When cool green beans are dropped into a hot roaster, something magical happens. The beans start to loose their moisture, the color begins to change, and exotic aromas spring from thin air. As the heat continues to build, the beans increase in size, sugars begin to caramelize, and the flavors unique to coffee are created. Two important milestones punctuate these complex transformations - First Crack and Second Crack.
Roasting great coffee involves more than simply applying heat to green beans; it involves knowledge about the vital points at which heat transforms the elaborate compounds within them. First Crack and Second Crack are critical signals that tell us if heat has, or has not, been successful in imparting those amazing flavors and aromas that we coffee lovers crave. As a result, successful roasting requires that those milestones be accurately recognized and carefully managed.
Ciel... seeking Heaven in a cup with......................................................... EXPOBAR Brewtus II - MAZZER Mini E - MAHLKÖNIG/Baratza Vario - GENE CAFE - HOTTOP KN8828P2 - HotTop KN8828B2
When I began roasting a few months ago, I looked for answers to two basic questions:
(1) At what temperature does First Crack begin?
(2) At what temperature does Second Crack begin?
These are pretty straightforward questions; however, I quickly discovered that precise answers were hard to come by. Like most newbies, I turned to the WWW for answers. I was disappointed to find confusing and contradictory information about the temperature at which these critical events occurred. Here are examples of the answers I found on popular WWW coffee sites like Sweet Marias and others.
Examples of Confusing First Crack Temperatures
On a given WWW site, one might read that First Crack occurs at 353F; on another site one might read that it occurs at 402F. Some large sites may even report several conflicting temperatures (temperatures varying by more than 40F) depending on which part of the site one is consulting.
Examples of Confusing Second Crack Temperatures
Here as well, we see a similar situation. One site might show that Second Crack occurs around 424F, while another site might say 454F - that’s a range of 30F. Further, it is disappointing to see that few sites provide any details about how the reported temperatures have actually been measured.
Having accurate answers to the above questions is important for many reasons; one of them being that most home roasters do not provide a means to monitor the actual bean mass temperature. A as a result, the operator must rely on other proxies, such as the roaster’s display showing the temperature inside the chamber. For example, what should the operator who wishes to minimize the chance of entering Second Crack (or perhaps wish not to enter Second Crack too quickly) do in the final stages of the roast? Should the operator consider lowering the temperature in the chamber to 454? … to 430F? …to 424F? In this regard, the wide temperature ranges reported above are sufficiently imprecise to make them of questionable use for a newbie roaster.
Ciel... seeking Heaven in a cup with......................................................... EXPOBAR Brewtus II - MAZZER Mini E - MAHLKÖNIG/Baratza Vario - GENE CAFE - HOTTOP KN8828P2 - HotTop KN8828B2
In light of this confusing information, I decided to take the matter into my own hands. In order to measure the actual cracking temperatures of coffee, I first began by modifying my home roaster.
To measure the bean mass temperature, I mounted a horizontal thermocouple (Omega K probe) through the back wall of the roasting chamber. This modification is explained at the following link: http://homeroaste...ad_id=2785
The K probe was then connected to an external thermometer (Hanna HI93530). These instruments were tested for accuracy in three distinct environments for which the actual temperature was already known. According to the manufacturers, both the probe and thermometer have a margin of error inferior to 0.75%. The probe was positioned inside the drum according to guidelines proposed by Davis and Ribich http://www.ambexr...robes.htmlFor example, the probe was: a) mounted on the side of the drum that is rotating the beans in an upward direction; b) extended into the bean mass at least one inch (a depth equal to 10 times the diameter of the probe); and c) centered within the bean mass, thus ensuring an even thickness of coffee beans above and below the probe, and thus keeping it submerged at all times.
All the roasts were conducted at an altitude of 230 ft above sea level using 250 grams of green beans. The roaster that I used was an electrically powered HotTop drum roaster with the above mentioned modifications.
Edited by ginny on 07/24/2012 06:57
Ciel... seeking Heaven in a cup with......................................................... EXPOBAR Brewtus II - MAZZER Mini E - MAHLKÖNIG/Baratza Vario - GENE CAFE - HOTTOP KN8828P2 - HotTop KN8828B2
I completed dozens of roasts while measuring the evolution of the bean mass temperature with the above setup. Here is what I discovered about the temperature at which First Crack occurs.
As a general rule, I found that First Crack begins at 384F
Since I have a quiet roaster, during the roast I frequently heard pops between 350F and 359F. On occasion, I also encountered an outlying First Crack between 369F and 374F, but that was rare. As a general rule, First Crack began at 384F. First Crack was typically quite active between 395 and 402F. There might have been small variations (one or two degrees) around those numbers from one roast to the next. I also obtained similar temperatures when roasting decaffeinated coffee beans. Those roasting at elevations superior to 230 feet above sea level should expect First Crack to begin below 384F.
Ciel... seeking Heaven in a cup with......................................................... EXPOBAR Brewtus II - MAZZER Mini E - MAHLKÖNIG/Baratza Vario - GENE CAFE - HOTTOP KN8828P2 - HotTop KN8828B2
Following dozens of roasts while measuring bean mass temperature, here are the measurements I found concerning the temperature at which Second Crack begins.
As a general rule, I found that Second Crack begins at 428F
I might on occasion have heard an outlying Second Crack occurring at around 424F or so, but that was rare. As a general rule, Second Crack began at 428F. Second Crack was typically quite active between 434F and 438F. There might have been small variations (one or two degrees) around those numbers from one roast to the next. I was not able to observe a specific temperature at which Second Crack actually ended. The highest temperature to which I roasted was 473F, and I heard cracks emanating from the roast until the final moment when the smoking beans were ejected. Those roasting at elevations superior to 230 feet above sea level should expect Second Crack to begin below 428F.
I would be interested in hearing from other HRO members who may have conducted similar trials, and in learning if their results confirm (or possibly challenge) the observations posted here.
Ciel... seeking Heaven in a cup with......................................................... EXPOBAR Brewtus II - MAZZER Mini E - MAHLKÖNIG/Baratza Vario - GENE CAFE - HOTTOP KN8828P2 - HotTop KN8828B2
I work in Celsius
My observations are based on several hundred roasts and several hundreds of kilos of beans
Growing region, growing condition, moisture level and elevation has some bearing on the temp of both First and second crack
My observations are (note we are talking bean temperature)
First crack is in the range of 195 C to 205 C
Second crack is in the range of 218 C to 223 C
As its a natural product it may alter by a degree or 2
KK
My web site > koffeekosmo.com.au I home roast and I like it :P
Blog - http://koffeekosm...gspot.com/
Bezzera BZ40:Bezzera Strega: BNZ MD74 Grinder: 6 Box hand grinders: Pullman Tamper Convex: (KKTO) Turbo Oven Home Roaster: CONA Glass Rod Syphon: Pyrex Brewer:
This has unfortunately been a confusing topic for as long as thermometers and coffee have been around. I can't seem to find the studies anymore that did exhaustive testing with in the bean probes but 1C came in around 380 to 385F internal bean temp.
As I'm sure you found in your research, when measuring bean temp in our roasters we're measuring a combination of beans and environment no matter how deeply planted the probe is in the bean mass. With this shortcoming, it is totally roaster and probe placement dependent for where we'll see 1C and 2C occur. OTOH, there seems to be a larger number of folks reporting 400F +/- 5 degrees for 1C which is the case for my fluidbed roasters as well.
As KK mentioned, different coffee's will experience some variation where 1C occurs. I have also noticed variations as much as 5 degrees F depending on how aggressively I pushed the roast on the way to 1C.
I guess the most important thing is where it happens in your roaster and probe placement so you can use that number for reference.
Allen
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
I work in Celsius... My observations are based on several hundred roasts and several hundreds of kilos of beans... Growing region, growing condition, moisture level and elevation has some bearing on the temp of both First and Second Crack...
KK, you have considerably more roasting expertise than I have, and I’m very grateful to you for sharing it with us here. First Crack and Second Crack are such critical turning points during roasting that I elected to focus on determining the most reliable temperatures at which cracking actually begins (excluding outliers). KK, I converted your results from C to F degrees, and I was amazed to find that our observations concerning starting cracking temperatures are almost identical.
Another amazing realization is the fact that actual cracking temperatures are considerable lower than what is generally believed by many home roasters, as John pointed out earlier:
In that post, John quoted results from a Sweet Marias “roast reaction chart” claiming that “First Crack occurs at 401F, and Second Crack occurs at 454F”. When John first noticed my lower cracking temperatures, he went on to say: “Yours is the first notes I've seen of First Crack taking place well outside of the norm practiced rather broadly.”
The magnitude of the difference between the relatively high temperatures that John quoted (from SM’s site), and the lower temperatures that I found using my home roaster are not trivial. Given the utmost sensitivity of the delicate and complex coffee flavors to extreme temperatures, exposing the smoldering beans to an extra 20F-30F of heat for an extra minute or two might quickly transform an exquisite roast into a unpalatable one.
Ciel... seeking Heaven in a cup with......................................................... EXPOBAR Brewtus II - MAZZER Mini E - MAHLKÖNIG/Baratza Vario - GENE CAFE - HOTTOP KN8828P2 - HotTop KN8828B2
Ciel, this has unfortunately been a confusing topic... it is totally roaster and probe placement dependent for where we'll see 1C and 2C occur... I guess the most important thing is where it happens in your roaster and probe placement so you can use that number for reference... Allen
Thanks Allen. Neal Wilson (RoastersGuild) agrees with you on this important point. He confirms that 1C and 2C occur at about 380F and 430F respectively. More importantly, he points to the critical importance of probe placement for accurately measuring BMT. He suggests that improper probe placement is responsible for the wide ranging (and confusing) numbers being reported about 1C and 2C. Ciel
Ciel... seeking Heaven in a cup with......................................................... EXPOBAR Brewtus II - MAZZER Mini E - MAHLKÖNIG/Baratza Vario - GENE CAFE - HOTTOP KN8828P2 - HotTop KN8828B2
For me 1st crack comes in at 395 to 405 F, 2nd comes in around 445 F. I do not roast past FC very often.
+1 on these measurements for my equipment. I also don't hit second crack very often (on purpose). I'll add that when I say 'First Crack' I do not mean the 2 or 3 outer snaps, but the moment there are several snaps in succession. When I say 'Second Crack' however, I do mean the outlying initial snaps. So the time between first and second crack is the time between the several successive first cracks to the moment of the initial second crack snap. It's important to note be cause as you notice, nomenclature can be different depending on the resource.
CHAD
Don't put the cart before the horse. Put the horse in the cart and listen to him say "weeeee" all the way down the hill.
seedlings wrote:
... +1 on these measurements for my equipment...
Thanks to Chad, Ringo and others for kindly sharing your 1C and 2C temperatures. Also, it was helpful of you Chad to take the time to explain how you define 1C and 2C for the sake of clarity – much appreciated!
Earlier, Allen pointed out that observed results are largely a function of roaster type and probe location. In order to maximize the potential usefulness of posted results, would it be possible for Chad, Ringo and others to include details about your equipment and probe placement please?
Examples of critical details about probe placement, accuracy and repeatability may be found here: http://www.ambexr...robes.html
Looking forward to learning more about your equipment, and probe placements...
Ciel... seeking Heaven in a cup with......................................................... EXPOBAR Brewtus II - MAZZER Mini E - MAHLKÖNIG/Baratza Vario - GENE CAFE - HOTTOP KN8828P2 - HotTop KN8828B2
rom a purely practical standpoint as opposed to a educational one, does the actual temperature at BT matter so long as your roaster and probulator hit and measure certain targets consistently and respectively?
In other words, if first crack occurs in X minutes, and your probe reads 1,000C +/1 a few C with each roast, just calibrate it in your head, no?
I'm on my third type of roaster, and each has had slight temp differences. My recent one measures in C, which I'm not as familiar with, so I just ran a couple throw away roasts to find out when the beans begin to yellow, and when 1st and 2nd cracks occur. It's been consistent, and turns out after doing conversions to F, I'm in the ballpark with everyone else's results here (cool beans for me!).
Jedovaty, you're right on. Variations in roaster type, bean type, probe type, speed of roast... will change temperatures. For me, repeatability is important. Heck, you don't need a thermometer at all if you can see and smell the beans. Physical cues may perhaps be the very best.
CHAD
Don't put the cart before the horse. Put the horse in the cart and listen to him say "weeeee" all the way down the hill.
Chad: "Physical cues may perhaps be the very best". Yes, but it not only takes time to develop the senses to read these cues, it's also still quite the subjective reading. For a beginner like me, this is hard, though I realize with time and experience, I'll learn. What surprises me is that everyone here seems to embrace both the science and art - but leaves a rather clear line between the two. In other words, we use all this great technology to track, log, even control the roasting, but at the end of the day, it's still an art to determine when to drop the beans.
I get all that. As a beginner myself, what I've struggled with a bit thus far has been when to stop, so this thread is very interesting to me. Given that we have variations in roasters, beans, probulator placement, ambient temps, etc, I can understand why the roast level is so sujective - city, city+, full city, full city+,etc.
However, I believe we can quantify this - instead of using actual temperature, why not use a % between first and second cracks, for example? This would eliminate the variable of differing BT #s across different roasters and TC placements. We could possibly take it a step further and measure the time it took to get to the %, and can go even more techincal by reading it as a differential based on the ROR from FC to SC.
The above two measurements could even be applied to drying phase and ramp to FC, though it'd likely require more math given charge temps and turnaround temps.
It's unlikely I'm the first guy to think of it this way - there's gotta be a reason why something like the above hasn't been adopted. Perhaps the "zen" aspect is too strong? Or there are too many other factors making it not as easy as it appears to me?
For a beginner, and remember coming from a real beginner here (I'm still learning to like coffee!), keeping it technical helps the learning curve; consider it like learning any art, whether piano, violin, kazoo, painting, sculpture, cooking, desserts, shaving, root beer, electronics - everything starts with technique, and ultimately transcends to the senses of the artist.
It's unlikely I'm the first guy to think of it this way - there's gotta be a reason why something like the above hasn't been adopted. Perhaps the "zen" aspect is too strong? Or there are too many other factors making it not as easy as it appears to me?
- everything starts with technique, and ultimately transcends to the senses of the artist.
My observation at HRO is that the drive toward the techinical is actually larger than the drive toward the zen.
I believe the problem is the 'other factors'.
A truly accurate bean temp is elusive.
Enviromental readings are variable as well. etc...etc ...etc
I think all these variables are misleading. Machines rely on hard imformation. I think we are the only machine (so far) that can correlate all the signals and variables, and then produce the moment of alchemy....
...a decision.
It's unlikely that any automated system will teach you to roast. It's more likely that you would teach your machine what to do.
Every discipline starts with technique....but we evaluate that technique with our senses.
Set some beans on fire...make some mistakes. It's a beautiful thing!
Have some fun and tell us about it! Happy roasting!! -Scott
However, I believe we can quantify this - instead of using actual temperature, why not use a % between first and second cracks, for example? This would eliminate the variable of differing BT #s across different roasters and TC placements. We could possibly take it a step further and measure the time it took to get to the %, and can go even more techincal by reading it as a differential based on the ROR from FC to SC.
You may not be the first to think of this but I haven't seen much of anything over the years.
I think you're on to something that could be very beneficial. How about instead of % of temp change between 1C and 2C and differentials we instead use RoR along side the number of degrees change from onset of 1C to finish. Eliminates translating into degrees. I would think even with sensor variations that change in temp would be a usable value.
For example, My Tanzanian Roast
- To plus 200F: RoR 75, Tr 175F
- To yellow: RoR 15, Tr 50F
- To 1C: RoR 22, Tr 100F
- 1C to end: RoR 9, Tr 34F
(RoR = rate of rise, Tr = Temperature rise)
This tells us you went from room temp to around 250F at a rate of rise of 75F/min, went from there to yellow at a rate of rise of 15F/min with a change of 50F, from yellow to 1C at a rate of 22F/min with a change of 100F, went 34 degrees past the temp of 1C at a rate of rise of 9 degrees per minute.
Totally goes around sensor differences between machines and gives you a usable value. Obviously we would need to agree on what would indicate the start of 1C (3 pops, 5 pops...) but this could be worked out.
Allen
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
Interesting thoughts! from jedovaty and Allen. Imma let it rumble in my brain a bit.
I consider first crack to have begun when I hear 3 or 4 snaps in rapid succession after the first few outliers have finished bragging about being in first place.
JD
Respect the bean.
John Despres
Fresh Roast 8, Gene Cafe, JYTT 1k, Quest M3, Mazzer Mini, Technivorm, various size presses and many more brewers.
Allen: we're on the same page, but I still think %s would work better instead of the temp differences, as they will eliminate any need to convert between units (e.g. C vs F), and help remove the partial variables such as thermocouple placement, accuracy, etc.
Here's my own example that actually demonstrates one end of the extreme, using FC to SC for simplicity:
- my now dead DIY roaster would show FC at about 400F, with SC around 425F. FC seems to be within norm of others, but SC is way off. It was consistent this way while the roaster worked, so I used that as my guide.
- my new roaster measures in C, and FC occurs at 200C with SC at 230C. Both these seem to be inline with what some others are finding.
I don't know where your SC occurs, but let's use an example that your FC occurs at 400F and SC at 440F; you ended the roast for that Tanzania at 85% of the way to SC. Based on the info you provided, it took about 3 minutes and 45 seconds after FC. You end of roast profile would then be something like 85% @ 3.75 minutes.
So if I were to try to duplicate that, my target temps for the DIY roaster would be 421F and the new roaster would be 225C, with a target to reach in 3 minutes and 45 seconds.
It's easier to do this for FC to SC, as the sensory cues are probably the most distinct and easiest to identify. Taking this concept to the other important areas of the roast is a bit more challenging, specifically the initial heat charge, drying time, and then FC. To demonstrate, it is understood that the general target "dry time" is 300F or 140C, typically 4-5 minutes. With the wonky behavior of my DIY roaster, is that 300F accurate? Using the bean yellowing may be more appropriate as it's easier to identify going from green to yellow. This then begs the question, "why not just go back to city, city+, full city.."; those are harder for a beginner to distinguish.
Or maybe the answer to that is, you don't really use this concept for the profile before FC.
Funny part to me is that this just seems to be nothing more than restating what we all already do, but it makes the profiles a little more portable and understandable for the beginners.
I can attest to folks being out and about. Me and Julie were out hitting the garden centers for loading flower pots all day. Nice weather! Tomorrow the Kamado/egg gets fired up!
ginny
05/18/2013 17:32
Jack, it is quiet today, I think the lovely day has folks out and about.
ginny
05/18/2013 17:31
sbonder
thanks for joining us...
JackH
05/18/2013 15:28
Quiet here today.
zombie coffee
05/17/2013 11:02
ricksroasters
thank you for taking time to join us...
allenb
05/15/2013 15:56
nikijack Howdy and Welcome to HRO!
freshbeans
05/15/2013 13:57
FataMorganaCafe
ginny
05/15/2013 03:04
blong2001 thanks for joining us...
ginny
05/14/2013 13:54
that is sweet... thanks for posting lylabrown...
for you
thanks for joining.
for you