The next level of nerdiness: Particle size variance in grinders

Note: I've made some updates to this article based on feedback. This is my journey, and I've changed the 'you' to 'I' accordingly. Hope it's useful. Thanks for reading!

"My coffee sucks." I thought. OK, not really, it was pretty awesome. But still, I have to admit, my coffee was terrible next to the coffee made with the same bean just one hour earlier at a high end cafe. It was a sad realisation that it did, and always will, until I invest in the heavy-duty grinders they invested in, and really dial them in. Why? The main reason: my grinder sucks. It sucks because it can't produce particles as evenly as the $1,000+ grinder (sometimes $3,000+) that they're using. (The second reason is I never go through enough coffee to dial the grinder in as scientifically as they are, and because I don't have a refractometer. But that's a topic for another day.)

Look at the picture below:

Coffee grinds prepared fine (left) and coarse (right)

What's the difference? Two main things. Firstly, the coffee on the left is finer than the coffee on the right. But there's another thing that's different: there are many randomly different sized chunks in the one on the right. Why is this the case? And more importantly, why does it even matter?

Why does particle size distribution matter?

Let's say you're making stir fry. You chop up the meat and vegetables and cook them, right? Right. But there are three things you can do that are wrong.

  1. The chunks of meat and vegetables are too small: You lose. Overcooked. The vegetables become too soft, and the meat is cooked through and too tough.
  2. The chunks are too big: You lose again. Nothing is cooked through, and the result is inedible.
  3. Randomly different sized chunks. Why would you do this?? This is the worst of both worlds. Now some is undercooked, some is overcooked and some is just right, but it's lost in the melee.

Now, think about that in relation to coffee. Instead of chunks of food, think about chunks of coffee. If you have chunks that are too small, too big, or randomly different sizes you're going to end up with coffee that's overdone, underdone or the worst of both worlds.

Coffee (the drink) is coffee that has been partially dissolved in water. You make it by exposing the coffee to water for varying lengths of time (and pressure and temperature). You only extract as much coffee as the water can touch and dissolve (so you can't make it with whole beans, because the bits in the middle of the bean would never get out to dissolve in the water). This is why we grind coffee: to expose it all to the water.

When we make coffee, we extract the flavours in a certain natural order: First, acidic, sour flavours, followed by sweetness and finally by bitter compounds. When you do it right, you get a pleasing balance/minimalisation of the flavours of bitterness and sourness that lets the characteristic of the coffee shine through.

But think about our stir-fry situation above. Three main things can go wrong when making coffee as with a stirfry:

  1. Grinds are too fine: You dissolve too much coffee, over-accentuating the bitter flavours. This is over-extraction.
  2. Grinds are too coarse: You don't dissolve enough, leading to a lack of (necessary) bitterness, and light/sour taste lacking natural sweetness. This is under-extraction.
  3. A combination of both: Your grinds are on average the right size, but because there are lots of small and big grinds, you get lots of bitterness and sourness mixed in and it drowns out the natural sweetness of the coffee.

Let's get mathematical (feel free to skip ahead if you don't want to).

Look at the following image of a hypothetical bean that's oblong shaped (to make calculations easier):

Theoretical oblong coffee bean ground into four even sized pieces

The bean originally has a volume of 8 units and a surface area of 28. The grinder grinds it into four parts. The four parts, when combined, have the same total volume as one bean obviously (because this grinder is perfect and has no waste). But now the surface area has increased: the surface area of each ground is 10, so total surface area is 40. Thus, grinding one bean into four (in this scenario) increases surface area by over 40% and thus increases the extraction rate by the same amount. The extraction time can be tuned to this extraction rate.

What if the chunks aren't even?

In the scenario above, all four particles have the same volume. But what happens if the grind size is less perfectly regular than the above example?

Theoretical oblong coffee bean ground into irregular sized pieces

In the image above, the volume is the same but now the surface area of the grounds is 38. So you might think 'great! Knowing that I can simply adjust extraction time slightly'. Unfortunately it's not that easy. Three things can happen, all bad:

  1. You leave extraction time the same as with the even chunks in the first example (all 1x1x2). In this scenario, you will correctly extract the medium sized chunk (the same size), but over-extract the smaller chunks (leading to bitterness) and under-extract the largest chunks (leading to sourness), thus overall delivering a suboptimal cup.
  2. You reduce extraction time to tune it for the small grind size: You eliminate the bitterness of over-extraction of any component, but under-develop the flavours overall and have too sour a cup.
  3. You increase extraction time to tune it for the largest grind size: You fully extract the large chunks but over-extract the smallest chunks, leading to bitterness.

So what are we looking for? Even grind size. An even distribution of particle sizes with minimal chunks and minimal fines.

What does an even grind distribution look like?

A perfect distribution would be one where every chunk is the same size. This is not possible of course, since coffee beans are not even sized geometric shapes! Thus, an ideal distribution of coffee particle sizes has a high peak around the target mean grind size and narrow sides (a small standard deviation). 

Different grinders are better or worse at producing a good distribution. The best are high quality flat burr grinders with the burrs mounted vertically. Next best are horizontally-mounted flat burr grinders, then conical burr grinders and finally blade grinders or cheap hand grinders.

See the illustrative chart below. This is a chart made up from looking at many other charts - it's not authoritative in itself, but a visual representation of other analyses done. 

Note: while charts (prepared by others) do show that basic hand grinders produce uneven grind distributions just like blade grinders, the hand grinders remain a far superior option for the much lower amount of heat they transmit to the beans. 

OK so what do I do now? What grinder do I get?!

It's easy to be misled about how 'good' your grinder option will be. Below is a general guide to the categories of grinder out there and how good they may be. Here's the harsh truth: Most of us can't afford to spend as much on the ultimate grinder as a cafe can. We can buy all the other gear cheaply and produce good coffee for under $100, but not the grinder. We can get close-ish though without spending much more. Don't think you need to spend $3000 or even $300 on a grinder in some vain quest, unless you really want to, as there's much more to coffee than just good equipment.

Blade grinder. Don't do this.

Not good: Blade grinders

  • Blade grinders (a.k.a. spice grinders): These are usually what most people think of as coffee grinders. They over-heat beans by processing them for too long, and in the end deliver inconsistent grind distributions similar to cheap hand grinders (although admittedly without the elbow grease). 

Baratza Encore

OK: Hand grinders, consumer electric burr grinders

  • Cheap hand grinders: E.g. those by Hario or Porlex. These are good for travel (I use one) and are far better than blade grinders or pre-ground coffee (or mashing your way through whole beans with your teeth out of desperation), but produce a huge amount of fines and coarse chunks that will not faithfully represent a coffee. One note - you can modify them to stabilise them, and they will perform a lot better if you do.
  • Conical burr grinders including anything by Baratza, Sunbeam, Bodum and many other brands. Many of these grinders retail for $100-$250 and differ in build quality and durability, adjustability (some are easy, some are hard; some have infinite variations and some have fixed clicks). Most of these will be perfectly adequate for many coffee afficionados, but unsuitable for espresso drinks. If you don't do espresso, these will likely do you fine (until your taste outgrows them). The best thing on a budget is to get one of these, then get it tuned/adjusted so the burr is perfectly centred.
  • Low-end flat burr grinders: Many companies like Baratza, Rancilio, Gaggia and others make low end (sub $500) flat burr grinders that work well with both filter drinks and espresso drinks. These are solid grinders and will last a while (particularly notoriously well-built models like the Rancilio Rocky). In general due to better build quality, this class of grinder will last longer and will perform well for espresso drinks. However for non-espresso drinks they won't offer much of an advantage over cheaper conical burr grinders (apart from durability).

Mazzer Mini. A workhorse that will satisfy almost everybody.

Good: Prosumer high-end hand grinders, medium-range electric burr grinders

  • Hausgrind or OE Lido 2 hand grinders. These new kids on the block typically sell for under $200 and can grind for espresso and make quick work of simpler grinds for immersion or pour-over drinks. They feature stabilised shafts and robust designs that mean they will last a life time, and can grind at a level of grinders that cost many multiples their price.
  • Medium range flat burr grinders: Most grinders by Baratza, Rancilio, Mazzer, Compak and many other brands that feature flat burrs. As with burr grinders, they vary in build quality and usability, and can have features including dosing aids, large hoppers, portafilter holders and others. These are already approaching the 'excellent' level and will likely satisfy all but the most discerning of connoisseurs, those who are trying replicate what they get in high end specialty coffee cafes.

The Mahlkönig EK43. A beast. Seen in specialty cafes everywhere these days.

Great: High end electric burr grinders

  • Mahlkönig, especially the EK43 (the darling of the specialty cofee industry). This bad boy sells for around $2,500 and is huge. It rose to prominence in the last few years as a specialty coffee grinder that is distinguished by the incredible consistency of its grinds and the efficiency - 20g in equals 20g out, with nothing left behind. It deserves a post of its own.
  • Mazzer Robur or Super Jolly. These are high end cafe-quality grinders were the standards for years, and were the yardstick against which every other grinder was measured. They still are superb performers that are hard to beat. They have proven that they can stand the test of time, unlike the Mahlkönig grinders which were only recently put into espresso service (and which for decades were spice grinders).
  • Bunn G series or Ditting grinders. Industrial strength batch grinders, suitable for most kinds of coffee but not used for espresso. The Bunn G3, a towering monster that is often retrofitted with Ditting blades, is a darling of many home coffee enthusiasts (who have a lot of room in their houses).