Low Temperature Roasting w/ Criollo

[Alan]

Dear John,

Recently I read your experimentation with low single temperature (250) roasting of nibs for about 15-20 minutes. Since I will be roasting nibs, this is a method that I would like to try, especially when I buy some criollo, as high temperature roasting of criollo makes me nervous . Is there any clarification that you can add to this process?

You mention that one should make sure that the nibs are in a very thin layer on a baking sheet and to put them in a preheated oven at 250. I don't believe that you stirred the nibs as they all had virtually equal access to the heat. Is this correct?

How thin of a layer would you recommend? Should each nib be touching the baking sheet and exposed to the air above, i.e., would the nibs all be in one single layer, or can the layer be a bit thicker.

Additionally, what type of baking pan do you recommend. I have a nice ceramic one that I have roasted with, but it takes a while to heat up and it sounds to me that this method of roasting requires a pan that heats up quickly. Is this accurate?

Would you change the temperature or the time for the Ocumare that I asked you about? For example, would it be reasonable to run 250 at 10 minutes or so?

Thank you for your clarifications.

Sincerely,

Alan

[seneca]

Not that this refers directly to your question, but I had the opportunity to spend some time roasting with a local chocolatier recently, and his target bean surface temperature was around 260F, whatever the varietal or origin. (Of course this is in a big drum roaster, so all us home chocolate makers certainly have to work with what we've got and customize temps from there...) We started sampling out of the roaster at around 210F, and it was really neat to taste the flavor emerging as we approached the target temp.

As another aside, I don't know if you've seen the magazine Cocoaroma, but there was a great little piece on pyrazines in their most recent issue and the effect of roasting on their production in cacao. Here's a link for more info on the magazine:
www.cocoaroma.com

[gap]

A. Patric, how did you go with roasting the nibs? I have just purchased some cocoa nibs and am going to experiment with various temps/techniques. Have you got any advice for a first-timer on what temps to start at or length of time (Forastero nibs)?

[Alan]

gap said:

A. Patric, how did you go with roasting the nibs? I have just purchased some cocoa nibs and am going to experiment with various temps/techniques. Have you got any advice for a first-timer on what temps to start at or length of time (Forastero nibs)?

Hi,

Sorry it has taken me so long to get back to you. I meant to do so yesterday but it slipped my mind. Well, the answer is that I haven't tried a slow roast yet. When I was planning on doing so, I was going to use the Ocumare that John was selling. But, he ran out. So, I had to wait, and though I've now ordered some, since it is back in stock, I have yet to receive it (should be soon though). Additonally, I'll be trying a 50% milk chocolate with Ghana Forastero before I move on to the Ocumare, so it will still be some time before I know. Also, I am still wondering whether I really want to use the slow roasting method as it seems that the layer of beans must be so thin, that in a normal oven, it would take a really long time just to do a small amount of beans like 2 lbs. (i.e., there would have to be multiple batches--probably 2-3). So, I am leaning toward the higher temp start and then dropping it after a few minutes. I still need to work out what temperatures that I 'll use for the Ocumare. I am going to use the same method for the Forastero that I did last time, though. You can find it here:
chocolatetalk.proboards56.com/index.cgi?board=techniques&action=display&thread=1139670968

I hope that this is helpful.

Sincerely,


Alan

[gap]

Thanks Alan, I appreciate the response

[Alan]

gap said:

Thanks Alan, I appreciate the response

Dear Gap,

I've finally tried the low-temp roasting. You can find the results here:
chocolatetalk.proboards56.com/index.cgi?board=roasting&action=display&thread=1142618876&page=2#1144207773

I hope that this is helpful!

Alan

[Alan]

Dear all,

Today I'm doing a low temperature roasting experiment on whole bean Ocumare in a single layer. I'm doing 15, 20, 25, 30, and 35 minute roasts at 225 F. I'm leaving beans of all sizes in each group so that I can compare the effects of the roasting across bean-size too.

I'll post my results.

Alan

[Alchemist]

Alan,

You may find that even for "low temperature" roasting, 225 may not be enough. All of my cool roasts were at the target temperature, usually 260 F or so as seneca pointed out. Good luck and see what you get.

One primary goal of roasting is to drive off most of the remaining moisture, and at 225 you are only a little above 212.

[Alan]

admin said:

Alan,

You may find that even for "low temperature" roasting, 225 may not be enough. All of my cool roasts were at the target temperature, usually 260 F or so as seneca pointed out. Good luck and see what you get.

One primary goal of roasting is to drive off most of the remaining moisture, and at 225 f you are only a little above 212 f.

Dear John,

Thanks for the tip.

The interesting thing is that I've come across so many opinions as to what temperatures are best, that there seems to be no consistency at all. Of course, everything that I'm reading is based on real roasters, and not roasting in an oven, and that is why I'm trying the single layer method. I'm attempting to approximate industry batch roasters in this way as much as I can. Another interesting thing is that I've come across numerous temperature ranges regarding fine flavor beans that start all the way down at 215 f and go up to about 260 f. This is especially common in the earlier industry books that I've been reading. It seems that perhaps the later ones don't even consider the issue as much...perhaps because at the time they were written (before the upsurge of artisanal companies) the companies were too worried about speed and efficiency to make these fine distinctions. That's my only guess.

Basically I culled together all the information I could regarding the temperature of roasting for fine flavor beans and decided to pick a relatively low temperature on the scale to see how that would work. Of course I can't say anything definitive at the moment because I'm still making chocolate from the nibs I roasted yesterday at 225 f, but as far as I can tell, they turned out quite well.

As for the whole beans, they are all out of the oven, and I'm waiting for them to cool completely. The 15 minute batch smelled far too much of acetic acid, the 20 minute was much better, and from there I wasn't able to smell a huge difference until the 35 minute batch which had no hint of acid at all, and may be over-roasted. I'll have to wait and taste.

What I'm trying to do by picking temperatures on the low end of the scale is evaporate just enough water and drive off just enough acid, without driving off too many pleasant volatile components at the same time. I'm also attempting to stop the roast before the sugars in the bean are heated too much and turn from sweet/carmelized/roasty, to bitter. This is another reason I decided to keep the heat low. If I'm happy, good. If not, then I'll turn the heat up some. I'll likely do that anyway just to compare the results.

I'm going to do some research today and I'll post the varying temperature guides various books that I've been reading give so that we all have the same info. I'm going by.

Alan

[Alchemist]

You are running into a lot of what I found when I first started and why I have attempted to give some better guidelines. The ranges are crazy. I think we will find that with this becoming a consumer/public enterprise much like coffee roasting, a lot of the old misinformation will start to fall by the way side.

More than anything, I think what your research and posts are telling me is that I need to go back and revisit my information on roasting. Update it. Consolidate it. Make it much more straight forward. I recall how much effort it took for me to roast cocoa in the beginning and now I can nail a roast with a couple of tries, and I want to be able to put that into words so everyone else can do it also.

For everyone, some random roast items to keep in mind.

• Moisture. You want to get rid of most of the remaining moisture so your cocoa does not seize when you grind it.
  
• Safety. Part of roasting is heating it sufficiently to kill surface contaminates. This could be an issue with low temperature roasting. You might consider 5 minutes or so at 300 for safety sake, then drop immediately to your target temperature. I don't know for sure if this will do anything, but it is worth pursuing.
  
• Flavor development. You are *roasting*, not just heating. You may drive off acids and moisture at lower temperature, but in certain beans, too low of a temperature and the flavor will just be "flat" as you have not made the appropriate flavor compounds.
  
• Time - whole vs nibs. This is just a thermodynamic issue. Some beans (Jamaican comes to mind) want a low short roast, but you can't get an even full roast in under 45 minutes as whole beans. This one really likes to be roasted as nibs so they can heat quickly, roast quickly and be done. Whole beans are going to need more heat and time. And note, I did not say more temperature - I said heat. This lead to a final item.
  
• Home vs commercial roasters. I had the opportunity to roast some cocoa on a commercial roaster and it was amazing but also very different. The is something you have to get straight in your head when roasting - energy and energy flow. Temperature is not energy - it is an indicator of how much energy is in the system. The big roasters can get away with talking about XX degrees F for 15 minutes as they have a lot of power to play with and a massive ability to get the energy INTO the beans. When they put the beans in, they are up to temperature VERY quickly - within minutes - and then hold there. This is because they have a large amount of power flowing into the system, and a lot of air flow (convection make a huge difference). We on the other hand have 120 V, 15 amps at most usually and no convection. We can only add energy so fast, and I personally have noticed that this equates to a higher ambient temperature than the professionals see, hence 300 to 350 F sometimes. The trick is that both of us (home and professional roasters) end the roast at about the same level - 220-260 F, actually a pretty narrow, well defined range, in a "reasonable" amount of time - 15-35 minutes. And more times than naught, this time is based on how fast you can get your beans to the temperature you want them at (without burning them)
  

OK, there, I have rambled enough. Does any of that help anyone or is all of that only clear in my own head with my base of knowledge?

I think what it ALL comes down to is the literature out there is of only marginal use as it is not home roaster based. We have to develop the home roaster base of knowledge (which is what you (alan) and others are doing nicely).

[Brad]

With regard to roasting Criollo, I've now done it several times with great success the following way:

Preheat the oven to 350 degrees.

1. Use a cookie sheet and spread the beans out so the sheet is fully covered.

2. Put the beans in the oven for 10 minutes at 350.
3. Remove after 10 minutes and lower oven temperature to 300. LEAVE DOOR OPEN AND FAN WITH A TOWEL UNTIL THE HEAT LIGHT COMES ON! (this only takes a minute)
4. Stir beans and put them back in the oven for 10 more minutes at 300 degrees.
5. Remove and stir well.
6. Put them back in for another 5 minutes at 300 degrees.

I'm very happy with this result and have had no beans burn.

I've now just received two other flavors of bean and am going to give them a try.

Any suggestions for Forastero???

Brad.

[Alan]

brad said:

Any suggestions for Forastero???

Brad.

That bean is really easy to work with. I have simply set the oven at 350 F, and roasted them (2 lbs. at a time in a glass casserole/lasagne dish) for 27 minutes, stirring every 5 minutes.

Works like a charm.

Alan

[Alchemist]

brad said:

Any suggestions for Forastero???

Brad.

For that particular Forastero (the Ghana),

1) use your same profile, but don't worry about the pulling and forced temperature reductions.
2) Go for a few snaps and cracks.
3) Go for an end of roast smell of baking brownies, not time. This is the bean I originally coined that indicator for.

It is a very forgiving bean.

[Alan]

Alright, here are my findings:

Inconclusive. I need to roast more tomorrow. My test batches were too small for me to really clearly and conclusively say which beans turned out best, and if the result was acceptable for chocolate making. I'm going to roast more beans tomorrow.

On the subject of roasting, I implore you, John, to put up your information about making a drum roaster:
chocolatetalk.proboards56.com/index.cgi?board=roasting&action=display&thread=1139009442

I would much rather build such a roaster and roast like that, and I would even buy a gas grill in order to do so.

I'll plead more if I must. Just let me know.

Alan

[Alan]

Alright,

I have just finished sifting through the various books I have at my disposal once again. Here is the filtered and refined gist of things:

With criollo used for fruity high notes and nuttiness to make a very good quality, high-cocoa percentage chocolate, one should roast low.

Commonly the ranges were anything from 205-230 F.

The most commonly targeted temperature was about 215 F and this is said to give these types of beans a sufficiently low roast...just above raw so as to preserve the high notes and allow other preferable characteristics to develop during conching wihtout developing a hard to get rid of bitterness. I found the most detailed information of what happens in roasting and conching regarding fine flavor beans roasted like this is given in the book "Chocolate Production and Use" by L. Russell Cook 1979.

The main points are evaporation of moisture and evaporation of acetic acid causing the "vinegary smell" but not enough heat to cause over-roasting bitterness, which is a quality that is far harder to rid the chocolate of than is too much acidity.

Something to point out here is the idea that the liquor will be conched at a temperature somewhere between 130 and 200 F, though 140-150 F were the most common ranges given.

I think that one problem here is that often we may be looking at chocolate not enough as a product in the works with serious interconnection of steps, and more as a series of things in and of themselves.

Roasting isn't necessarily perfect when the bean tastes good, it is perfect when the roast performed takes the bean to such a place that when refined with sugar and conched at heat for a certain length of time, the finished product tastes incredible. How is that for being general. One must have an idea of what that finished product should taste like, and some idea of the qualities of the beans that one doesn't want in the finished chocolate, as well as the ones that one does want.

I'm going to go out on a limb here and forward a hypothesis that will be tested, in part, today. I think that the Ocumare needs to be more de-hydrated than full-on roasted. I believe that the temperature must be dropped. My hypothesis is this:

Roasting Ocumare on a thin baking sheet, in a single layer, at a low temperature around 215 in the oven until the bean is sufficiently dry (husk is dry and crackly) and absent of almost all the smell of vinegar will render a very lightly roasted bean that will then develop into a very interesting chocolate when refined and conched at around 150 F, probably for about 72 hours.

Now, this hypothesis is not only born from my reading, but from my thoughts posted yesterday about cookie baking temperatures and how they need to be lowered and not raised when more batter is added so that they aren't burned on the outside and raw on the inside, and the fact that by beans roasted yesterday were not consistently (i.e. from the outside to inside) roasted, but were either better on the outside than the inside, or over-roasted on the outside and fine on the inside.

I think that this lower temperature might just be the answer.

Today I will do part of the testing to see if my hypothesis is worthy or just a load of you know what. If I come up with beans that I find to be workable, given what I have described above, then I will roast a whole batch like this and make chocolate, thereby fully testing my hypothesis.

Finally, one might ask how it could be that Brad roasted his Criollo at such a high temperature and had no problems with it. Right now my answer is, and I may be completely wrong, that he was probably making milk chocolate and that the additional cocoa butter, sugar, vanilla, and milk, and the lack of conching allowed for a chocolate where any bitterness caused by over roasting that would be very noticeable in a 75% conched product, was fairly well balanced by the lingering acid and other added products.

Like I said, I could be completely wrong here, but if I am, then at least everyone can learn from my gigantic mistakes. That's better than me keeping my mouth shut to save face.

Alan