topbanner.gif
Login
Username

Password




Not a member yet?
Click here to register.

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

JackH
12/14/2019 4:07 AM
I read somewhere that higher elevations have faster roast time and require somewhat lower temperatures. air is thin and has less Oxygen

allenb
12/14/2019 12:44 AM
Yes, 1st off, you must use only Panama Esmeralda Geisha beans and be sure to only roast on Saturdays. Actually, this isn't completely true. Please post in all about roasters forum. Thx!

wjohndon4566
12/13/2019 10:36 AM
I’m at 9,000 feet elevation, is there any special adjustments that need to be made to roast using an SR 540 at this elevation?

snwcmpr
12/07/2019 9:29 AM
roar

snwcmpr
11/27/2019 11:44 AM
greenman

Users Online
Guests Online: 2

Members Online: 0

Total Members: 6,297
Newest Member: jmartin2119
In Memory Of Ginny
Donations

Latest Donations
Jared Brunette - 5.00
Daniel Boozer - 5.00
renatoa - 2.00
JitterzZ - 2.01
renatoa - 2.00

View Thread

Who is here? 1 guest(s)
 Print Thread
How to control a fluidbed roast profile?
tamarian
Thanks Allen. The PXG4 just arrived. Since I knew it has nothing included, no manual or cables, I ordered a serial db9/audio cable from eBay, but the audio plug turned out to be too big. Will look for an adapter, but in the mean time will go through the PDF files and see how this thing runs.
 
http://english.varietalcafe.com
tamarian
The db9-audio cable from eBay did not work after adding a 2.5-3.5 audio plug adapter, so I ordered the Fuji cable and tried seting it up manually. The PXG4 I got has output1 as SSR drive, and seem loaded with other options I don't need, output 2 voltage, two digital inputs and alarm output.

First thing was resetting to factory default, and setting it up for heating (rv/reverese). The voltage for output one from settings only allows 0-10 or 2-10VDC or current, while SSR requires 4-32 vdc (even by the Fuji specs PDF), but can't find it in channel 6 settings, so I selected 2-10 vdc. Ran a test on halogen lamp and the thing goes on/off quite fast and goes well beyond my manual SV, so something isn't right with the setup. Same halogen test works fine on my other PID (Autonics), so I'm guessing it has to do with the output 1 SSR setup? Anyone encountered that?
 
http://english.varietalcafe.com
allenb
tamarian wrote:

The db9-audio cable from eBay did not work after adding a 2.5-3.5 audio plug adapter, so I ordered the Fuji cable and tried seting it up manually. The PXG4 I got has output1 as SSR drive, and seem loaded with other options I don't need, output 2 voltage, two digital inputs and alarm output.

First thing was resetting to factory default, and setting it up for heating (rv/reverese). The voltage for output one from settings only allows 0-10 or 2-10VDC or current, while SSR requires 4-32 vdc (even by the Fuji specs PDF), but can't find it in channel 6 settings, so I selected 2-10 vdc. Ran a test on halogen lamp and the thing goes on/off quite fast and goes well beyond my manual SV, so something isn't right with the setup. Same halogen test works fine on my other PID (Autonics), so I'm guessing it has to do with the output 1 SSR setup? Anyone encountered that?


Give these guys a call: http://www.instrumart.com/

They talked me through my setup and are very knowledgeable with the Fuji line.

I didn't have any issues with the SSR outputs on my Fuji.

Allen
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
Dan
Using the 2-12VDC output to drive a 4-32VDC SSR is common practice.

Did you autotune the PID after resetting to factory defaults? That would explain the overshoot.

What is your Cycle setting? If set very fast it will cycle the SSR quickly. Normally, this is a good thing, especially if you are using bare nichrome heaters.
 
http://www.intactamerica.org
allenb
tamarian wrote:

The db9-audio cable from eBay did not work after adding a 2.5-3.5 audio plug adapter, so I ordered the Fuji cable and tried seting it up manually. The PXG4 I got has output1 as SSR drive, and seem loaded with other options I don't need, output 2 voltage, two digital inputs and alarm output.

First thing was resetting to factory default, and setting it up for heating (rv/reverese). The voltage for output one from settings only allows 0-10 or 2-10VDC or current, while SSR requires 4-32 vdc (even by the Fuji specs PDF), but can't find it in channel 6 settings, so I selected 2-10 vdc. Ran a test on halogen lamp and the thing goes on/off quite fast and goes well beyond my manual SV, so something isn't right with the setup. Same halogen test works fine on my other PID (Autonics), so I'm guessing it has to do with the output 1 SSR setup? Anyone encountered that?


Shoot me the PXG4--------- number from the label. The 2-10 vdc and current output should be voltage or current proportional and not time proportional according to the manual but we need to know the model # designators to be sure.

Allen
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
tamarian
Sure, it's PXG4CPT1-FVYG4, MFD 2009-04, NO 9420969 TC

Dan, attempted auto-tuning and it overshoots. It turned out I have the connection to SSR reveresed. Minus to plus and plus to minus. Setting them correctly results in no SSR signals. Even though mv shows valuse going out, but no reaction on the SSR ot halogen lamp.

Also, channel 2 settings are masking the setting for tc1 to set the proportional time, even after rest, it's masked, somehow....

Still digging in to see what's going on, and I appreciate all the help :)
 
http://english.varietalcafe.com
allenb
Sure, it's PXG4CPT1-FVYG4, MFD 2009-04, NO 9420969 TC

According to the manual you've got Output 1 is setup for SSR 4-32 VDC. This is a cycling variable time base output to your SSR, not a variable voltage output.

Output 2 is setup for proportional DC voltage output (0-10).

I wouldn't be trying to power up a halogen bulb with the SSR output from your Fuji. Halogens, if my memory serves me correctly, have an inrush current that the controller may not like. Have you tried cycling an SSR with it?

Allen
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
allenb
Go to pages 134 to 140 for info on masking/unmasking. Takes some reading and playing around to figure it out. The option to just unmask everything is an easier route but clutters up the list.

http://www.instru...Manual.pdf

Allen
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
tamarian
allenb wrote:
Have you tried cycling an SSR with it?


Oh, yes, I've only been using the SSR output to an SSR and the SSR to a halogen lamp. It works great with my Autonic single SV pid, but not with Fuji.

Thanks for the unmask all tip, will check it just to see TC1 setting,.
 
http://english.varietalcafe.com
tamarian
I think it's working, just overshoots a bit. After unmasking TC1, it was set at zero. Tried setting it to 1 second, it won't accept it, and returns back to zero. Not sure why, but went ahead with auto-tune and set sv to 80, it goes to 83 and backs down to 81-80-81. The halogen never goes off, just blinks faster. My other PID will shutdown for a second or so, while Fuji seems to work on split seconds.

I'll take that and try to construct a reasonable profile. What is the best function to choose? PID, Fuzy, self or PID2?

Any profile pointers? I'm guessing a preheating segment to 150, drying ramp to 140 in 2 minutes, , ramp to 150 in 3 minutes, ramp to 205 in 4 minutes and final ramp to 225 in 4 . minutes.. 5 segments in total.
 
http://english.varietalcafe.com
allenb
tamarian wrote:

I think it's working, just overshoots a bit. After unmasking TC1, it was set at zero. Tried setting it to 1 second, it won't accept it, and returns back to zero. Not sure why, but went ahead with auto-tune and set sv to 80, it goes to 83 and backs down to 81-80-81. The halogen never goes off, just blinks faster. My other PID will shutdown for a second or so, while Fuji seems to work on split seconds.

I'll take that and try to construct a reasonable profile. What is the best function to choose? PID, Fuzy, self or PID2?

Any profile pointers? I'm guessing a preheating segment to 150, drying ramp to 140 in 2 minutes, , ramp to 150 in 3 minutes, ramp to 205 in 4 minutes and final ramp to 225 in 4 . minutes.. 5 segments in total.


I've tried fuzzy and straight PID and didn't notice any difference so I've always kept it on PID.
When you do the autotune you'll have to select a setpoint (SV). I've found setting it to a temp just prior to 1C by a few degrees gives the most stable results. When autotuning, it will go full throttle up to SV then drop back some, then over shoot a bit then back then end the autotune session.

For profile pointers,

It's good to go fast to 120 C (2 min) then slower to 150. Try and end up at 150 by the 5 min mark. Then pick a ramp time/temp that will get you to 1C in another 5 or 6 minutes. Program another ramp segment that takes you to your finish temp so you achieve somewhere around a 3.9 to 5 C degree/min rate of rise. Make sure your last segment temp setting is well above your finish temp or it may end the roast before you're ready. I use a small micro switch between the controllers SSR output and the SSR so I can kill the roast when I'm ready. Obviously you'll need to play with different profiles to find the sweet spots.

To keep from going too slow from room temp to the half way point of the first segment (120C), I ended up just programming this segment with the temperature setting but 0 for time. This allows the roaster to go full power getting to 120 quickly but it backs off and takes control getting to the setpoint of the 2nd ramp segment.

I've found the best way to program the ramp segments is to first determine what bean temp rate of rise I want during the different phases of the roast and then translate this into time/temp segments to achieve it.

Allen
Edited by allenb on 07/12/2012 8:23 AM
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
tamarian
Thanks Allen, good stuff!

For auto-tune, I'm guessing it needs sacrificial green beans? Auto-tuning with beans differs from auto=tuning an empty roaster.

Great tip on the zero time initial segment!
 
http://english.varietalcafe.com
Dan
Autotuning always results overshooting a few times until it learns how not to. I've always autotuned with an empty roaster, or empty oven at work, then left fuzzy logic turned on so it would keep learning from thereon.

I believe there are some settings that limit overshooting the SV. What it does is stops short and then creeps up to the value. But this is more valuable in a process oven than a roaster. Hovering a few degrees around the SV is plenty good enough for a roaster. BTW my Love fuzzy logic PIDs at work keep a restaurant-sized 15kW oven at SV or one degree below, that's ±0.5° deadband, and that's using mechanical relays on a 40 second cycle.
 
http://www.intactamerica.org
allenb
Dan, great info on the autotuning functions! I didn't know that Fuzzy was continuously updating the PID, sounds like a great feature.

As far as autotuning with or without beans. I've always found tighter control when I autotune with beans versus without. This allows the controller to see actual conditions. I realize that the PID settings cannot be good for all phases of the roast since the thermal conditions change constantly throughout the roast.

I've tried autotuning during the drying stage, browining stage and post 1C and the time just prior to 1C seems to be the most representative of the whole.

I made a mistake in my previous post suggesting a 7 to 9 degree rate of rise after 1C. This is in degrees F, not C, so you'll have to translate this to C to be valid.

Allen
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
allenb
Remember that the PXG series has the option of minutes/seconds for ramp/soak time units which allows much improved profile design. Go to section 6-4 in the instruction manual under "ramp soak time units" and change it to minutes/seconds instead of the default hours/minutes.

With minutes/seconds you can break the first segment (ambient to 120 C) into 2 or 3 parts to taylor the slope instead of going full power which might be too fast to 120C depending on how many KW you have available.

If you end up with too much KW, you can limit the output % (secton 6-2).
In additon, I had to set an output lower limit to keep ET from dropping too low when the PID over corrected during some phases of the roast. I think I set it to around 25%.

Allen
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
tamarian
Thanks Allen and Dan, you made the learning curve much easier :)
Wa'il. 1 Kg PID'ed gas-fired fluid bed roaster, GS/3MPS, K10F
 
http://english.varietalcafe.com
tamarian
Completed auto-tuning on the roaster on the Lo setting and set to Fuzzy, entered my preferred profile for Tchembe in 4 segments (ramp to 120 in 2m, ramp to 155 in 3m, ramp to 205 in 4m, ramp to 220 in 3m, soak for 1m) soak is just to do the drop routine.

It worked almost perfectly. Only surprise was themocouple seem to differ by 10 degrees than my plug-type thermcouple. Usually 1st crack at 205c, it's now at 196c. I'm gussing it due to being shorter than my plug-type thermocouple, So the roast was darker than planned, but looked, smelled and tasted (by chewing) great.

Pics: (The temps on artisan graph are not from the PID, since I still don't have a loader cable, so used my Extec to datalog.
tamarian attached the following images:
tchembe-pid.png 20120713_233323.jpg

Wa'il. 1 Kg PID'ed gas-fired fluid bed roaster, GS/3MPS, K10F
 
http://english.varietalcafe.com
tamarian
Just added an exit chute, from the left side, above the sloped plate, where the beans are in downward motion. Works great. The 90% of the beans exit right away, then some popcorning action, turn down air flow and they are out in 10-20 seconds.

forum.homeroasters.org/forum/attachments/20130409_122804s.jpg

tamarian attached the following image:
20130409_122804s.jpg

Wa'il. 1 Kg PID'ed gas-fired fluid bed roaster, GS/3MPS, K10F
 
http://english.varietalcafe.com
allenb
Wow, very nice exit door and chute! That must have taken some precision curve bending to get a good seal.

I like this a lot being able to dump in literally a few seconds and begin fast cooling.

I haven't seen many examples of exit doors/chutes on fluidbeds.

Allen
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
tamarian
Thanks Allen. Not much precision. I used a 3 inch hole saw bit, it eats a lot of metal from the edges, so the gap was about 4 mm all around, between the cut round piece and the edges cut on the RC. But I installed a cut rectangular piece of the pipe (left over from cutting the section for the glass) behind the cut circle, and padded it with a piece of cooking silicon rubber. The weight of the tamper was enough to push the door in enough to seal the door. Will add detailed photos later, along with a new charge hopper/chute I'm almost done with.

Next will work on

1. designing an air-tight trier, to draw in about 15g samples for cupping various stages of the roast when sample roasting.

2, Find a spot for BT probe that works for different charges, form 150g or 200g to 1kg
Wa'il. 1 Kg PID'ed gas-fired fluid bed roaster, GS/3MPS, K10F
 
http://english.varietalcafe.com
Jump to Forum:

Similar Threads

Thread Forum Replies Last Post
Breville Dose Control Pro GRINDERS 23 12/03/2019 11:07 PM
Gas Control advise needed Drum Roasters 5 11/26/2019 12:28 AM
Fluidbed Roaster project Fluidbed Roaster 12 11/07/2019 5:51 AM
Automated propane gas control? Electric and Gas Heat Sources 5 11/03/2019 10:53 AM
Propane Flow meter for fluidbed roaster Building a Coffee Roaster 17 10/21/2019 3:39 PM
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