Level: Apprentice
Read Time: 12 minutes
Thanks for all the great posts on roasting, and specifically the Behmor. Thanks to Uncle Sam and tax returns, it's time to upgrade from oven roasting to the Behmor.
I feel like I have a good grasp or how to adjust the recipes if I was using a non-black box Drum Roaster, but given the Behmor is a black-box, I was hoping for some tips on how to control the Behmor to take everything you've taught and be able to self-adjust recipes as desired. I'm trying to get to the point where I can make a test batch and say "ok, I want to change XXX, I know how to do that", especially since I don't want to have to rely on you for a recommended starting place on Behmor settings for a given bean.
1) In the Behmor, what kind of thoughts go into deciding which programmed mode to use? You mention P2 is your favorite but use P1 for all the roasting samples. As I'm trying to make my own decisions for how to roast and how I want to change the flavor of my chocolate, I'd love any tips you have for how to decide which one to use. Do you use the scientific method and just make a lot of chocolate with the various modes, or do you know something about each mode that helps you hone in on one?
2) I love the idea of making small batches and iterating to get to a final preferred recipe; can I use, say 1lb of beans in the Behmor and if so, does it simply scale for doing the larger batches?
I ask because I know you specifically called out using 2 lb of beans with the Behmor. My other half prefers we eat all the bars before trying another batch and obviously 1lb of beans would mean less bars and a faster chance for me to iterate. I know you also mention in ATA 251 that using >=2lbs of beans makes things more reproducible.
3) Throughout the ATA's you mention how you can alter a roasting profile to bring out different flavors but since the Behmor is a black box, do you have any tips on what to change in the Behmor to bring out different flavors? You mention roasting longer can lead to a heavier roast, and in ATA 244 you mentioned manually dropping power could avoid making the Nut component go bitter by using manual mode to drop from P5/100% to P4/80%. Any other tips?
4) In ATA 276 you touch on cooling and mention the Behmor. It didn't occur to me to think about how to cool the beans until that post. For the Behmor, do you let it run its cooling cycle and then take the beans out? As soon as the roasting is done do you remove the beans? (Yes. I'm being the person that is worrying about all the details. Yes, I know I'm not going to ruin my chocolate either way).
After last time’s rant, this is a wonderful and welcomed set of questions.
With all the changes to the Behmor roaster, it is about time I give some updated notes. First off, you’ll note I called it the Behmor roaster because I’m hearing many people asking about the ‘new’ Behmor 2000. For clarity sake, the Behmor 2000 is the Behmor AB (stands for All Black) with some updated options. The Behmor AB is the updated Behmor 1600. At this writing the Behmor 2000 is the only version in production and the one I’m going to discuss here.
So when the 1600 first came out it had 5 automated pseudo profiles. I’m making the distinction because they are not quite what I mean by a roast profile. For me a profile MUST contain the bean mass temperature. Anything less than that is not a profile but more a power setting. If you must, I could be convinced to call it a power profile, hence ‘pseudo-profile’ but it is no more a profile than the medium setting on your stove top. For that matter, it is not a lot different from trying to use profiles in an oven. They really just don’t work in that they are not repeatable and translatable. I’ve given my thoughts on that in the past so I’m not going to hash it out again and instead hop into directly answering the questions.
I used to advocate P2 as the application of heat was (and is) similar to the general way I roast manually. That is to say the initial application of heat is high then transitions to less heat and finally bumps back up slightly to keep from stalling the roast. And you can STILL use P2 if you like. It works just fine and as it always has. But the biggest change in functionality between he 1600 and the AB and 2000 is the inclusion of some semblance manual control. In short, once a roast is started, the Profile buttons (p1-p5) change purpose and are then Power buttons, each one being 25%, so 0, 25, 50, 75 and 100% respectively.
What that means is that I no longer use P2. I start EVERY roast on P1 and then manually control it based on experience, smell, and in MY case, the temperature of the beans. For that last part, I can not stress enough that I mean that literally. I have modified my Behmor 2000 to accept a rigid thermocouple. You can see how I do that in THIS VIDEO. That brings up another feature the 2000 has, namely the ability to see a couple chamber temperatures. I’m not even going to bother to tell you how to access them because they (I ask your forgiveness for yelling) HAVE NO CORRELATION TO BEAN MASS TEMPERATURE. There are thermodynamic LAWS that show they cannot have a usable correlation. You can read all about that HERE. What this means is that unless you are willing (and able) to modify your Behmor, your ability to really control the roast for maximum effect is going to be severely limited and teaching yourself is going to be much more challenging……but not impossible.
With all that, here is how I roast and how you can try to follow suit.
First off, I roast 1 kg. Yes, I know I sell in 2 lb increments, not 1 kg, but I have the luxury of being able to load 1 kg every time. 2 lb and 2.5 lb can and will work. I just like the performance at 1 kg. Next, I pre-heat the roaster for 1 minute AND I verify my beans are between 65-75 F. There is nothing magical or special about this, I’m just working to keep my variables to an absolute minimum so that my roasts are more consistent and predictable. The more you can control your variables, the more success you will have. I mean, what would be BEST is that you bring your beans to a constant temperature (say 75 F) and pre-heat the Behmor to 100 F. But I can’t see many people doing that…but it would make things way more consistent, reproduceable and predictable. I fear if I said that though people would assume it was too fussy.
With that in place, I load the beans, insert my thermometer and start the roast. Another perk of the 2000 is that you have the option of two drum speeds. Personally I see zero reason for the lower default speed but Behmor likes to give users options, so there it is. From a consistency standpoint I 100% suggest the higher speed. Once you have started roasting, just hit P5 (this puts it in Manual control mode), D (think drum) and the 1 lb (this takes it back out of the Manual mode). That increases the drum speed and will give you more convection which leads to a more reproducibility. There is a misconception that slowing the drum down will slow the roast down…..and from a pedantic stand point it is correct BUT is not predictable and really is just introducing a variable that you have no need to introduce. It is slowing down convection and that can lead to a less even roast. After that I roast until the bean mass is around 200 -205 F and then I turn down the power to 75% by pushing P4. Now, if you have not modified your Behmor, what are you to do? That is difficult. The answer is that you need to learn what a roast smells like when it is around that temperature….but without knowing the temperature, being 100% sure is harder. What I can say is that during the Drying phase (ambient to 212 F) there is very little roasting aroma. All that is there is a dry, hot smell. Once you get to the end of the Drying phase other aromas start to come out. That will be somewhere around 14-16 minutes in. If you are way before that and smell something more that, then that is not the droid you are looking for….er, that is not the aroma I’m talking about. How can you be surer? You really can’t which is why I like a thermocouple in there. All you can really do is play the long game and use your best judgement, make the chocolate and see if worked. If it did, don’t fix what is not broken. If it seems not quite right, do everything again, dropping power a little later (one minute is a good place to start) and seeing how that is. If that tastes better, then you are dialing in what the aroma should be.
Moving forward, I then let the roast proceed 2 more minutes and then turn it down to P3 or 50% power. This should nicely get you through the Development phase and assuming you were at 200-205 F or so two minutes before, then you should now be closing in on 232. At this point I push the C button. This will reset the timer to 3:10 which is the minimum time you want to in the Finishing phase. That will assure you have full heat penetration and the beans are not under roasted.
After that we are at a judgement call again. Depending upon the bean, you can stop the roast there or go up to another 2-3 minutes. Your benchmark as always is waiting for the aroma to change to slightly acrid. Some people note a bite at the back of their throat. This isn’t a sign of the beans being over roasted but much more (another minute) would be too much. At this point I hit Cool, open the door and allow the roaster to cool the beans. You really don’t need to pull them out for extra fast cooling.
That is how I roast. I know you asked how I decide how to treat each bean and since we are talking about beans I have already vetted, it is way more simple than many people want to believe. What I’ve told you above is it. Hit the 2-2.5 minute development phase, and 3-6 minutes finishing phase, pulling the roast when it gets a little sharp and you are done. After that it is just fine tuning as I’ve laid out.
Can you do 1 lb? Based on everything I gave you the answer is yes and no. Pedantically the answer is yes. Sure you can….but it will be even MORE of a black box because you now don’t have a sufficient thermal mass to help regulate the ramp and the two turn down setting are going to give you too much power. I’ve tried roasting on Manual at P4 (75%) and P3 (50%) power but the time at which you hit 200-205 was a mile wide and you are fully dependent on your sense of smell, and then the ramp rates are also all over the board and your options for turning the power down are reduced. When you were roasting 1 kg at 100% you have 4 other power settings to pick from, but if you are starting at 50% then you are down to half those choices and 25% might be too much still but 0% (p1) will clearly stall the roast. So basically, no, you can’t roast 1 lb if your goal is consistency and dialing in a profile methodically. I can’t even do it with my thermocouple in place because it doesn’t touch the beans.
And that is it. I covered the other parts of this question above all at once.
I’m going to end on this with a few caveats.
Please try really hard not to ask me about other thermocouple placements because you heard about it on the internet. This includes the temperatures that are available to you on the roaster that I’m still not going to tell you about because they are of no use. They MIGHT work in slightly isolated cases, where you are roasting one bean, over and over BUT you are 100% on your own and since I don’t own your roaster in your conditions I literally won’t be able to assist you.
Next, no, you can’t use a flexible thermocouple. The danger of having it get pulled into the drum is way too high.
And no, you can’t use and IR thermometer. The drum will 100% mess up the readings.
I think that is about it. Happy roasting and chocolate making.