the espresso blends that i've been happiest with over my career all have one thing in common and that is they've all had some coffee from papua new guinea in them. not every coffee from there is going to be appropriate for an espresso preparation but if i canfind one that is that's what i'd like to buy.
difference between french and italian roast coffee, anything that i'm going to use in espresso is something that i'm going to roast a lot of so if i can figure out if i'm going to use the coffee forespresso that can influence how much of the coffee i want to buy or if i want to buy it at all.
i have a few samples of things that are available and i'm starting by roasting 100 grams of each sample. basic protocol cupping roast, light roast 8 to 12 minutes. with the ir-1 i find that the longer side of that range works best for me. with batches this small temperaturemeasurements aren't accurate especially toward the end of the roastbut the error consistent enough that it's reasonable to abuse this is asample roaster. the main drawback is that the sampletrowel becomes useless so i can't smell the coffee and visualmonitoring is just through the sight glass.
some people say they can evaluate a coffee for suitability for espresso at the cupping table i've never been able to figure out howto do that reliably but it is possible to eliminate some coffees from consideration at the cupping table. there are two things that i consider absolutely critical in a coffee that i'mgoing to use for espresso. first, it needs to be free of defects. one bad defect can make a cup of coffee taste awful and when that's further concentrated intoa shot of espresso it doesn't get any better.
second, the coffee needs to be uniform. five cups in a protocol cupping. for the size of cups that i use that's 12 grams of coffee per cup. most espresso tends to be made as a double which will be made with a little morecoffee one in five cups not being uniform onthe cupping table can easily translate to 1 out of every 3or 4 espresso drinks tasting different so good luck dialing that in. cupping is a really good technique forevaluating those properties
and that's also going to give me a goodidea of what i have available to work withwhen i'm deciding how i want to try roasting that for a drip preparation. you can already smell the lack ofuniformity on that one phenolic no one of these samples is pretty bad not uniform, has defect the other two are better and similar toeach other
i'm getting pineapple, coconut, cherry, there's a bit of a paper-like taste that i don't like and this is a concern because that's a tastethat sometimes goes away for production roasts but sometimes it doesn't. these aren't the greatest coffees from papua new guinea i've ever had on the cupping table and if this were the end of my samplei'd probably pass on all of these but 2 of these aregood enough that i'd like to see how they might do as part of an espresso blend.
the only way i know how to tell how a coffee is going to taste as an espresso is to try it as espresso. but running a cupping roast through the espresso machine isn't going to get me anything palatable so something that i like to do is take another 100 grams from the sample and do what i call an espresso sample roast. just like sample roasting for cuppinglets you evaluate the quality of the green coffeeand gives you an idea of what you can do later during product development tobring out the best in that coffee for a drip preparation,sample roasting
for espresso isn't aiming to necessarily be the best expression of that coffee in espresso but gives you a ballpark idea of what'savailable to work with in that coffee. the same concerns about using the ir-1 for this apply. temperature measurements are consistentbut pretty far off from what they would be with a largerbatch and the sample trowel doesn't work. a traditional sample roaster wouldprobably be better suited for this but i make do with what i have andwith some practice it's possible to do this reliably. my preferences for espresso tend to traditional tastes.
i want something that's smooth with excellent body. i want something with good intensity of flavor. the coffee should be sweet but neveracrid or astringent a good espresso sinks into the tongue and if you don't rinse the taste out ofyour mouth it can stay there for a couple hours. i want something that has enoughcomplexity that it stays interesting to taste during that timeand remains pleasant so you don't want to wash the taste out of your mouth. my espresso sample roast is geared toward those preferences
but there's a lot of room in espresso todo very different things and lots of people are making amazingespresso that i can taste and enjoy but that i personally wouldn't want todrink every day. if you're looking for something that'svery different from what i like i think the general technique i'm using is still applicable but youmight want to experiment with different roasting parameters with coffees youalready have and see what gives you the best predictive ability for the sort ofespresso you enjoy. what i'm shooting for here is a roast
between 16 and 17 minutes to end with acoffee that has between 17 and 18 percent weight loss my temperature readings will be off atthe end but if they weren't i'd be at about 450 degrees fahrenheitby the end. that's about 20 degrees hotter than thestart of second crack so i'm firmly in dark roast territory butstopping well before i'll end up with something that's going to taste like afrench roast. with a batch this small you don't getmany shots so it's also nice to roast a slightlylarger batch of something similar
just to get the grinder and dose set about right. if you're not consistent at making espresso you'll either want to practice that beforeattempting this or get a good barista to help you. that's not a bad idea even if you can pull the shots yourself and some help also makes it easier to dothe evaluation blind. here i have a couple samples that i'm considering. the first one istart and sweet decent body, the acidity is a littleoverpowering and there's sort of a paper-like tasteto it that was
also observed on the lighter cupping roast.it's not very well balanced and it's the sort of coffee that in myexperience isn't going to be well structured. an espresso using this would likely comeacross as sort of a muddled experience thatisn't very good. between the poor suitability for espresso and how this performed on the cupping table i'm not interested in buying this one. the next one i'm immediately getting an aroma like cloves before even tasting the coffee. the cup is sweet and well-balanced there's a good intensity of flavor andit's really interesting.
there's a lot going on with this shotand it's all delicious. it's a lot better than i expected. here is something that i haven't doneany product development with and it's still one of the best singleorigin shots of espresso i've ever tasted. it can probably benefit from blending but it's exceptional as is. i'm definitelyinterested in buying this one considering this technique moregenerally it's common to find coffees that aregood but not excellent as espresso on their own and it'simportant to keep in mind how you might
combine a sample with other coffees in a blend. it's also important to remember thatthis isn't tasting a finished product. maybe acoffee is good but it needs a different roast to control how the coffee expresses itself inespresso. the more practice you have tryingdifferent roasts in blends as espresso the better your intuition will be in connecting what you're tasting here to acoffee buying plan. now that i have the coffee it's time to decide how i really want toroast that.
and i'm using the same procedure asusual. try something, pull a bunch of samples, see how those do on the cupping table. here i have a roast that's slower than my cupping roastand my first sample from that is a little darker than that butstill prior to the start of second crack. the nearest to my espresso sample roast is the seventh cup but this roast getsthere a little faster. the first cup isn't bad on the first sip but as that coffee cools the acidity starts to dominate in unpleasant and unbalanced way.
the 2nd through 10thsamples are all pretty good. that represents a very broad range wherethis coffee is good though there are large differences inhow these taste as the coffee gets darker. the main problem with cups after the 10th one is that the intensity of flavor drops off. what this means is that if i need to adjust the roast for what iwant in espresso chances are good that this will stillwork well on its own in a drip preparation, though that would need to be confirmed through tasting.
considering other ingredients that ihave to work with i decided to start by replicating the second sample at the production roaster with the ideathat i would use this to bring up acidity and intensity in an espresso blend. trying that as espresso the coffee isbright and sweet medium body. on its own it isn't as goodas the espresso sample roast from earlier but as a blend component it'smore interesting i can use this to punch up the intensityand complexity of the blend. the next step is blend construction.
i might have been tempted to try toreplicate the darker sample roast from earlier to get that neat clove note but iobserved that in an espresso prepared from another coffee. this is a coffee from sulawesi that i'veroasted to a french roast. as espresso it's smooth and really sweet. the roast might look unusual to some.this particular lot is very susceptible to roast defects anduneven roasting unless the earliest portion of the roastis taken slowly. that's not true of coffee from sulawesigenerally
but it's an issue with this lot. i found that i liked the coffee best when moving throughsecond crack to the end relatively quickly. the spice characteristics needed highertemperatures to develop but not a lot of time. for a third coffee i have one from brazil. as espresso this one is smooth and chocolatey. it's lacking in intensity but other coffees inthe blend can compensate for that the finished blend with these threecoffees turned out very nice sweet well-balanced and interestingcharacteristics from all three coffees come together nicely.
the last thing to check is that this remains delicious in a milk drink.
sometimes you can get a coffee that makes a nice espresso but when you make a cappuccino out of ityou can no longer taste the coffee and i would consider that to be unacceptable but this blend passes the milk drink test.
