Homeroasters - Home Roasting Coffee Community » BUILDING A ROASTER » Fluidbed Roaster
Who is here? 1 guest(s)
 Print Thread
Measuring Bean Temp in a Fluidbed
allenb
Top  #1 Print Post
Posted on 11/02/2023 9:17 AM
I continue seeing posts where a builder or user of a fluidbed is trying to find the right spot for the bean temp probe. While it's true that controlling via environment temp (hot air entering the inlet of the roast chamber) beats trying to control via bean temp, you still have to measure bean temperature if for nothing else, to know it's rate of rise during the crucial development stage. Getting that stage correct is a must or your roasts will be hit or miss and probably a lot of misses.

So, where do I place my sensor for BT? I've tried all the usual places in or near the bean mass and never had much success there. Then, one day when my thermocouple wire had slipped out of the bean mass and was dangling 1/2" or so from the top of the roast chamber, I noticed it read around 400F (204C) at start of first crack and showed a steady readout through the rest of the roast. Almost all of my roasts from then on, read close to the same so I found a way to tie off the TC near that location which isn't any special place but just somewhere near the top of the roast chamber. You may be thinking, that can't possibly be accurate because it's sensing too much convection air and not the actual beans. I thought the same thing but time after time, when placing the sensor in the calmest part of the bean mass, the reading is almost identical to the leaving air at the top of the chamber. So, I don't fret anymore about where to stick that TC in the bean mass as it is not necessary. And, even if your setup makes the reading come out higher than 400F (204C), it doesn't matter. All you want is to see a representation that can allow you to follow rate of rise.
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
HarryDog
Top  #2 Print Post
Posted on 11/02/2023 11:43 AM
When I first started roasting I slid the probe in from the top in the middle just over the green bean mass, this was the most accurate that I remember.

Now in the bean mass just off center it's not as accurate, I try to ignore it and just mark FC to complete the roast.

I don't remember the thread but a member reported a good spot in the bean mass. He had 3D printed the base using Black and grey plastics if I'm remembering correctly? I know this is not much to go on but does anyone know where that thread is?
Edited by HarryDog on 11/02/2023 2:32 PM
 
renatoa
Top  #3 Print Post
Posted on 11/02/2023 1:36 PM
For an asymmetrical build the best I found was in the middle of the slide/chute plane.
The probe is buried into beans and protected from the hot air stream.
 
btreichel
Top  #4 Print Post
Posted on 11/02/2023 11:34 PM
With my P1's and a hurricane lamp I found the best spot was towards the bottom of the return flow of the beans.
 
renatoa
Top  #5 Print Post
Posted on 11/03/2023 8:29 AM
General rule is: least airflow and bulk of beans.
Location is different for poppers (vortex along the wall), or for fountain (central air column), or for asymmetrical designs, but the goal is the same of all.
 
oldgrumpus
Top  #6 Print Post
Posted on 11/04/2023 5:33 PM
Just a dumb question from a drum roaster... I'm sure someone has used an infrared thermometer? Seems like it should work as the reflected reading is bouncing back after contact with the beans.

Don't mean to take this discussion off-topic... I'm just curious.
 
renatoa
Top  #7 Print Post
Posted on 11/05/2023 7:12 AM

Quote

Can infrared thermometers work through glass?
No. infrared thermometers cannot measure the temperature accurately if there is glass, liquids, or any transparent surface between the subject and thermometer.
The temperature values displayed in such conditions are of a transparent surface and not the subject.


https://www.omega...20subject.
 
Spyderman24-7
Top  #8 Print Post
Posted on 11/08/2023 2:24 PM

Quote

oldgrumpus wrote:

Just a dumb question from a drum roaster... I'm sure someone has used an infrared thermometer? Seems like it should work as the reflected reading is bouncing back after contact with the beans.

Don't mean to take this discussion off-topic... I'm just curious.


Tried using a decent IR thermometer a handful of times and it was a bit inaccurate as in the motion causing fluctuation. Also strongly considered a thermal imaging camera and contacted FLIR to ask about using one just for roasting. Problem is the frame rate (lower) of most affordable units is just to slow to really accurately display temp changes with the agitation involved.
 
oldgrumpus
Top  #9 Print Post
Posted on 11/08/2023 11:05 PM

Quote

Tried using a decent IR thermometer a handful of times and it was a bit inaccurate as in the motion causing fluctuation. Also strongly considered a thermal imaging camera and contacted FLIR to ask about using one just for roasting. Problem is the frame rate (lower) of most affordable units is just to slow to really accurately display temp changes with the agitation involved.


What if it could be fed into Artisan and have it smooth out the fluctuations?
 
Spyderman24-7
Top  #10 Print Post
Posted on 11/09/2023 2:00 AM

Quote

oldgrumpus wrote:

Quote

Tried using a decent IR thermometer a handful of times and it was a bit inaccurate as in the motion causing fluctuation. Also strongly considered a thermal imaging camera and contacted FLIR to ask about using one just for roasting. Problem is the frame rate (lower) of most affordable units is just to slow to really accurately display temp changes with the agitation involved.


What if it could be fed into Artisan and have it smooth out the fluctuations?


Honestly no clue as I don't rely on any type of software and control all things manually/on the fly. Actually was looking into that possibility to give me real time feedback due to being profoundly deaf for 5 yrs now as I relied primarily on hearing the development before that suddenly transpired. The thought of being able to point the device at the bean mass and get thermal feedback within a few seconds seemed doable, but will stick what currently works as is.
 
renatoa
Top  #11 Print Post
Posted on 11/09/2023 3:33 AM

Quote

oldgrumpus wrote:

What if it could be fed into Artisan and have it smooth out the fluctuations?


Or by selecting a longer sample interval (default 3 seconds), and check oversampling.
 
Jump to Forum: