Homeroasters - Home Roasting Coffee Community » BUILDING A ROASTER » Fluidbed Roaster
Who is here? 1 guest(s)
Page 4 of 6: 123456
 Print Thread
Fluidbed Roaster project
jessep
Top  #76 Print Post
Posted on 09/27/2020 3:39 AM

Quote

cdrake39 wrote:

Should I wire an inline fuse? Or an on/off rocker switch?


You could stick a circuit breaker in there at the AC side just to make sure nothing happens, or just rely on the house breaker smile

Looks good to me. I also grounded all the metal parts so if the element burns up or a wire gets lose and manages to find it's way out of the element/mica and touch the housing it should blow the breaker before doing too much welding.

I currently rely on the relay to be my on/off switch and leave the AC plugged into the machine. As long as the USB is disconnected, it "should" stay disabled. I feel like a physical switch could add another level of safety and also a physical "kill" switch if the controller get's a bit too aggressive.
 
allenb
Top  #77 Print Post
Posted on 09/27/2020 7:30 AM

Quote


I currently rely on the relay to be my on/off switch and leave the AC plugged into the machine. As long as the USB is disconnected, it "should" stay disabled. I feel like a physical switch could add another level of safety and also a physical "kill" switch if the controller get's a bit too aggressive.


A physical, manual kill switch for the heating element is a must on any roaster where there is any chance of air flow failure regardless of software hi limit controls. Many have had good luck with their electronics keeping the roaster safe but there always seems to be that 1 out of X that destroys the whole works.
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
jkoll42
Top  #78 Print Post
Posted on 09/27/2020 8:58 AM

Quote

allenb wrote:

Quote


I currently rely on the relay to be my on/off switch and leave the AC plugged into the machine. As long as the USB is disconnected, it "should" stay disabled. I feel like a physical switch could add another level of safety and also a physical "kill" switch if the controller get's a bit too aggressive.


A physical, manual kill switch for the heating element is a must on any roaster where there is any chance of air flow failure regardless of software hi limit controls. Many have had good luck with their electronics keeping the roaster safe but there always seems to be that 1 out of X that destroys the whole works.


I'll second Allen on this. If you can run an inline switch to the element only it would keep airflow but the easiest is just to run the entire setup off a power strip that has an on/off switch assuming you are under a standard wall outlet load like 15A in the US. Also with any roast setup keep a fire extinguisher close. It doesn't even have to be a large canister although that's what I keep around for my gas setup.
-Jon
Honey badger 1k, Bunn LPG-2E, Technivorm, Cimbali Max Hybrid, ECM Synchronika w/ Flow Control
 
cdrake39
Top  #79 Print Post
Posted on 09/27/2020 11:20 AM
Greatly appreciate the feedback everyone! I will add in the kill switch (and use a power bar with a switch in the meantime) and keep a small extinguisher on hand just in case.

Would you say there are any issues with how the hot/neutral wires for the heating element are run through the DC PSU? The AC Main wires split when they come in (using a T Tap splice), one end going to the DC PSU and the other end going to the heating element. I assume the DC PSU won't see any loads higher than what the blower motor connected to it is drawing? (ie 200W). The white dots represent where the AC Mains are connected
 
jkoll42
Top  #80 Print Post
Posted on 09/28/2020 6:29 AM

Quote

cdrake39 wrote:

Greatly appreciate the feedback everyone! I will add in the kill switch (and use a power bar with a switch in the meantime) and keep a small extinguisher on hand just in case.

Would you say there are any issues with how the hot/neutral wires for the heating element are run through the DC PSU? The AC Main wires split when they come in (using a T Tap splice), one end going to the DC PSU and the other end going to the heating element. I assume the DC PSU won't see any loads higher than what the blower motor connected to it is drawing? (ie 200W). The white dots represent where the AC Mains are connected


The only thing I see that I really don't like is the T splices. I would be OK with having them temporarily to verify everything is working correctly but inline splices tend to be unreliable and are not efficient. Soldering up a T splice is quite easy

Jump to 6:29 for a T splice solder example

-Jon
Honey badger 1k, Bunn LPG-2E, Technivorm, Cimbali Max Hybrid, ECM Synchronika w/ Flow Control
 
JackH
Top  #81 Print Post
Posted on 09/28/2020 8:51 PM
Nice link jkoll.
I agree, soldering is best as long as you can insulate the junction properly. I use shrink tubing for straight connections as in the video.
We have also used insulating tape that grips itself, almost molds together. I am not sure if it is the same type used for sport grips.
---Jack

KKTO Roaster.
 
cdrake39
Top  #82 Print Post
Posted on 09/28/2020 10:07 PM
Greatly appreciate the feedback! I'll swap those connections out for some soldered ones.
Bit of a busy week ahead so I plan on doing a quick test with a lamp wired in instead of the heating element to verify functionality. Fingers crossed I should be up and running this weekend.
 
cdrake39
Top  #83 Print Post
Posted on 10/05/2020 7:24 PM
Wiring is just about sorted. Verified the operation with a lamp and everything seems to be working as it should. Power to the lamp cuts out when the fan power drops below 30% as specified in the sketch. My only issue right now is the 'Heater Off' button doesn't seem to work. I will drop the value to zero in the slider, but doesn't actually turn the power to the heater off. Any suggestions?

Edit: May have just answered my own question after looking at the image I posted - 'Action' wasn't specified in the dropdown...oops
cdrake39 attached the following image:
heateroffbutton.jpg
 
cdrake39
Top  #84 Print Post
Posted on 10/06/2020 5:13 PM
Quick question regarding wiring of the heating element - are there specific Live and Neutral connection points, or can it be wired either way? The element in question is linked below.

https://www.maste...udes-mica/
 
jkoll42
Top  #85 Print Post
Posted on 10/06/2020 7:23 PM

Quote

cdrake39 wrote:

Quick question regarding wiring of the heating element - are there specific Live and Neutral connection points, or can it be wired either way? The element in question is linked below.

https://www.maste...udes-mica/


Heating elements are a form of resistor so are blind to polarity. They do not care which terminal live/neutral connects to.

Go wild and flip a coin Grin
-Jon
Honey badger 1k, Bunn LPG-2E, Technivorm, Cimbali Max Hybrid, ECM Synchronika w/ Flow Control
 
cdrake39
Top  #86 Print Post
Posted on 10/06/2020 7:56 PM

Quote

jkoll42 wrote:

Quote

cdrake39 wrote:

Quick question regarding wiring of the heating element - are there specific Live and Neutral connection points, or can it be wired either way? The element in question is linked below.

https://www.maste...udes-mica/


Heating elements are a form of resistor so are blind to polarity. They do not care which terminal live/neutral connects to.

Go wild and flip a coin Grin


Good to know! Thanks
 
cdrake39
Top  #87 Print Post
Posted on 10/10/2020 11:26 AM
Does anyone have an idea where to find very thick mica sheet that I could use to make an insert to go over the spot with the green arrow in the photo below? I currently have a threaded rod through where the red arrow is pointing but due to the arrangement of my heating element it's tough to keep it tight there and there is a little side to side movement. I think a slotted mica insert at the top would keep it from moving laterally in that direction. Thanks in advance.
cdrake39 attached the following image:
20201010_102028.jpg
 
jkoll42
Top  #88 Print Post
Posted on 10/10/2020 3:49 PM

Quote

cdrake39 wrote:

Does anyone have an idea where to find very thick mica sheet that I could use to make an insert to go over the spot with the green arrow in the photo below? I currently have a threaded rod through where the red arrow is pointing but due to the arrangement of my heating element it's tough to keep it tight there and there is a little side to side movement. I think a slotted mica insert at the top would keep it from moving laterally in that direction. Thanks in advance.


Yessir! https://www.mcmas... Extremely easy to order from, warehouses on both coasts and central and they charge actual UPS charges and typically ship the same day.

As far as thickness, even the 4 mil (.004") sheets will give you over 2,000V insulation so thickness would only matter if you need it for structure.
-Jon
Honey badger 1k, Bunn LPG-2E, Technivorm, Cimbali Max Hybrid, ECM Synchronika w/ Flow Control
 
cdrake39
Top  #89 Print Post
Posted on 10/10/2020 4:31 PM

Quote

jkoll42 wrote:

Quote

cdrake39 wrote:

Does anyone have an idea where to find very thick mica sheet that I could use to make an insert to go over the spot with the green arrow in the photo below? I currently have a threaded rod through where the red arrow is pointing but due to the arrangement of my heating element it's tough to keep it tight there and there is a little side to side movement. I think a slotted mica insert at the top would keep it from moving laterally in that direction. Thanks in advance.


Yessir! https://www.mcmas... Extremely easy to order from, warehouses on both coasts and central and they charge actual UPS charges and typically ship the same day.

As far as thickness, even the 4 mil (.004") sheets will give you over 2,000V insulation so thickness would only matter if you need it for structure.


Thanks for the feedback! Unfortunately it would need to have some sturdiness to it, so would need to be thick. For now, I've joined multiple pieces together with some Kapton tape (here's to hoping it holds up to the temps).
cdrake39 attached the following image:
20201010_152902.jpg
 
jkoll42
Top  #90 Print Post
Posted on 10/10/2020 7:27 PM
Oooh - didn't realize it was that thicc! That would be some expensive mica.

I bet the tape will hold but if it doesn't you could experiment with making a laminated mica. Possibly thin coatings of JB Weld Extreme Heat (good to 2400F) between the thin mica sheets and vise clamped to dry? Almost like you would do with a vacuum press for fiberglass/carbon fiber. If nothing else it would be cool to see if it works!

Also rigid calcium silicate sheets are good for those temps, don't conduct and aren't real expensive.
Edited by jkoll42 on 10/10/2020 7:32 PM
-Jon
Honey badger 1k, Bunn LPG-2E, Technivorm, Cimbali Max Hybrid, ECM Synchronika w/ Flow Control
 
cdrake39
Top  #91 Print Post
Posted on 10/10/2020 7:35 PM
Thicc indeed! Interesting idea with the JB weld high heat. I'm crossing my fingers this holds up since removing the element is a bit of a pain with my setup...

Noted about the calcium silicate sheets!
 
renatoa
Top  #92 Print Post
Posted on 10/11/2020 2:43 AM
Kapton has several ratings, but the most common is 260 C, too low for inside of the heating element, where temperature exceeds 350 C.
I measured 320-330 C at the exhaust of my hotgun, in reduced power mode. Full power is 600 C, never used.
There are also adhesive alu/copper tapes, but the adhesive is not rated to such temperatures.
I would try to clamp that area with a piece of tin.
 
cdrake39
Top  #93 Print Post
Posted on 10/11/2020 5:55 PM
Thanks renatoa! I'll likely swap that out for something different in the very near future.

So I'm just about ready to roast - any recommendations on last minute wiring checks before I start it up?
cdrake39 attached the following image:
20201011_155117.jpg
 
jkoll42
Top  #94 Print Post
Posted on 10/11/2020 6:32 PM
Other than run a heating/cooling cycle without beans as a shakedown and keep the power kill/extinguisher close..... let 'er rip!

Great looking unit!
-Jon
Honey badger 1k, Bunn LPG-2E, Technivorm, Cimbali Max Hybrid, ECM Synchronika w/ Flow Control
 
cdrake39
Top  #95 Print Post
Posted on 10/12/2020 12:33 AM
The first roast was okay, but man was the profile tracking a mess! I ended up removing the cyclone chaff collector and swapping it for a mesh collector that sat on top of the roast chamber and created less back pressure. The beans themselves seem fairly uniform, dropped them around the 14 min mark (despite what the plot says). It seemed as though the BT thermocouple flatlined around the 179C mark. The heater was running at max capacity for the majority of the roast. Perhaps it's just an issue with the thermocouple giving a false reading and the heater can't keep up? Any feedback is much appreciated!!
cdrake39 attached the following images:
20201011_232702.jpg 20201011_232638.jpg 20201011_233108.jpg
 
renatoa
Top  #96 Print Post
Posted on 10/12/2020 3:43 AM
Driving a roast based on feedback from BT sensor is a very questionable task, the beans-air mix is too turbulent to have a reliable reading, even with huge averaging.
After some study I developed a different control approach, based mainly on desired setpoint, giving a significantly less weight to error between measurement and setpoint.
Named this new approach P_on_S, proportional on setpoint.
Combined with a new integrative term computation, that take into account the RoR trending (2nd derivative of temperature), to compute the integrative sum, this approach give very good results (for me) in a popper based sampler.
If you are interested to test this approach on your machine, I can share the code changes with you.
The above are valid if using a TC4 ecosystem, or other software solution where you have access to PID sources.
 
renatoa
Top  #97 Print Post
Posted on 10/12/2020 9:41 AM
An alternative to the above approach... you could consider measuring the exhaust temps and profile based on this parameter instead.
Rob Hoos wrote an article on this subject, related to Ikawa, but it is valid for any FB:
https://www.ikawa...mperature/
 
cdrake39
Top  #98 Print Post
Posted on 10/12/2020 10:14 AM

Quote

renatoa wrote:

An alternative to the above approach... you could consider measuring the exhaust temps and profile based on this parameter instead.
Rob Hoos wrote an article on this subject, related to Ikawa, but it is valid for any FB:
https://www.ikawa...mperature/


Thanks! That just reminded me that another user mentioned he uses ET for PID tracking as well. I'll play around with that today. If unsuccessful, I'll let you know if I want to go about modifying the code as you mentioned.
Just happy to be roasting at this point. Looking forward to the experimentation!
 
cdrake39
Top  #99 Print Post
Posted on 10/12/2020 8:36 PM
I've put together a new profile to control via ET rather than BT. But I'm noticing an issue now - when I plug in the main power cord, my BT thermocouple reading is all over the place (shows temps as high as 200C without the heat even turned on). When I unplug it the power cord, the reading goes back to normal. I tried swapping the wires on the board but it didn't fix the problem. Any suggestions as to why this might happen? There's a definite correlation with when the AC cord is plugged in..
 
Coffee_runner
Top  #100 Print Post
Posted on 10/12/2020 10:42 PM
Looking good. Trying a new heating element. And maybe a new fan. Don?t think I had enough air flow through my heating element.
 
Page 4 of 6: 123456
Jump to Forum: