Not a member yet?
Click here to register.

Forgotten your password?
Request a new one here.
You must login to post a message.

01/04/2019 10:55 AM
2 years today...Miss you Ginny!

01/04/2019 4:24 AM
A little shout for the G'ster. 2 yrs. roar

12/27/2018 5:14 AM
Hi Brandon, please disregard the "please update". I think we're all up to speed now. The need for cache clearing get's me every time! limb

12/26/2018 11:36 AM
Update? pouring Merry Christmas back atcha

12/26/2018 3:07 AM
Merry Christmas Brandon. Please update soon! christmas tree

Users Online
Guests Online: 2

Members Online: 0

Total Members: 5,766
Newest Member: cimbalino
In Memory Of Ginny

Latest Donations
John Despres (Scene... - 25.00
snwcmpr - 10.00
Ozo - 20.00
Josh Woodrow - 10.00

View Thread

Who is here? 1 guest(s)
 Print Thread
Getting Accurate Temperature readings with Fluid Bed
Mine is in Artisan Designer, not a real world plot, is VERY hard to get smooth continuous RoR with hot air roasters, smooth is specific to machines with a lot of thermal inertia.
You can see an example of oscilating RoR here, in this recent post:
In hot air machines ET is the tractor for BT, you have no thermal inertial mass to afford a moment of rest. Keep ET ramping permanently and should have a nice BT/RoR curve too.

Another example of one of my recent roast , check attached.
Did a mistake at about 11 min, when I seen RoR start rise instead decreasing and stopped ET increase, that little flick above 196 marking. As you can see the effect was more dramatic than expected, should keep ET increasing but reduce the ramp.

My Etiopians (Kochere) have clear centerline, very little tan.
renatoa attached the following file: [6.13kB / 31 Downloads]
Thank you for that curve. I will try to use that as best I can for a guide. What do you mean by "clear centerline"? Sorry I am still a newb at this and learning all of the terminologies.
As a side note, I brewed that Ethiopian this morning and it was not bad at all. I am sure it can be much better but at least it didn't go to waste. Thank you for taking the time to help this newb!
Want to Help?
By donating you can ensure that HR will continue to be open for years to come. Expenses are 100% transparent and can be viewed publicly at all times. You can find more information on donations and why we request them by reading up on the forum topic here. Thank you.

clear centerline: not charred, very visible.
As in this pictorial:
Brewin Bruin
[quote]ChicagoJohn wrote:

Another thing to take into account is that TC's do not necessarily measure temperature at the tip -- the EMF they generate is from the hottest point in the TC circuit in relation to the cold junction (which is electronically determined in practice). If, for example, your RC wall gets hot and transfers that heat efficiently to the TC wires, the TC output would be affected by the RC wall temperature and less sensitive to changes at the junction / tip. So with that in mind, I use a piece of high temperature silicone tubing to limit thermal conductivity between the RC material and the TC wire.


I’ve been lurking on the fluidbed roasters threads for a while, and thanks for the wealth of information! I’m a little puzzled by this explanation here (if I recall, the author is a retired chemist). 🤔 Coming from an excessive amount of chemistry myself and having taught this topic a few times, there’s a problem with what was said above. The EMF ( also called potential or voltage) arises ONLY at the junction, not anywhere else in the wire.
I think the observed issue Chicago John is pointing out is more a matter of the wires conducting heat to the cold junction, throwing off the reading. The voltage read is the difference between the hot and cold junction potentials, (the junction is where the two different metals of the thermocouple are joined, the cold junction is built into the readout device, in ancient days it was another thermocouple sitting in an ice bath, but that was a pain). The temp of the wire in other places doesn’t matter because there’s no junction there, unless it’s throwing off your cold junction reading by conducting heat. So, if you keep the thermocouple wires reasonably long, that should leave a bunch sitting at room temp before you connect them to your readout.
Edited by Brewin Bruin on 01/02/2018 2:30 AM
Should this mean that all commercial roasters that keep at least 10 cm of wire inside the drum in the beans mass, kind of fluid bed... delivers wrong temperatures... :confused:
Brewin Bruin
No, that’s not correct. The length of the wire by itself does not change the reading. It’s the possibility that the wire is short enough to conduct heat from the roaster where the junction is to the readout (the second, cold junction) that is being considered. If the wire is pretty long, say a few meters is hanging out in room temperature air, then the wire is cool where it contacts the readout, the heat it picked up in the RC has been given off to the room.
The trick about thermocouples is that they are always read with essentially zero current in the wires. Sounds goofy if you only consider Ohm’s Law, but it’s quite possible. This means that things which affect the wire resistance like its diameter, length, and temperature can be ignored, they don’t effect the voltage reading.
That being said, the longer the wire is, the less heat it can conduct from the RC to your readout. Generally this is not a problem, but since someone was noticing odd behavior, I’m guessing that this is the source of the problem.
So, to sum up, length of wire inside RC doesn’t change your reading by itself (although where you place the junction, that is, the tip of the thermocouple, certainly will). If you’re worried about problems, you can run 6 feet or so of wire in a location at room temp, away from the RC, before you connect it to your readout.
“Learn by doing” —definitely including doing it wrong.
Got it, but again you puzzled me with the zero current statement :)
Thermocouples generates voltage, not variable resistance as a Pt element... measuring a voltage does not need a current pass through the junction!

The difficulty is to measure the (too low) voltage, which has variations almost radio signal levels... 40-50 µV/°C => 4-5mV for 100 degrees, an awful low voltage to be measured using ordinary measurement equipment. Thus, special noise requirements.
As a comparation, an Arduino board ADC resolution is 5mV for a single bit ! Connecting a thermocouple directly to an Arduino board will read 0, 1, 2... for ambient, 100C, 200C degrees, and nothing between. :)
Brewin Bruin
Yes, the voltage is in the tens of millivolts at a few hundred C, and the changes with temp are much smaller yet. You can’t measure that with an Arduino ADC, but Adafruit sells thermocouple readout boards which talk to Arduino and other similar products. They do this reading certainly well enough for coffee roasting, with a resolution of about 0.06 degrees C; the accuracy with a type K thermocouple is much worse, about 1-2 deg C typically, but that’s fine for this purpose. This device includes a very high impedance low voltage ADC and the cold junction compensation, all built into one teeny chip on a small board. You can use two of them for two thermocouples, no problem. There’s good info on the Adafruit site. They sell two boards, one for type K only, one for multiple types, for a couple bucks more. I use type T in other contexts, but for coffee roasting, K is a good choice simply because they are the most popular, and plenty adequate. You can even get a platinum RTD board as well, they are all about $15 each, if that floats your boat.

Since the TC4 seems to be unavailable at the moment, this is another option people could pursue, though the Artisan software may not work with it; it would probably require a different sketch. If all you want to do is log data, that’s quite do-able if your mind bends that direction.

The zero current statement was just to point out why wire resistance doesn’t matter, it’s just reading that very small voltage which has always been the trick. You are right, measuring voltage does not require current.
Edited by Brewin Bruin on 01/02/2018 3:03 PM
“Learn by doing” —definitely including doing it wrong.
No, that’s not correct. The length of the wire by itself does not change the reading. It’s the possibility that the wire is short enough to conduct heat from the roaster where the junction is to the readout (the second, cold junction) that is being considered. If the wire is pretty long, say a few meters is hanging out in room temperature air, then the wire is cool where it contacts the readout, the heat it picked up in the RC has been given off to the room.

I can't tell you how many builds I've worked that errant temperature probe heat conduction didn't cause lots of heartburn until I figured out the problem. Glad you addressed it in your post. Many builders aren't aware of the potential temperature read anomoly and don't realize the problem till way after completion.

1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
Brewin Bruin
One other source of erratic readings is something you mentioned (post #150, Brewer to roaster - blower problems, 9/7/2011). Placement of the probe can give wide variation. In the region prior to the perf plate, it might make sense to use a thermocouple probe specifically designed to measuring air, for example:


This might even work in the RC. I’m not advocating any specific brand, but in general these have a tube with lots of holes in it surrounding a bare thermocouple junction. The tube can protect the junction from flying objects (like beans in the RC),but it also serves as a means to average out temperature variations. Your airflow may be not very homogeneous in temp, but it flows through the tube, which will end up smoothing thing out a bit, I suspect. The solid SS sheath more typically found on the $5 thermocouples sold widely don’t do this job quite as well, stainless steel is actually just a so-so conductor of heat. That’s why snotty cookware uses copper or aluminum cladding on the bottom of SS pans. I think I may try rolling my own using some copper or brass tube around a bare TC junction on my build. Both of these are very good heat conductors. You could use gold if you want to get flashy, up to you. 😜
Edited by JackH on 01/06/2018 9:12 AM
“Learn by doing” —definitely including doing it wrong.
Jump to Forum:

Similar Threads

Thread Forum Replies Last Post
Transparent Fluid Bed Design Building a Coffee Roaster 50 01/18/2019 7:28 AM
Modified SR700 300g Fluid-Bed Roaster Building a Coffee Roaster 4 01/15/2019 6:06 AM
3d Printed Fluid Bed Roaster Building a Coffee Roaster 51 01/14/2019 12:29 PM
5+lb Gas Fluid Bed Roaster TC4-> Artisan Issue Building a Coffee Roaster 4 12/30/2018 9:06 PM
sinner7's 5-7kg Fluid-bed in the works... Fluidbed Roaster 38 12/14/2018 10:06 AM
Homeroasters Association Logo, and all Content, Images, and Icons © 2005-2016 Homeroasters Association - Logos are the property of their respective owners.
Powered by PHP-Fusion Copyright © 2019 PHP-Fusion Inc
Released as free software without warranties under GNU Affero GPL v3
Designed with by NetriX