Hmm, Mexican Altura sure roasts fast.

  • Thread Starter
  • Thread starter
  • #31
The picture below is the beans being roasted today. This is the second time for the Kenya Kiriyaga Kamwangi Peaberry bean. It roasted lock-step identical to the first attempt last week. It was selected because it has a loud and distinctive first crack. I wanted to see if the exhaust temperature took a dip (due to the beans suddenly releasing water vapor) at first crack. It rewarded me by showing a 13 degree F drop in exhaust temperature which aligned with first crack. Until then, there had been a steady increase in exhaust temperature the entire roast. I believe I have a relationship in exhaust temperature and first crack. This will really help me with the quiet beans which are devilishly difficult to hear.

A total of 5 beans will be roasted today, four of them are new to me. See picture for what they were. One is the Kenya Nyeri Gatomboya Peaberry bean, which I expect to roast very similar to the other Kenya bean already roasted today. There is a new Ethiopia Decaf I am doing for my wife as well. The numbers in the bottom left are the subjective Cup of Excellence numbers assigned by the bean seller.


  • 30 July 2022 two kenya peaberry Rwanda Ethiopia and decaf ethiopia.jpg
    30 July 2022 two kenya peaberry Rwanda Ethiopia and decaf ethiopia.jpg
    225.5 KB · Views: 38
Last edited:
  • Thread Starter
  • Thread starter
  • #32
Roasting summary: I ended up doing a total of 6 batches of a half pound each. I did two batches of the Kenya Nyeri Gatomboya peaberry beans so I could play with different roasting curves to see which gives the best flavor. The rest of the beans were roasted based upon what those "family of beans" normally like. That determination is based upon bean type, region, altitude grown, pulping/processing method, etc. The method is adjusted by what the bean does during roast (how long till they turn Full Yellow, How long before the beans enter the "first crack", shedding of the "silver skin chaff", etc).

All of the beans produced good "numbers" (moisture loss, development time, color, appearance). Roasting makes coffee beans plumper, so a picture of them would look like I added extra beans to the bags. Some of the Peaberry beans overachieved in this area. The beans from Rwanda probably had the greatest diversity in size and other characteristics (they don't have sorting down to the art which Kenya and Ethiopia exhibit). Now after roasting, they need to be "rested" for a couple days before they are tasted.

The lack of the sorting of the Rwanda beans really showed itself in color variations in the roast. The Rwanda beans are the ones in the bottom right of the photo. They have the greatest diversity in color.


  • Post Roast 6 half pound batches.jpg
    Post Roast 6 half pound batches.jpg
    167.2 KB · Views: 24
Last edited:
  • Thread Starter
  • Thread starter
  • #33
Brief cupping notes on the Rwanda. It has only rested 28 hours. From the picture you can see it was not a darker roast. (it is the bottom right bag in the previous picture). It was only taken to 22.2 percent development. The chocolate flavor is bigger than I expected, but (thus far) any fruit/citrus is detectable but quite light. It does produce a very clean cup, with a well presented body. It is less bright and acidic than expected.
Last edited:
  • Thread Starter
  • Thread starter
  • #35
Those pesky beans from Bali finally bent to my will, or perhaps rather... I started listening to what the bean was trying to tell me. It required a more gentle roast, as the beans were very low density. One of my first two batches actually had some tipping. This surprised me, because everyone had said the Gene Cafe had such a gentle roasting cycle, that tipping was nearly impossible.

Copy of my posting on the coffee bean corral site:
Bali Organic Kintamani Natural review

Cupping flavors: Very Solid body with Fruit very present. This bean has a lot of different fruit flavors present, and quite frankly they are hard to separate unless you let the coffee slowly roll down your tongue from the tip, to the base. The first impression was a confusing mix of several fruits in a coctail for this reason.

Statistics: Roast level midway between City to City+, Moisture loss 14.09 percent, Development 19.2% Green Bean description:Fairly consistent in size, with a few more quakers than what would be normally seen in a Kenya or Ethiopia coffee.

This is a low altitude bean, the first attempts to roast this bean were too aggressive, and not tolerated well by the beans. I really had to tweek my curves a bit (both length of temperature steps and actual temperature) to squeeze the best out of this bean. Previous roasts had a problem with run-away at the end, as well as second crack beginning as first crack was ending.

Roasting challenges: Lower than expected ideal roasting temperature (similar to, but a bit more gentle than a typical lower-altitude Kona Roast).
Roaster Gene Cafe CBR101 2016 model. Roast description: (uniquely for Gene Cafe CBR 101 models).
*Preheat the empty drum to 392 F (200 C) for 5 minutes.
*Rapidly stop and remove the drum. Quickly fill with 8 ounces (227g) of green beans. Re-insert the drum, start the roast at 392 F (200 C), start kitchen timer.
*At 4 minutes on the timer, raise the temperature to 450 F (232 C).
*It is worth noting I normally have the first (dehydration) phase run 5 minutes, but I was wanting to progress from dehydration to roasting phase a bit quicker to preserve the fruity esters this bean has.
* Continue at this temperature until the first pops of First crack PLUS 60 seconds. (Normally, this temperature change happens more quickly after first crack starts, but due to the lower roasting temperature, a longer period of higher heat was needed to ensure enough heat was in the beans to complete first crack quickly).
*On this roast, first pops of crack began at 10 minutes 30 seconds, temperature was lowered at 11 minutes and 30 seconds (on the kitchen timer) to 440 F (227 C).
*Continue to roast at 440 (227 C) until roast is complete.
*The goal on this roast was midway between City and City+, this was reached 13 minutes 00 seconds into the roast. Virtually all of the beans had light cracks at the tips.
*The beans were dumped and force cooled. This was before Second crack began by a clear margin.

Post Roast Notation: Sometimes I sample the raw beans after a roast, just to get a preview of where the roast ended up. I don't make coffee with them until they have "rested". I did pop a bean in my mouth, and taste the flavor, as well as the bite-through of them to get a feel for how dense and brittle/firm the beans ended up. These beans definately have a lighter density. For espresso drinkers, this will mean your usual Grams of grounds may be excessive with this bean.

For Gene CBR 101 users, the best results with this bean is a light roast, as given above. I attempted darker roasts with this bean (into second crack), and they were a bit too savory and tannic for my liking. Fruit flavors were very diminished by darker roasts. But, your milage may vary.
  • Thread Starter
  • Thread starter
  • #36
Yemen Mocca Khulani Natural review

Cupping flavors: Clearly defined Chocolate with very faint berry in the finish. Strong and well developed body with a full mouth-feel.
Statistics: Roast level City+, Moisture loss 13.66 percent, Development 23.1% Green Been description: Lots of variability in size, with a surprising amount of peaberries mixed in. This has the greatest variation in size/type of any previously purchased from CBC. This is a higher altitude bean, so an agressive roast cycle was tolerated well by the beans.

Roasting challenges: The varability in bean size meant there was a potential for variation in final roast level. Roaster Gene Cafe CBR101 2016 model. Roast description: (uniquely for Gene Cafe CBR 101 models).
*Preheat the empty drum to 402 F (206 C) for 5 minutes.
*Rapidly stop and remove the drum. Quickly fill with the green beans. Re-insert the drum, start the roast at 402 F, start kitchen timer.
*At 5 minutes on the timer, raise the temperature to 473 F (245 C).
*Continue at this temperature until the first pops of First crack PLUS 30 seconds.
*On this roast, first crack began at 10 minutes 00 seconds, temperature was lowered at 10 minutes and 30 seconds to 437 F (225 C). *Continue to roast at 437 until roast is complete.
*The goal on this roast was City+, this was reached 13 minutes 00 seconds into the roast. About 25% of the beans had light cracks at the tips.
*The beans were dumped and force cooled. This was before Second crack began by a large margin. Post Roast Notation: There was some varability in roast (due to difference in bean size), but not as significant as I had feared. It was not notable in the cupping.
When you roast several batches back-to-back what exactly do you do in between? Do you let the roaster cool down completely or…? I mean longer than the automatic cool down?
  • Thread Starter
  • Thread starter
  • #38
Billy Banana, great question. I always run a cool down cycle between beans. I do this for two reasons.

1. It extends the life of the heating coil on the Gene Cafe.
2. It returns my starting temperature to a point which helps make the roasts more "repeatable in outcomes".

I have always insisted giving the 20 dollar plus beans, a longer cooling rest on the Gene Cafe before beginning them. Many of them are lower altitude beans. Repeatability is important to me.

Based upon my measurements, the cooling cycle automatically stops when the exhaust air temperature drops below 150 degrees F (65.5 C). If all your roasts begin with a 5 minute pre-heat, this gets you at a fairly consistent point. For the pricier beans, I give it the time to get a bit closer to ambient temperature, to try to further constrain the outcomes. I use that time to catch up on my roasting notes, and do calculations for moisture loss and development.

A couple roasts sessions ago, I was doing some Jamaica Blue Mountain, Kona Peaberry, Panama Finca Lerida Geisha, and the inexpensive Bali Kintamani beans. All of them got a longer than normal pause between roasts (after the cooling cycle). The pause for most of them was an extra ten minutes. The reason why the Bali beans got some "extra love" was because the first two attempts with that bean were disappointing.

A little information on what is collected in my roasting notes:

Beginning weight of roast, for me it is 8 ounces (227 grams).
Ending weight of roast is recorded, and used to calculate moisture loss.
Moisture loss calculation.
Time which the beans turned Full Yellow.
Time which the first crack was heard from the beans. This is used as part of the calculation of development time.
Times which temperature changes were made on the roaster (up or down, and the degrees at those steps).
Time the roast was completed, and the beans were dumped into the bean cooling device. (second factor used to calculate development time).
And Calculated Development time.

Notes may include those values recorded from previous notes, to see how consistent my roasting is. It may contain how a value is selected to be changed, to improve the outcome.

It also includes a once a minute note log on the progression which the bean is roasting. It includes factors like smells, color, non-color changes in the bean (puffing, swelling, seam spreading, chaff releasing, cracks forming etc). Also notes of when smoke is seen leaving the chaff collector.

Finally, I also take two temperature measurements each minute. Once is via a thermocouple located in the chaff collector, and the second is via optical pyrometer which is measuring external drum temperature at a specific spot on the drum as it rotates around.

The goal of all this note taking is to improve my understanding of the roast process, and the small details which can affect the outcome. It also keeps me very busy during the roast. Occasionally, if I am roasting a bean which I am very familiar with, which gives repeatable outcomes, I will dispense with all the note taking, and simply read my notes from a previous roast of the bean, and repeat the recorded steps. I may note where I want to tweak it a bit, and make minor changes to reduce moisture loss, or decrease development time (if it was previously excessive).

Sometimes I incorporate cupping notes at the end of the roasting notes. This can help me figure out how to tweak the roast to modify the outcome. i.e. if the bean is supposed to be floral and fruity, but ends up with those characteristics diminished in the cup... it helps me realize my roasting curve needs adjustment.
Last edited: