Home roasting of coffee beans is a funny business. Indeed there are many reasons why people take this task upon themselves and some of these reasons hold more weight than others. I know a lot of people state cost as a factor, that buying green beans works out cheaper than commercially roasted beans. That may be true to some extent, but there is often a lack of consideration for factors that reduce the apparent cost benefit, such as the difference in weight between green coffee and roasted coffee and the inevitable roasts that will go wrong from time to time, not to mention the cost in terms of time to do the roasting (how valuable is your time?), and the cost investment of a roaster. So while cost may not entirely hold water, it can be a useful tool to convince a loved one that the introduction of a coffee roaster to the home is a sensible decision. Another factor that is often claimed is the ability to always have the freshest possible coffee. To a certain extent this is true, but many home enthusiasts quickly get over the buckets of gassy crema produced by a one-day post roast bean, and realise that you can have coffee too-fresh. With that realisation, there certainly are options for getting adequately fresh coffee outside of home roasting. Quality and variety of bean is one I actually tend to agree with to a larger extent. At least in Ireland, variety can be somewhat lacking. Hasbean for example offer 50-odd different beans as greens, ranging from CoE winners all the way down to robusta filler. Sweet Marias in the US and kaffeespezialitaet.at in Austria also offer excellent selections of greens, and while many of these are stellar quality greens, home roasters must concede that while they can probably do these beans justice on a good day, the product will probably never be as good as a top quality commercial roaster at the top of his or her game. A reason that is rarely considered or given by someone entering home roasting, but is often given by those experienced in it is knowledge. If you devote the time and attention, you can learn so much about the coffee bean, about particular origins and varietals, how they react to levels of roast, how it influences the cup. Whether by design or not, fresh insight, and a deeper connection to the process is learned.
My home roasting journey started about 2 years ago, with a small fluid-bed air roaster called the iRoast2 (from hereon referred to as the iRoast). Probably my primary reason for jumping-in was freshness. I knew that the crema created by stale beans tended to be thin and pale, and crema being one of the measures I was trying to improve in my espresso, I took action. This isn’t intended to be a review of the iRoast, but it fell down in certain areas for me, chief of which was the taste. Everything seemed to be marked towards a bitter, burnt flavour, even if the roast was stopped at a medium point. Only being able to roast 150g of green coffee was another nuisance, especially when you consider that weight goes down to about 120g after roasting. If you dose 20g of coffee each time you make an espresso, one roast gives you about 6 cups. The noise from the iRoast was also nothing short of deafening, like 5 hairdryers on the go at once. It was just possible to hear first and second crack, but it was a strain. On the plus side I think my hearing has improved immensely with the routine of this aural workout.
Upon realising that professional roasts of much lesser beans were resulting in a more pleasing cup, the iRoast got put to the wayside, along with several kilos of green beans. For a number of months, I exclusively used commercially roasted beans, and I had some great cups, and nearly gave up on home roasting. I knew of course that there are better roasters available than the iRoast, but at the time for the price it was the right roaster for me. Had money been no object I would have bought a Hottop roaster, and while I don’t doubt the quality of craftmanship that goes into a Hottop, the high price coupled with the small (although better at 250g than the iRoast) batch size wasn’t ticking enough boxes for me.
In the summer of 2007 the US home roasting community got a new champion – the Behmor 1600, based on a Ronco rotisserie oven, extensively modified by creator Joe Behm for coffee roasting. It was to be a full 1lb drum roaster for only $299. In the interim period much has been said about this roaster, the vast majority positive, though some have questioned its ability to roast a full 1lb as well as drawing unfavourable comparisons to the programmable Hottop in terms of customising the temperature profile. On this side of the Atlantic, we could only sit back and watch in envy as no european model was available. However, in late 2007 / early 2008 it became apparent that Hasbean Coffee were in talks with Behmor to bring the roaster to market here, with the major obstacle being a long CE certification process.
About 4 months have now passed since Steve Leighton of Hasbean sent me a preproduction model of the EU Behmor for testing. In that time I have roasted more coffee than in the previous 18 months with the iRoast. I’ve roasted pre-blended espresso beans, single origins for espresso and for filter, peaberrys, pacamara, monsooned, small beans and big beans. My aim here, following this obscenely long-winded introduction is to distill my impressions of those 4 months with the Behmor.
Taking the Behmor out of the box, you can’t help but be struck by the aesthetics, and without wanting to be too unkind they are not it’s strongest feature. The Behmor looks somewhere between a microwave oven and a toaster oven, though both longer and shallower than the former. It’s also lighter than it looks. The upside of the reasonable weight should mean shipping costs remain low, but also moving it in and out of a cupboard, or off of a shelf becomes a minor issue. Having been accustomed to pictures of the US version, one obvious difference was also immediately apparent – the door handle. The US handle is more flush with the door, while there is a defined separation in the EU model. This had to be changed in order to meet CE certification. I prefer the looks of the US model, but more importantly the pronounced “jutting-out” of the handle makes it vulnerable to abuse from less than careful delivery men. All in all though the looks don’t bother me, it’s a utilitarian device, form meets functions. I wouldn’t buy a coffee roaster as an ornament for my kitchen.
The Behmor could (but shouldn’t) be operated by a child. Open the door, remove the chaff collector, remove the bean cage, open the cage, put in the beans, close the cage, insert the bean cage, insert the chaff collector, close the door, select the weight, press start. The roaster will default to profile 1 unless instructed otherwise. This profile delivers 100% power for the entire cycle, and I’ve found it to be the cycle I always try first starting with a new bean. By and large, keeping an eye (and ear) on the progress, and ending the cycle at the desired point will deliver good to great results. You could easily and happily only ever do 1lb, P1 roasts on the Behmor. The one drawback with P1, however, is that it gains an enormous amount of momentum, and you can easily jump from 1st crack straight into 2nd crack, and before you know it you’ve passed your desired roast level. P2 attempts to address this by reducing the power to stretch out the gap between first and second crack. This is achieved, but perhaps not to as great a magnitude as the profile graph might suggest. In roasts where 1st and 2nd ran into one another on P1, on P2 1st crack tended to be quiet and fleeting, with 2nd appearing about 90s to 2mins later. All of this is somewhat academic, and really is only important the first time roasting a certain weight of a certain bean on a given profile, because with accurate weighing repeated roasts hit 1st and 2nd within very close times.
Eventually I felt I need a more gentle roast profile, and moved on to try P4 and P5. These profiles gradually build the intensity. My early roasts using a full 1lb on these profiles hit first crack at 20mins in. While I was aiming for a light roast on these particular beans, this kind of length was approaching baked territory. So I simply adjusted the weight from 454g to 350g, and first crack hit about 17mins in. It’s a simple solution, though some might argue unnecessary, the 454g batch came out well. The particular bean will govern whether this is of benefit or not.
On a taste basis, I tended to prefer P1 roasts for beans intended for espresso (again – it depends on the blend profile), whether pre-blended (Hasbean CoE / Brazil Perfeito) or single origin (Brazil Fazenda Cachoeira / Cuba Turquino Lavardo etc) this profile quickly became my go-to profile for espresso. The Fazenda Cachoeira in particular highlighted this for me, on P1 it was nutty, chocolaty, lots of body – superb SO espresso, on P5 though there was a sour note, the nuttiness was diminished, it was all round a poorer cup. The Cuba Turquino Lavardo on the other hand highlighted how good the Behmor could do a roast compared to the iRoast. The Cuban actually was one of the more reliable beans for the iRoast, it coped with the abuse from the iRoast better than most. However, out of P1 on the Behmor, it quickly became clear that the product of the iRoast was only an imitation of what Turquino Lavardo could be – sweet, syrupy, deep, an immense SO shot. Strangely, not something I found with this bean on the iRoast, out of the Behmor, Turquino Lavardo needed a whole week rest to come into its own.
For french press / filter roasts, I initially had a little difficulty due to overshooting the roast level. I love lots of acidity in my filter coffee, so bringing for example Kenya Gethwumbini into 2nd crack was really muting that. This is one of the down sides of doing 1lb roasts, if you don’t hit your mark the first time, it can be a long week of drinking something you know is not as good as it could or should be. Luckily for me, the Gethwumbini brought to that level was quite nice in a 50:50 with the Brazil Fazenda Cachoeira as an espresso. Subtracting 30-90s from the total roast time of the previous roast in these cases, brought me to where I wanted to be. I also found P5 to be quite useful with these lighter roasts, there was a more pronounced gap between first and second. In the end I had no problem getting to my goal of big, juicy filter coffee bursting with acidity. I knew air roasting like on the iRoast tended to pronounce acidity to a big degree, so I was a little concerned that the Behmor (a radiant heat / drum type roaster) wouldn’t bring out this quality, but other than overshooting the desired roast, this just wasn’t a problem.
Moving on to more practical matters, and this is a big one for a lot of people – smoke. The Behmor handles smoke venting quite well, I found 1lb roasts on the Behmor to have a similar if not lesser effect on my kitchen than 150g roasts on the iRoast. For most of the roast in fact, there is no apparent smoke, only the smell of roasting coffee. Once you start heading towards 2nd crack, however, the smoke starts to appear. For me, it means still having to crack open a window, and having to close the door to the hall where the smoke alarm is. Opening the door for quicker cooling, also releases more smoke into the room than leaving it cool with the door closed. The cooling is not immediate, not like a Hottop where the beans drop out of the drum and are cooled very quickly. As you can imagine, opening the door speeds this process somewhat, but releases more chaff and smoke into the room. Once I was familiar with the roast progression of a particular bean, I was able to allow for the slower cooling and keep the door closed.
For the most part I had no problem with the bean cage, in the US a smaller grid cage is offered separately. The standard one I received allowed chaff to exit readily, and only with particularly small beans (Yirgacheffe for example) or broken beans did a small number of beans fall through the cage. When I say small, I mean less than 1%, a non-issue for me.
Monitoring roast progress on the Behmor is really about audible cues. While you can see into the drum, the colour of the light, as well as the grid of the chaff collector make it difficult to determine roast level on appearance alone. Really it’s about identifying those 1st and 2nd cracks. As there is little noise from the Behmor during roasting (I’ve had PCs that made more noise) this is easy.
The Behmor profiles can be adjusted to a certain degree, though nothing like the programmable Hottop. There is a somewhat convoluted explanation given on how to alter the percentage of the overall roast given to the different legs of the profile, but as far as I was able to ascertain, the first and middle leg durations are set based on the time when the roast is started, and any additions or subtractions of time after the roast has started are added on or taken away from the end leg. A great tool to visualise this somewhat abstract idea is BehmorThing. It allows you to make these additions and subtractions virtually, and see the effect on the profile. The use of the BehmorThing goes far beyond that as a tool for keeping track of all your roasts, using it alongside the Behmor it’s a superb (free) addition to the roaster. My only problem is that it’s Windows only, and having recently migrated to Mac, it’s one tool I sorely miss.
What is and what isn’t the Behmor? The Behmor is practical, affordable, capable of producing excellent results, and importantly capable of roasting a full 1lb of green coffee in one batch. It is a tool to learn about different coffees. The Behmor is not pretty, not fully programmable and not a tool to play at master roaster at home, if that’s what you want this is not the tool for you. It’s not without its flaws, such as the slower than ideal cooling. However, for the amount of coffee I consume on a weekly basis, the ability to roast a full 1lb, coupled with the relative affordability make it the roaster that best meets my needs. If there was some glaring flaw with the beans produced, that would override all other considerations, but there isn’t. You may not fool a seasoned cupper, but it’s definitely capable of producing roasts that do not stand out from professional roasts like a sore thumb, such as those from the iRoast. CE certification has been granted, so it won’t be long until the 220-240V version is available to buy, no final price has been set yet, the current exchange rate fluctuations, particularly the weak pound are playing a role in this, but the hope from Hasbean is that it will come in under £200. If that is achieved it will be a great price for those of us in the eurozone, equating to an outlay of less than I paid at the time for the iRoast against a strong pound, when you consider the programmable Hottop goes for £530 the message comes into sharp focus.
It has been a long time in coming to these shores, there were times in that period where I considered importing the US version and trying to arrange an adequate power supply (there’s a whole issue beyond mere voltage that becomes potentially very important with a roaster – cycles or hertz of the power supplied). I also considered going for the analog Hottop, but I knew I was still going to want to be able to roast greater volumes. There is enough tweakability in the Behmor to allow room for experimentation with particular beans, but there is also enough simplicity, to just start a roast and hit cool at the desired point. Four months in, I haven’t come close to trying out all the permutations the Behmor has to offer, I’ve had some batches that weren’t great, either the roast went too far, I overcompensated and cut it too short, or the profile didn’t suit, but at no time did I feel I couldn’t make the necessary changes for the next roast. Until the time when I feel I cannot make those changes, the Behmor is my roaster.
For a video run-through of a roast from start to finish check out the Behmor Video.
Sincere thanks to Steve Leighton from Hasbean for providing the testing unit.