I thought I was pretty clever, coming in to work before 5am to swap out the burrs in our main coffee grinder. One hour and a number of odd complications later, I was starting to sweat at the thought of all those coffee orders just over the horizon...fortunately, I got everything brewing in time. Our fix-everything-man came by and helped me figure out the problem: turns out I was just being too gentle in trying to loosen a bolt. Gentleness aside, I did manage to stab myself pretty good with a screwdriver:

Lesson learned.

(By the way, I am very pleased with the Salter kitchen scale I've been using to set the grind. It's only accurate down to about an eighth of an ounce, but that is reasonably okay for the volumes I'm doing. The long on-time is great, and anyways the digital is a great improvement over “Old Bouncy” the mechanical kitchen scale. I heard about it over on Barista Exchange in this thread.)

I have to say that my appreciation for the Fetco Grinder/Brewer set-up has gone up a lot lately. After monkeying around with some brewing ratios & parameters, I have our house blend at a place that I really like. It's not radically different, just a little better—I don't have a refractometer (yet), but a TDS meter, brew chart, and some math put it right where I want it. And my palate agrees! I've even got a few good comments from customers, which is nice.

However, working on the grinder today, I couldn't help but notice the way that grinders are lagging behind other coffee equipment. Any coffee geek can tell you the primary things we want out of a grinder. Extra kudos might be given for good dosing mechanisms (particularly on espresso grinders) and limiting the amount of grounds that stick somewhere other than the dosing target, but really we just want it to perform well on the basics: it should be fast, quiet, and cool, the range of particle-sizes should be fairly small for a given setting, and it should be finely adjustable (ideally stepless).

The thing is...those qualities are expected out of any decent commercial grade grinder today, whether it's for espresso, brew-to-order, auto-drips, or bulk grinding. I think that there are a few really significant improvements that could be made to all categories of coffee grinder. And the big one is, you guessed it:

1.) Grinding by mass.

That's really what we want when we're grinding: a specific weight of ground coffee, to brew with a specific amount of water. (Thank you, universe, for keeping water's mass pretty much stable to its volume as far as us baristas are concerned.) The problem is, most of us coffee-brewers are grinding by volume or time. At home—well, most people at home probably aren't even grinding, but either way they're probably just using a scoop of some kind—measuring by volume. The only automatic grinders for drip coffee that I've seen use a timer to allow beans to fall into the grinding chamber; you just have to use trial and error to get the right mass of coffee—beans vary a lot, and as the burrs age they don't grind as fast. When using an espresso grinder, one doses by eyeballing the basket (volume), except for those fancy new electronic doserless models (timed). The only time I see baristas consistently dosing and grinding by mass is in shops with brew-to-order methods—except for those Mallyke press-pot grinders (timed). The standard method seems to be just scooping some beans into a cup on an accurate scale before grinding.

So, what's wrong with these methods? Screwdriver-wound in my hand still throbbing, it's mainly those timed auto-drip grinders that are bugging me. They work pretty well—if you have someone that knows how to set them, and is using an accurate scale. If the beans don't vary for any reason—that is, if they are all of uniform size and density. If the grinder is regularly re-programmed as the burrs wear, and finally if you don't mind grinding a ton of coffee while you zero in on that perfect time.

Too many coffee shops don't have the regular equipment maintenance that they need. That's a separate and much larger problem, but it's a fact. The more coffee equipment that will make good coffee in between service visits, the better the coffee will be in the average cafe. Personal experience with our grinders suggests to me that quarterly re-calibrating is the absolute minimum, and unknown events can create a need for adjustment more often than that. I don't have the statistics, but I doubt the average American coffee-shop is getting thorough quarterly maintenance.

Beans vary. Using different blends? Using the same grinder setting for a French-Roasted Indo as you are for that lightly roasted peaberry? The mass : volume ratio for these beans is significantly different, as is bean size, and that plays out in noticeable differences in dose when using the a timed auto-grinder. So what do you do? My grinder has 2 hoppers with 3 settings each, so in theory I can work with my roaster to find a mass:volume ratio for each bean and blend he sends my way, break those into 3 (or 6) time:mass profiles, program the buttons to fit those profiles, label each bag accordingly and train my staff to grind that way, and then replicate that system at all our locations.

I am far enough down the rabbit-hole that I am giving this real serious thought, at least for 3 mass:volume categories: the thought of a 3-liter brew varying by as much as an ounce or two in the dose kinda gives me shivers. But it's not the most elegant solution.

For shops that are doing brew-to-order of whatever kind, pour-over rails, Clovers, etc., the issue isn't correct dosing—measuring onto a .01g accurate scale handles that. It's training, and time in coffee prep. Now, granted, cutting-edge shops seem to have these kinds of systems down. For a shop that is very interested in the all the benefits of brew-to-order, but is considering starting such a system from scratch (*cough*), the element of training everyone to dose correctly and the time spent on each step of brew-to-order looks a little intimidating. A programmable grinder that worked by mass would knock out a huge amount of concerns for these operations, and would probably be welcomed by the cutting-edge as well.

Espresso—well, on the one hand espresso grinders work pretty well. Eyeball-dosing or using timed doserless controls both seem to work pretty well. If you're using multiple beans for espresso, you almost certainly have a grinder for each, so that's not a problem. On the other hand, if someone did invent an espresso grinder that could quickly grind mass-accurate down to a hundredth of a gram or so, I'm sure it would not be left out in the cold.

So, let's invent this thing. I don't think it would be too crazy. Basically you just need to insert a step in between the hopper and the grinding chamber, where the coffee is measured by weight before being ground. I can think of a number of ways to do this—from a paddle-wheel or gum-ball-dispenser-type device to doll out the beans at a measured rate, or just a “stutter” of timed releases. A smart enough machine should be able to learn a mass:volume/time ratio from one partial dose, and then achieve the desired full dose with subsequent cycles. It would be very important for this machine to “finish clean”, with no grinds or whole beans left in, since you're brewing different beans and don't want to contaminate. Perhaps a “bypass shoot,” where leftover whole beans are shot out unground? As always with coffee equipment, compactness is important, a minimum of moving parts is a plus, and the ability to clean it easily and often is paramount. Bringing us to the next item on my wish list:

2.) Cleaner insides. This is a peevish comment, maybe, but I can't believe how gunked up grinders get inside. Microscopic coffee particles, in all their oily goodness, get everywhere, and congeal into this weird clay-like substance all over the place. I can't believe that they're not affecting flavor as they're knocked loose with each grind cycle. There has got to be a way to seal up the area around the burrs better, and to prevent oil/particle build-up. While we're building this dream machine, let's also put on a “purge” button, that grinds a very small amount of beans to make sure the residual grounds aren't from the previous batch.

3.) Since we're being whimsical, let's also tack on a computer brain that analyzes particle-size (maybe even surface area), probably by some kind of optical measurement, and

4.) Electronically controlled grind adjustment. Automatically adjusts for burr wear, can be programmed to provide different grind settings for different coffees.

Taken all together, this grinder will allow us to program a dose, by mass. That's the big thing. Add on 3 and 4 and we can make a one-touch button that grinds your coffee exactly how you need it, which would be ideal for medium-to-high volume shops using multiple brew-to-order methods. Sounds great, yeah?

Please steal this idea if you're capable of making it happen. I have a long history of thinking up stuff that I don't or can't actually make (laser bore-sights are common now, NASA and the Navy are using my 3-axis gyro system for steering in frictionless environments, and those nail-clippers that catch your nails are everywhere. That last one was going to be my fortune, I tell you.) All I ask is for one of these grinders in my shop, and maybe my name in the credits somewhere.

Seriously, though, I think we are creeping up on some big changes in the coffee scene. 3rd wave shops are experimenting with all kinds of brew-to-order, and that's great...but 3rd wave shops are still in a very, very small minority. What we need to see are some brew-to-order machines that work great with relatively low fuss, so that more coffee shops can get in on the action. The cutting-edge shops will always be great, but I'm also very interested in the rest of the coffee scene—the big & little chains, the independent shops that don't necessarily have the money, training, background, or clientèle to go to the extreme innovative edge—but they still have customers that could appreciate a truly great coffee were it prepared right. For that, I hate to say it, we're going to need really good brew-to-order machines, Clover or Trifecta or some future-machine glimpsed but dimly, that are more automated, requiring less complications from the perspective of the buyer. By eliminating waste and allowing a per-coffee price-point, without the big scary variables of training that manual brew-to-order methods require, these machines could bring quality single-origins to every corner, much as lattes et al. were brought to the masses by the 2nd wave (and its reciprocal relation with more consistent/more automated espresso machines).

And these future-machines will need a future-grinder.

Somebody get on that.