Roasting Woes

Spent many years roasting cacao bean on pan. Years without success on uniformity. I can safely say every roast was unique without any chance of repeat. A pan over lpg flame and constantly stirred by manually with large spoon. At least I can make the flame strength constant. I also experience the woes of wood fire stove. The flame was simply too unpredictable. Constant stirring? No chance. I can make a rhythm for about a minute to five. That’s all. Uncomfortable heat, ringing phone, person questioning and boredom always get in the way. Mechanical stirring is being adapted by few but implementing it never once crossed my mind.

I mustered enough guts and shifted to oven roasting. Bought myself a low cost locally fabricated version. Switched to fermented beans. Then used about one full jute sack, 63 kg estimated, before getting the roast that suited my senses. That’s a lot, I guessed. But nothing compared to amount I used with pan roasting. Three years flew like a breeze and never had any failed roast. Now the question is. Am I an expert cocoa bean roaster?

The answer is yes and no. I am an expert when if comes to the very specific equipment at my possession. I need time and test batches to adjust for every equipment I handle. I did a demo in Marinduque once. The first result was a total failure because of a defective thermostat. It was burnt. Second and consecutive trials were okay.

I did a same demo in a different place. Again, using a different roasting equipment. First trial was success because I monitored it closely. Consecutive trials were successful with minimal supervision.

Then. We fabricated our very own drum roaster. It was a huge gamble because it was our first and I had no experience on drum roasting before. We built it based on theory and observations gathered from other working roasters. It is two-in-one. Adjusting my skill to use the machine and making modifications on roaster to make it more efficient. First run was successful. Second was a bit of failure. Third and consecutive were okay.

Maybe the next step should be modifications.

Marvin is the lead chocolate maker of Ben and Lyn Chocolate Inc. Has strong background in food research and development. Occasionally conducts training and lectures. Lecturer of Cocoa Foundation of the Philippines. Do coaching and consultancy services on his free time.

A Batch of Beans In Question…

I have been tinkering with this batch of beans for quite a while now. It is by no doubt fermented but has no acidic smell. I hate to describe it but it is slightly similar to dried cow dung.

I never like its unroasted flavor either. It taste like chocolate but the fruitiness is missing. The acidity, nuttiness, almond, cashew nut and barbecue that I’ve tasted from the previous deliveries. I love discovering new flavors. This one disappointed me. It is simply flat with disagreeable odor.

FYI, good beans, like I described, taste good already in its unroasted state. I assumed that was true to all until I got a not so good batch. The taste improved after roasting and even got better after refining. Chocolates made out of it got sold rapidly to the point of me wanting more of the specific beans.

First test processing.

Roasting. The typical brownie smell during roast is not evident. I could hardly smell it. If my only roasting indicator is odor, then I am sure to fail. No significant improvement in bean taste either. My last option is the previous time and temperature records. I applied what is common to all and just accepted the result.

Winnowing. I noticed, more shells were sticking to nibs. More hand picking is needed. Longer time turning the nibs again and again and more eye pain. Pan roasted beans usually has this characteristics due to uneven heat distribution.

Finished product. My partner liked it, but I don’t. The decision is continue with the rest of the batch but never buy similar beans again. Some of our patrons might have taste buds similar to me. That sure is loose on our side. This is such a risky decision.

I never do bean cut test, the appropriate bean sampling method. All I do is get some beans, peel and chew it. Then visually inspect the rest. If the noticeable defects are negligible then I go for it.

Maybe now is the time to implement proper sampling.

There are still few sacks left. I am trying to figure out what happened to this beans so I did a cut test and compare it to good batches. To my surprise, the appearance is far superior than good batches.

bean cut test bad good

Now we suspect the bean is washed. I mean washed after fermentation in attempt to remove to much acidity. We heard some buyers don’t want acidic beans. It is done by shortening fermentation period or washing after. Maybe what we got is not a bad batch. It could be our standard is set to specific.

Marvin is the lead chocolate maker of Ben and Lyn Chocolate Inc. Has strong background in food research and development. Occasionally conducts training and lectures. Lecturer of Cocoa Foundation of the Philippines. Do coaching and consultancy services on his free time.

Try: Skip Bean Roasting

For the reason so many beans are wasted. Many of them actually have good nibs (broken, halves, shells off). However, they are likely to get burnt when roasted together with selected batches. I am throwing them all away with little hesitation.

cocoa-bean-rejects

For the reason, there are small chocolate makers who are successful in winnowing raw fermented beans. By the book, the practice winnowing-before-roasting requires pre-heat treatment to ease shell separation off nibs. These chocolatier I am speaking of winnow raw beans directly.

For the reason some bean to bar makers are omitting one essential step, the roasting process. I want to try it too, but never wanted to use the good selected beans. I will be using the rejected bean with good nibs for this experiment.

For the reason the idea is new and different. It may attract new adventurous buyers.

For the reason of less heat application. There are roaming claims, skipping roasting preserves more nutrients and antioxidants. The resulting product would be attractive to health conscious customers.

For the purpose of using less resources. Roasting requires several things. The roasting machine itself, electricity for for turning the roasting drum, electricity again for electric fueled type, liquefied petroleum gas, and labor. Without roasting, these resources will be lessened.

Marvin is the lead chocolate maker of Ben and Lyn Chocolate Inc. Has strong background in food research and development. Occasionally conducts training and lectures. Lecturer of Cocoa Foundation of the Philippines. Do coaching and consultancy services on his free time.

Making New Cacao Bean Cracker Prototype

I have been noticing a serious problem with my latest cocoa bean cracker prototype. I feel it is getting slower and slower. I confused if it is slow or I just want it to be faster.

Here is the situation. Most of the time the roller runs contininously but there are no cracked beans falling. Falling down and stop for 10 to 30 seconds and start falling again. If it comes to worst, none will fall until I gently pound the hopper with spoon. The ten second stall are big deal when accumulated. What more for a complete stope.

Thing gets fast if I baby sit the motorized bean cracker, otherwise slow. I feel frustrated cause I can’t do other things while winnowing. If I left it unattended, it will take considerable time to finish the batch. I could have use the wasted time to grind it.

Once, I watched the the cracking motion over the hopper. I saw roasted beans tumbling between roller and grinding surface for about five seconds before breaking. This needs a serious improvement that means a complete machine remake. I can meddle with the existing but it may halt production if something goes to worse.

So I built a new one and this timeĀ  I made it more like the cracker from thechocolatelife.com post. The crude DIY Bean cracker made of HDPE board, screws, wooden roller wrapped with knurled black rubber and a wooden handle.

My previous kinda follow its princible but the built style was very different. It is hard to revise and take apart when I need to.

For ease of construction and assembly, I chose ten mm plywood as main frame and hopper. I leave it as is for the test run but will be coated with polyurathane before doing a bigger test batches.

new diy bean cracker

The roller and grinding surface are steel. Choosing wood for this is a no for me. Cacao is abrasive and will eat away wood in no time. Ground wood will go straight to chocolate liquor if missed by vacuum power.

Roller axle is 3D printed PLA. Again, this is for testing only. I assume it is not going to last larger batches. I am setting a trip to machine shop once I am satistfied.

new diy bean cracker with hopper

Parts are made for quick assembly and dismantling. Parts could be replaced with stainless steel one at a time.

————–

Initial testing phase showed very promising result.

Marvin is the lead chocolate maker of Ben and Lyn Chocolate Inc. Has strong background in food research and development. Occasionally conducts training and lectures. Lecturer of Cocoa Foundation of the Philippines. Do coaching and consultancy services on his free time.

A Very Slow Bean Sorting Process

While I am struggling to speed up most of the bean to bar chocolate process, this very first step is a long shot. The bean selection.

I thought a bean quick scan was enough. I was gravely mistaken. A noticeable effect was on grinding equipment. It was creating a noisy irritating sound every time a stone passed in. Causing misalignment and damage to burrs. If the noise ceased after few seconds. Then the trapped stone was soft enough to crushed by… If not, I have to disassemble the machine to find the culprit. A waste of time. Still a good thing if there was no damage.

Started looking into the beans deeper. Tossing them one by one. A rather slow process just to get rid of stones, soft or hard. With the process I was also removing other foreign objects. Pieces of plastics, woods, coffee and others.

To speed things up a bit. I set up a sorting screen. Shaking a bunch of beans on top was quickly removing dust particles, tiny stones and some broken beans.

It became faster and slow again afterwards.

Cocoa beans comes in assorted sizes. Not unless I ordered it presorted, which I guess no cocoa farmer are willing to do that for me. Different bean sizes roast differently, so sorting according to size is needed. I always taste what I roast so my senses know it.

I made three sorting screens. Not standard screens. Just what I found on hardware store and tested for myself.

The current sorting process is done in three stages. First, using the biggest screen. Placing a bunch of beans on screen top and looking to it one by one. Tumbling them over several times for better result. Beans smaller than screen holes automatically falls and destined for stage two. Those that stays are called big (by me) and ready for roasting after getting rid of all defects.

The process is the same as stage one except for the bean sample and screen. The beans have already passed the first stage and the sorting screen has smaller holes. Those that stays are medium and the rest are for third and last stage.

The third stage. Percentage passing through the second screen is low but is always taking most of the time. It has too many defects to remove. These are small beans.

I am still thinking how can I speed this without spending a fortune. However, for now, hiring labor as part-time sorters is the best I can do.

Marvin is the lead chocolate maker of Ben and Lyn Chocolate Inc. Has strong background in food research and development. Occasionally conducts training and lectures. Lecturer of Cocoa Foundation of the Philippines. Do coaching and consultancy services on his free time.