nicolai wrote: It was in response to post no 16

Yes, but I can't resist a good straight line. And this site has a very high information to chatter ratio, so I shouldn't dilute it with bad jokes. Apologies all around.

But today I roasted some Ikawa Vietnamese to a nice medium roast level. Since that is semi-washed, I tried some Costa Rica black honey gesha with the same profile. FC came even earlier. I thought it would burst into flame. Nope - it came out with an Espresso Vision score of 24 - Light. I'm very bad at anticipating the results of non-Ikawa coffees, and they tend to be MUCH lighter than I expect.

I've been able to get some Dark and Medium-Dark roasts by using a created profile that ends at the melting point of aluminum (slight exaggeration).

GDM528 wrote:
1) Bean temp is about 40C lower than the setpoint temperatures.
2) Take at least 30 seconds to ramp between setpoints.
3) Allow at least 2 minutes for the beans to fully 'soak' at the target temperature.

I have a few more experiments in mind, and open to any suggestions from the 'bean gallery'

Hi. This is a fantastic experiment you made. What could be extremely interesting is to test two beans at the same dose.
Maybe a very small dense bean like an Ethiopian and a bigger lower elevation bean like a Brazilian.

If they are more or less identical you can really use the thermal stability of the unit to design profiles.

Alas, I don't have a wide range of greens to work with, but I'll keep it in mind - maybe source some peaberry to exaggerate the effects.

Bigger question is what "more or less identical" means - roast color? brewed flavor? espresso tasting notes?

I was actually referring to the temp curve - On the other hand I think we have already established that the "bean-probe" is actually not a bean probe, but more the environment after the bean has taken in the energy that was applied?
So based on that I do not expect a great variation between a small bean and a bigger bean as it would be similar to a smaller load or larger load.

Btw...do you mind getting me some more information around your bean probe setup - Quite nifty that it is wireless and with a small display.
And where do you log?

May not have a probe buried inside an actual bean, but I can infer a lot based on how it influences the environment around it. For example, I can get pretty close to the temperature of an ice cube by stirring a bunch of them in water and measuring the water temperature. I make a similar inference by comparing the temp curves between a chamber with and without beans. When the temp curves overlap, I argue that I can pretty closely infer the bean temperature. I took care to ensure the tip of the temperature probe was thoroughly buried in the spinning bean mass, to maximize the influence on the air temperature local to the beans.

Parts List:
https://www.adafruit.com/product/3400 - good luck sourcing one of these
https://www.adafruit.com/product/5165
https://www.adafruit.com/product/4484
Randomly chosen K-thermocouple from Amazon - then stripped off insulation
Kapton tape from whoever via Amazon

All told, less than $50 in parts. Adafruit provides really useful example code that jumpstarted the code I developed - they're awesome. Programmed the two buttons on the display unit to start/stop recording, which is stored on the Raspberry Pi processor board. Thanks to the Raspberry Pi and its built-in radios, it sits on my local Wi-Fi network, where I can log into it with my PC, transfer over the recorded data, and paste into a spreadsheet.

Hey gang, I love reading all these experiments! I'm a little less DIY than probably most of you are unfortunately. I'm curious to know what green sources people are buying from and how the Ikawa profiles are handling them in general? I've heard Ikawa green beans are very pricey compared to other sources.

IMO the Ikawa-sourced greens are unusually spendy, and I, like many other peeps that purchase a prosumer coffee roaster already had a supply of greens on hand. There's another thread that supposed to cover roasting non-Ikawa beans here: IKAWA Home - profiles I'm currently working on recipes for greens from Klatch.

About that fan...

In the following experiment I step the fan speed between 80% to 60% while holding the temperature steady. This will test the Ikawa's ability to hold temperature despite changes in airflow.



The app doesn't allow enough fan setpoints to pulse the fan speed both up and down in a single recipe, so I had to run a second test:



Ikawa's thermal control impresses me again. The temperature wobble is only a couple degrees and settles in 10 seconds as if it never happened.

I'm still inclined to pin my fan speed at 80% and forgetaboutit - unless there's some tactical reason to vary it during a roast, which I can't think of. Still good to know that I can set a temperature target and not have to worry about the fan speed affecting it.

Again, extremely relevant research. Have you tried with beans in it? That would also be very interesting.
I also dont think that fan-speed will impact the flavor (At least not my palette) - But it may be relevant if you are roasting peaberries and they are pushed into the chaff container during the roasting process - Otherwis I dont see any major reason why not just going for 80% flow

I need to drink down all the experiments I've been running, before I sacrifice more for science.

A few other fan experiments I'm thinking about:
1) Effect of fan with beans in the chamber
2) Spinning bean mass 'altitude' versus fan speed
3) Effect of fan speed on the exothermic phase, e.g. 'flick and crash'