So, Counter Culture is hosting these little show-and-tells for the new Marzocco Strada EP. So, Mike & I went to NYC to check it out!
First time on a Greyhound bus, incidentally, not sure how I dodged that in the past. Wasn't too bad! Particularly since I already spend 10 hours+ in the company of my head roaster. Mike & I were keeping a mental tally of how many people we thought had us figured as coming in to the city to get hitched, but that's another story.
Hardest part of the trip was NO CAFFEINE. Made our way to the Ace Hotel before the event, fortunately.
The Strada! The Strada EP! What's the big deal here?
If you haven't been following it, the big deal is Pressure Profiling.
Most espresso machines you'll meet use an electric pump to create pressure while pulling a shot--usually about 9 bar.
(A "bar" is a little more than an "atmosphere"; 9 bar is about 130 psi if you're interested in that kind of thing. Unit conversion!)
Lever machines create pressure differently, using the released force of a spring that the barista sets with a big-ol' lever. Instead of a flat, on-off pressure of 9 bars, lever machines create a curving profile, almost but not quite a bell curve.
When Faema started using electric pumps in 1960/61, they simply took the average of the entire duration of a lever-created pressure cycle, and came up with: 9 bars! Which works pretty well for espresso, let's be honest here, and it became more or less the standard for all machines.
(But by the way, as a guy who works on weird espresso machines sometimes: check your pressure! I run into a lot of older and even newer machines that for whatever reason are pulling low or high, which can do weird things to shots. Most machines have a pressure gauge on them somewhere, but if you don't have one, or don't trust it, you can use a SCACE device (or a technician who has one) to check the pressure right at the head. Most pressure adjustments is done mechanically, right at the pump--you just keep your eye on the gauge and turn a screw. Make sure you pull your test shot through a packed espresso basket, otherwise you won't adjust the pressure correctly.)
So, the electric-pump machines create the same pressure consistently, they're easy to use, great. But there is still among our kind the Legend of How Lever Machines are Great. I can't speak to this personally, not having had a lot of experience with them, but it seems like I've seen periodic discussions/arguments about if and how and why lever machines do some things a little better than pump machines, in terms of how the shot actually tastes coming out.
So, some effort has been given to replicating aspects of lever brewing in modern pump machines. In my understanding, this is part of the reason that espresso machines have started incorporating "pre-infusion": a short period of un- or low-pressurized water contact before the full brewing pressure is applied. This starts saturation/extraction, and causes the bed of espresso to expand. The latter reduces the impact of small bed density irregularities, in theory, while the former changes the brew process slightly and seems to improve flavor.
But the big difference between lever machines and, say, that classic Faema E61, is the pressure profile during extraction. Flat plateau vs. almost-a-bell-curve.
(By the way, there's an attendant theory that bean selection, roasting, and blending practices for espresso have shifted since the introduction of pump machines, much as they probably did to accommodate the drastically higher pressures of lever machines in the '40s. How true this is I don't know, I don't know if the data exists even anecdotally on that (I don't know if the various roasters' organizations have been around or organized long enough to have secret vaults full of that kind of info), but it does kind of make sense. If a roaster is tasting finished, brewed product and then going back to tweak selection/roasting/blending, than a consistent shift in one brewing variable (like everybody switching to flat 9-bar espresso brewing) should in theory create a change in that philosophy. How many roasters actually do that, and how well the habits of great roasters are noticeably successful enough to inspire imitation by the larger community, I don't know.)
Specialty Coffee in all its guises right now is all about difference. Terroir. But the intensity of espresso, the way it exaggerates coffee, and (maybe) the way that the brewing process kind of bottlenecks what tastes good, have encouraged a less-flexible, less varied approach to espresso until recently. That's one of the reasons that espresso has been the Bastion of Blending in the specialty world: pretty much any bean worth the money will taste good brewed up by itself, but not necessarily as espresso. Many is the single-origin bean that just tastes gross as espresso. And so, espresso is blended, for balance, and ideally blended to the (very similar) brewing profile of modern machines.
(Incidentally, I wish I were placed higher in the echelons of specialty, because Michael & I both are waiting, maybe mistakenly, for the Great Blending Renaissance. That's a future post, maybe.)
Regarding espresso, I'm fond of putting it this way in my training classes: the old (Italian) idea was that there was One Perfect Espresso, kind of a Platonic Demitasse. You get your beans right, your technique right, you climb the mountain, and there it is, radiating light, the single (well, probably a double) Paragon Shot.
But now, of course: I don't quite believe that, any more than I believe there is a Platonic Cheese, or Platonic Beer. There are many, dozens or hundreds, of mountains, with phenomenally good and significantly different espressos sitting atop them.
But to get there, up those mountains (anyone familiar with Dennett's probability-space?), we need to be able to explore a lot of different variables. SO Espresso is hard.
Arguably, better temperature control has already started to affect this. The shift here--from One Blend To Rule Them All to the Many-Mountains Paradigm-- is only possible if the brewing variables are both a.) stable enough to be trusted and b.) able to be modified. Temperature instability through the brew cycle creates inconsistent extraction--so you want a blend that tastes okay even if a little over-or-under extracted.
In particular I think this has caused many espresso blends to be hamstringed in terms of acidity, because the beans that create awesome incredible lick-the-bottom-of-the-cup brightness when pulled right are also the ones that create PHREOW bad flavors when over- or under-extracted, so most espresso blends play it safe and don't go broke for tasty acidity.
That problem has been steadily chipped away at by better temperature control technology, everything from the dual-boiler design, better heat exchange machines, to the PID heat controls (arguably the biggest thing since the move from lever machines).
So stable temperatures ought to open up the range of things-that-will-taste-good as espresso. But, point b.), how easily is that temperature adjusted for different coffees? The newer PID machines can be (fairly) easily adjusted, and the highest-end machines now have separate boilers & PID controls for each head. I've even heard some rumblings about machines with dynamic PIDs that allow for temperature to be programmed through the extraction cycle.
Which may be something that Marzocco looked at, I'm not sure. Scott Guglielmino from Marzocco, our narrator for the evening, told us that the Strada resulted from a long period of seeing what variables could be controlled at the machine level, and what had the biggest impact on flavor. And the answer they came up with was not modifying the heat throughout extraction, but modifying the pressure--being able to recreate different profiles.
I've been seeing the first incarnation of the Strada, the MP, around for a bit. At trade shows, at a shop in Montreal, and so on. The MP uses a paddle-group and creates true variable-pressure profiling by (in my understanding) a really tricky system of pressure release valves. Fun to use, but of limited use in a shop setting in this kid's evaluation, both because the profiles are hard/impossible to recreate, and because I really can't imagine not being able to multi-task milk while pulling: you need to have your hands on the paddle the whole time.
(By the way, I really don't know enough, anything at all really, about how the other pressure-profiling machines there, especially the Slayers, compare technically or performance-wise to the Stradas. Comment if you do; I'll have to look into this.)
The EP, now... I admit that I walked into the room (Counter Culture's training center) with a lot of scepticism. The MP didn't impress me, and given how tricky I think espresso already is--so sensitive to the variables we do control--I wasn't sure how important adding another variable would be for on-the-line baristas.
Consider me woo'd.
The EP allows you to manually create a pressure profile using the paddle on the group, or on a computer and then transferred by USB. No, really. After Scott gave a brief background and technical overview on the machine, he turned us loose to play with it. Many variably-profiled shots of Apollo 5.0 later, I have seen the light.
Didn't even intend that joke, I swear.
You can make shots drastically better, worse, different by controlling the pressure. There, that is my position.
The shots that are better, here's my theory, gathered from my experience and everyone in the room and THE INTERNET:
- Infusion vs. percolation brewing. Or saturation vs. washing, however you want to put it. In an on-off pressure situation, you are slamming your coffee with a lot of high-energy water, probably getting most of your extraction yield by physically washing stuff off the particles. Lower pressures at the beginning or end of the extraction should allow more time for the coffee to soak up water and lose solubles through diffusion, possibly changing the composition of the end yield--higher percentage of things that dissolve more easily than they wash and vice versa. I'd really like to see the science on this one.
- Fines migration. This one is so complicated my head hurts already--fines location indirectly affecting a lot of things about the extraction. PROBABLY IT IS AFFECTED BY DIFFERENT PRESSURES.
- Variable extraction rates. Scott put this one forward as one of his main theories, and it makes a lot of sense. We're looking to extract a certain percentage of the bean, not all of it, and we know a lot of the undesirable flavors come from the latter part of the extraction--that 22-35% range that takes a little more time/energy/surface area to extract. So, by ramping the pressure off from a high peak, we can slow extraction and effectively lengthen the window within which we can kill the shot while it tastes good, before it's overextracted.
The Strada EP is loaded up with other design features, all of which are selling points:
- All-stainless PFs. Yes, your PF is actually chrome-plated brass, that's why it's so hard to clean. These suckers should be REALLY easy to keep shiny, which keeps your shots tasty.
- VST (or whoever) baskets (and screens?) I will not flip out about this right now but this is MAYBE A REALLY GOOD THING.
- PFs have a flat bar along the front--they were specifically designed to be tamped with the edge on the counter, the way most good baristas do. The spouts are removable for ease of cleaning, which has the wonderful side-effect of causing the spout to pop harmlessly off if you try to tamp with the spouts on the counter.
- Adjustable drip tray. I will here refrain from slandering 20oz cups.
- I asked Scott about the possibility of volumetric display with this machine (I remain interested in volumetric control and display) and that's when he broke out the big guns. Built-in scales for each head with auto-tare feature tied to the shot ARE YOU SERIOUS. Yes. They're talking about tying that into the programming so you could set a dispensed-mass condition in addition to the pressure profile. (As a tiny niggling point, I'd still like to see a volumetric read on what's going into the shot, just because that's more accurate a way to work with brew ratios than what's coming out. But I digress. Built-in scales of this nature are freaking ridiculously cool and useful.)
- They've switched the way the steam-wand valves work to a solenoid, which means (gasp!) people like me won't have to spend the better years of their lives rebuilding steam wands. Also, the steam wand (works on a lever) has really nice variable pressure itself because of the new controls, which is very cool.
It can work like an Auto-Volumetric.
Meaning, off a reasonably consistent dose/grind, you can use it as push-button machine. OR geek out with it the way you're supposed to. But for the kind of shops I know, where the ability to multi-task can be paramount, and where in all honesty not all baristas (not all owners) are as quality focused as we would like--this ability is a life-saver, because it means that the machine is capable of the Highest Geekery without sacrificing its utility for sheer make-drinks mode.
All told, very, very impressed, and surprised: I wasn't expecting to be sold this hard. I may angle for this machine at our next location.
Otherwise: it was very good to see coffee people, haven't been to any events since Houston. Counter-Culture crowd & NYC folks are very awesome as always. Wound up not hitting as many coffee-bars as I'd planned in New York but that was okay too, had a nice getaway.
I did make it into Everyman, had an incredibly grapey Aeropress of CC's Finca El Puente. Also apparently I mainly take pictures of Sam with his hands on a paddle am I right?
Blog has been dead! And I apologize. But I have a backlog of topics, so we'll see if I can't crank a few more out before the seasons change.
Sounds like a great trip. How was the Stumptown cold brew in a bottle?
ReplyDeleteVST Baskets their own variable/
ReplyDeleteMy reaction about pressure profiling:
Here's a link to an article about the baskets:
http://coffeegeek.com/opinions/markprince/04-29-2011
I consider use of the VST baskets to be a VERY VERY important variable in itself that absolutely needs to be controlled for independently. The reason is that I believe the use of these baskets can make the difference between being able to achieve a desired extraction yield or not.
I recently learned that typical contemporary specialty coffee espresso doses in traditional double baskets produce extraction yields substantially lower than the baskets were designed to produce, because these doses are not what the baskets were designed for.
The reason for dosing more coffee in these baskets is probably mostly that it was a way to help facilitate dosing when the function of the doser mechanism was circumvented by grinding to order, and perhaps a preference for low extraction yields because variations in extraction yield around a low value are less likely to produce over extracted flavors if brewing parameters are not consistent.
I have recently gotten very surprising espresso results with light roasted coffee by dialing in with doses around 14 grams in a traditional double basket, specifically a La Marzocco double.
Dosing this way I've been able to produce palatable espresso that features low toned sweetness from light roasts that would not have produced palatable results using 18 grams or more in the same basket under any circumstances.
I don't have specific numbers on this right now, but it is my understanding that extraction yields considered low (too low) in practice are not uncommon because of this.
This may also possibly explain excessively acidy results with some coffees/roast levels that may typically be considered inappropriate for espresso*.
*It is my understanding that it was the intention of the Slayer Espresso people to attempt to get good nuanced results using more acidy coffees they enjoyed as not espresso for single origin espresso by manipulating pressure profiles.
VST Precision baskets are designed to offer a reasonable range of extraction yields from the doses that contemporary specialty coffee people are actually using.
ReplyDeleteThe VST 18 gram double basket is designed for 17 to 19 grams of coffee. The 22 gram triple is designed for 21 to 23 grams.
In situations where doses are large enough for a given traditional basket that it is impossible to produce a desired rate of extraction from the grounds, fixing this problem by some combination of reducing the dose and/or using VST baskets would be a great priority to "fix" the shot.
I don't know, but I strongly suspect that the trend of doses larger than traditional baskets were designed for is a very important problem responsible for unintended consequences in our whole concept of how espresso works, how roasting results work in the context of espresso, and also perhaps the preference for drum roasters as opposed to fluid bed*.
*Note: drum roasters also look cooler and have much more romance and superior aesthetics.
Pressure profiling is very interesting to me, and I have no doubt that manipulating pressure this way influences results.
However, I suspect that if pressure profiling is being pursued as a possible solution to certain problems with espresso, that there is a very good chance that these problems/challenges/assumptions are a result of the too large doses for traditional baskets problem, and if that is the case the solution is not pressure profiling, even if pressure profiling is interesting and virtuous in itself.
I believe that a very large extent of the benefit of the VST baskets is "fixing" this dosing problem by making large doses work in these baskets.
I think the VST baskets perform better overall, not only by "fixing" this dosing problem.
It worries me that the influence and benefits of these baskets may be confounded as results of other variables, such as pressure profiling.
If Strada EPs are distributed from the factory with these baskets, it would be very easy for different and better results to be attributed to the fabulous expensive new machine, not to $25 VST Precision baskets that will fit in any 58 mm portafilter.
If both baskets and the machine are changed simultaneously it is impossible to determine what is responsible for any different and better results.
These variables must be controlled for separately to actually assess either one.
Here's a link from James Hoffman's blog where he describes this dosing problem and its influence on roasting.
http://www.jimseven.com/2011/05/22/a-lovehate-relationship-with-espresso/
About volumetric dosing buttons:
ReplyDeleteI fully believe that when all other variables are consistent, the amount of water to produce the desired shot in terms of sensory qualities, extraction yield, and brewing ratio, is also consistent.
If on a shot to shot basis the flow rate in a shot varies fast or slow, it is most likely caused by variations in dose size.
A shot with a high or low dose will run fast or slow and will be different whether it is brewed with the desired amount of water or a different amount of water.
Shots that run slow or fast will be more similar in terms of brewing ratio if they are made with the same desired amount of water instead of some other amount of water.
If a shot runs fast or slow the desired volume of water will flow in an amount of time that is also fast or slow, and so stopping a shot early or late to compensate will tend to result in smaller differences in the amount of water used compared to the desired amount than may be expected anyway.
I use the auto volume buttons. There is no reason to me that I should attempt to use some other amount of water when making a shot than the amount I know is what is ideal for the shot I am trying to pull.
If shots runs fast or slow they are faulty, most likely because of an inappropriate dose for that shot (I think this is the overwhelmingly most likely cause), and stopping it early or late is not necessarily better or even much different from letting the machine do the same or nearly the same automatically.
The La Marzoccos include an extremely useful shot timer for reference. To me I'm not saving a shot by trying to use different amounts of water for shots that run slow or fast. Seeing shot times vary when I know everything else has been controlled is diagnostic.
I don't know a better way to facilitate consistent shots in a cafe than using volumetric auto dosing controls.
I can't imagine being able to produce more consistent results in terms of sensory qualities, extraction yield, or brewing ratio without this variable being consistently controlled by a mechanism that does so perfectly.
Most of the important details behind my opinions are a result of the extremely sensible, rational, and instructive teachings of Scott Rao.
ReplyDeleteEvery person who is concerned with making coffee should absolutely read both of his books.
I also recently purchased but haven't yet received or used Extract Mojo.
I think every organization concerned with making coffee should have the ability to measure concentration and calculate extraction using a refractometer and Extract Mojo.
I believe being able to measure and calculate these things is extremely important diagnostic.
Whatever machines a cafe has, cafes tend to have tens of thousands of dollars invested in coffee equipment. The money spent on Extract Mojo to measure its extractions should be a simple choice.
Sorry if this is a double-post, couldn't tell if the page actually submitted my first one.
ReplyDeleteGreat article, a very good read, and I'm right with Dennis' comments.
One question, are you saying the current Strada EP has per-group scales built in, or that this is a planned feature from Marzocco?
Cheers,
Dan
I love my new espresso machine, its my favorite thing in my kitchen!
ReplyDeleteOn scales per group: from the Coffee Geek review:
ReplyDelete"Additionally, there'll be a new option available on the Strada soon from La Marzocco, something that many professional baristas have been begging for: built in scales that are right in the drip tray! At the show, La Marzocco had a preproduction version of the scale (designed by Marco of Uber Boiler fame) at the show built into one Strada, showing one scale on the multi-group machine)."
Lovely.