I tried to roughly place the bead probe in the same location as yours, after multiple attempts I settled on slightly closer to the ramp as that was the only position that didn't interfere with the flow.

The wife has relegated all roasting activities to the tool shed which is a bit nippy, but the exhaust keeps my hands warm

Here's a thermal image of the IKAWA after five runs. Used a linearly-declining RoR profile with a finish inlet temperature of 270C (which equates to about a 225C BT in the chamber).



All that blue can be deceiving - I wouldn't recommend running the IKAWA in the playroom of a daycare center. But still, kudos to IKAWA for minimizing direct exposure to the hazardous temperatures within. Not giving out any litigable advice here, but it looks like the IKAWA crosses the burn threshold about halfway up - YMMV.

Also deceiving are some of the objects in the image: the control button is black, and the polished copper tubing I use to redirect the exhaust output has a blue, ghostlike image. It's because polished metal has very low emissivity, which flummoxes the IR camera. Everything else: the white powdercoat paint, glass, and cork all have relatively high IR emissivity for more accurate readings.

I was interested in attaching 3D printed parts to the surface of the IKAWA, so it was useful to know what areas of the roaster will play well with the plastic parts. Some 3D printing materials (like PLA) can start to soften in the mid 40's. So, bad news for PLA prints: the upper shell of the IKAWA can reach 50-60C. Good news is other materials, like PETG, ABS, ASA, Nylon, should be OK.

I've installed four thermocouples in my IKAWA - is that enough?

If you're thinking about placing a thermal probe in your IKAWA, the following charts may be helpful. Also, instructive when trying to compare or copy someone else's roasting results.

One is located low in the chamber, just 2cm up from the bottom of the chamber wall. I'm calling that my "BT" probe because it sits in the beans during the roast.

Another is located at the top of the roasting chamber, kinda-sorta in the same general area the Pro models place their in-chamber probe. IKAWA calls their probe "ET", so I'll do the same.

Next downstream is another probe in the trough (or tunnel) that connects the roasting chamber to the chaff cyclone thingy. I'm calling that probe "TT".

Finally, I've located a probe where the exhaust is just about to exit the roaster. I use some copper tubing to redirect the exhaust, so I ran the probe through that tubing and into the top of the chaff cyclone. That probe is called "XT".



I hope you've memorized the probe names, 'cause I'm gonna use them on the graphs.

All the thermocouples are narrow-gauge bare-wire probes, and I checked that all were within about 1C of each other.

Then I ran some profiles that represent the edges of my known roasting universe, to see if/how the thermal profile affects the thermal readings. After diligently pre-heating my roaster (always do that!) I ran once with an empty chamber, then again with a 65g dose of the same greens for each profile I tried. I also kept the fan speed fixed and flat at 80% for the entire roast.

The difference in readings between empty and loaded is often quite instructive, so I'll start there. This next chart shows the inlet temperature settings at the top of the chart, followed by the empty/loaded temperature readings of the "BT" probe. At the start of the roast the probe sits at the top of the greens, and by the end of the roast the beans have lofted enough for the probe to sit at the halfway point in the spinning bean mass.



Interesting how closely the empty/loaded curves overlap. I had previously installed the probe about 1cm higher, and the loaded readings were consistently a bit lower in that case.

Next up is the ET probe, located about 2 cm below the top of the chamber:



The ET readings are demonstrably cooler with greens added - because it's further away? Maybe, maybe not. My pet theory is pV =nRT, aka the Ideal Gas Law. Air pressure drops in the chamber volume above the beans, so lower pressure --> lower temperature. The beans act like a 'plug' in the airflow, raising the pressure underneath the greens, which the BT probe picks up as an elevated temperature reading. When the chamber is empty there's less resistance and therefore less drop from BT to ET.

The pressure effects really kick in for the TT probe. This appears to be the most constricted portion of the air path, which compresses and heats the air temperature above the ET readings. Neat.



And finally, the exit port thermocouple, XT. Just guessing, but I'm thinking the pressure here is somewhere between the ET and XT probes.



When you put all the loaded-chamber readings together, you get this chart:



So yeah, the ET readings are lower than the TT readings further downstream - physics!

Same all-probe test, with a light roast profile:



Good news: the characteristics are virtually the same for linear RoR profiles, from light to dark roasts. It gets weirder for a stepped profile, but the trends are the same:



Some key takeaways (at least in my opinion):

If you want to avoid the hassle of placing a probe directly in the beans, placing it in the exit trough could be a reasonable alternative.

Sticking a probe in the exhaust port is not that far off from putting it in the chamber.

If you want to skip adding a thermal probe altogether, you can estimate the temperature readings just from the inlet setpoints.

Some big caveats:
Airflow resistance plays a key role here, so batch size will shift the curves, as will fan speed. The greens themselves may also affect the results due to size grade, density, moisture, etc, Topics for another study...

One other thing I learned:
At the end of the cooldown cycle, the bottom of the roasting chamber is actually cooler than the rest of the path. I'm feeling far less urgent to swap jars and eject the beans quickly. IKAWA's cooling cycle is super-effective.

thanks for doing the work so other can learn. im loving this roaster more and more as i go along and your tests are only strengthing my thoughs

Hello,
I am learning a lot about the IKAWA HOME roaster on this forum.
Today I would like to share information about a project I tried.
I installed four K type thermocouples in my IKAWA HOME and connected them to a Phidget USB 1048 Databridge 2xTC.






I use Artisan roasting software to record roasting data. Data can be obtained from IKAWA HOME via Bluetooth and from Phidget via USB connection. (to look at a picture)
The names of the thermocouples to be added are "ET2", "BT2", "EXH", and "INT".
IKAWA HOME is not equipped with a "BT" sensor. Therefore, I set "ET2" to "BT" (Symbolic Assignments). This is because it will be easier to compare with IKAWA PRO's data and profile.





Data logging really helps my roasting experience.
If you are interested in more information, please let me know. I hope this information is helpful.
(I study English every day, but if you find any inappropriate expressions, please let me know. I will respectfully update it.)
Have a wonderful coffee life!

Nice!

I am setting up my IKAWA Home with thermocouples for Artisan. Can you please show us screen shots of where you created the Artisan settings?

How are you fastening the wires to the exterior? What are the black squares? And how did you fasten the Phidget?

Gary,

The black squares is Zip Tie Mounts - Small Cable Tie Adhesive Mount, you can find on Amazon easily. And I put Scotch Mounting tape on between Phidget and Ikawa.
I'll post the screens shots which you asking soon.
By the way, I can connect Ikawa with MacBook pro. But I couldn't do it with Windows. I'm not sure why?

Thank you, Yuzo.

For your Windows computer, did you download the Artisan Windows version?

Gary,

Yes, I did. I tried V2.10.0 and V2.8.4. It seems like Artisan needs permission to communicate via Bluetooth.
I set up the permission on MAC OS (see screenshot) But I don't know how to do that for Windows OS.


I'm happy with the Artisan Mac OS version and nice to have the Artisan Windows version.

Here we run into version issues. I am running Sonoma 14.3.1. My Bluetooth settings don't present the option to provide access to an application. For Windows, maybe someone familiar with that can post settings.