New Supplier, New Coffees, and More Data

Over the past few months I have changed my coffee buying habits in a lot of ways. For a while I had been buying things in bulk (5 pounds or more). The end result is that I ended up with a LOT of coffee. It is more cost efficient to deal with some of the green coffee buying clubs compared to retailers, but I had been finding the coffee more hit or miss than I would like. In the end, I decided what I really like is VARIETY and really high quality coffee. I would rather have 10 one-pound lots of really individual and fantastic coffee than 10 pounds of a mediocre coffee, even if it costs me significantly more. Net net: I switched to buying most of my coffee from Sweet Marias, much as I had done when I started out in this hobby years ago.

Anyway, fast forward to a few weeks ago, and I came across a posting from the Green Coffee Buying Club about a discount that was being offered from Roastmasters.com. Happy alarm bells started ringing in my head for a reason that took me a day or two to work out. Turns out that Roastmasters is the green coffee/home roasting endeavor of one of my favorite coffee shops - Willoughbys.

You see, my infatuation with great coffee started at Willoughbys in New Haven, Connecticut during my years as a grad student. I had drunk coffee all my life, but finding Willoughbys was like finding an oasis in the desert. They roasted all their own coffee, had tremendous variety, and the brewed coffee they offered was like being in a different galaxy compared to what I had been drinking. In short, I was hooked, and stopping by the cafe became part of my daily ritual. Some of my nicest memories are of me and my wife hanging out outside the cafe, drinking coffee together, enjoying the sunshine, and having a bagel at breakfast time. Even after I graduated and moved to New York, mail-ordered coffee from Willoughbys was my modus operandi.

Realizing this, it wasn't hard to convince myself to buy some green coffee from them. What is really nice, though I haven't done it myself, is that you can buy the same exact coffee from them roasted, as well as the green beans, to try your roasting chops against them. This is a really nice option for those who are learning to roast.

So far I have tried a few coffees from their selection, and they have been really good. They are a bit on the pricey side, but the shipping from Connecticut is a whole lot faster then buying from Sweet Marias. I will definitely be a repeat customer.

The first coffee I tried was an Ethiopia dry processed coffee, Amaro Gayo. These beans have made the rounds lately, gaining a name for themselves as a really nice cup. I have to agree- a City roast I did was packed full of berry aromas and tastes, with a solid cocoa background. I'm glad I bought a couple of pounds. The next coffee I roasted was a lovely, light Guatemala Antigua by the name of "Finca Columbia". As usual, I was playing around with roast profiles, and I managed to come up with a temperature progression I really liked. Here is the derivative curve, as I discussed in my last post. I should also fess up that I did a rolling average to clean up some of the noise in the data.



The roast profile for this coffee was 80% power to start, dropping to 70% at 360 degrees. I have been looking at these derivative curves for a while, and something jumped out at me suddenly. There is a common trait in this temperature data - the rate of temperature change drops consistently at first crack. Thinking a little more, this makes perfect sense; at first crack the water in the beans is powerfully boiling off (hence the cracking sounds). The heat being applied to the beans at this point is going into this reaction, instead of driving the temperature of the beans up. Similarly, at about 400 degrees or so (about 13.5 minutes in this graph) you see an uptick again. This is the point at which you have a City roast - first crack is ending, and the beans start to expand more before eventually reaching second crack. I have overlaid a lot of roast data, and these characteristics are very consistent. It points toward some interesting ideas in terms of roast control, and I see that other folks have also picked up on this in cyberspace, designing controllers and real-time data collection devices to track this rate of temperature rise.

Here is another example I added to the same chart:

This is a Kenya I roasted, and you can see how these derivative charts are useful. I started using the same exact profile as the Guatemala in the previous example, and noticed that it was taking heat slower than I expected. In retrospect, it isn't surprising since Kenyas are such high-grown, dense seeds. I won't bore you with any specific heat calculations, so never fear....

Anyway, you can see where I increased the heat flow and the subsequent rise of the curve, and again the drop when first crack arrived. These diagnostics are very interesting indeed, and I have some other ideas to make them better. By the way, that Kenya Muranga'a was super-intense, sweet, and fruity. Great stuff.

Engineers Gone Wild

One of the things that has been troubling me for a while is better understanding the differences between roast profiles. A change in how a coffee has been roasted, even when the final temperature is the same, can be dramatic. However, looking at the temperature data doesn't always tell the story. Here is a good example:


This is two Kenya beans that were roasted differently. The differences in the curves are quite subtle even though what I was changing in terms of temperatures, times, and fan speeds were very different. Even roasts that were absolutely terrible, like the experiments I did on high drop temperatures, show barely a difference in the roast curves. Some of this is due to the fact that the scales are large since you cover a sizable temperature range in the roast, but some of it is just subtlety in the temperature profile producing large effects in the taste of the coffee.

I was playing around in Excel the other day trying to determine if I was just fooling myself that I have any control over the roasting process. I decided that I needed to look at the data another way, so I decided to take the derivative of the temperature data and see if anything illuminating happened.

Without getting into the mathematical details, the 1st derivative is essentially the rate of change of temperature as a function of time. Taking the same two coffees, plus a third example, you can see the data this way:


The x-axis is still time, but now the y-axis is the change in temperature per unit of time. You can see when the coffee is dropped in at time=zero, there is a huge temperature drop (this is why it is a negative number), and eventually it recovers to something between 10 and 50 degrees of temperature increase per minute until the end of the roast.

One interesting fact immediately popped out after looking at a few charts like this. On the experiments I did with a higher drop temperature, the rate of temperature decrease when the beans are added is substantially higher than when the coffee is dropped at a lower temperature. I am not sure why this is (or if it matters), but I could imagine that that heating element is still on and going strong for the lower drop temperature, and in the other case the environmental temperature and thermal mass have stabilized such that the heater is not on as much. Anyhoo....

The more interesting data is when you zoom in around the region from first crack to the end of the roast:


Don't worry too much about the noisiness in the data; that is typical of this sort of data analysis and the sample size I have used. A couple of things are interesting here. First, I strive to keep a roughly 10 degree per minute temperature rise from first crack to end of roast. You can see that these roasts are doing pretty well. Looking at the earlier times, you can see my attempts to ramp down the heat as first crack approaches in order to have a nice slow, steady progression through that phase of the roast.

The dots on the graph indicate where first crack began on each roast. You can see that the roast of the Kenya Tambaya had a slower ramp rate at first crack than the other two. This is exactly what was intended. I am trying to develop a roast profile that very gently enters first crack. By looking at these curves, it is evident that it is working. The other two Kenya roasts were roasted with the same profile, hotter going into first crack, and the data shows this clearly.

I'll be looking at these derivative curves more as time goes on. It really hones in on important parts of the roast. Amusingly, I came across some other posts on the web where people are actually building micro-controllers for their roasters that track this same 1st derivative data as a hands-on tool for roaster control. Mediocre minds think alike and all that.

Mathematics aside, you are probably wondering how the coffee tastes with the slow entry into first crack. The initial data is extremely promising - very sweet coffees with well integrated acidity and no grassiness. Stay tuned for more coffee reviews.

Drop Temperature and Ethiopia Birbissa again

If you have paid any attention to the roast profiles I have been posting with the Hottop, you'll see that the coffee is typically getting to about 300 degrees in 5-6 minutes. This drying phase is important, as I have talked about before, and really sets the stage for what the coffee is going to taste like.

There are some hints out there that drying the coffee a bit faster can have positive qualities in the cup (more sweetness/acidity, less woodiness), so I started thinking about how to do this.

Unfortunately, all electrically powered roasters are very underpowered. You really need a gas-fired tool to have a significant effect on the ramp speed. Since a gas powered roaster isn't in the cards for me any time soon, I decided to play around with the temperature at which the coffee is added to the roaster. Letting the roaster pre-heat more, I surmised, would speed things up. It did indeed, but with less than stellar results, I am afraid.

Here is a graph of Ethiopia Birbissa with the two roasts:

You'll see that the roast that had the coffee added at 400 degrees did recover faster, and the earlier part of the roast was overall somewhat shorter. The total roast time and final temperature were the same. Comparing my notes from the original roast to this new one was like reading about two different coffees. While the first one was a fruit bomb, the higher drop temperature roast was woody, flatter, and much less interesting. I am not sure if there was a bit of scorching going on or something else, but it wasn't a positive improvement. I did a second test with some Red Sea Blend at the higher drop temperature and had a similar result. Rather than increasing sweetness and/or acidity, these coffees were flat, uninteresting, and overly woody.

It looks like it is back to the drawing board a bit. I have had some really excellent coffees lately with a more traditional roast profile which I hope to post about soon. Also, I may play around with smaller roast loads to see if the ramp speeds can be modified in that way instead of with the drop temperature.

Columbia "Dos Payasos de Tolima"

Well, the few readers of this blog won't be surprised that life intruded once again on poor Scotto, making blogging very infrequent. Rest assured that roasting and drinking coffee has not gone by the wayside; in fact, it has gotten even more interesting. More on that later.

I have been in stash-reduction mode for green coffee for quite a long time. Recently, I finally got around to ordering some more coffee from Sweet Maria's after a long hiatus. Boy, did I forget how stellar the beans are. I buy a fair amount of beans from places like the Green Coffee Buying Club and such, largely because it is inexpensive. It is possible to get good, and occasionally, very good coffees from there, but the coffee from Sweet Marias is uniformly suberb. It is a bit more expensive, but you are essentially paying Tom to be your personal taster and quality-control guy.

This Columbia is a revelation for a fatigued palate. I roasted it inside (Yay! No more freezing garage with the Hottop - more on that another time) and produced a roast curve as follows:

You can see the power levels (the fan was set at 25% for the entire roast) and the boxed area where 1st crack was.

Roasting inside is taking some adjusting in profiles, as the behavior is different from the cold garage. Things were a bit quick at the end, but the coffee was still excellent. It is sweet, sweet grapes all the way in this one. There is a perfect, sparkling acidity, raisins, and an unctuous quality to the coffee that lingers forever. If you like Kenyas and such, you should definitely check this one out. It is an intense coffee, and one of the best Columbias I have had in a while, perhaps since the Cerulean Warbler.

Dinner at WD-50

I’ve been to every 4 star restaurant in New York City, and dozens of other restaurants in the area over the years. I am fortunate enough to be able to afford such pleasures now and again (that's the humble way of saying I worked my ass off to get where I am), and as a foodie and a serious home cook, this is one of the things that I really like to do. WD-50 has always been on my mind to go to, but in the back of my head it was relegated to second-tier status. I finally remedied the situation over the weekend, and now I am kicking myself that it took so long. In a nutshell, the food is thought-provoking, and more importantly, delicious.

WD-50 and its chef, Wylie Dufresne, have gotten lots of press over the years, chiefly because of chef Dufresne’s use of what is called molecular gastronomy. Personally, I think that is a stupid term, but what it describes is the use of science to present food in a different way. Chef Dufresne has been featured on Top Chef as a judge and contestant, and has a way about him that is certainly intriguing. But what about the food?

You see, gimmickry gets you nowhere unless the food tastes great. Any idiot can throw unusual random flavor combinations together, or use liquid nitrogen or gelling agents to create odd textures. At its heart, this is what I feared about WD-50 and what kept me away so long. I don’t want to go to a restaurant and be shocked. I want to eat great food, period.

My wife and I showed up on Saturday evening for our reservation. The staff were attentive and engaging, and we were never left hanging. So far so good. We quickly decided on the tasting menu, and then waffled on whether to get the wine pairing as well. In the end, since the restaurant is offering the awesome deal of 50% each bottle of wine when you order the tasting menu, we quickly decided to go it on our own. We had a bottle of Pichler Riesling to start, which turned out to be a fantastic combination with the first half of the menu.

Note: There are pictures of many of these dishes at the WD-50 website. They are worth a look!

The amuse, “bay scallops, white chocolate, mustard, beer” was the weakest dish of the night, but was still interesting. Since the various items on the plate were so small, I had a hard time to figure out what each individual thing was (besides the scallops). There were curlicues of some sort of crispy thing (mustard flavor, I think), bits of a “cake” (beer flavor?) , a couple of tiny scallops, and a white puree (chocolate?) on the bottom. Taken all together, there was an interesting combination of the scallop with a taste of what evoked for me horseradish cheese spread, of all things.

Next was “everything bagel, smoked salmon threads, crispy cream cheese”. As a Jew who has eaten bagels and lox his entire life, this one really tickled me. It is also a great play on the restaurant’s location, the lower east side of Manhattan, which was once one of the biggest enclaves of Jewish immigrants in the US. What threw me for a loop, though, was the fact that this perfect little “bagel” was.... ice cream! The smoked salmon threads were a flossy/cotton candy texture, but with intensely smoky salmon taste. There was also crispy cream cheese (no idea how they did that), and some really nice pickled onions as well. Again, this was a taste from my childhood transformed into something very unique texturally.

Things were getting interesting after the bagel, and next came “foie gras, passionfruit, Chinese celery”. I am a foie gras nut, especially when cold, so I was pretty excited about this one. The cylinder of foie, when cut, released a lava flow of intense, sweet/acidic passionfruit puree. It was sitting on top of crystallized celery which has sort of a granita-like texture to it, with a slightly sweet taste. The richness of the liver with the sweetness of the passionfruit was absolutely delicious. What was even better was the pairing with the Riesling we were drinking, which itself had intense passionfruit notes. They both played off each other, creating what I would say was one of the top ten combinations I have had in my restaurant dining career. This was an utterly fantastic dish.

Next up: “Scrambled egg ravioli, charred avocado, kindai kampachi”. This was essentially two dishes in parallel, each excellent in their own way, one very homey and traditional, and the other more stripped down and modern. The egg ravioli was a cube of scrambled egg, fluffy and tender, and there was a “gravel” of potatoes. Taken together, you could have been eating classic US diner food – eggs and homefries. At the same time, there was a strip of perfectly fresh, raw tuna and a cylinder of avocado puree that had been charred with a torch. This was a very nice combination as well. I liked the play on classic and modern with this dish, and the way they played off each other.

The egg dish was followed by “Cold fried chicken, buttermilk-ricotta, tabasco, caviar”. This was a cylinder (Chef Dufresne seems to be a big fan of this geometric shape) of cold fried chicken, cooked to perfection, accompanied by fresh ricotta made from buttermilk, thickened Tabasco sauce, and caviar. There were many groans of pleasure during this one. The chicken (not white meat, but not sure exactly what cut it was - perhaps a terrine?) was flavorful and crisp, the soft, tangy ricotta and spicy/vinegary Tabasco cut through it, and there was a subtle brininess from the caviar. Again, a very homey dish spun in a different way. I think the use of Tabasco with caviar is very clever, and something I wouldn’t have thought made sense before I tasted it. The unctuous quality of the fish eggs is often paired with creamy, tangy sour cream, but I think that detracts from their delicacy. Is there a play on the classic blini/sour cream/caviar dish going on here, with the sour cream being replaced by the ricotta? Or perhaps a "Chicken McNugget" provocation? Who knows, but it was very good.

“Perch, kohlrabi, 'dirty' grape, cocoa nib” followed the egg, and I was slightly apprehensive. I really don’t like kohlrabi much; it has a very earthy and musky quality that I find off-putting. This dish was a piece of sauteed perch on top of a verjus (an acidic juice from unripe grapes) gelee, with halved red grapes and small baton of raw kohlrabi on top. There were small shards of cocoa nibs scattered about, which I presume is why the grapes were “dirty”. Fish and cocoa sounds like an odd combination, and perhaps it is, but overall this was a very successful dish for me. The tender fish, sweet acidity of the verjus, and the grapes are a natural partner. However, the musky/bitter quality of the cocoa nibs and kohlrabi kept the dish from being too 1-dimensional. Without the nibs, I think the dish would have been unbalanced. I found that the cocoa created a lingering quality in the mouth as well as an interesting textural counterpoint.

Next was the mysterious “beef and béarnaise”. I had no idea whatsoever this dish was going to be, and it surprised me when a soup bowl was placed in front of me. There were several “béarnaise gnocchi” floating in an intensely-flavored roasted beef consommé. The texture of the gnocchi was yielding and soft - I believe it was "encapsulated sauce", and the broth was unbelievably powerful. When combined in the mouth, it was like eating the very essence of this dish, but in a new way. This dish brought back some intense taste memories for me, since eating Chateaubriand and béarnaise sauce at a local restaurant with my parents growing up is something that has stayed with me. This dish just blew me away.

“Lamb loin, black garlic romesco, soybean, pickled garlic chive” was another dish that got a stunned “wow” from me after the first bite. The Colorado lamb was cooked to perfection, and the garlic Romesco sauce was intensely flavored. However, the thing that took it over the edge for me was the edamame, which were dried into a very crunchy pebble form. That intense vegetal bite balanced out the richness of the lamb and created an unusual and delicious combination

“Caramel apple” was an apple sorbet filled with a center of liquid caramel. A refreshing palate cleanser after the meat dishes

“Hazelnut tart, coconut, chocolate, chicory” had a soft consistency, with the bitterness of the chicory complementing the chocolate.

Looking at my copy of the menu, “Carmelized brioche, apricot, buttercream, lemon thyme” came next. Oddly, I have no memory of it whatsoever. Perhaps too much wine?

Lastly, there were "Cocoa packets, Chocolate shortbread, milk ice cream". Little balls of chocolate that turned out to have ice cream insie, and odd rectangular chocolate packet with some sort of crunchy chocolate in side (the shortbread?).

If there was one annoyance, albeit completely expected, it was the coffee. Why oh why do fine restaurants spend so much time and effort on the food only to serve barely passable coffee? I had the same experience at Per Se where an absolutely transcendent meal was followed by an average Cuppa Joe. A couple of people, including my wife, have suggested that these restaurants have a "coffee Sommelier" or some such, and it makes sense for me. New job for Scotto anyone? After all Thomas Keller serves three different salts and multiple butters with some dishes; are you telling me he can't offer a few different coffee origins to complement dessert? Time to get with the program, guys!

Throughout the meal, I was facing the kitchen watching Chef de Cuisine Jon Bignelli work the pass. I was disappointed that Chef Dufresne was not there, but lo and behold he appeared at the bar later in our meal. Our server took us into the kitchen to look around after, which was neat. I could not believe how calm it was given the intricacy of each plate that had to be put together.

Lastly, I couldn't resist interrupting Wylie at the bar as we left to thank him for a wonderful meal. He was very nice to us, and took the time to chat for a moment. I am regretting not getting a picture with him, but that would have been a bit geeky. I know how hard these guys work, and I felt a little guilty about bothering him when he was trying to relax a bit.

Anyway, sorry for the long post, but I needed to write it down so the memory would linger longer. It was a fantastic evening.

Ethiopia Dry-Process Birbissa

As I think I have mentioned, there is very little information on "optimum roast profiles" out there. Searching the internet will find tidbits, but it is still a DYI affair for the most part. If I had to sum up the conventional wisdom such as it is, it would be this:

• Keep roast times below 14 minutes or so
• Slow start/fast finish is better than the reverse
• There are three main segments of the roast:

1. The drying segment, from start to about 300 degrees, should be about 3-5 minutes (longer for light roasts and shorter for dark roasts)
2. The segment from 300 degrees to the start of first crack should be as rapid as possible to avoid woody flavors
3. From the onset of first crack to end of roast should be 3-5 minutes. Shorter for drip and longer for espresso

Got all that?

Let's have a look at this roast of Ethiopia Birbissa:


Under 14 minutes? Check.
5 minutes to dry? Check.
Fast ramp to first? Maybe.

Pretty good against the "standard" if I do say so myself. Too bad I totally spaced out during the roast and forgot to lower the heat when I originally wanted to, but in the end it seems to have worked out. I wished I had pulled it a bit sooner since the roast got a bit darker than I had intended, but so be it.

The aroma in the cup was amazing. Pure, pure blueberies all the way. This is really the first coffee I have had that has been so distinct in one berry note, though I jave had many others with grape, strawberry, etc. Very nice. The taste follows suit, but the acidity was lower than I wanted, either due to the coffee or the roast - I am not sure. The body was syrupy and overall this was a nice dry-process cup, but nothing that exciting, IMO.

Colombia Finca Buenavista -Carlos Imbachi Microlot

Boy, did I forget how hard stainless steel is to machine with home tools. Over the weekend I set out to install a thermometer probe into my Hottop, following the basic method outlined here. Sounds easy, right? Well, let's just say that much cursing ensued as I tried to drill through things with my electric drill. Anyway, its in there, giving me access to the actual bean mass temperature instead of just the Hottop's sensor which is bolted to the back wall. And a good thing, too! See:

This was a roast of a Columbian bean that I had ruined (and never blogged about, of course) with my Gene Cafe when it was having troubles. The blue line is from the Hottop readout and the pink from my new thermoprobe. Obviously, one should pay attention to the pink line, which is a fast acting K-type thermocouple buried in the rotating beans. You'll notice the temperature drop when you add the beans which is not sensed by the stock thermocouple, and the eventual crossover of the curves.

I highlighed the times where I played with the heater power to slow things down, and I also boxed out the area where first crack started and ended from about 10:45 to 13:30. What I was attempting to do was slow down the rate of heat transfer going into first crack in order to improve flavor development and have enough time after the end of first and before the onset of second crack. You can see that I stalled things out a bit where the pink curve flattens around 11 minutes in, and eventually it ramps up again later until I dropped the beans into the cooling tray. I need to play with the heat and fan settings more to have a slow but steady temperature increase instead of this stalling behavior. This roaster is like driving a Ferrari compared with the very slow reacting Gene Cafe. Still, you clearly need to anticipate things by a minute or so given the electric heating on the Hottop. Lots of room for improvement, that's for sure.

Anyway, the coffee was a nice City+ roast and was actually a decent cup. There were very intense fruity notes both in the aroma and taste. I called it tamarind in my notes, but I see that Tom from Sweetmarias called it guava. Your choice. The acidity was nicely controlled, and the cup was quite refreshing. As it cooled, there was a bit of a flavor change that is hard to describe, but I attribute it to stalling out the roast a bit and having things drag on too long. Let's call it "flat". Most people wouldn't pick up on it, but after roasting a lot of coffee I am starting to be able to tell what went wrong.

My next few efforts will be to speed things up a tad, and keep playing with the heat settings.