Skywalker roaster... | [292] |
Skywalker, the AL... | [214] |
Skywalker Roasts | [94] |
My first popcorn ... | [47] |
War on Farmers by... | [39] |
Page 1 of 2: 12
|
Automated airflow/blower control
|
|
tamarian |
Posted on 11/25/2013 9:33 AM
|
1/2 Pounder Posts: 320 Joined: January 21, 2012 |
In pursuit of full automation, I've been lately wondering if blower control can be fully automated to maintain minimal airflow to ensure smooth agitation without the need to constantly adjust blower as the beans gets bigger and lighter. It would be similar to the function of a PID, an Arduino/controller will read from a sensor the digital value (X not sure of what, airflow, air pressure, air speed?), and the Arduino device/controller will adjust the blower higher or lower to maintain a "constant rate" of X. What is this X, that can be maintained constant (or reliably calculated) that wil allow the beans to be adequately agitated from green to fully roasted, without overdoing it with too much airflow? Wa'il. 1 Kg PID'ed gas-fired fluid bed roaster, GS/3MPS, K10F
|
|
|
oldgearhead |
Posted on 11/25/2013 10:39 AM
|
1 1/2 Pounder Posts: 1128 Joined: February 10, 2011 |
My DC drive has an analog input, I think it's 0-10V, for PLC or whatever control. It would be pretty easy to do it on my machine. but different: __ beans size __ load weights __ test-weight (% moisture) __ desired profiles ..tells me it's not worth the effort... No oil on my beans...
|
|
|
coffeeroastersclub |
Posted on 11/25/2013 4:19 PM
|
Pounder Posts: 535 Joined: May 26, 2009 |
I wouldn't make it too complicated. I would just find a happy set medium for air flow and just let the PID control your heat adjustments. It really doesn't hurt to have more bean activity towards the end of the roast; it helps loosen stuck on chaff. Len "If this is coffee, please bring me some tea but if this is tea, please bring me some coffee." ~Abraham Lincoln
|
|
|
oldgearhead |
Posted on 11/25/2013 4:32 PM
|
1 1/2 Pounder Posts: 1128 Joined: February 10, 2011 |
Actually I keep the heat the same and adjust the air-flow rate, as follows: T = 0 Temp = 550?F Air-flow = enough to just keep the motion fluid. T = 3 min. - Lower air-flow to slow moving T = 6 min - Again adjust air-flow to slow moving T = 9-10 min (First crack) Increase air-flow just enough to slow the BMT to 1?F/ 5 seconds... T = 12-13 min - Air-flow to max for cooling No oil on my beans...
|
|
|
eric |
Posted on 11/25/2013 5:48 PM
|
Newbie Posts: 17 Joined: July 13, 2012 |
I have tried this - problem is locks in your batch size and then the changing bean volume during a roast makes it difficult. I used a 10g load cell (used in those cheap .01g jewelry scales) connected to a spare input on the TC4 - there is a PTH with 3.3V next to the terminal block and GND at the other end, then the diff output of the load cell connects to the thermocouple inputs. Res is quite good, though the levels are even lower than thermocouples. Put a paddle into the the RC picking up the top of the plume in a spouting bed.... but not worth the trouble really. Now I just use a fan "profile" that is keyed to the bean temp profile, currently simple linear works fine and just winds the fan down as the bean temp rises. Put beans in, adjust fan for correct flow, hit start and no need to touch another control till cooling trigger (if stopping before end of profile) - if you want to roast that way. Eric
|
|
|
allenb |
Posted on 11/25/2013 7:46 PM
|
Administrator Posts: 3858 Joined: February 23, 2010 |
Eric's fan "profile" that is keyed to the bean temp sounds real interesting. I'd like to hear more about how you implement this in software. Another, although crude, method that should give fairly good results when keeping batch size and total roast time fairly close to the same would be to base fan speed off of elapsed time. With my fluidbeds I've found I'm making close to identical fan reduction tweaks at nearly the same points of time during most roasts. This would require using a TC4 with code that would allow applying variable power via phase angle mode to an SSR. You would need to run several roasts to find out what fan power levels to drop to at the usual 4 or 5 stages of reduction and program into the TC4. I have no idea how much of a challenge the code writing would present since I've never done this but I wouldn't think it would be real tough. Interesting topic and discussion! Allen 1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
|
|
|
allenb |
Posted on 11/25/2013 8:01 PM
|
Administrator Posts: 3858 Joined: February 23, 2010 |
After thinking more about my "elapsed time" fan power level scenario I see some issues already. Without being able to set a startup baseline flow as Eric is doing with his bean temperature based fan adjust, you would have issues when going to beans with major differences in density. This needs more work! Allen 1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
|
|
|
Airhan |
Posted on 11/25/2013 8:42 PM
|
1/2 Pounder Posts: 228 Joined: September 29, 2013 |
Perhaps you could use a webcam or microsoft kinect to have a computer automatically judge the amount of bean loft? I do not have have any programming knowledge of how to implement this, but I do know the kinect has pc drivers and development material out. The kinect is also really good at distinguishing objects. (here is a example of face tracking with, surely it could track beans? https://www.youtu...VZ5_TZbldc) Or maybe a photoresistor in the RC? If it is covered by slow moving beans less light hits it and if the beans are flying around then more light will hit the photoresistor. placement would be really tricky... if this is even possible... Edited by Airhan on 11/25/2013 8:51 PM Aaron
"Grind it like it did you some great injustice!"D.L.Clark |
|
|
Airhan |
Posted on 11/25/2013 11:11 PM
|
1/2 Pounder Posts: 228 Joined: September 29, 2013 |
This is the video I meant to link, the other one is just a ball. https://www.youtu...cKTCF0sAfc Aaron
"Grind it like it did you some great injustice!"D.L.Clark |
|
|
coffeeroastersclub |
Posted on 11/26/2013 1:20 PM
|
Pounder Posts: 535 Joined: May 26, 2009 |
Quote allenb wrote: After thinking more about my "elapsed time" fan power level scenario I see some issues already. Without being able to set a startup baseline flow as Eric is doing with his bean temperature based fan adjust, you would have issues when going to beans with major differences in density. This needs more work! Allen Allen, maybe just using a variable rate fan switch and do an initial config with 1 pound increments (or smaller), mark the dial where each works best and stick with that for the whole roast while letting the PID controlled heat source do the rest. Seems like a good balance. Len "If this is coffee, please bring me some tea but if this is tea, please bring me some coffee." ~Abraham Lincoln
|
|
|
oldgearhead |
Posted on 11/26/2013 1:56 PM
|
1 1/2 Pounder Posts: 1128 Joined: February 10, 2011 |
I guess it would not be too hard to accomplish with industrial video, like that used in 'color sorters. Just hold an idea 'spout'. The same video scheme could also control the heat as well as when to stop the roast. The high cost might be a problem. No oil on my beans...
|
|
|
tamarian |
Posted on 11/28/2013 8:58 AM
|
1/2 Pounder Posts: 320 Joined: January 21, 2012 |
Thanks for all the input folks, Seems the straight forward option is to key the blower to the temperature profile, starting from a set point to start the movement. I'm still looking at the fluid dynamics aspect of this, to see if there is a more elegant way to do this. I'm thinking the differential pressure or flow from inlet to exhaust might give clues to how the blower should be adjusted to maintain balanced movement. Wa'il. 1 Kg PID'ed gas-fired fluid bed roaster, GS/3MPS, K10F
|
|
|
alexcampbell |
Posted on 01/08/2014 4:35 PM
|
Newbie Posts: 42 Joined: December 16, 2013 |
I would use a beam breaker type setup that sends out the frequency that it is broken at. They send a digital one or zero to a controller of some sort. Once the frequency that the beam is broken goes above a threshold (to account for chaffe discharge too) you could have the fan speed decrease in some sort of stepwise fashion. I would agree that you could most likely set the fan speed according to temperature if you take a nice slow ramp. If your ramp was too fast, you could have a positive feedback cycle where the temperature increase would cause a fan speed decrease, which would cause a temperature rise etc. You would most likely have to integrate a time constant of some sort. |
|
|
snwcmpr |
Posted on 01/08/2014 5:03 PM
|
1 1/2 Pounder Posts: 925 Joined: March 03, 2011 |
Would the exhaust air flow be a way to register how 'restricted' the RC is? I mean as the beans get lighter then more air flows through. Seems that all you need to do is 'read' the air flow at the exhaust and find the desired flow rate and as it rises, it could be reduced to maintain that desired flow rate. Or not, Ken in NC --------------
Backwoods Roaster "I wish I could taste as well as I wish I could roast." As Abraham Lincoln said "Do not trust everything you read on the internet". |
|
|
JETROASTER |
Posted on 01/08/2014 6:29 PM
|
Administrator Posts: 1780 Joined: March 06, 2010 |
Quote Would the exhaust air flow be a way to register how 'restricted' the RC is? I mean as the beans get lighter then more air flows through. I'll bet it could be useful in some way. The expansion from heating has to be accounted for though. -Scott edit to fix quote only Edited by ginny on 01/08/2014 6:53 PM |
|
|
lmclaren |
Posted on 01/11/2014 12:32 AM
|
1/4 Pounder Posts: 159 Joined: March 20, 2011 |
Hi Tamarian, I am going to implement with the following: http://www.parall...duct/28317 Edited by JackH on 01/11/2014 4:03 AM |
|
|
tamarian |
Posted on 01/11/2014 3:04 AM
|
1/2 Pounder Posts: 320 Joined: January 21, 2012 |
Quote lmclaren wrote: Hi Tamarian, I am going to implement with the following: http://www.parallax.com/product/28317 You're the one person here I had in mind, to be tackling this Edited by tamarian on 01/11/2014 3:16 AM Wa'il. 1 Kg PID'ed gas-fired fluid bed roaster, GS/3MPS, K10F
|
|
|
oldgearhead |
Posted on 01/11/2014 8:05 AM
|
1 1/2 Pounder Posts: 1128 Joined: February 10, 2011 |
Yes, we know the beans get lighter and the volume of air increases as the process progresses. Here is a little more info I discovered, from experimenting, about this subject: If the area of the RC air inlet is 0.3009 sq" and if the area of the RC air outlet is 0.3068 sq", and the ET (environmental temperature) is 32?F, then the blower inlet air temperature will be 150?F at T+10 minutes. ...but... If the area of the RC air inlet is 0.3009 sq" and if the area of the RC air outlet is 0.1964 sq", and the ET remains 32?F then the blower inlet air temperature will be 131?F at T+10 minutes. It would appear that over-restricting the RC's air outlet starts to limit the amount of heat available for re-claiming.
oldgearhead attached the following image:
No oil on my beans...
|
|
|
coffeeroastersclub |
Posted on 01/11/2014 1:05 PM
|
Pounder Posts: 535 Joined: May 26, 2009 |
Quote oldgearhead wrote: Yes, we know the beans get lighter and the volume of air increases as the process progresses. Here is a little more info I discovered, from experimenting, about this subject: If the area of the RC air inlet is 0.3009 sq" and if the area of the RC air outlet is 0.3068 sq", and the ET (environmental temperature) is 32?F, then the blower inlet air temperature will be 150?F at T+10 minutes. ...but... If the area of the RC air inlet is 0.3009 sq" and if the area of the RC air outlet is 0.1964 sq", and the ET remains 32?F then the blower inlet air temperature will be 131?F at T+10 minutes. It would appear that over-restricting the RC's air outlet starts to limit the amount of heat available for re-claiming. Is that a picture of your halloween mask? Len "If this is coffee, please bring me some tea but if this is tea, please bring me some coffee." ~Abraham Lincoln
|
|
|
allenb |
Posted on 01/11/2014 1:56 PM
|
Administrator Posts: 3858 Joined: February 23, 2010 |
Robby?
allenb attached the following image:
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
|
|
|
coffeeroastersclub |
Posted on 01/11/2014 2:06 PM
|
Pounder Posts: 535 Joined: May 26, 2009 |
Quote It was Robby's great grand dad. First it was just a robo head on a table; next generation they attached legs and feet to the head (robby's grand dad). However the concerned public thought he just looked too ridiculous (just a head with legs attached), so they put a torso in and "Presto" robby was born. Its called robolution (similiar to our evolution). Len "If this is coffee, please bring me some tea but if this is tea, please bring me some coffee." ~Abraham Lincoln
|
|
|
lmclaren |
Posted on 01/11/2014 5:38 PM
|
1/4 Pounder Posts: 159 Joined: March 20, 2011 |
thanks Tamarian, The module linked (thanks JackH) is like a line of photo interrupters, it should be easy enough to look for blocked light (floating beans) above clear air, eg: x x x x x x x x x x x where x's are light readings. Probably pick on up soon, working on the auto doaser for my grinder atm using a load cell. best regards Lee |
|
|
coffeeroastersclub |
Posted on 01/12/2014 6:57 PM
|
Pounder Posts: 535 Joined: May 26, 2009 |
Quote Of course there was "Robby" the robot from Forbidden Planet. However we should not forget an equally famous (or infamous if you will) robot from "Lost In Space", simply named "The Robot": Len "If this is coffee, please bring me some tea but if this is tea, please bring me some coffee." ~Abraham Lincoln
|
|
|
coffeeroastersclub |
Posted on 01/12/2014 7:32 PM
|
Pounder Posts: 535 Joined: May 26, 2009 |
I find that The Robot's formal name was, in typical grandiose Lost In Space fashion, "B-9, Class M-3 General Utility Non-Theorizing Environmental Control Robot". WOW! Len "If this is coffee, please bring me some tea but if this is tea, please bring me some coffee." ~Abraham Lincoln
|
|
|
JackH |
Posted on 01/12/2014 7:35 PM
|
Administrator Posts: 1809 Joined: May 10, 2011 |
Len, get some motors on the ears so they spin when it runs. |
|
Jump to Forum: |
Thread | Forum | Replies | Last Post |
---|---|---|---|
Another Propane Gas Control Method | Electric and Gas Heat Sources | 31 | 10/14/2023 6:48 PM |
Air blower vs vaccum cleaner pump | Fluidbed Roaster | 4 | 08/21/2023 10:52 AM |
Renatoa can you walk us through a roast based on Heat power control vs PID? | Roasting Coffee | 32 | 08/21/2023 12:41 AM |
Unable to get Artisan Scope's PID to control heating | Dataloggers/Controllers/Rate of Rise Meters | 4 | 08/01/2023 9:11 AM |
Problem with electric control Quest M6 Coffee Roaster | Quest M3 Roaster | 27 | 06/20/2023 3:39 AM |