#42: The Best Way to Roast for Espresso | Panel Discussion | SCA Lectures 2018

It’s time to dive deeper into the kinds of questions many roasters have mulled over, but not discussed enough: Should a roast be designed according to flavor profiles, should you roast with a specific brew method in mind? To what extent should a roast be designed for a specific dose, grind, and brew ratio? Should you roast differently for milk-based espresso?

A panel of industry leaders in espresso (Ben Put of Monogram Coffee; Geoff Woodley of Ikawa; Jen Apodaca of Royal Coffee; and Tony Querio of Spyhouse Coffee Roasting Co.) share their theories and approaches to improving espresso coffee in this panel hosted by O. M. Miles.

Related Links
Table of Contents

0:00 Introduction
1:45 Introducing the panelists
3:30 Each panelist was given a Colombian Caturra from the Nariño region and roasted it using an Ikawa. They take turns explaining their roasting philosophy.
18:30 How each panelist incorporates cupping when roasting for espresso
29:45 A discussion on why it’s relevant for roasters to know how to make espressos on bar in order to be better roasters
39:45 Is airflow a primary variable to keep in mind when roasting?
44:30 A discussion on the style of roasting where you keep the same starting and end time and adjust everything in-between instead
46:00 Ways in which having moisture content, density or water activity metrics helps when profiling
51:00 Panelists offer general recommendations for roasting for espresso

Q&A with the audience
60:00 What the roasters think of blends vs single origins and pre-roast blending vs post-roast blending
1:09:00 How to translate Ikawa roast profiles into production roast profiles and opinions on drying vs Maillard vs development on the sample coffee.
1:13:00 Outro

Episode Transcript

0:00 Introduction

Heather Ward: Hello everyone, I’m Heather Ward, Senior Manager of Content Strategy at SCA and you’re listening to the SCA Podcast.

Today’s episode is a part of our SCA Lectures series, dedicated to showcasing a curated selection of the extensive live lectures offered at SCA’s Specialty Coffee Expo and World of Coffee events. Check out the show notes for relevant links and a full transcript of today’s lecture.

This episode was recorded live at the 2018 Specialty Coffee Expo in Seattle. Visit coffeeexpo.org to learn more about this year’s schedule of lectures – and get your tickets!

In this episode, industry leaders in espresso, including Barista Champions and award-winning specialty roasters, share their theories and approaches to improving espresso coffee. Together, they dive deeper into the kinds of questions many roasters have mulled over, but not discussed enough:

Should a roast be designed according to desired flavour profiles? Should you roast with a specific brew method in mind? To what extent should a roast be designed for a specific dose, grind and brew ratio? Should you roast differently for milk based espresso? How to best preserve the terroir and origin characteristics? How should the roast differ depending on rest period?

Now, our panel – Ben Put of Monagram Coffee; Geoff Woodley of Ikawa; Jen Apodaca of Royal Coffee; and Tony Querio of Spyhouse Coffee Roasting Co. – will kick things off for us, led by moderator O. M. Miles.

Also, to help you follow along in this podcast, I’ll jump in occasionally to explain who’s speaking


1:45 Introducing the panelists

OM Miles:  I’m with IKAWA coffee and I’m moderating the panel that we’re all here for this morning. It’s the best way to roast for espresso. I’m a business development manager with a collar for the US. If you have more specific questions just in general. We’re going to be going over an outline for the lecture soon and doing our presentation and then having a really open dialogue. So, we’re going to have some questions at the end. But if you all wouldn’t mind since we’re starting off the lecture here just making sure the cell phones and whatnot are all quiet so we can be a part of the discussion. So, roasting for espresso, bad news, a shocker for us here. There’s not one best way. But in the session though, we’re going to talk about a few different things around this topic like looking at what the folks in the industry are doing and different approaches that we have for roasting for espresso.

We’re going to take a closer look and unpack why, the why behind that why leading professionals are approaching roasting for espresso looking at these ways and we’ll actually, it’s going to be really great because we have some real tangible examples about that why and about how we’re roasting for espresso. And then in regard to future trends will be able to talk about some insights here with our roasting pros and look a little bit more at Innovations and roasting for espresso as well. We’ll also get to do a Q&A at the end. So, if you’ve got some burning questions that are building up as we talk through the panel, we’ll have some time at the end to be able to talk through those.

We’ve got our panel here of our roasting pros and we’ll start with Ben Put. Actually, we should have well, we didn’t have to sit in this order actually but Ben Put he’s down at our end here and Ben is the co-owner of Monogram Coffee based in Alberta and in addition to being the co-owner at Monogram, he’s a world Barista competitor and Champion, has gone to World level with the Barista competition from Canada about three times now and is also an IKAWA user and was one of the folks involved in this experiment that we did with roasting for espresso.

Next is Jen Apodaca, she is the Director of Roasting for the Crown at Royal Coffee and Jen’s got about I want to say 13 years of roasting under her belt. She has roasted for companies like Intelligentsia and Blue Bottle and actually does a lot of educational programming at the Crown with Royal. Next up is Geoff Woodley, he’s the Marketing Manager at IKAWA right over there second-to-last. And before coming to an IKAWA as our marketing manager Geoff was the Director of Coffee for Detour Coffee in Canada. And next up, last but not least is Tony. He is the Director of Coffee as well at Spyhouse Coffee and also super experienced roaster and uses the IKAWA as well at Spyhouse.



5:15 Each panelist was given a Colombian Caturra from the Nariño region and roasted it using an Ikawa. They take turns explaining their roasting philosophy.

OM Miles: So, just a little bit of background on what we did in preparation for this lecture. We all had a kilo of a wash Colombian coffee. It was a Caturra variety from the Nariño region. And this coffee was donated by Royal Coffee Thanks Jen and what we did was take that coffee and you know, we had a very open-ended challenge to roast it for espresso. Next, we each shared those roasted profiles. We roasted those espresso profiles on and IKAWA sample roaster. We shared those profiles with primarily are our filter Alex who I’m going to shout out. He’s also part of our marketing team and we all are going to explain each of those different approaches that we had, why we profiled a specific way on the IKAWA sample roaster with the intention in the mindset of using that coffee for espresso.

So, before we dive into the specific individual profiles that each of our roasting experts here created. I’m going to talk a little bit about what we’re looking at when we when we look at these profiles. So, the red line up top data is being collected. So, this is the Ikawa Pro app and basically, the data that were collecting from the roasting chamber, there’s a temperature sensor actually collecting that information and plotting it in an XY graph that if the majority of us are roasters here I’m sure this looks familiar. It’s going to be based on the exhaust temperature. So, it’s helpful to keep that in mind when we’re talking about these different roast profiles and how they were designed. Below that, the white line is going to be air flow and the air flow specific to the roaster that the panelists were using.

Think of it less as a damper, sort of letting cool air in cool air into the roasting environment and think of it more as actually a fan. So, in the roaster were using a fan and pushing hot air into the roasting chamber and we can control how fast or slow that fan is spinning through that air flow line. So, that’s an illustration of the specific roaster that the panelists were using. To put it into more perspective. So, below is the fan I was referring to, the heating element above that and then the exhaust temperatures being collected by a temperature sensor inside of the roaster.

So, we’re going to dive right In and break down some of these individual profiles that each panel is created again with the intention of roasting for espresso with the Colombian coffee that we all had. So, to dive right In Ben, I’d really love to hear what your what your mindset was for this profile.

Ben Put: Yeah, thanks. So, the way I sort of approach roasting the Ikawa is I think one thing that it’s really highlighted for me is you roast your grinder. So, basically what I tried to do with this profile was approach it in a way that I knew that I was extracting a certain percentage and I think you also really roasted to your brewing parameters and I think this is something that isn’t often highlighted on bigger roasters because they’re harder to recreate execution of stuff. So, I think what it’s really brought up for me is that I know this profile and how it roasts for both my grinder and for my brew ratio. I think that’s been huge. So, often when I roast. I’ll create profiles of templates that have worked in the past. So, the way I did this is I had I’ve roasted a lot of Colombian samples as sample roasts. And what I initially did is I just focused on the development time to see when I hit first crack, can I just extend that a little bit and will that give me enough development?

And usually what happens is as I roast if the coffee still sort of has too much acidity. I’ll start to lengthen the roast, to play with Maillard. I think that often this is another thing that Ikawa I think works really well because you can look in and actually watch the coffee roast and change. You have a really good sense of when you’ve hit Maillard and how Maillards sort of developing. So, I’ve had some espresso roasts where acid isn’t a problem and they’ll be a lot shorter than this. I found the Nariño had lots of nice acidity but had a lot of acid. The other thing I noticed with this coffee is some of the coffees cracked at very different points than others so I tried to come into first crack slower because my concern was if I had a steep rate after first crack that some coffees would develop very quickly, and some would still be underdeveloped. So, I tried to come into first crack slower, so all the coffees had a good amount of time to develop but still go through first crack.

OM Miles: Awesome. Thanks. So, I was able to explain specifically the red and yellow lines a few minutes ago. I want to point out another color that we see in this graph and it’s a yellow line of basically what happened in real time with the specific roast that Ben performed on his roaster. Also, he’s referring to development and first crack as well so I think it might be helpful for me to point out that that line down the towards the end there with the lightning bolt is first crack. So, the lightning bolt just below that will have a timestamp for when first crack was marked and then all the way at the bottom there is DTR which stands for development time ratio. Basically, just taking the time after Ben marked that first crack there in comparison to the entire roast and having that expressed as a ratio percentage inside of the app. So, Jen

Jen Apodaca: Yeah.

OM Miles: Curious to hear your process with this espresso roast.

Jen Apodaca: All right, so I just recently got an Ikawa, like last November and so I’ve just been going nuts playing with it with lots of different ways like playing with air flow and playing with length of roasts and playing with different links of the of the development stages. I apologize that my voice is a little rough. And so, this, I kind of did the same kind of starting point. I had a very successful sample roast that I really truly enjoy, and I decided to instead play with, I really like lighter roasted espressos. I like them to be sweet, but I really do also like a vibrant acidity. I’m more of the type of espresso fan of a more of a long shot. So, I roasted for that type of espresso. I decided to shorten my drying stage and move my Maillard time up by 30 seconds compared to my sample roast to give it more time for Maillard to happen to occur. One thing that I also did is my air flow.

I’ve played a lot with the airflow. It’s been really fun. I’ve been having descending air flow. I’ve had rising air flow. I did this like V format thing has been really fun, I’ve been really liking this one and then like even like a flatlined airflow. But, in this particular one, I decrease my fan speed about the same time that you will notice that yellowing stage would occur and at that point the fan speed it slows down. So, you when you slow the fan speed down, you have the heat that you’re able to attain a little bit more heat. But also, potentially you’re having, I don’t know, maybe we’re having a little bit of conduction heat transfer at that moment. I’m not exactly sure if that’s truly happening, but I wanted to see if it would work and then rising towards the end for smoke abatement as well.

So, this yeah, I don’t know what else. I wanted to keep the same amount of post crack development time as my typical sample roast though. I wanted like a really bright acidity and that most of that comes from the fact that I wanted more of a long a long shot espresso, which I think that those go really nicely together.

OM Miles: Awesome. Geoff, curious to hear about your profile as well.

Geoff Woodley: All right. So, I approached this a little bit differently. I thought it would be interesting to approach this from a perspective of maybe a retail bag being taken home for a home barista or milk drinks. So, I approached this to create an espresso that was quite easy to work with and a little bit lower in acidity with lots of body, lots of syrupy, sweet sort of sticky. Great base for an easy espresso or a milk drink. So, there’s a significant amount of development on this one. So, I approached this wanting to make sure that there was it was still retaining the character of this coffee. It’s really nice bright Colombian coffee, but with the goal of reducing the acidity down to a point where it was easy to work with and at the same time balancing out some of that sweetness in character that’s in there in that coffee naturally. So, I spent quite a bit of time in the Maillard stage with a lot of heat to really try and build some body and then, after first crack keeping a fair amount of heat on there. I’ve found it wasn’t just time after first crack that was relevant to how that coffee tasted. I found actually like a declining or an even richer, the exhaust temperature sort of flattened out some of the character in that coffee and I found that actually gently increasing that exhaust temperature after first crack counter-intuitively actually retained more espresso of the acidity and character in the coffee while balancing out, like I said, some of that acidity lengthening that time after first crack.

OM Miles: Great. Thank you, Geoff. Tony. Curious to hear.

Tony Querio: So, for mine, I’ve worked with the Ikawa up for two or three years now and do all my sample roasting on it. So ultimately, I looked at the coffee. It was a Colombian. It was Nariño, medium-high elevation. It was, you know medium-high density. So, I treated it the same way I treated most of my samples and just ran my sample profile on it. This profile is one that I’ve adopted initially from the one Ikawa published by Rob Hoos a couple years ago on their blog and I worked it. It’s kind of a variation my standard sample profile kept with my team and we kind of looked at the three major factors we look at and copied, total line temp, sweetness, acidity. What flavors are we seeing in this coffee? What do we want to maximize? What do we want to change? We found this coffee had a lot of sweetness and the acidity was very nice. Like poking on that edge of Clementine that like sweet orange with just a little bit of sparkle So, what I did to do that is the same way you profile any other coffee and working with your Maillards, your end temps, your That’s what transfers heat. So, what I do is get my fan speed up where I want it and then. drop the fan right away and sometimes if it’s a denser coffee, it needs more energy. I’ll actually shake the roaster a little bit to get the fluid bed moving but I keep it as low as it can go to keep the beans moving so we’re not just like having them sitting on one spot and getting uneven roast. That’s just been kind of a little trick I learned. So, go ahead and take that home with you guys.

19:45 How each panelist incorporate cupping when roasting for espresso

OM Miles: Thanks, Tony. I haven’t heard of that trick before but I’m going to give it a try. A few observations on my end just before we dive more into the specifics of espresso and it’s relevant as well in roasting. I think it’s just again just an observation. It’s kind of fascinating that this is maybe one of the first conversations that I’ve had with roasters and I consider you all some of my close colleagues at this point and we have these conversations so often and it’s kind of frustrating sometimes because when we talk about roasting, it’s so easy to get into that area of well, you can talk about what you experience and you can talk about it, but there’s no way to really know because we’re all in a different page, the temp sensors were different or different thermocouple, different placement, different size and different airflow things like that. But I’m pretty proud that this is finally, you know one of those discussions where we can all be on the same page. We all used the same green coffee, we really are on the same page with a lot of the variables that were using to have this conversation. So, just wanted to share that as one of my personal observations. I don’t know if anyone here has had that same experience with roasting conversations. You want it to be so productive but it’s hard because you were all working with so many different machines. The other observation that I had is that it’s really interesting that a lot of you were saying that you started with sort of a sample roast profile and then built up from there. And whoever wants can start with this, but I’m really curious to hear actually about how you kind of step to your process from roasting to cupping and then to espresso or did you skip out on the on the cupping portion and jump right into tasting the coffee that you roasted in that specific brew method. I feel like a lot of times a lot of things can get lost in translation. I’ve definitely tasted a coffee that cups really well, but then we throw it throw it into the grinder and pull it as an espresso. It has a whole other personality.

Geoff Woodley: I did a classic like my one of my go-to sample those profiles for the for this coffee just to get a sense of what it tasted like before I profiled for espresso and I did cup that out next to some very obviously very different-looking espresso roasts, and I kind of got a sense of what those coffees tasted like, but I didn’t make any final decisions before pulling them for espresso.

Ben Put: For me, a big thing about Ikawa is that your if you take good data and save profiles, if you delete the profiles that are bad. I think that’s one of the keys. Essentially your profiles evolve, and I think that’s something really interesting is if I go through mine and I look at the first profile I used to the last, they have changed and evolved. And so, you sort of go down a path. I think if you look at all of these. It actually is very similar to evolution. They’re all very different and I imagine, I’m very curious to try all of them because I mentioned they all make a coffee, but I think what’s happened is we’ve all started down a path where we’ve had, for me like it is based on a cupping profile, but obviously it’s a fair bit longer. And then I have some espresso profiles that are shorter that I think the big thing is if you are careful with your iterations your profiles especially on a special will slowly evolve and change and so for me this profile slowly got a little bit longer just so I get a bit more Maillard, but I think that’s a big thing for me is that they sort of evolved from cupping to something else as based on like you deleting the ones that didn’t perform well and then it’s very much like natural selection for me.

Jen Apodaca: I also, so I did four roasts. I did like the one that I said the sample roast that I really enjoy. I did a colleague’s sample roast that’s always performed really. I did a longer roast that was a little bit more developed to accentuate more of like a really good espresso for milk drinks, kind of what you were talking about earlier. And I did one other one. I can’t remember exactly what I did but I tried those four and I tasted the one that I thought that had the best flavor development which is the one that I ended up modifying to get what I have now. And that one, although this coffee has a really nice bright acidity. It’s really juicy. In that particular roast. I got more toasted marshmallow like that vanilla and that’s actually one thing that I really enjoy in espresso, that’s sort of perfectly toasted vanilla marshmallow. So, knowing that I had that I was able to go from there and I cupped again. I also really think that when you cup coffee, whether you’re going to use them for espresso or for filter, you need to have a lot of, not really tackiness for me. But I really enjoy when it feels like raw sugar, that you can feel the sugar on your tongue. And I think that that means that when you then transfer that roasted coffee to the person who is going to brew it, they have a little bit more to manipulate with their brewing equipment and on the espresso machine. Now, I am not a barista, so I don’t know. I can pull a shot if like nobody else is around and I know how to clean a machine and I know how to change a group head, but I don’t know how to, I have never worked at a specialty coffee shop.

I worked at a pre-specialty coffee shop, but that was the Dark Ages. So, I, fortunately, have a colleague who helped me pull shots and she helped me work through knowing what kind of espresso I was trying to go for. And, so again, like my profile really is for that more long shot that I really like because it was really sparkly but it still had enough toasted brown sugar so we worked through some recipes that we thought would work really well with this profile. But, one other thing I wanted to say is my profiles really short and I know that it’s short and so one thing I noticed too Ben that there was a lot of variance in roast degree, especially if I’m going to pull that up. I have a very short profile. So, I just did a pick afterward. I picked out anything that looked a little underdeveloped or quakers. So, if you roast my profile you have to do maybe a couple batches and sort through.

OM Miles: So, we need to buy an optical sorter to test out.

Jen Apodaca: There’s a little more than just roasting. Yeah.

Tony Querio: For me. I really approached this how I would approach it for my Cafe purposes. So, we don’t often have coffees that are dedicated single origin espressos. We have a coffee that when we have it in on the company table, we have it in brew.

This is going to be a great espresso and espresso is a brew method and every coffee will shine better on some brew methods than another. It’s not a hard black and white statement. If you do an immersion brew of certain coffees, they are going to lose their delicacy. You do large batch brews of a very dynamic coffee and some of those dynamics get muddled. You do a small batch of a very dynamic coffee and every cups different. Coffees are unique in themselves but at the end of the day we treat coffee like coffee and kind of see it’s uniformness in a positive manner we don’t change as much. We try to make our jobs easier and I tried to make my jobs for my baristas easier. So, I was looking for something that would make this coffee the best it would be but also not an outlier. That it would be uniform, that if I were to pass this across the bar to a barista and say hey dial this in for me. They would be able to start with our starting specs and kind of be on spot. I think there’s an overarching thing as roasters, your job will be a lot easier if you try and make the people who serve the end products job easier. If you have really crazy out there specs like your coffee may taste amazing that way but you have to do a lot of reteaching and then how do you if you are a wholesaler? In that kind of situation, how do you sell that to an account? Because they probably even treat it the way they treat most of their coffees. So, yes, we need to advance things but be true to yourself, develop your signature and do your signature and consistently do that.

OM Miles: I feel like some people could argue that espresso when you when we look back, you know all the way back to the origin story of the drink that maybe it wasn’t even initially designed to be very tasty. It was definitely a need-based quick drink, you know a very quick way and an express way, if you will, to have your cup of coffee in the morning or throughout the afternoon.




29:45 A discussion on why it’s relevant for roasters to know how to make espressos on bar in order to be better roasters

OM Miles: I’m interested to hear from you all. Why do you think that as roasters especially, so I know some of us have had you know experience being on the barista side of things? But why for us as roasters might just be relevant to understand how to roast for espresso? And I know the word intention has come up a few times and we’ve talked about reverse engineering something so starting with the idea of what we want from our espresso and then working back from there. Why would it be relevant for us to understand how to create that end product? It’s a big one. It’s a thinker. I know.

Ben Put: Why delve into espresso?

OM Miles: Why? Yeah.

Ben Put: I think espresso often is the worst drink we serve. I think it’s so strong. It’s the acid is off. It’s so sour.  I often think of baristas and roasters. It’s like the people that drink whiskey. They can drink whiskey and it doesn’t really affect them and then you give whiskey to someone that’s never had whiskey before and it’s like they hate it. But, people like whiskey. But in our industry, sometimes I drink espressos that I can barely stomach and it’s not because like anyone’s bad at roasting or brewing. It’s just because it’s so hard. And I think more to me like filter roasting is also very hard but espresso like if it doesn’t go well, you’re making like the strongest potentially grossest drink that we charge people money for and I compare buying an espresso in a shop to buying a lottery ticket. You’re spending about the same amount. It’s like $2 to $3 and your odds are one in a couple million. I totally understand and I’m not throwing shade or anything but it’s just such a hard brew method. If the roast is off it’s going to be really tricky if the grinders not behaving you get sort of one shot, you don’t get to taste it. If I have batch brew, I know what I’m serving. So, I just think the reason we need to focus is because it is so stinking hard.

Geoff Woodley: Yeah, I think I’ve definitely been discouraged with espresso in the past like roasting for espresso. And then you do finally hit that roast you feel like I’ve nailed this and then the interesting thing is if you go to a wholesale account, you’ll see the lineup and then 19 out of 20 people are ordering a latte to go or something like that as well. So, it is, I think something we can benefit from paying more attention to and making it better throughout. I’m guilty of definitely ordering a filter more often than an espresso and because of exactly what you were saying there, but things like this. I think this really is fascinating. I’ve learned a lot already today on this and the more we can like dive into the specifics of roasting for espresso. I think the better we can make everybody’s experience from that home Barista to wholesale account or flagship cafe. Regarding roasting for espresso or to drink it black or for a milk drink I think is even a separate discussion altogether. But yeah.

Tony Querio: For me, like totally I could have thought of espresso scary and learning to roast for espresso is very hard because your variables are magnified so much, you know and what is espresso to one person is not espresso to another person. You come down to filter coffee and yeah there’s a range in, there’s a range in times but like with espresso or putting the weight of a kindergartener through a little puck of coffee for 30 seconds. Like it’s so intense and then you have your old school espressos, your new school espressos, espressos can be anything right now. That’s a world we are in so it’s scary and if you’re doing them if you have a multi-beam blend and espresso, you’re saying it’s a ratio of two to one or whatever, but you put in like nineteen beans in that grinder like it’s never shot, that’s not the same idea. So, it’s hard. It’s scary. And that’s where I kind of echo what I said before is knowing what you do and how you do it is going to narrow that so much and understanding that like the espresso you may get in cafes, I work in it’s not going to be the same as you’re going to get in Ben’s cafe or if you go to the Crown. We have our different approach and different styles, but we’re making our product for that and that end result helps a lot.

Jen Apodaca: Which one can we do? So I’m in the unique position right now because I work for an importer and I teach and I have people come up to me, often customers of ours will write to me and they want to they want to get better at their craft and I’ve worked for a couple companies where when you work for a company, you know what kind of menu they want, you know, what kind of coffees they buy, you know what kind of flavors they want to give to their customers and so, as a roaster your job is to put your ego aside and roast how you’re supposed to. You are supposed to deliver on flavor. The flavors that they want to promote, the flavors they want to sell, the flavors that they have their brand behind. But now in the capacity that I’m at now my first question for whoever wants to ask me how can I get better at what I do? It’s like well, what do you want to do? And I consult and speak with people who roast well after second crack to people who maybe pull even before second crack has happened. So, it’s a wide range of possibilities let me tell you.

We’ve never even talked about whether they have washed coffees or natural coffees and what they believe about milk drink building and there’s there are variables. But one thing that. Is been said to me a lot which I find really interesting is that folks who roast light always think that it’s so easy to roast dark and those who and they’re like it’s so easy to roast dark. You just have to burn the coffee, it’s not even a big deal. Roasting light is what’s really hard and then I talked to folks who really love dark roasted coffees and they’re like gosh it’s so easy to be a light roaster. I mean you just like here one snip and just pull the coffee on into the cooling tray. I mean, are they even trying.

And it’s interesting that I can hear from both sides and what I can tell you is if you spend a lot of time and in roasting for espresso, roasting for whatever you are roasting for is that it’s really hard to do all of them. And then if you go into like the different processes of the coffees and different origins, it’s even harder. So, speaking towards roasting. On the dark side since I did not give you my dark side roast. I will say one thing specifically about espresso that I think was a misconception when I was a younger roaster was that we would extend development time quite a bit trying to build more sugar because we’re caramelizing, we’re reducing sugars. We are reducing sugars, but we are also creating a lot of dry distillate flavors. We’re also losing a lot of our organic compounds in there. So, I would suggest if you are roasting you want like a darker roast for your espresso.

I would suggest, no problem reach that high-end temperature, but maybe even shorten your post crack development time because you’ll be able to retain, you’ll be able to get that caramelization that you’re looking for with that final high in temperature, but you’ll also retain a lot of the organic acids that will like also keep that acidity and just structure.

You’ll also be able to yield more after you roast the coffee and you’ll probably be able to use a smaller dose. So, you’ll save money, but I think it would taste better too.

OM Miles: Jen, I’m actually, I want to know if there was motivation to use a shorter roast?

Jen Apodaca: You only asked me for one profile. I could have given you three.

OM Miles: Okay. I just bring it up because in terms of just some of the stuff.  I’m looking at each of these profiles and there’s a few different differentiating factors with each one. When I look at yours. I see the shorter time and of course like in comparison to the longer profiles. Ben, I see your increased airflow, especially at the end of the roast so when that cooling time has started Tony, I think one of your standouts of your profile is definitely the way that you’ve designed the airflow line and then Jeff I think, for me it’s that your profile has that “S” shape that we’re all really familiar with.




39:45 Is airflow a primary variable to keep in mind when roasting?

OM Miles: I know we can’t jump into all of those all at one time but if you had, if each of you had sort of a driving variable that you were adjusting specifically for espresso whether it be… I feel like we’ve talked about Maillard a bit so whether it was manipulating that or really manipulating more so the airflow. I think there’s airflow is really curious it’s an area where a lot of people have questions. So, I think if someone had that as a driving factor. I know it’s hard to pick one because it’s roasting. It’s working with a lot of variables putting them together in a way that works specific for that coffee and its intention. But would you say anyone here had really the air flow as one of the primary variables in mind when you were making your profile? I’m seeing a yes from you Jen.

Jen Apodaca: I think that that’s like a primary concern on this roaster.

OM Miles: Yeah.

Jen Apodaca: It’s everything to do with your heat transfer. I mean if you’re talking about a drum machine then your air flow could be a damper at the back what you’re what you’re doing is if by opening your airflow you are releasing smoke and chaff but you’re also reducing your heat. So, a lot of times if you roast on a machine like that, then when you increase the airflow you might even bump the heat before you increase the airflow to make sure that you have enough heat. If you imagine like opening an oven before you open the oven door all your heats going to go out of your oven when you’re baking those cookies or whatever.

So, it’s kind of like that on a drum roaster where it’s much different here? Here the way the machine works is like you set where you want your temperature to be and then you are adjusting your fan speed and the machine will determine how much heat it needs to get to where you want to go based on those two parameters. So, it’s like driving a completely different car and so you have to think about it in those terms. I would feel like, some part of me still thinks about like smoke abatement because I’ve been roasting on drum machines for so long, but it might not actually be a very big concern on the Ikawa.

OM Miles: You say that because of the cyclone

Jen Apodaca: Yeah. Because the cyclones is always pulling. I’m under the impression that your machine will not let us have zero airflow because it will probably catch on fire.

OM Miles: Yeah.

Jen Apodaca: Yeah.

OM Miles: Exactly. That’s exactly so.

Tony Querio: It will go to sixty.

OM Miles: Exactly!

Tony Querio: Sixty to zero.

OM Miles: Sixty to a hundred.

Tony Querio: So, for me, it’s that same kind of concept. One thing that Miles addressed in the beginning is you only have an exhaust probe on this roaster. So, these curves may look very different than what you’ve seen in other devices or it’s not that S shaped curve that we’re used to seeing because we’re only looking at the exhaust. And like Jen is saying it’s like you are driving a different machine but it’s also driving that machine backwards than what you are used to. You are saying I am used to manipulating my gas. If I want my bean temperature to go up, I give it more gas. But this machine it’s all driven off determining how to keep an exhaust airline. So, if you have a higher differential in your exhaust line, then it’s going to put more heat into it. And the same way with it being fan driving hot air versus fan driving cold air. If you can see in my profile where I have increases, they’re working together because if I’m increasing air and I’m blowing hot air and pushing more energy at the same time. If I go to change the other one but I keep that constant I’ve not only manipulated that factor. These are factors that work together even in a drum roaster. It’s less of a factor drum roaster because you have more conductive energy and you have direct flame elements as well. But those two factors are working together at all times in all roasters.



44:30 A discussion on the style of roasting where you keep the same starting and end time and adjust everything in-between instead

OM Miles: I think it was Tony, you were talking about how you actually kept the same starting and end point with your profile and adjusted basically everything in between with that. I see you’re with your profile you have more of a sharp incline happening after a certain period of time. I’m sure you might have touched on this already just a little bit, but I’m curious to know.

Tony Querio: Why?

OM Miles: Yeah, why that increase at that time, so much later in the roast?

Tony Querio: Yes. So, this profile which I adopted from another roaster you can go back and find it on Ikawa’s blog. It was designed to replicate the standard SCA cupping profile. Being in a roaster things are going to move faster. There’s just more energy transfer. So, the reason why the exhaust temperature is so low for so long is I’m trying to extend my drying period as long as possible and get things more into a realm that I’m more familiar with off of coming, like I came from a Probat PRG 1 sample roaster. So, when I had to switch and [45:47inaudible] based on another machine, well, I’m going to make this easy for myself. So, using a low temperature, low airspeed. I’m trying to retain moisture inside, do energy transfer.

Water transfers energy better than plant cellular structure. Cellulose doesn’t transfer energy while water does so I’m keeping as much moisture in the bean as long as possible and being that this is a fan-driven roaster, drying wants to happen faster. You’re blowing away the moisture. You put a fan over a puddle, and you put a fan in just a puddle there and the puddle with the fan is going to dry out faster. So, that’s kind of the point of that. So, it’s right at right when I’m coming out of drying. So, the beans are starting to turn yellow. That’s when I wrap the heat up because it’s giving off most of its moisture at that point. There’s moisture still created through the chemical reactions, but I’m trying to get drying as long as I can to get back into a more normal situation.



46:00 Ways in which having moisture content, density or water activity metrics helps when profiling

OM Miles: And you touched on the topic of moisture briefly. Does anyone here use variables from any metrics that you’re getting from the green coffee specifically? Things like the moisture content or density water activity. We’ve seen a lot of different profiles that are based on that. Again, not usually for the context of espresso, but did you all find that with this coffee measuring those things helped you create these profiles for the espresso?

Jen Apodaca: Yeah. One of the ways that I adapted it is that there’s three points before the before the end of the roast and the V. I have the V going down, it bottoms out right where my Maillard is or right where yellow stage begins, but I decided to increase the fan speed just before first crack a little bit and that was because this is a dense coffee, and I know that if I increase the fan speed that it will have to kick my heater on harder. And so that way I can give it a little bit more power going through first crack, which will kind of maybe make it crack a little bit closer to each other because there was some variance.

Ben Put: I didn’t use many metrics in terms of measuring, but I think naturally you tend to. This is an aside, but I think it’s one of your colleagues, did a study on water activity versus Maillard. So, if anyone’s curious in that, I think it was like one of the best articles written about it, which is super, super interesting.

OM Miles: Awesome, and where can folks find it?

Ben Put: Is it on your blog?

Jen Apodaca: Yeah, royalcoffee.com. It’s on our blog and just search up water activity. His name is Chris Kornman, you could also search up. He’s written a bunch of stuff about density as well. It’s really good stuff.

Ben Put: It was like one of the most eye-opening things. I think we have all used water activity to give us an idea of how long green will last but the article really dives into how to roast based on water activity, which is super interesting.

Jen Apodaca: So higher water activity, more potential for sugar browning that’s happening during the Maillard stage. So, you might see accelerated Maillard on higher water activity coffees.

OM Miles: And I know we’ve been talking a lot throughout the lecture specific to the Ikawa too but things like that as well. These are general green and roast concepts as well that exists both with a primarily convective roaster and also roasters with different ratios of convective, conductive heat transfer or direct heat or with an electric burner or with an atmospheric burner. These are kind of universal ways to approach how you can profile for espresso or otherwise as well.

Tony Querio: One change I made coming off the previous question. I mean, it’s a Colombian Nariño, had normal water activity normal moisture. It’s pretty dense coffee. So, I knew I had enough cellular material and water still present inside that bean to be able to use a lower heat application, lower-density coffees comes with lower moisture. You have to hit them harder. I have to hit them harder. As you can see there’s four roasts up here that are all probably very good because I look up to all three of you a lot and this is really intimidating. These are probably all goods so take that as a positive when you go home that there’s a lot of different ways to make good coffee. But, if this was a lower density coffee at lower moisture, the temperature where I do that first bump would be higher. I would be giving the roaster more heat to get into my drying phase just because there’s less material that I want to transfer energy there to do it.

51:00 Panelists offer general recommendations for roasting for espresso

OM Miles: We’re going to be in a few minutes here switching into a Q&A and we’ve slated outed a decent amount of time to cover a few questions that anyone here might have for the panelists. But before we get to that though, just general, if we can go down the line with general roasting recommendations for espresso. Maybe some of, I think specifically it helps to maybe demystified a little bit that process for folks and what would be one of the biggest recommendations at that you have for anyone here? Another open thinker question.

Ben Put: So, I think Jen hit the nail on the head for a lot of it. Is that people have sort of glorified first crack quite a bit. And I think it’s super important. I’m not saying don’t worry about first crack, but people will often only adjust first crack and I think a part of it is because it’s super easy to measure, you hear the coffee crack, you can start timing that. And I think especially for espresso all the stuff that happens before first crack is super important. I think, it was Ben Kaminski who once you see Maillard happened on the coffee technically it’s developing. And I think that is a huge part of figuring out your espressos, is how long is your Maillard time based on that, not just based on first crack.

And I think lots of like some other roasting companies are starting to focus on that as well. I think the other big one for espresso. So, when you cup filter or brew filter, we’re all in very similar pages, similar brew ratios. I think the big thing. I think something good and bad. EK espresso has allowed us to evaluate coffees very easily and very quickly, but I think they have made it, so they don’t always transfer well once you put them in a shop. So, if you are profiling your, let’s say I’m sitting in a roastery, I have an EK. I have like perfectly formulated water. I’m extracting to 23%, the shots are big and then suddenly you go to a cafe that using an old an old Robur or something and its pulling18% and you don’t know why it tastes good. You should know, so I think that’s the other big thing is really understand what are your customers pulling your espresso on? Are they using a grinder that’s giving them 19% and if they are then you should be tasting your coffee at 19% or working with them to do that? I think that’s the big thing about espresso is that recipe is so key that it if you’re profiling for in an environment that is different than the actual service environment. Your profiles will never work.

Geoff Woodley: But less on the technical side, but I think for me I would say that it should be very purposeful. There’s like I think it should be a lot of intent with a result when chasing a good espresso because we talked about how difficult espresso is. I think that there has to be some versatility in the result. So, you have to understand that the end result isn’t going to be the same every time that we want it to be but as Ben said the grinders are going to be different, the environments are different, the waters going to be different. And creating an espresso roast in a very confined environment where there’s only one grinder, one person.

Developing that is not going to give you a result that’s maybe going to work as well for everybody as one that you test in different environments, have different people using it. Only cupping espresso roast, I think is challenging. I think pulling it on different machines and making sure that if the intent is to wholesale or to sell this coffee, then it should work well for as many people as possible.

Jen Apodaca: I’m not really sure what I should say. Overall, in roasting which is probably the only advice I can give is just always remember that time, as well as temperature, is affecting your coffee. The longer the times are, your coffee is going to be losing the organic compounds because it’s the longer exposure to heat. If you want to have a really like high acidity coffee, then find a way to extend your Maillard time so that you can create more aromatics and development time before that. And, if you want a longer post crack development time and you want more time for the caramelization to occur, even like shortening your Maillard because they both kind of go hand-in-hand. You can’t really lengthen one without lengthening your total roast time.

So, if you’re going to lengthen your drying stage then you have to Also it’s going to affect how long everything else is. So, all three of those stages, drying stage, Maillard stage and your post crack development time. They all play together. It’s more of I have a 12-minute roast and where am I going to move my sliders? You can’t just, if you lengthen one the next thing, you know, you have a fourteen/fifteen-minute roast, and it’s going to dramatically affect the flavor of your coffee.

Tony Querio: Patience. One of the easy things with profiling for filter is you can turn around in a couple days and make your change. Espresso tends to meet often in classic forms. It gets a little bit more rest to let some of the gases come off and stuff and it can be really hard as a roaster to, after you’ve done this first few batches, you got a good idea of what’s going on. To have to wait that time period, you’re going to want to do what you feel like you should do but then it takes a lot of data tracking and stuff to remember. You can’t not make a change for the ten days cycle that it may run on a pretty normal espresso.

If you know something needs to change, it needs to change but then you have to evaluate that first one based on what you do the first time, not what you’re doing right now because you may have made changes and that’s hard. Yeah. Write down as much stuff because. you think you may remember but you’re not going to remember. Someone’s going to ask you another question in the middle of you doing that batch, and you’re going to get flustered and stuff happens. Record everything and then like Jen has been saying profiling is profiling and know how to profile. The skills for profiling espresso, there’s a little bit further remove that we all talked about with recipe and everything. But, if you know how to adjust sweetness, if you know how to adjust acidity, if you know how to adjust roast level, if you know how to adjust Maillard, drying times, post crack development. Those factors transfer. The one thing that I always try to push to people is coffee science is a science. Coffee is a plant and with roasting and the heat application plants should behave like plants. Food should behave like food and we mystify a lot of this, and we come up with theories that don’t align with other. Watch cooking shows, talk to chefs, see how they do things to get the same results because we’re all dealing with organic matter and flavor compounds.

OM Miles: Yeah, I took a few one-word takeaways from all of your answers and I think some big things to remember in, I guess the overall roasting for espresso process would be data like you’re mentioning whether it’s digital or manual. Capture data take notes that you can refer to later. Be patient, modulation, the little things and the roast. Again, we were talking about first crack a lot. But there’s a lot of other things involved in that process that are really important milestones and things that take note of intention, was one I wrote and then communication as well. I think collaboration could kind of be under or related to that communication topic. I think we covered a lot of ground and it was really great that we actually had some profiles to refer to with each of our individual processes.


Q&A with the audience

60:00 What the roasters think of blends vs single origins and pre-roast blending vs post-roast blending

OM Miles: Want to open it up a little bit and if anyone here has any questions or just generally wants to share their experience with roasting for espresso or found that anything that we were talking about resonated. Please feel free to share.

Question: A member of the audience is asking what do you think of multi-bean blends compared to single origin coffees?

OM Miles: Great question.

Jen Apodaca: Blends are great. I’m a fan of blends. I’m blend positive. I did not create that as such a cool word though, but I got it from those guys at modern times, so I give them credit. But I would just say be wary of your blend components. If you’re creating a flagship blend, then it needs to be consistent year-round or you need to maybe address having seasonal blends. Make sure that not every coffee that you put in your blend is going to be available at its prime flavor year-round. You may have to change every six months or three months. So, make sure that your flavor profile that you’re trying to create is a sustainable flavor that you can create year-round, that’s my advice. And then also I would recommend not getting too crazy. I personally wouldn’t do more than a five-bean blend.

If you’re doing something for like large batch brew and you’re putting in like 90 grams or 200 grams in for a big batch brew, then there’s a really good possibility that your blend percentage ratio will be the same in that big batch Brew, but maybe not in a portafilter when you’re doing 17 grams. So, 17 grams is like. I counted it out with like three or four different types of coffees, but dark roasted, light roasted and Ethiopians. I wanted to know but it’s 130 beans. So, if you have like a seven-bean blend, you might not have consistent shots, that’s what I, by math I would say that that’s probably true but from experience, I think these guys might know better.

OM Miles: Probabilities probability

Question: A member of the audience is asking what are your feelings on post-blending vs pre-blending?

Ben Put: Like post-roast and pre-roast blending. I think the big thing is when you’re doing a blend you have to make sure that your matching solubilities across the coffees because you’re going to put, all that coffee is going to get ground at the same grind setting and it’s also going to get brewed at the same brew ratio unless something really weird happened.

So, I think that’s a big thing. I think if you were going to pre-roast blending you would have to make sure you’re very similar solubility coffees. I think the advantage of blending afterward is you can roast to the correct solubility. I think that’s huge. You want to make sure that when you’re pulling espressos that are blends, that they’re both extracting it percentages that you feel are making the coffee sing and that will also allow you to go back and adjust those roasts accordingly. And I also think it’s important to taste the components separately.

Jen Apodaca: I’m going to disagree with you.

Ben Put: Oh, really?

Jen Apodaca: So, I think that you can do both, and I think that they’re real both are really hard. It can be really easy to do post-blending, but you can actually, and you have to work really, really hard at it. You have to completely change your profiles. If you’re going to try and pre-blend. It is accomplishable. but it’s really hard and you need to be honest and have like a really good blind panel to make sure that you’re creating the same, that they can’t tell which one is which. So, just be honest and like let people give you the criticism but you have to do a lot of hard work to get there. So, it is possible. You should do what you what you think is best. I mean coffees roast in different ways. So, some coffees with certain heat application will roast quicker than others. With solubility, this is the one thing that I think is kind of weird and from my perspective, darker roasted coffees or coffees that have had longer time in the machine usually have a lower TDS. They probably also need more, so you need to dose higher if you’re trying to accomplish a certain extraction percentage. So, if you’re going to post blend and you have a dark, a medium, and a light then that’s why blend ratio is super important. That’s why, actually like your light, you’ll actually have a smaller percentage than your darker roasted base where you have like a higher percentage if that makes sense. That way you’re getting the proper extraction of each. What I think is a great thing to do is if you have multiple blends in your post blending because you want them to be roasted at different levels. Then I would recommend pulling shots of them as components to see how they extract and then using that data in order to get a more cohesive blend based on their how much you add of each component.

Tony Querio: I do both, depending on the context to be honest. For me, part of the initial thing was scale. If you’re doing even at a one to one to one ratio on a three blend, just example. If you have an order come through for 35 pounds of coffee and your roaster outputs 30 pounds, you have a lot of coffee. And this is a business, you got to, if you have the means where you can balance that flux great, if you don’t then you don’t and for me. It was something I did for multiple reasons, but one of the things that gave me the confidence in doing it was that product is going to be ground together. It’s going to be served together. I’m going to evaluate in a roasting together. When I was learning to roast it was before we were really talking about solubility. And the reason people talked about post-roast blends was maximizing each individual component. That’s not the point of a blend. The point of a blend is to have a cohesive whole and if you’re maximizing all of the flavors of all of these, they’re not going to be cohesive. The acidity of one might clash with the other that in a way could work better being treated uniformly. The other challenge with it is to get solubility has to do with how long you’re in a lot of the phases of roast and coffees behave differently. They go through their different phases at different times. But there is research being done by a few people that’s showing if we extend out the drying time, we’ll actually make it so that all the beans in those pre-roast blends go through those phases much more uniformly. So, you do get similar end results.

Geoff Woodley: Yeah, I’ll just quickly add. I think some of my favorite expressos, not all of them, but a lot of them have been blends and I think they have the potential to be really versatile and complex. But I think they are also the more some of the more challenging to do well because of all these things so it because they are sort of seen as a staple. They can sometimes be less efforts put into developing them sometimes and it’s easy to because they’re always there to forget to check back in on them. But I think that a good blend requires constant evaluation and feedback in iterations as coffees change, but also, they can be really great, and a good blend as different coffees rest will behave differently and taste a little bit differently. And a good blend can also kind of rest on each other and extend the sort of sweet spot for when that espresso is really good. So, you can have two coffees sort of morphing together as they rest and bringing that sweet spot out a little bit further and sort of making it easier for baristas or home user, whoever’s pulling that shot to get a good shot over different dates that the espresso goes to roast.

OM Miles: Sounds like you can almost split the answer into like sensory and QC considerations and then also business considerations.

Jen Apodaca: Yeah. We’ve done both in the same blend where we took two components and rested of them together, but we wanted to keep the high component that was only like 15 percent that was also offered as a single origin and we wanted the maximum acidity and floral notes in that coffee. So, it was kind of a little bit of both post- pre blend, pre-post, a mashup.

1:09:00 How to translate Ikawa roast profiles into production roast profiles and opinions on drying vs Maillard vs development on the sample coffee.

Question: A member of the audience is asking how do you approach translating Ikawa roast profiles to production roast profiles.

OM Miles: Hot topic. We can move through this one. This one is probably going to be one we have to move a little bit more quickly through, but something tells me there might be some post-lecture conversation. So, if you all are okay with that some people can ask questions depending on each person’s individual schedule here.

Ben Put: I can speak really quickly to it. I think a lot of people want to be able to take a sample roast and easily apply it to larger production. And I am of the opinion that this only kind of works. To me the way it works is comparatively within coffee. So, if I know that I roasted this Colombian and the profile worked this way and then I roasted another Colombian and required more development time or dried differently, it’s probably going to do that on a larger scale as well. So, I’ve done that before where I can say, oh this coffee did need more heat in the Ikawa. It’s probably going to need more heat on our larger roaster. So, the Ikawa captures a lot of data but if you think about how much is chemically happening in a roaster capture a very small amount of data. So, I think the ability to transfer them is difficult. I do more as comparative.

Jen Apodaca: I think it depends on what you’re trying to transfer it to. Upstairs in The Roaster Village I have my Ikawa roast that was successful and then I do it on the Probotino, that’s a one-kilo. So, it’s going to be pretty similar. It’s going to be a little longer because it’s a little bit larger machine doing a larger batch, which is normal. You’re roasting on a 70 kilo or 120 kilo, your roast times are probably a bit longer than somebody roasting on a 12 kilo or a 5 kilo. And that makes sense because there’s a whole lot more that you’re roasting and it takes so much more thermal energy to produce. But, the Ikawa is interesting to me because if you’re trying to translate it to go to Probot, that’s one thing. If you try to translate it to go to a Loring that’s probably easier actually. Trying to translate it to go to a Diedrich, that’s also very different. All of these different types of roasters have different types of heat transfer, different types of manipulations that are possible and not possible.

So, one thing that I noticed on the Ikawa, just very shortly. I’ve just been playing with it and I still feel like I need to do it a little bit more to see if I have anything conclusive but playing a lot with the airflow. You can really move around when you want first crack to happen. So, a lot of air roasters first crack happens very late, like 400 degrees Fahrenheit or like to 200 Celsius. But on a drum machine, you might be looking, depending on your product placement, you could be looking at anywhere between like 380 or 390 for first crack, maybe even 395, so much lower. So, there’s ways that you can do it to mimic when first crack happened. So, then you can kind of look at like maybe stages of the roast and then trying to translate that but that’s kind of really the first step in and it’s not going to be conclusive with every coffee and just let yourself be wrong and say well that didn’t work. Let’s try something new.

OM Miles: Another takeaway for everyone here. Anyone can head to the app store, be able to download the Ikawa Pro app and be able to try out these profiles for yourself and if you do have any questions for me specifically as someone that is based here in the States here with Ikawa please feel free to let me know and then if you have any espresso questions as well, you guys can come and chat with the panelists here. Just please there’s some evaluations on your chair. If you all could fill those out it would be much appreciated.

Thank you for attending our Roasting for Espresso Lecture.

1:13:00 Outro

Heather Ward: That was O. M. Miles, Tony Querio, Jen Apodaca, Geoff Woodley, and Ben Put at Expo in 2018. Remember to check our show notes for a full transcript of this lecture and visit coffeeexpo.org for tickets to this year’s event.

This has been an episode of the SCA Podcast. Thank you for joining us!


Subscribe to the #SCApodcast on iTunes, Stitcher, Soundcloud, Pocket Casts, or RadioPublic.