Ask the Alchemist #108


How do I measure the temperature in the Behmor and match them to your temperature recommendations?

You don’t.

Caution, rambling Alchemist lecture ahead. There is no good, reliable, repeatable and consistent way to roast by any temperature measurement you can get from the Behmor. Over the years I have tried to tease out useful temperature data from the Behmor, and eventually determined that due to a host of reasons, it is not possible to get data you can actually use to roast with. Root Chocolate  ran a bunch of roasts with the Behmor 1600, took a bunch of measurements and the results look basically like what I found. Unfortunately, ‘basically’ is the key word here as they did not look exactly like mine. Nor will theirs look like yours or even their own if they change beans, weight or even ambient temperature. And that inherent variation is what makes any temperature data less than helpful.

Then what are you to do? As outlandish as it seems, what I suggest is using the Behmor as it was designed and intended. Call me crazy. The whole point of the Behmor is to give you a way to roast in a simple and repeatable manner and of course, have properly roasted beans at the end. And that is what it does. Your only need is to find the program(s) and setting(s) that work for your tastes. Notice I didn’t say ‘find the program and setting to roast properly’.

You may have noticed I don’t give many specific profile recommendations for the Behmor. The reason is they all work. All of them. I have done hundreds of roasts in the Behmor with dozens of beans and by following one set of rules, I have had 100% successful roasts. And what are those rules? It’s nothing more than I have outlined time and again.

Load: 2-2.5 lbs cocoa bean

Profile: Any

Time: 16 to maximum time

That’s it. For some reason people want more. They want me to say that they should roast 2.2 lbs of Nicaragua on P3 for 17:15 mins. Unfortunately, it doesn’t work that way because my tastes are different than yours and I don’t know what you want or if we perceive the same flavor the same way. Only you can work that out. The truth of the matter is that you can roast 2 lbs on P1 for 21 minutes (the hottest, fastest roast you can do) or 2.5 lbs on P5 for 16 minutes (the coolest, shortest roast) and both roasts (and all the ones in between) will be acceptably roasted. The key is that some may or may not be to your tastes. It’s up to you to zero in on what you like. And pretty quickly you will come to find you probably like a relatively narrow range of roast profiles, regardless of which bean you use. It’s that saving grace that will keep you from having to fine tune each and every bean.

So how do you do that? I’ll admit. I’ve be negligent here. I’m too close to it. I’ve been roasting too long. I feel into thinking that it was just an intuitive process that everyone naturally knew how to do. My apologies. Here is how to go about it without a bunch of temperature measurements and complicated plots. And it is worth noting that you can apply this iterative process to your entire chocolate making endeavor.

I want to talk a moment about what I am calling the iterative process. It’s just a fancy way to say that you do something (roast, winnow, refine, add cocoa butter, add sugar, etc) a certain way, evaluate it, and then change ONE item and do it again, noting the difference. You do multiple iterations. It’s by this process that you can learn rather quickly how a given change affects (or doesn’t affect) your overall product and which direction it affects it (do you like it more less or is it just different).

It’s said that it takes 10,000 hours of doing something before you become a true master at it. We aren’t talking about mastery here, but the concept holds. You can’t make the perfect chocolate the first time out. It takes time to learn what you like, what you don’t like and how different parts of the process affect the outcome. But there is something implied in those 10,000 hours that isn’t said out right (I guess that is why it’s only implied). You can’t just do something for 10,000 hours without trying to get better and expect to get better. You have to actively try. You have to pay attention. You have to do it methodically. Doing one iteration after another with intent in mind helps build that mastery. If you spend 10,000 hours throwing darts at a dart board without trying to hit the center you won’t get any better at hitting the center. You have to modify what you are doing with each try (iteration) if you want to have a hope of improving. That’s what the iterative process is all about. And one final point about this. It is critically important to not change too many (the best is 1) things at once. When I was in the lab I watched many very intelligent people fail to understand a system or fix a problem because they changed lots of things and hoped for the best. Every so often it works, you make things better, but you don’t learn anything. And without learning you are no better off than you were before. Those that don’t learn from history are doomed to repeat it an all that. So that is my goal here. Not to teach you how to roast, but to teach you how to teach yourself how to roast (or do anything else) based on a solid process that works.

So let’s do it.

Roast 2 lbs of cocoa beans 18 minutes on P2. It’s that simple. I recommend the profile, weight and time for a specific reason. They give you a direction to take depending on what you think of the chocolate. More beans, hotter or colder profile and shorter or longer profiles.

A quick side note. I make all my test chocolates at 70% nibs, 5% cocoa butter and 25% sugar. Vary it to your taste but having one recipe you use for learning is very helpful.

First off, did you like it and do you think it could be better? Let’s ignore that you got lucky and this is the best chocolate you have ever had and don’t want to change anything. It’s probably not going to happen and isn’t helpful. So what didn’t you like? Let’s go through some of the most common things you might taste and the direction you could try.

Is it green and grassy? It’s possibly under roasted. You have two options.

Increase the time or go to P1.

If the chocolate flavor is there, I would just increase the time by 2 minutes. If it has not really formed, I would increase the heat by going to P1 or reducing the load (not in this case but in further iterations since you started this iteration at the lowest weight recommended).

Is it to a bit acrid? It’s possibly over roasted. It could also be that it was roasted too strongly.

If there is a good chocolate flavor but it has that burnt edge to it, you probably need a more gentle profile. Just go in order. P3, P4 or P5. If there isn’t much chocolate flavor, you may have just roasted too long. Stay on P2 and roast 2 minutes shorter.

Is it too fruity? Increase the roast time a couple minutes or go to a hotter profile.

Does it just not match my tasting notes? Throw a dart or flip a coin to pick what to do next. If in doubt, just change profiles and keep weight and time the same.

I clearly can’t give you every possible taste combination. But hopefully this gets you going. Look for patterns. Start to understand what happens during a roast. Look for some of these items.

Volatile acids are initially driven off. These are not fruit notes. These are like vinegar and other sharp to the nose aromas. Next chocolate flavor develops. Also fruit flavors start to form. At some point fruit flavors start to change and reduce. Nutty and savory flavors can become more noticeable at this point if not before. Finally roast flavors start to dominate. And if you continue, chocolate, fruit and nut flavor is burned away and you are left with a powerful acridness.

The REALLY great thing about the Behmor is that it is nearly impossible to under roast IF you roast not less than 15 minutes. Likewise, as long as you have 2 lbs in there, there is basically no way to burn your beans. You can get to the level of roast flavor becoming noticeable, but I have made perfectly good chocolate on P1 with 2 lbs of beans roasting for over 20 minutes.

Finally, I want to touch on bean type and paint with a very broad brush of generalization. Keeping firmly in mind that what I have said above holds for ALL beans. This next part is really about fine tuning and choosing whether to go longer vs hotter, or shorter vs cooler.

Criollo. Longer is better than hotter, shorter is better than cooler

Trinatario I tend to like longer over hotter and cooler over shorter.

Forastero. Hotter over longer and cooler over shorter.

Notice the pattern?

One other thing to remember. Your bean mass can be used to affect the profile. If you are roasting on P1 and you want it hotter, then reduce the weight of your beans. And you can do it just a couple ounces at a time. But don’t drop below 2 lbs generally speaking. If you are at 2 lbs, then increase the time as it’s your only good option. In theory you could drop the weight even more but you start to defeat the purpose of the very nice profiles if you do. Trust me and all the testing I’ve done and keep to those initial ranges I gave. 2-2.5 lbs, any profile, 16 minutes and up.

One last thing about iterations. If you make a decision (I over roasted, I have to roast less) and take it all the way in one direction (you decide cooler vs shorter) but you have made it all the way to P5 (the coolest profile) and you can’t go any shorter because you are already at 15 minutes with 2.5 lbs of beans, then you need to re-evaluate your initial decision. Maybe you didn’t under roast or maybe you are mistaking the given flavor profile of a bean for under roasted. Even though it doesn’t feel right, try roasting longer or hotter and see what you get. Learn. Oh, and take notes! Don’t trust your memory. After a few roasts you will forget what you have done and that is as good as throwing darts randomly.

Finally, please keep in mind these are guidelines and not rules. If you find that taking Criollo hotter instead of longer give you the taste you want, then that is fine. Do it! These are just what I have learned that work for me over some 2-3000 hours of roasting. Your mileage may and probably will vary. And if you get totally stumped, my door is always open.

1 Comment