all for the great conversation

Patrick Grzelewski

#6 Fermentation, drying and age

Hey Patrick

There are a lot of people more qualified then me to talk about this but I'll give some of the topics a go.

Fermentation is to my very basic understanding the break-down of sugars into alcohols and organic acids. From limited experience it can change mainly the type of acidity perceived. Its a tricky topic which should definitely be studied further. So much of what happens is relative to the environment in which it is done and the climate at the time of processing that it is impossible for us to make generalizations. Each farmer has to find a way that works best at his property.

In wet processing: dry fermentation is where the coffee is pulped and then left to ferment exposed to air with no water and wet fermentation is where the coffee ferments under water.

This can be greatly influenced by ambient temperature or water temperature. Hotter temperatures speeds the process up.

Soaking can be good for evening moisture content and leading towards a more even-proper drying.

As for aging some people are doing experiments testing water frequency. The idea is that properly dried coffee should be within a certain range. If coffee dries too fast than water will be leached out faster and fragment the cell. This leads to faster oxidation. This is not scientifically proven but more theories based on observations.

Regarding dry-process naturals. I think people's opinions of dry processing or naturals will change as they taste cleaner coffees. Each process has and should have its influence on whats in the cup. This doesn't by any means keep us from tasting varietal characteristics or terroir.

As for scoring, there is just too much subjectivity to cupping. Peoples tastes are based on their culture, experiences, and personal preferences. Can you say that a Norwegian, an American, an Australian, and a Japanese person will fill out the boxes acidity, mouthfeel and flavour the same way?

Scoring is a good way of measuring what the preferences are and hopefully rewards profiles that we think are better. Hopefully this will lead to rewarding coffees that we want to see more of. We just have to be careful what kind of coffees we are rewarding.

#7 Dry Processing - Hitting the right note

Hi Felipe,
Great to find this discussion on processing. I'm looking at improving my coffee processing and found some of your comments interesting. I currently process using a semi-washed/pulp natural style plus a Tree dried natural (we mechanically harvest) What intrigued me most was your claim to be able to control fermentation in any (Natural processed?) coffee to develop "strawberry yogurt, bubblegum, candy-like flavour"

My experience with processing naturals is that it can be very hit and miss. Some batches score off the chart whilst others are down right bad. To better control this product we have incorporated a sizing screen that comes off the washer separator, sorting the floaters by size into
1) Small, fully tree dried coffee typically peaberry
2) Larger partially dried, blacks
3) Single embryo reds, Green bean - basically waste and/or repass coffee.

We have found this separation has already improved our quality, as before everything that floated went straight in to the mechanical drum dryer. (sun drying is not an option due to high labour costs and frequent harvest rain)

Last season I had a drier full of this partially dried natural when the gear box and motor failed. My first thoughts were that the coffee would be spoilt - over fermented and defective. In fact it was the opposite and it displayed some of the above mentioned flavours (which I personally find good) and was clean.

Interestingly we have even rehydrated some of the partially dried coffee and then pulped it. We found this to be a very clean coffee, with slightly less body but with similar yet muted flavours to coffee I described above. All batches were very consistent - which was more than I could say for my natural processing over the years. I guess it comes down to dry times, humidity and temp. In the perfect world I'd love to be able to manipulate the fermentation process to develop those "good" flavours. I'd be keen to hear your thoughts on how I can achieve this with my drying profile.

#8 sorting and processing

Hey Greenc,

Those are good questions and it seems you are on the right track.

When it comes to nature it is impossible to control the exact result every time so I didn't mean to claim I can do this every time. Although I have seen trends of our coffees dried within a certain range of days, layering on the cherries a bit more thick and allowing the temperature to rise a bit causes these types of flavors. none the less they must be cleanly picked and sorted or else you get aggressive undesirable notes.

What you are mentioning though is more related to separation or sorting before drying. This to me is critical. If we want to evolve from producing good coffees to truly special thorough-bread coffees this is what must be explored more. As you know well the washer does not separate every product but only separates in two categories. Your screen sizing (and some producers here do it as well) I am sure works well as the coffee probably dries more evenly then if it was all mixed together.

As for processing, we still don't understand much in terms of what is happening chemically. Definitely the amount of sugars and the environment and time will cause things to happen. Some of these can be very desirable. Mechanically drying as well as pulping will limit the amount of reactions occurring. Pulping will take off the sugar and mechanically will speed up the drying before cultures can be developed. When your gear box shut off and your coffee sat there (presumably in the shade) it allowed the coffee to ferment and this influenced the flavour.

Just curious where is your farm?

I think it is important for us to keep controlled data of temperature, humidity, and wind so that we can relate what is happening in one-anothers properties.

Natural processing is hard especially when there is a huge variation in ripeness and humidity of the cherries but should not be written off as a bad process.

Probably the most limiting factor for us to really study these methods is the fact that the cost of doing these experiments or drying methods doesn't make fiscal sense. To begin we could try and understand what part of drying influences what in the flavour so we could perhaps combine processes like you are doing with rehydrating your naturals. I'm not a scientist so my method is just try it out and cup it.


#9 Brix

I'm loving where this discussion has gone, and i appreciate your comments, Felipe, about keeping controlled data on these experiments. Along those lines I was wondering if you could speak a little more about how you're using Brix readings both in harvesting and perhaps also in monitoring the drying process?