Scott Rao - ROR rate of rise "the flick"

[kenneroast]

My experience with "the flick" has been dependent upon the coffee. I find that some coffees I can avoid it all together without much caution once first crack kicks in, but other coffees it's feels like an impossibility. There are so many variables about a coffee to consider: how it's processed, where it came from, what elevation did it grow at, what variety is it, how old is it, etc. All are things that I imagine making it's contribution to "the flick." I believe in the possibility that one can avoid the flick with most, if not, all coffees. I'm too inexperienced to know how to manage it, and also don't feel I can agree or disagree with Rao's commandments.

This Ethiopia Hambela has been quite cooperative through most roasts, but this Ethiopia Agaro Nano Challa always puts up a fight by smacking me in the face with "the flick." This Kenyan is just like the Agaro Nano Challa. In fact, they usually team up against me. I have tried several different things in effort to avoid "the flick," but haven't found anything that works reliably for me.

Note: All roasts below are 454g / 1lb

[thusband]

@kenneroast, on your profile you have that top line called, is it ET-BT or is that your MET?  That's where I see a flick but what is it measuring?  I'm colorblind so I struggle identifying the lines. In the legend the color for both MET and ET-BT looks the same to me.

[Daave]

I did a few roasts tonight. Verdict: DeltaBT is extremely helpful to view during a roasting session, and I need to roast some more to get more control over it.

My roasts seem more controlled this time now that I'm doing a slower warm-up with the pipe plugged in (I previously warmed the Huky up without the J on the unit... that was a mistake), but I still had a flick at the end of them. It's definitely better though.

I'll get some more roasts under my belt first and we'll see if I can get more control over the ROR.

[kenneroast]

@kenneroast, on your profile you have that top line called, is it ET-BT or is that your MET?  That's where I see a flick but what is it measuring?  I'm colorblind so I struggle identifying the lines. In the legend the color for both MET and ET-BT looks the same to me.

The top black line is my MET / DT (Drum Temperature). Ignore the ET-BT reading. I forgot it was on the graph, and I should probably remove it.

Here's a key for each line that you'll see on the graph:

Starting from top and working your way down (ignore the solid line at ~437F):

• MET - Black
• ET - Red
• BT - Dark Blue
• BT Delta / (RoR (Rate of Raise) - Blue
• ET Delta (RoR) - Orange

The flick that I'm referring to happens with the BT Delta line (#4).

[thusband]

OK got it, thanks.  And you can also see the hint of the coming flick in your ET Delta.

[edtbjon]

This is becoming a really interesting thread. This far in my roasting "career" , I have paid little or no attention to the ET value, except for checking that the roaster is properly pre-heated. I will take a look at my previous profiles to try to see if the flick in the delta-ET curve in some way coincides with the point where I usually turn the gas down and sometimes add even more fan/air. If so, it would be really useful, as it would take away that "I guess now is the time to turn down..." and I could read it out from the TC data.
Edit: I checked my latest roast profiles/logs and it seems like on my config (solid drum/72rpm/IR heater), there is no such "warning", i.e. the delta-ET starts to rise before the delta-BT. If that just happens to Kenneth or if that is normal behaviour of a perforated drum I don't know, but if I had that "pre-flick" I'd certainly make use of it.
Anyhow, this (probably) only applies to the Huky, as we have a very effective roaster (heater), but with a relatively very low thermal mass. If this same phenomena would occur on e.g. a North TJ-067 (which weighs in at some 200+ # and other larger drum roasters, someone would have figured this out a long time ago.
The way I plan my roasts today is to avoid that flick. I've done that more or less ever since I felt that I had some control of the roaster (instead of the roaster controlling or playing tricks on me... ). My learning strategy have been to follow Mr. Rao's advice, as these "commandments" seems sound to me. If I'm able to roast according to those commandments, I should also be able to apply other roasting strategies, if I'd like to try that.

[SusanJoM]

Kenneroast, can you post what your gas and fan and grate settings were in those two roasts? 
And, what the charge weight was?
Please ?

As to Scott Rao's rightness or wrongness,  it seems to me that learning how to control the roaster to comply with his observations is as good a lesson plan as any.    If we can make the roaster do his tricks we can probably move on and get it to do the tricks we prefer for our own tastes.  It's an exercise which is probably worth pursuing.  Kinda like all that stuff we learned in school just to teach us to think.

I have no idea whether eliminating the flick will improve the taste of my roasts.  I do think that figuring out how to do it reliably would be a benchmark for me, maybe a final exam for passing HUKY novice 101....

[kenneroast]

Kenneroast, can you post what your gas and fan and grate settings were in those two roasts? 
And, what the charge weight was?
Please ?

Added the charge weight to my original post (454g).

I didn't write down all of the settings I used for these roasts, but I can ballpark the plan I followed based on what I had written on my whiteboard. Roasts were almost consecutively from what I posted, but I had maybe one or two between them. I made subtle adjustments to hone in particular parts of the curve I was hoping to improve.

Here's a rough layout:

Event	Env Temp	Bean Temp	Heat	Air	Direct Flame
Charge	460F		1.5kPa	40V	Open
Turning Point		~210F to 230F	3.75kPa
Dry End		~300F	3kPa	50V
		320F	2.5kPa		Open
		340F	2kPa
		360F	1.5kPa
Pre-First Crack		380F	1.25kPa		Close

I haven't really been writing down much information about how I manage first crack. Most of the time, I'm quite trigger happy on the controls and things move at a speed I'm overwhelmed by. Notes end up suffering from it.

Here are some other notes to go with the outline above:

• For the profiles I posted earlier in this thread, I was cranking up the heat when BT was around 235-240F while it descended. The idea was to see if I could get a sharper curve at the Turning Point.
• I seem to get a later first crack with the settings above. There have been roasts where I'm more aggressive with heat between 300F and 350F, and first crack comes sooner than I'm ready for. I don't understand the science behind it, and I haven't spent much time reproducing this scenario.
• This depends on the bean I'm roasting, but usually at 380F, I'm anticipating first crack within 5-15 degrees. I've had beans not start first crack until ~400F. A pretty wide window to work with.
• Between 390F-395F, if I haven't hit first crack yet, two things I sometimes: bump the airflow to ~60-70V for 10-20 seconds, or I just settle on an increase in airflow on a set setting.
• I have not yet determined the amount of heat to apply once first crack hits, but I tend to drop the heat down to ~1kPa, and sometimes below 1Kpa. I believe that part of the reason I get a dip on the RoR line is because I take away too much heat and don't properly manage the descend. For some beans this seems to be ok, but for others it definitely is not ok.
• You'll see in the profiles that the times started to speed up (specifically the mid phase). It's partially because I started roasting earlier in the day when it was cooler, and I also was late on some adjustments a couple times. As the day started to warm up, timing was off. Roasts were moving faster than I wanted.

Also note: For these roasts posted, I haven't tasted them, and won't be able to. All of those coffees were given to my mom and uncle. They're not overly picky about their coffee, and the most feedback I'll receive from them is: "this tastes really goooood!," or they'll find some other kind way to express they're not fond of it.

[SusanJoM]

Thanks.
I am reminded of "Pin the Tail on the Donkey".

Someday one of us will nail it.
And that will be exciting :-))))

[edtbjon]

Well, Hank seems to nail it, judging from his graphs, but they are (in my mind) a bit too coarse, at least the RoR curve.
Finding that nice decending curve is easy if you are planning for going into 2C. But I personally almost never go there, but tend to stop in the 210-215C (410-422F) range. That gives me a very short time for the development stage in which I have to try to drag out the roast without stalling it. So, from this nights session (Nynäshamn, Sweden, Europe...) with 400g loads of beans, I usually got into 1Cs with 50-60% air and 40-50% gas. As I already knew about "the flick", I had planned for a decrease in heat, which normally is to halve it after 50-60 seconds. When I approach the finish of 1C I usually halve the heat once again (landing at say 10%, just above keeping the heater alive... and maybe I increase the air to 80% or so. At this point there are only seconds left before my dumping point and I'm just hoping for the delta-BT not to go negative...)
I was on my way to say "I grew up with...", but at least I learnt how to see and differentiate different roasting grades from the illustrated guide over at SweetMarias. That was the best way for me to learn roasting on a Gene Cafe roaster. (It's still a very good reference...)  As we all are using the same TCs, I guess it's safe to say that we are some 5C/10F below the given temps for anything from 1Cs and above.

[hankua]

I just started out with Artisan; this chart is set with 3 sec. sampling vs 6 sec. on the previous one; I guess there is a small flick at the end. I'm still seeing an offset between the TMD-56 meter display and Artisan. One of my goals is the BT should always be ticking up if possible.

[SusanJoM]

I just starting out with Artisan; this chart is set with 3 sec. sampling vs 6 sec. on the previous one. I guess there is a small flick at the end. I'm still seeing an offset between the meter display and Artisan. One of my goals is the BT should always be ticking up if possible.

Hank, I have my sampling rate set at 1 second;  you can get it much tighter.
What is your smoothing set at?  (Tools > Extras)

[hankua]

curves-1 deltas-2

[SusanJoM]

Interesting.  I'll have to see how that looks.  Mine are 20 and 20

[SusanJoM]

Here is the same profile with smoothing set to 1 and 2 (first) and 20 and 20 (second).  I guess my smoothing has been covering up a multitude of errors.  On the other hand, it sure is a lot easier to read.  Hmmm....