A grinder or a set of scales?

In recent weeks there has been a degree of buzz in the coffee world about some research carried out by Ben Kaminsky, a grinder and the concept of ‘coffee shots’.  This all followed a lecture by Ben in August, which we were unable to attend.  However, we’ve kept a close eye on the resulting online debate which developed for a short period of time (links to some of these posts are included at the end).  This post is a reflection on these new developments from the point of view of our business and starts with a set of scales . . .

We have a philosophy of continuous development and learning – in fact it’s in our blood!  We’re always looking for ways to improve what we do, and this is especially true when it comes to roasting.  We’ve spent a lot of time setting up monitoring systems on our roaster, which has given us a wealth of data; we’ve added dials to the gas flow to improve the accuracy of this data.  We’re now at the stage where we’re setting up better systems to analyse the data.  The systems are quite simple and are based around the question, ‘What will happen if?’.  We’ll look at our roast profile and the data we have, and then we’ll create a little hypothesis, a small tweak to the roasting which we hope will result in improved flavour, or perhaps the highlighting of a particular aspect.  We’ll then apply the changes and then cup (or taste) the resulting coffee to try to determine the impact of the tweak.

It is through this approach that we’re developing a better understanding of the roasting process, and particularly how our roaster functions.  More importantly we’re cementing a philosophy of continuous learning.  In recent weeks we’ve been looking closely at the weight loss of coffee during the roasting process – we’re interested in using this as one method for determining roast consistency; we also have a hypothesis or two about the links between weight loss and espresso extraction.

There seem to be various ways of determining roast consistency – the look of the beans is important, the degree to which it followed a previous roast profile, and then there’s weight loss.  Beyond this we could also look at using colour cards and colour meters, which are more accurate but also more costly!  At the very least it is the coming together of different pieces information which provides a rounded picture, rather than relying on one particular aspect.

At the start of every roasting day we have the small task of weighing out the batches of beans which we’ll be roasting that session.  We currently work on 10kg roasts and therefore at the start of the day we’ll line up all the roasts and then schedule them onto our roasting software.  After each roast has been completed we’ll weigh the beans again to determine the weight loss – we record this as a percentage.  For example a 10kg start weight, which results in an 8.6kg end weight, would achieve a 14% weight loss.

This all sounds simple enough until we factor in our scales.  We purchased our scales when the business started, and when we were very conscious of managing costs.  The scales have an accuracy of 50g, which on the face of things seems pretty good.  However, this means a reading of 10kg could be anything from 9.950kg up to 10.049kg, and an 8.6kg end weigh anything from 8.550kg up to 8.649kg.  If we play with these extremes a little we find:

Start weight

End weight

Weight loss

9.950

8.550

14.1%

9.950

8.649

13.1%

10.049

8.550

14.9%

10.049

8.649

13.9%

These figures clearly show that the margin for error on our weight loss is around 2%.  This doesn’t seem huge but we’re only roasting coffee in the 13-15% range.  A 15% weight loss, with a 2% margin for error is not accurate enough for where we want to take our roasting.  And so we’re now on the look our for a better set of scales.

A year ago this accuracy didn’t matter; we were considering different aspects to our roasting and this data wouldn’t have the same impact it does now.  I do remember a conversation with a fellow roaster who commented on the lack of accuracy to the scales – the comment was lodged in my memory banks but at the time it ceased to have meaning.  I am sure there are roasters out there who’ve been through exactly the same process; I’m sure there are others out there who will smirk to themselves, wondering why we didn’t know this already; there may be one or two who haven’t got this far yet.  Our roasting has come a long way in the last 18 months and our commitment to learning and education will help us develop this even further.  In education circles this is referred to as ‘process learning’ where the outcome is seen to have equal or even less value than the path taken to get there.  There are some who might remember exam papers stating ‘Show your workings’, which allowed you to receive marks even if your answers were wrong.  It is a powerful approach to learning and one which I believe is undervalued in both society and in our education system (I am a trained teacher by the way ans still spend part of my week in school).

And this then brings us back to the recent debate in the world of coffee.  I’m not going to comment on the actual work undertaken as I haven’t yet had the privilege of attending one of the lectures.  Instead this is an observation on the resulting discussions.  The one thing that struck me most was how people seemed to latch onto what I would call the ‘outcomes’, primarily the Mahlkonig EK43 grinder and the concept of a ‘coffee shot’.  Mahlkonig as a company must be thrilled with the number of sales, and particularly the great product reviews!  What seems to be discussed less, or enthused about less, is the ‘process’.  This, to me, is the exciting bit.  Ben Kaminsky is at the stage of his development which makes his work ground breaking in the world of coffee; it makes our research into weighing scales seem almost insignificant.  And yet at this point in our development, it is our work on weighing scales which will have the most impact on our business.  If everyone chased the ground breaking stuff, we would never refine and improve the day to day stuff.  I only hope I get the chance to listen to Ben talk first hand, to hear how he went about his work, to understand the challenges he faced and also to get a better understanding of his findings.

I can imagine in a year or two people discussing the ‘Mahlkonig EK43’ and ‘Coffee Shots’, trying their best to remember the name of the person who first brought them to their attention.  I’d rather a scenario where people discuss the work of Ben Kaminsky and then ask, ‘Which grinder did he use?’

At a personal level I am intrigued with these developments and a part of me (the ‘boys with their toys’ part) wants to get an EK43 and pull some ‘coffee shots’.  The other part of me realises that our business will benefit more if we invest in a better set of scales.

Other Posts

Other posts on these developments in coffee include:

Jeremy Chandler: http://www.prufrockcoffee.com/2013/08/the-kaminsky-report/

Colin Harmon: http://colinharmon.wordpress.com/2013/08/13/ek43-tales-from-the-bar/

Dale Harris: http://bar1sta.com/?p=164

James Hoffmann (in response to Dale Harris): http://www.jimseven.com/2013/08/21/alright-lets-talk-about-ek43s-then/

http://sprudge.com/ek43-pin-up-girl-photospread.html