Chemically Imbalanced (espresso-jogged screeds) » playing with (roaster) fire

coffee, spread out

bz — Tue, 27 May 2008 03:32:49 +0000

the jebena boil: “at the still point, there the dance is.” … pic by jake

it shook all the wobbly, week-long sickies right out of us. just clunked ‘em right out, and ended too the related seven-day spro fast with a sock to the jowls …

here we were, sitting in a rough semicircle at the fine coffee establishment that by now you all know sits by the perfume counter at belk’s department store, and we’re getting a live ethiopian coffee ceremony. with an OPEN FIRE on a sunday afternoon. in the middle. of. the mall.

hudgens, it seems, has a new-ish employee, a gliding, gracious ethiopian, named tigest, who did the honors — nearly two hours of swirling and smiling in regal native garb for one batch of coffee grounds. this blog (which has never been to ethiopia, but assumes, like all americans, that what little it knows about a single african country surely applies to all african countries regardless of their placement on the continent) believed the process to be Profoundly Indicative of the Cultural Pace and Attitudes. it was peaceful, firstly. also, a very obvious emblem of daily life.

the roasting process, being the phase most likely to attract security or set off electronic shopping mall detection systems, took place swiftly and with a symbolic measure of green coffee tigest had picked up somewhere in north carolina. using a traditional long-handled roasting pan, she tossed and swished over a propane burner for what seemed like mere seconds, when the chaffing and cracking and heavy smoking commenced in short order and the batch showed up hissing under our noses for a whiff, gaudy n’ shiny, a mixed roast ranging from still-brown to a hearty french. a blend.

sometime near this point, we agreed this indoor inundation of roasting aroma made fine, fine payback for the constant olfactory affliction shannon has long suffered at the hands of the belk perfume counter. they couldn’t compete with this waft, and NO ONE would be asking for any acqua di gio today. stetson, maybe.

raw incense curled away from a dish. the traditional companion snack — popcorn, in this case — was fire-popped and passed around. the coffee grounds went into the bulbous clay “jebena,” and began to bubble. of interest to this blog: tigest’s practice of pouring off a bit of the liquid in a cup, then waiting for the coffee to begin to boil, at which point she’d reintroduce the poured-off stuff as a coolant to keep the brew in the right range, and also from getting too bubbly. pour, replace, swish about. adjust flame. glance over shoulder to see if anyone in belk has called, “fire!”

each demitasse was warmed with a small slosh of initial coffee, then each fully filled and passed around. as you might imagine, it was one-note strong. nothing overly bitter, or rank. just a clean, constant streak of BLAM, right in the middle of your tongue. hot straw and other earthy materials, maybe. wake-up coffee. black, somewhat crisp and loooong in the finish. being fresh from the throes of a coffee-hating, weeklong illness, we drank all of ours.

they reuse the grounds. ah yes, only three times here in our modern era, although the older ethiopians still do a four-rounder with the same soggy mound. so, then. settle in for round two (“huletegna”), which was more like cloudy drip coffee. then round three (“bereka,” or good luck), which came in like a very earthy tea. as a sort of ceremonial dessert, she did a new set of grounds — the mall-roasted ones from minutes earlier — with a bit of powdered ginger.

potent, almost chewable, and probably the best drink of the day.

what you got, though, was the grace of the thing. the low-key, slice-of-life realness of sacramentalizing through every one of your senses the bread of a culture’s existence. with friends, in the evening, a ceremony can take three hours, tigest said. a day can hold four of them. shannon, doing the math, asked if there aren’t other activities that they might sometimes enjoy.

in truth, the approach seems just woven into life’s fabric, which made it particularly relevant to solis jake, he being about to adopt from the country. whole new meaning to the term, “trip to origin,” eh? even on the mall tile, the ritual was enough to stop plenty of mallers for a gander. a clothing store owner inquired about the mystical circle of sippers. tigest never stopped gazing around and grinning. and security, they never showed.

covert-mall-fire-pan roasting

ritual

one coffee, but a blend.

one theory for the brew’s smoothness: the ‘cense had saturated our nasal cavities.

tigest, at her craft and lifestyle
all pics by jake.

“entertaining” at the bloghouse

bz — Sat, 19 Apr 2008 05:20:23 +0000

it’s a facile, fool-proof way to tell you’re having a really swell time with fabulous barista persons:

you’ve forgotten the pizza crust. this blog gravely apologizes to the traveling atlanta threesome, all of whom were forced to return home with carcinogens in the barista hair.

CI’s favored roast: fifth crack

bz — Wed, 30 Jan 2008 19:05:32 +0000

alas, the hazards of winter home roasting: (a) the cold. (b) the hotness.

we do, however, like the arty kitsch that is the glass fracture pattern splintering upward from the roaster exhaust. very urrrrban.

gog is dead. long live magog.

bz — Thu, 27 Sep 2007 14:36:52 +0000

back to the roaster’s lid-juggle — now curvaceously illustrated and actually tasted!

it’s pretty simple, really. we used our wbc fanboy lot of green kenya gethumbwini peaberry (scroll down) and charted two separate air roasts — let’s call them gog and magog — back to back, in the same weather and electrical contexts.

controlled for sameness:
* the programmed roast profile — 335 degrees for four minutes, 410 for three and 465 for four.
* 125 gram microbatches
* the finished roast — 10 seconds past the first hint of second crack.

the difference between the two as charted below is how we managed the airflow through the roast chamber, or, if you want to be all ghetto and accurate about it, the i-roast’s double-layered removable lids.

essentially, gog was allowed to progress further before the top (second) roaster lid was applied, which traps a lot of air and quickly forces finishing temperatures into the 455-dgree range. in neither case did the roasts “stall,” or flatline, but in magog’s case the second lid dropped much quicker, as soon as temperatures appeared to begin levelling off. thus, magog’s second crack (**) and the finished roast were achieved about a minute earlier.

you’ll note the looooong and gradual warm-up, the barest temperature escalation after first crack and the basic, major variable — you can drop the second lid pretty much whenever you want and fairly dramatically alter the amount of time between first and second crack, which is crucial to your citrus and caramel flava flavs. ultimately, with air, you’re pinpointing the ideal time gap for the beans in question, no?

we’d be interested to know if that final escalation is, in anyone’s estimation, too sudden …

in the cup, gog gave us toasty marshmallow and fresh bread and dried apricot. but oh, magog. she brought us blackberry cobbler and ginseng. it’s not “screams blackberry,” but it’s also non-predictably kombucha-like and sublime!

deidrichizing, now more cerebrally

bz — Fri, 14 Sep 2007 05:18:44 +0000

warning: wee-hour wonkiness follows.

casting about the interconnected webs for some sort of roasting standard whereby one can be sure he is “properly drying” coffee without baking them thar beayns, we stubbed our shins on a handy reminder that the audible first crack in the roast process is an exothermic reaction, meaning th’ beayns generate heat themselves, thus adding to the temperature in the roast chamber.

very quickly thereafter, they revert to the endothermic status of absorbing heat, which is a crucial time to make sure enough outside heat is being applied to avoid stalling the roast. (jaime offers a parallel text). it would seem to this blog that even with a finely tailored air-roast profile, ambient temperatures could easily alter when this switch takes place during a given roast. all the more reason to hover, then, and manipulate the airflow as needed to avoid temperature plateaus.

we’re calling our solution the finger’s lid juggle!

ultimately, “baking” would appear to occur any time temperatures flatline after beans begin to caramelize — or sucrose begins to melt. gurus having been consulted, that benchmark would appear to be roughly 370 degrees.

bonus inquiry: if, as our lady of wikipedia observes, an example of exothermy is “mixing water and strong acids,” then wouldn’t that sort of qualify coffee brewing itself? a gin and tonic with lime? just asking!

the finger sez: deidrichize your roaster!

bz — Thu, 13 Sep 2007 02:58:49 +0000

roasting log: written proof this blog has no idea what it’s doing.

let’s ignore the incoming new york times traffic, shall we? and insist that this is still some hack blog …

they laughed when CI vowed to “simulate the beneficial aspects of the cold” weather on hot-air coffee roasting. well nigh two years later, we might be getting close! handicapped as we are with roasting machines that huff and puff in small batches, it was nate the finger’s idea to deidrich-ize these babies, stealing a page from the drum-roasting juggernaut by controlling the airflow in the roast chamber — instead of temperature itself — to help sculpt an ideal, curvy roast progression.

longtime blog sufferers will recall that numerous attempts to lengthen the amount of time it takes to bring a third-pound of green coffee to a palatable finish have been jettisoned, in part because hot air is so efficient, and can nicely do the cooking in much less time that it takes a rotating drum and a stationery heat source.

imagine the scoffage, then, when the finger told us he was executing 10-minute air roasts. this blog’s exact taunt, it seems, was something along the lines of, “you think you’re a finger, nate, but really you’re just a pinky with a mood ring.” or something to that effect. we can remember our 10-minute roasts producing totally, er, unheard-of coffee flavors. like, say, newt tail.

but here’s the key: we had tried to slow down a roast only by lowering the programmed temperature. nate has altered only the speed at which heat exits the roast chamber. he also takes rather seriously the advice to warm up the granules slooooowly. to do this, this blog has begun using the lowest temperature setting on the roaster and leaving the lid off for four minutes. then, with a fairly standard temperature escalation (400 degrees for three minutes, 455 for four minutes), we’ll ease into first crack — with the lid still off. watching the actual chamber temperature carefully (as opposed to what is programmed), we’ll wait until the temp stalls, not escalating for 10 straight seconds. then we slam on the “first” lid, or the chaff collector (lid diagram).

this has the effect of nudging up the temperature in the chamber by trapping more heat, and since the i-roast has two layers to the top lid, you get to do it twice! only the second time, this blog waits until after the first crack has died, the pause has stretched to 30 seconds or so, and the temperature again threatens to flatline. throw on the second lid, and the batch skips into second crack, which is somewhere near our usual finish line, but the arrival a lot slower and easier to replicate.

in other words, the finger has found a way to program a general roast curve, then tweak it on the fly by restricting airflow whenever a batch needs “help.” it’s like a ghetto, hot-fan roastmaster trick — without the industry jealousy and secrecy! unfortunately, this requires actual personal attention and precludes leaving a roast in progress and, say, completing a private wax. but the results are startling and somewhat mystical. will that deter this blog from far-reaching, overswift and faux-conclusive homilies? it will not.

homily the first: there’s chaff all over the laundry room. that’s what those lids were for!

homily the second: these beans don’t taste chalky, brick-like or overbaked like previous long air roasts. we have no idea why.

homily the third: using the lids — or playing with airflow — allows a home roaster to more deftly manage the length of time between first and second crack. as the rule of thumb goes, the shorter the pause, the higher and brighter the acidic citrus notes in the finished cup. the longer the pause, the more caramel and chocolate shows up. presumably, you could even find hot tar.

homily the fourth: the longer profile, and the freewheeling use of airflow restrictions, seem to greatly negate the profound influence of ambient temperature. previously, a roast curve was only good as long as the weather stayed the same. changes in room temperature and humidity have caused this blog to lose all interest in formerly exalted coffees. lid-juggling is a great way to compensate.

homily the fifth: nate the finger is a studly boar-pig.

CI does the extraction factor juggle

bz — Sat, 02 Sep 2006 19:30:06 +0000

the sulawesi, it tasted of baked hash. maybe some newt tail. extraordinarily gritty.

thus, conjuring the notion that brew temperature is how one buys espresso solubles, the conclusion was reached that we were, ah, overpaying. that is, the extraction of oils was heavily tainted by the extraction of bean infrastructure itself — ribs and spines and ligature. this was not burnt. this was baked and dry. the thinner the swill and the more wretchedly brackish the cup results, the more plain it became that the issue was not of dose or machine anomalies. indeed, close examination revealed beans issuing from the roaster with the taint of mud on their breath.

why should you care? this was a mildly surprising development since, in the summer months, i never have problems with overly baked bean interiors of the kind to easily sell themselves for a bit of hot water. indeed, hot ambient temperatures typically force me to lengthen roast profiles to avoid fast cooking. but this discovery — that mother sulawesi had been crock-potted to death — led to a handful of clarifying barista moments.

elucidation the first: stalling interior bean temperature during the roast — which bakes dry the innards that then (a) taste like terra cotta, and (b) more easily break down under hot water pressure — is closer at hand than one thinks. in this case, the lack of meaningful chaff on the sulawesi left the chaff collector above the roast chamber mostly empty, allowing more hot air to escape the chamber toward the end of the profile. result: long gaps between first and second crack, a glaring sign that the steady temperature climb had plateaued. i should have noticed. this fine line, between a steady rate of increase and a stalled profile, demands an attentive roaster of course — particularly when using hot air. i, frankly, was not very. in fact, the evening roast period is typically when i wander off to brush my teeth. bad roast artisan. bad.

elucidation the second: the most reliable method of covering up undesireable solubles in a coffee seems to be a low dose, cooler temps and short shots. the low dose, of course, demands a finer grind, which would seem to expose MORE of the defective solubles for extraction. instead, i now believe that what’s more important (than previously thought) is the temperature of the water — and its dwell time in the espresso puck. thus, a low dose and a short shot, combined with a cool brew temp, leave the MOST amount of bad solubles in the puck while extracting the necessary oils … which are less plentiful, probably, because we’ve baked them all away. a short shot is all you’ll want!

elucidation the third: this also informs non-defective coffee extraction. with the sulawesi — and most other single-origin espressos i brew — roasted optimally, the increase in water temperature, combined with an already fine grind, is enough to extract a more rounded result. but what if the coffee is, say, rather acidic on the tongue and the goal was to apply a wee damper? then i’d go back to the low-dose, low-temp shot — but lengthen the extraction time.

this is a departure from my previous assumptions that when playing with the flavor of a well-roasted coffee, you just adjust the dose/grind see-saw until you get something that tastes good, then zoom in on the optimal brew temp. instead, i’m starting to think that the brew temp is a better slide rule by itself — and that longer-than-average shots are acceptable too. essentially, the positive traits of the coffee are there for the taking. the low dose is your clarity benchmark. why not let the water temp do the work? it’s a more valuable currency than i thought!

we should note: this blog could change its mind tomorrow.

me. esmeralda. lunch.

bz — Thu, 06 Jul 2006 17:48:28 +0000

annnnd, we’re live-blogging the lunch-hour roasting of a quarter (to the gram) of our esmeralda spoils. because, you know, this blog’s coverage of the runup to the actual consumption of said delicacy has been way too low key.

1:44 p.m. – the one good thing about small-time air roasting devices: it allows us to divide each pound of premium panamanian decadence into four uber-mini batches. i split them last eve, like a mother sea otter counting whelps. nursed them into separate and pristinely clean baggies, i did. it can be confidently stated that only three whole beans were spilt in the entire transfer process. all three were retrieved.

1:56 p.m. – hmmmm. some serious parchment remnant on a coupla these babies.

1:59 p.m. – we detect serious citrus roast smoke. or maybe that’s the nearby reservoir of laundry detergent.

2:02 p.m. – rather large beanage. roasting very evenly, btw. first crack should be fun.

2:04 p.m. – hints of first crack. now a roll to it. very clear and snappy, and rather early in the process considering the lengthened roast curve i chose for this batch. some serious large cojones, these suckers.

2:05 p.m. – as the smoke drifts outdoors, the clouds break for the first time all day and the sun busts a grin. as if to say, “ah, whoa. what IS that groovy waft?”

2:09 p.m. – nice, distinct rest between cracks. nerves are taut for hints of second, before which we will — under great stress — press the “cool” button.

2:10 p.m. – cooling. whew. looking very nice and city-plus-ish, this roast.

2:11 p.m. – prepare the vac seals! also: the blogwife can turn the dryer back on. it’s in the same room and all, you know. must have an unadultered voltage source for the roast. stability being of paramount importance…

2:15 p.m. – only 48 hours’ rest now, precioussss.

2:21 p.m. – join us tomorrow, when this blog will hammer out a gripping real-time account of coffee cherries in the midst of the dramatic patio-drying process — before your very eyes.

2:57 p.m. addendum – arrrggghhh. it’s a good thing we used this cast iron skillet under the gram scale. we dropped a bean! thankfully, it has been recovered and returned to among its degassing brethren. this whole roast-a-rare-coffee-during-lunch thing is just way too stressful.

CI’s logistic-free journey to source

bz — Thu, 22 Jun 2006 03:01:09 +0000

this blog had fomented evening plans including the consumption of the latest netflix installment, deep thoughts on the subject of pending panamanian goodness and perhaps shots from the long-neglected levered device.

alas, it is becoming evident that when innocent bystanders realize you’re the sort of blathering nutcase who would drop a C-note on three pounds of green coffee, they apparently surmise that you’re also up to any caffeinated task. this is pretty much true. thus, a brief encounter with a workplace superior who has an affinity for coffee foilage has left this blog in possession of a wee crop of home-grown coffee cherries and only a cursory clue as to how they should be steered toward eventual roasting.

as we fire up the browser, this blog somehow suspects it is going to have to learn about “de-muscilage” and other gruesome-sounding terms. to which this blog says: bring it on. we also thumb our noses in the face of others’ heady trips to origin. who needs ‘em, when you’ve got plump-ish cherries emanating from local living rooms? it’s like our own transcendent journey to source, but within the comfortable confines of suburbia!

UPDATE: and how, exactly, would you tell your boss that he had, ah, picked his cherries too early?

WBC: afar-blogging

bz — Sun, 21 May 2006 04:15:36 +0000

james hoffmann’s el salvador finca la fany bourbon used in his competition blends.

with james safely in the world barista championship finals, it falls to this blog to make wildly unwarranted performance extrapolations based on a single component of his competition blend.

stunning, really, how a crashing hoffmann had the time to share some greens with an amateur junkie an ocean away. (Wild CI Extrapolation: he manages time well!) the international parcel arrived shortly after the el salvador trip, and constitutes, as i recall, one of that country’s two contributions to a competition blend that consists entirely of central american coffees. (WCIE: he can take a risk!) the roaster, a u.k. stalwart if i’m not mistaken, even sent roasting notes so that this home-bound hack in the southeastern u.s. wasteland mightn’t overly abuse said gift. to say this blog was uber-stoked to be playing with such coffee would be an understatement. (WCIE: that hoffmann can enamor an audience!)

cooked ‘em to a stout full-city plus, one batch just after second crack, another about 15 seconds in. the beguiling result on the buds was a strong sour cherry up front, and an easily marked lingering hickory on the back end … which was fine and good — intriguing even — but not the sweetness our u.k. champeen had noted or the caramel his roaster had suggested. hoffmann’s response: “nitwit! be ye daft?”

actually, no. recommended a longer roast profile, he did, and a hearty updose (17 grams in the double basket) brewed at about 200F. (WCIE: he’s entrenched himself in roasting theory!) this blog, not being a fool, did as it was advised. today’s result, after a dozen or so good shots: ahhhh, intense light-toasty sweetness. a rather resilient light syrupy linger on the palate, and a never-ending subtle dance of secondary notes and flavors. (WCIE: if this be but a single component, that blend must be a nuanced whirlwind!)

ended with a double macc on the porch, reflecting on the soil conditions that must have contributed to this very not-african crisp sweetness, and the u.s. and u.k.’s first-time status as wbc finalists. also: how hoffmann’s height must affect his tamping method. (WCIE: he overcomes challenges! genetic challenges!)

CI: turning over new leaves — again!

bz — Thu, 27 Apr 2006 02:45:01 +0000

in an attempt at transparent self-effacement, this blog prints the following self-effacing snippet — self-effacingly.

“Bready. A bready taste manifests in coffees that have not been roasted long enough or at high enough temperature to bring out the flavor oils.”
— “Home Coffee Roasting,” updated edition

see also, “ploof reduction, simplified.” does this make the just-blogged roast-lengthening exercise one of overwrought discovery-mongering where there was already long-recognized guru evidence showing the same basic thing? did this blog just turn simple bread-mitigation into a three-color chart and a set of pretentious taste notes? does anyone do any preliminary research any more?

grave questions. but i say no! for one, the blend in question had already been roasted to a stout, mouthfeel-reducing vienna, making the detection and elimination of but a single unpleasant component an exercise in sophisticated guesswork. for another, certain east-coast shock-jock podcasters (motto: one airwave is not enough. three waves for us!) have recently claimed to expose the above author’s espresso methods as, er, “less than snobby enough,” if i recall the show correctly. shameless, gotcha, trade-floor journalism, that.

but, in a stunning about-face, this blog endorses this journalistic achievement and insists, in true third-wave fashion, that what i claim to have discovered only seems old — you just “don’t get it“!

oh yeah: actual cup improvements

bz — Tue, 25 Apr 2006 15:25:20 +0000

thought about wading into this nonsense, numchucks afly. but no. perfectly adequate meaty rebuttals have been filed. it’s the new blog! no more superfluous echo-punditry here, no sirree. [my only succinct (!) take is here]. and so, we dragged ourselves back to the bar, where this blog is assiduously working to earn its third wave merit badge — platinum edition.)

temps in celcius

and you thought we had forgotten about the amateurishly dubbed ploof reduction. basic reprisal gist: a (slightly) longer roast profile (yellow line) is indeed indispensably useful is eliminating robusta/malabar taint in the otherwise seductively dark and moody puro scuro. more husky spice, less fruity beer.

the idea: stick with an uber-small (135 grams) batch , swill a good capp to aid in meticulous note-keeping , and wire your own thermoprobe — never trusting the built-in model. reduce the entire roast progression by five degrees, beginning to end. rest five days (the minimum for this blend, in this blog’s opinion). taste.

the blogwife, who as has been noted far eclipses this blog in matters of pure, intuitive taste differentiation, helped pull, slurp and spit the dueling batches five days later. herewith, the unified production of notes:

the shorter roast (red trendline):
* strawberry up front, followed by pineapple. bittersweet cocoa, then very slight caramel. tobacco. beef. then an annoyingly airy beer-head taste, and the oat-yeast unpleasantness at the back end.

the longer roast (yellow trendline):
* huskier raspberry up front. caramel and chocolate. sage and thyme. a much simpler and clearly delineated progression, in other words, and one i’m much more prone to soaking over.

why blog? the production of random taste notes on a coffee few find interesting and still fewer will ever experience is a worthless and overblown coffee-blog filler tactic — agreed. the regurgitation of basic roasting principles has the potential to bore the more seasoned and professionally noted readership — true. just call it public service journalism — the kind of thing nobody reads but everyone likes having the option of reading. besides, this being a blend produced in perpetuity (not seasonal), the notes have some shelf life. it’s a post for the google-monkeys!

besides, this blog was just dog-tired of bloviating about other people bloviating. remarkable, really, how small the changes in temperature must be to produce a taste leap. remarkable, also, how little extra time is required at the reduced temperature. end of remarkableness.

USBC: the orgy of allusions drags on…

bz — Mon, 17 Apr 2006 02:40:55 +0000

rocks, sticks, bling … now popcorn. a kernel showed up in a puro scuro roast today, oddly enough. which sounds sort of like a caragay special, when you think about it.

blundering toward ploof reduction

bz — Thu, 23 Mar 2006 03:04:23 +0000

warning: vast, overreaching conclusion-making follows.

a sign of this blog’s recent troubles: the only source of caffeine for the past 36 hours is excedrin migraine. (not, mind you, that this blog was breaking any federal laws by using medication for non-legitimate uses … the headache, it was real — and probably stems from the lack of caffeine!) recent dawn wakings have prevented industrial espresso-machine clattery that might wake the young-uns, and the late re-entry leaves time only for a modest facial scrubbing and a reacquaintance period with the blogwife.

silver lining: the puro scuro has rested longer before use, which, it turns out, is a good thing. not just because the lengthy degassing happens to lend an even deeper, richer texture to the toned-down mouthfeel and a not-unlike-sidamo pepper finish … but because, this being a dark italian blend of unspecified components, it almost certainly contains portions of either (gulp) robusta or malabar coffee that need quite the long rest period to rid the shottage of overly yeasty qualities. me, i’m a stripped down purist — single-origin arabica coffee, please, or else a modest blend of same. the puro scuro, however, grows very intriguing with the yeast-modifying long rest, and i have grown to deeply appreciate a shot — or seven — late in the evening, on five or six days’ rest, when i get the chance.

but! might a similar effect be had with a darker roast, you ask (gratuitously showing off your knowledge of roasting basics)? why, yes. seemingly. by accident, two darker batches — one roasted light french and another that went way long due to temperature instablity — seemed to not only contain fewer flavor trails in general but also lacked the ploofy robusta taint. (side note: to this blog’s great chagrin, the burnt product produced the closest thing to date of dead-on parisian poo mimicry! join us next week for more frenchie espresso secrets revealed!) obviously, the long rest time is preferrable if yeasty ploof is not your thing. but by tweaking the roast profile from my current baseline to go slightly longer and finish slighty darker, then perhaps the puro scuro will get more alluring still!

or not. trouble is, these subtle improvements seem to take me longer and longer. 10 pounds of beanage, just to eliminate the irksome ploof. *sigh*

updosing is like a cheap date!

bz — Thu, 26 Jan 2006 04:00:36 +0000

the resident home barista over at the third-wave echo chamber restates what some believe is the test case for aussie-style updosing. no one, however, seems eager to do the actual test of instaurator’s hydraulic theory so succinctly extrapolated by george sabados in barista mag and so recently wielded by this blog with all the efficacy of duct-taped numchucks. the further idea that two-dimentional roasting — where the internal bean color closely matches the outside — goes poorly with this dosing method is a novel concept, for me, and one also in need of a healthy poke.

but! this updosing fiddling has at least offered this blog an interim way to sweet-mask the flaws in its less-than-optimal roast batches (a key asset for any snob on a budget). sabados said the technique came about because he wanted an equalizing force to cover barista inconsistency in a pressure environment … so what about an equalizing force to cover roast inconsistency in an amateur environment? so far, i’ve used th’almighty updose to turn barely scorched mexican beans into a passably pungent double latte. and the mud-brick zim became a merely educational sip of crabapple sunflower surprise! call it puro scuro on the cheap! call it espresso in vitro! call it the phantom additive! the result is like the barista’s cheap date — tolerable, easily arranged and with all the trappings of the real thing!

CI clumsily wields the hydraulic principle!

bz — Tue, 24 Jan 2006 04:24:15 +0000

there is a dramatic difference in cup character between SM’s zimbabwe aaa+ and the award-winning leopard forest zimbabwe aa+ bizarrely imported directly to tiny travelers rest, s.c. (the background) — a taste divergence easily explained by numerous differences in the cultivar, processing and import route of the respective coffees. but what struck me was the difference in the way it handled (yes, this subject again … don’t accuse this blog of abandoning a topic some find less than scintillating in the name of eye candy and page views).

the similarities: dry, earthy/grass aromatics. chaffy. extremely even roast finish with both. extreme balance in the cup (extending, in the case of the local import, to near-boring blandness). the tactile experiences were wholly apart, though, and unexpectedly yielded another facet to the theory of what makes a coffee that handles well in the shot-building process. the original intent of the comparisons was to simply get my bearings on the zim, since i’d really become befuddled with the odd balance and “baggy” taste of the local stuff. it was supposed to be award winning, supposed to have been cupped at 87 by the east african contest judges. but, to my palate, it wasn’t in the same league as the kenyas and ethiopians. thus, a direct comparison over various roast profiles, rest times and brew conditions of the aa+ with the expertly vetted aaa+ from my usual supplier.

what i found was that it took a much coarser grind to get the aa+ to pull espresso (2 oz. double shot in 27 seconds). there was a springy/gummy feel to the grounds under the tamper, a very poor cup when down-dosed, and my best example yet of a coffee that brews best with the aussie method of leveraging hydraulics via the up-dose. the aaa+, on the other hand took a bit longer to reach a comparable roast at the same profiles (i started with the exact same weight in green coffee), took a much finer grind to pull as espresso, packed much “cleaner” and “snappier” in the portafilter and brewed very thin in body, though nice and balanced in the flavor profile. there’s no doubt that the darker i roasted the better each bean handled, as theorized earlier. comparatively, however, there was a major difference in how the blander, “wetter”-feeling aa+ handled from the more complex-tasting, “drier”-handling aaa+ in the shot build … which suggests that a shot that feels better may tend to brew better the american, down-dosed way while the more demanding, erratic behavior of a coffee in your hands could indicate a prime candidate for australian up-dosing**. fight the more difficult shot-build, in other words, with the hydraulic virtues of a tightly packed portafilter! use the force! instaurator is your father!

**mind-blowingly counteractive caveat: at least in the case of the twin zimbabwes. and really, who drinks zim? the land of dictatorial power grabs offers far less opportunity for excellence in the cup than, say, its better-known east african neighbors. speaking of which … a foursome of us has just purchased a hefty lot of the wet-processed ethiopian sidamo that has gotten some recent raves. you will likely be force-fed more random and spontaneous quasi-theorems as we delve into it!

your mother lode of tactile adjectives, right here

bz — Fri, 13 Jan 2006 04:59:53 +0000

“this is no game,” he said. “you will never know what it’s like to work on a farm until your hands are raw, just so people can have fresh marijuana.”

which is, like, totally true duuuude. i won’t know. it’s also like saying one’s barista skillz may never be taken seriously because one operates an espresso machine with a meager vibe pump and doesn’t have a budget that affords the luxury of, say, throwing away a third of all brewed shots. also true: i don’t and i don’t. but, ah, the syllogism mixed all up is, and does not produce the stated conclusion. (sorry. i’m not knocking anyone specific, and i guess you gotta be seriously cranial about new yorker satire and to see a link to some of the espresso gurus around whom all of caffeinated cyberspace now seems to orbit). the point: let’s explore more gratuitously the episodic observations of amateurs! or, uh, not. bail now, if it ain’t worth it.

with that for context, let’s review: a recent flurry of working with various beans and blends has led me to the observation that some coffees work better in your hands than others. they don’t necessarily taste better in the cup, but they do tend to more obviously foretell what that taste experience will be like. how? they grind more “cleanly,” sounding like clean snaps in the mazzer, not an overly wet crush or a drier splinter. they distribute more keenly, spreading noticeably easier beneath the finger and offering silky-uniform resistance — unlike the clumpy or more “gummy” feel of some other blends. they tamp with a single distinct “thump” or “crunch” at which point you can literally feel that the grounds have tangibly locked into a single, uniform puck under pressure. when this happens, as it has most recently with the code brown espresso, the barista’s confidence soars and the shots tend to directly reflect what has felt like, in your hands, a sharp, precise and uniform shot-build.

am i describing this phenomenon better than last time? i doubt it. regardless, i’m going to officially retire the issue after this post (i’d prefer to research cup results any day). again: what i’m calling “workable” coffee doesn’t necessarily taste better than other coffees that feel noticeably more uneven in the distribution and more “springy” in the tamp. often, these coffees tend to sound rougher in the grinder as well, though to no apparent detriment in taste. eventually, the shots can be every bit as desireable, but the shot-building process is more demanding, less strewn with quality “flags” and other things that mean the build simply doesn’t feel as good in the process.

i should note some parameters here: we’re talking both home-roasted and commercially roasted beans. we’re talking highly repetitive, long-refined barista techniques here (you’ll have to take my word on it — not pro skills, but attentive and always-improving amateur hands). we’re talking a lot of instances across long periods of brewing all kinds of blends and single-origin coffees. we’re talking very subtle physical hints in how a coffee handles and what causes that in a bean or blend.

now let’s float a morsel of a theory. the roast pictured below, a deliberately drawn-out batch of sulawesi toraja grade 1, was done in a hot-air roaster over a much longer time period than is usually experienced in such a machine (rambling bloviations on why short profiles are ok here), yet carefully roasted at a very steady rate of increase. nonetheless, it tasted horrible:

the catch: the ground beans handled like silly putty fresh from the microwave. numerous roasts at similarly long profiles suggest that, in fact, roast profile has a lot to do with bean workability, and that slow-cooked beans handle extremely well — almost as if the gradual nature of the roast has created more of a fine-sand texture and consistency to the ground product. these grounds produced highly replicable and predictably spot-on pucks that brewed evenly over and over, with fewer occurances of distribution and tamping error than i’m used to and more predictable consistency from shot to shot. unfortunately, roasts this long in a hot-air machine leave the aforementioned tastes of baked brick in your molars.

numerous variations on shorter roast profiles with factors including rest length and ambient temperature maintained comparably, showed that faster-roasted coffee (at least faster-roasted sulawesi) tends to handle more inconsistently. clumpy, harder to distribute perfectly, more springy in the tamp. here’s a more typical sulawesi profile that’s probably a bit quick (i’m still tinkering with moderately longer stuff for my own tastes):

does this mean, if the results of this comparison hold out over a large scale, that longer roast profiles are better? not at all. it means they handle differently, and can sometimes even produce worse espresso. does this mean a hot-air roaster is bound to be produce harder-to-handle espresso? uh-uh. but there are far better places than this blog to find newly emerging hot-air apologetics. obviously, it would stand to reason that the typically longer roast profiles of a drum roaster are what would create highly workable espresso that still tastes good. also, i suspect that roast profile isn’t the only thing that goes into workability. origin would have to play role, and perhaps the grinder too. it could be the grinder, in fact, that holds the key to how espresso feels, and some beans and roasts could simply perform better between the blades! who knows?

am i worn out on this? you bet i am. there was a lot of poo swilled in the knock-around sessions for this screed. i now have loose conclusions biffing around my brain but no empirical results — neither the marketplace of theories that i asked for nor the detailed research others requested, but an intruiging concept nonetheless. that is, if you’re one of those cerebral sorts who might ponder the creative process of off-the-wall new yorker satirists. and i do.

cliffy’s notes

bz — Mon, 09 Jan 2006 03:38:53 +0000

findings in a nutshell: i don’t know what makes some beans workable … but i think roast profile is a major factor. i also know that stretching a batch into 12 minutes on a hot air roaster is a bad idea. makes for general mud-brick characteristics with hints of mud, aftershocks of brick and mud aromatics. also: brick notes in milk! a good idea, on the other hand, is mother sulawesi, roasted extremely evenly (read: slow rate of increase) right to the cusp of a steady second crack, rested two days, with a bit of textured unpasteurized whole milk:

thus ends the short version. when i collate roast curve charts and research more appropriate adjectives, i’ll follow with the unabridged version. oxford edition. with a foreword by the bioluminescent cypriot (who will, of course, take all the blame for newly discovered past inconsistencies).

actual studious follow-up

bz — Thu, 05 Jan 2006 04:18:07 +0000

the aforeshadowed research has consumed days of data-logging, acute grind-performance observation and vocab-busting soul searches for more apt terms to describe the conundrum of what makes some beans and blends more tangibly “workable” in the shot-building process … yet is still wholly unlikely to satisfy the CG curmudgeons formerly loathe to proffer hypotheses or admit to similar lines of inquiry (speaking here of all but the two who stabbed). on the bright side, we may soon be able to blame this blog’s entire array of consistency conundrums on the bioluminescent cypriot! just five more batches of beans to rest and swill, and this blog will offer its own caveat-laden, bum-protective semi-conclusions in this very space. i know: you could cut the apathy with a pallo tool.

overcoming, accidentally

bz — Tue, 06 Dec 2005 01:41:18 +0000

you didn’t want this sulawesi, all muddy and dusty like the falloff from a wall of poorly mixed bricks — the kind that used to make up the crumbling exterior of the church basement of my childhood. used to dig my fingers in the stuff and think i was Super Extra Strong.

anyway, i was mystified — and still am — enough that this makes the second self-effacing blog post this week. seriously. who’s going to read if i’m always dawdling around with unanswered questions hung out on the innerneck like so much soiled linen? still: my suspicion is that the roast finish had something to do with this. namely, the attributes of the roasted sulawesi were light, dry and — hey! — brick-like, even though the total roast length and temperature profile was the same as other, darker-looking roasts, and, more importantly the duration was the same, as measured by the second crack on my 125 grams. my further suspicion is that local voltage is to blame. the supply of electric current, in other words, tends to fluctuate through the back porch outlet and adjoining hom extension cord (imagine that), thus unevenly affecting what might otherwise seem like comparable roasts. bring on the hideously priced variac voltage regulator, then, and companion t-shirt featuring obscure roaster-geek imagery! truth be told, i wanted one of those shirts anyway and an accompanying excuse to hail the might variac depicted thereon. pitiful.

the problem reminds me that i need to finish what i started on the roast curve project, where i posited that ambient temperature is clearly a larger factor in roast times and the ability to “control” a finish than previously thought. duly noted.

after suffering through the masonry macchiatos, another batch — heartily dark, with just a touch of oil after five days’ rest — offered a respite. a deeply spicy, somewhat nutmeg-laced respite. under-dosed. brewed cool (197 F, maybe). you wouldn’t care about this anecdotal delight except that, as illustrated again here, there are a still greater number of frustrating factors to be managed by a home barista intent on sublimity: roast fickleness, grinder limitations, temp flushing, etc. it all makes the GS3 and its stunning “transparency” look more enticing than ever. but this blog is going to france in february, so how could it afford la marzocco’s quasi-home machine of the minutely programmable PID temp control and sand-blasted side panels? dispatches from the country that produced faure’ and 9,000 charred renault carcasses will follow, though. if it’s any consolation.