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.

CharcoalRoaster
11/04/2019 1:58 AM
+1 snwcmpr

snwcmpr
11/03/2019 2:16 AM
Can we make the shoutbox UNAVAILABLE until a member has a certain number of posts?

allenb
11/01/2019 2:20 AM
Funopt, please post in the gas and electric heat sources forum

Funopt
10/30/2019 5:17 AM
Can someone help me for using forced propane burner as my heating element. I rather want to use lpg than electric. Do you think it would work

snwcmpr
10/22/2019 5:31 AM
Thanks to you all....... I was not sleeping ... I stayed awake worried about it all. :)

Users Online
Guests Online: 4

Members Online: 0

Total Members: 6,246
Newest Member: cabbar
In Memory Of Ginny
Donations

Latest Donations
renatoa - 2.00
JitterzZ - 2.01
renatoa - 2.00
allenb - 25.00
snwcmpr - 10.00

View Thread

Who is here? 1 guest(s)
 Print Thread
Bake-a-Round build with heat recovery
Brewin_Bruin
After doing heat gun roasting for 6 years and 80 lb of Sweet Maria’s finest, I decided it was time for a fluidized bed roaster. This build owes a great deal to oldgearhead’s work. Here’s what’s similar:
Ametek Lamb with DC speed controller (KBIC-125, slightly higher current model than used by OGH). The blower never runs much over half speed, the DC controller is super.
single 1700 W heat gun element (Master HAS-043K, 120 V)
Bake-a-round RC--What a deal!
hot air recycling
Differences:
I sandwiched the heat gun element between two soft firebricks (the type made for kiln construction, not the fireplace variety). It was easy to cut a channel in the brick, and it’s an excellent insulator for electric current and heat. I still used the mica insulation that comes from Master, but it’s not strictly necessary. (picture 1, 1a). The bricks are mortared together as the next pictures show, and the recess for the 3.5” SS tube is cut with a hole saw (top of picture 1, picture 3b)
I used an Ametek Lamb two-stage 2-stage blower (97 CFM, 8.3Amp, 120 V, [1000W electrical] model 116392) vacuum blower with a tangential bypass. I like to keep the motor cooled by ambient air, and the tangential bypass makes it easy to make connections to the heater assembly. (picture #2).
No cocktail shaker. I connected the Bake-a-Round RC to the heater assembly using 3.5 in stainless steel tubing (0.065” /16 ga wall thickness) with a stainless V-flange to allow very easy disassembly for dumping coffee and cleaning. (picture 3, 3b) The Bake-a-Round is affixed using a flexible layer of high temp silicone sealant (Permatex 81878 Ultra Copper) nearly 1 mm thick. This compensates for the big difference in thermal expansion between the stainless steel and the Pyrex, the glass would crack if it was tightly sealed. The thermocouple positions are visible here, the one in the RC (BT) is about 2” above the perf plate, the ET is just below the perf plate. I use the same 19 hole perf plate pattern as OGH. Nothing is machined or welded, the 20 ga. SS perf plate is silver brazed to the SS tube, and even that isn’t really necessary, you could just use the silicone sealant (soldering is a pain, the thermal distortion works against you). The flanges are a tight fit over the SS tube and stay put (I silver brazed the lower one, again, not really needed.)
The heated air recycling is done using 4” metal ductwork form the local home supply, I don’t like to heat PVC routinely, especially around food (picture 4). The chaff collection is easy with the wye-connection to return the air to the vacuum intake. The straight arm dead ends, and the chaff collects there (an internal screen keeps most of it from the intake box) above a gate valve from a shop dust collection system; just open that when done and the chaff dumps out. The air return uses a standard shop vac filter element to prevent chaff from being pulled back into the heater (pictures 4a-4c). I plan to insulate the duct with some old quilt or bath towel at some point. The whole recirculation duct just lifts right off, which I do when I’m cooling the beans (picture 5). The box is just 3/8” scrap plywood, there’s an opening on the side not visible in picture 4 for the motor cooling flow. Noise is quite tolerable, the firebrick insulation means the wood doesn’t get heated notably.
The system is working nicely with 350 g loads at about 1500 W on the heater (measured with clamp-on ammeter), hitting FC around 9-10 minutes. I have separate cords on the heater and blower, but the combined current is less than 20 A at 120 V, so I could run them off the same circuit (I don’t).
I’m beginning to grasp how to control my profile better. My setup doesn’t recycle heat as efficiently as OGH’s, but it suits my needs, I prefer somewhat smaller batches. Insulation on the hot air return may lower the heater power somewhat. I will add in an Arduino with three thermocouple mini-boards (Adafruit MAX 31855, the third one is to monitor recycled air) with an LCD screen and a datalogging board when I next have the time to build, replacing my current thermocouple readout. Eventually I may let the Arduino try to control the fan speed, but I’m not in a rush there, I am my own PID for now.
Final cost US $500-600, including tax & shipping.
Edited by Brewin_Bruin on 03/10/2018 3:14 PM
 
Brewin_Bruin
These are 1, 1a & 2, showing details of heater and blower assembly.
Brewin_Bruin attached the following images:
picture_1_2.gif picture_1a_2.gif picture_2_2.gif

Edited by Brewin_Bruin on 03/06/2018 8:31 AM
 
Brewin_Bruin
OK, here is the RC #3
Brewin_Bruin attached the following image:
picture_3_5.gif

Edited by Brewin_Bruin on 03/06/2018 8:37 AM
 
Brewin_Bruin
Overall view (#4a & 4) The heat recycling tubing is visible here (on right) and the shop vacuum cleaner filter element is shown on left.
Brewin_Bruin attached the following images:
picture_4a.gif picture_4.gif

Edited by Brewin_Bruin on 03/06/2018 3:09 AM
 
Brewin_Bruin
pictures 4c & 5 details. The heat recycling tubing simply lifts right off.
Brewin_Bruin attached the following images:
picture_4c_1.gif picture_5_1.gif

Edited by Brewin_Bruin on 03/06/2018 3:09 AM
 
specialreserve
Brewin_Bruin wrote:

These are 1, 1a & 2, showing details of heater and blower assembly.


very nice ...how did you make the hole in the fire brick?
 
Brewin_Bruin
This sort of firebrick is extremely soft, ordinary woodworking tools cut it easily. In order to make a pilot hole for a spade bit to follow, I first scored a shallow cut down the center of each brick on a tablesaw; a number of simpler tools such as a hacksaw could do the trick.
The two bricks are clamped together, with the scored channel in the middle, this guides the drilling. On one side I used a hole saw to cut a channel for the 3.5" stainless tube to rest in, guided by the center channel. The hole saw cut doesn't need to be perfect, you can slide the tube into the groove and wiggle it back and forth to seat it nicely. (The SS tube gets sealed into place when mounting the RC using the same high temp silicone sealant). Then I used a spade bit on a drill press for the center hole, and flipped the bricks over to drill from both sides. The pilot channel down the middle helps ensure the bored holes meet nicely, it keeps the point of the spade bit from wandering. On the opposite end from the SS tube, I rebored the center hole a bit larger for a short distance to fit the copper adapter that connects it to the blower, it just happened to be the perfect size. You can see the profiles in picture 1. After fitting, firebrick mortar is used to join the bricks permanently.
The firebrick feels comfortably warm to a bare hand with the heater on, and I don't have to worry about shorting the electrical connections. The insulated wire enters from the cool end, and has not been bothered.
 
Brewin_Bruin
Having roasted a number of batches with this unit, the performance is meeting all my hopes. Today I did a 450 g batch of Nicaraguan with 1550 W on the heater (Sweet Maria's site suggests that's probably wet processed, but I'm unclear). That means I with my 1700 W heater I can readily do 500-550g of beans, about the limit for the Bake-a-Round RC. The roast profile is attached, first crack was at about 9 minutes, 2nd crack at about 14, I let that stretch a little for development. Still experimenting but I'm enjoying what I'm producing and so are my friends (not an objective criterion, but ultimately what matters most, I think).
The unit behaves very much as oldgearhead describes: it IS the PID. I played with the air very minimally on the roast below, not touching it at all until 9 minutes, and then very small tweaks. The heater power was just left constant. I'm getting tuned to how much power is needed for different batch sizes, basically 50 g change in green beans is about a 25 W change on the heater, over the 300-450 g range.
Preheating to 140 C seems to be doing the job nicely: set the power at the estimated value (using a clamp-on ammeter) for the bean load, and then use the air to make the temp stabilize in the RC at 140 C (takes just a minute or two) ; this is slightly less than the amount of air to get the bean moving properly once loaded, but it in the ballpark. Once the beans are moving then the profile turns out pretty much as desired with minimal intervention.
Chaff collection using the wye junction at the bottom of the air return as a "dead end" (assisted by a screen in the wye) works really well. There is no discernible chaff in the beans themselves, and it's simple to empty because there's a gate valve right there.
Bean cooling is extremely easy in the RC, and removing the heat return duct (it just lifts off) speeds this a great deal, note the graph. No need to play with the air flow, it speeds up modestly with the return removed, and no beans get launched into orbit. No need for a dump mechanism. I can easily remove the RC because of the V-flange connection to empty out the beans. I may implement a J-tube as OGH has shown in his "Brewer to Roaster" thread to pull the beans straight into a container. The vacuum intake on the unit itself will do the job, no need to hook up the shop vac also.
I don't see roast defects now: no under-roasted quakers, no tipping, no divots (got some tipping when I preheated too high, got divots by cooling too aggressively). Terrific uniformity.
Multiple profiles recorded this month also posted for comparison.
Brewin_Bruin attached the following images:
450_g_roast_profile.gif roast_profiles_mar_2018.jpg

Edited by Brewin_Bruin on 03/16/2018 1:50 PM
 
greencardigan
Nice work. Any videos of it running? I'm interested in seeing the bean circulation with the flat base roast chamber.
 
Brewin_Bruin
I'll be putting a video together soon. The skinny on bean circulation is that it works fine, the uniformity is the strongest indicator of that. The beans at the very outer edge of the perf plate move more slowly, but they keep moving, because the center is constantly forming a fountain which subsides down all sides evenly. The 19 hole plate is shown in the attachment (I pasted this drawing onto the plate to drill it), the outer ring is the edge of the RC. Kudos to oldgearhead, this is his design, I hope he's smiling where he is now and roasting something delicious. The BT thermocouple is located in the outer, blank area of the perf plate, about 3/4 inches from the edge, in the descending bean column.
I'm running the blowier at maybe 40% or less power, so getting motion is no problem. When I tweak it, I'm turning down the air and just keeping the fountain in a steady "simmer" as the beans drop in density.
Brewin_Bruin attached the following image:
perf_plate.jpg

Edited by Brewin_Bruin on 03/19/2018 6:19 AM
 
greencardigan
I was worried there might be a dead spot in the circulation on the base at the corners. However, with the glass design you can see the base there.
 
Brewin_Bruin
So far this thing has been performing really nicely. I did a nominal 1 lb/475 g load of Tanzanian from Sweet Maria's the other day, at 1650 W it reached first crack at about just over 8 min (I had preheated it thoroughly to 150 C before starting). That's about as big as the Bake-a-Round can manage without blowing beans out the top at the end of the roast. So, this unit clearly has sufficient power. The fine tuning is determining how much power to use for a given load size, which connects to whether I want to stretch out the development phases, and depends on the bean origins to some extent.

The paper vacuum filter just before the vacuum intake has turned a toasty brown in places but isn't charred, so I'm thinking that's okay. Using thin metal duct for recirculation means more heat loss than PVC such as used by OGH, but I'm not at all keen on the subtances that hot PVC can release, and there's enough heat to get the job done. The heat recirculation tubing also traps the chaff extremely well just above the gate valve on the "Y" connector, very easy to dump it when done.
Following up on greencardigan's comment, I installed a tiny "funnel" using beveled silicone rubber strips glued with the same high temp silicone rubber adhesive mentioned above around the bottom edge of the RC (photos to follow). This has visibly helped the bean recirculation, they don't slow down when they reach the bottom plate now. The only slow spot is right next to the thermocouple, it impedes flow a little but nothing serious, the roast uniformity is very good.
Edited by Brewin_Bruin on 05/22/2018 11:14 AM
 
AMRoberts
Brewin_Bruin wrote:

... Preheating to 140 C seems to be doing the job nicely: set the power at the estimated value (using a clamp-on ammeter) for the bean load,
...


Great roaster. Physically bigger than I can manage to store, but I'm trying to learn things as I think about where to go beyond my modded popcorn popper (my batch size is just too small).

Is your preheat done for verification that your power estimate is correct for current conditions (ambient temp)? My understanding is (or was) that fluid bed roasting transferred energy to the beans via convection from the hot air, so there wasn't any need for preheating.

Cheers,
Alan
 
Brewin_Bruin
You are correct, the primary heat transfer is via convection. However, there are always other things going on. Simply preheating the heating element means the air flow is up above 150C from the beginning of the roast. The heat transfer rate depends on the temperature difference between the beans and the air. Especially in a recirculating unit like this one where you're using less heater power, that helps the drying portion (BT below 150C) take off. I found that preheating too high (even 160 C) can lead to tipping, notable charring on some beans, but a 140-150 C preheat means that by the first minute the beans are nearly at 100 C, and the roasting profiles are quite similar to stuff I've been looking at (charts on right, still trying to learn how to judge these things, but it's useful to start with what others have found to be good).
So is it necessary? Probably not. Neither is roasting your own beans necessary, it's a matter of what's been working well for you.

Also, I enclose the pictures of the silicone rubber used for the "tiny funnel" modification to help bean recirculation. This rubber is rated for high temp, and hasn't suffered at all in the RC. However, due to its flexibility, it's hard to cut a nice beveled section as required. I ended up sandwiching the rubber between plywood to make it hold still on the bandsaw (photo on left). You can hand cut that, if you have the skill and patience.
Brewin_Bruin attached the following images:
silicone-rubber-cut.jpg silicone-rubber-funnel.jpg profile_3.jpg 4_22.jpg

Edited by Brewin_Bruin on 05/23/2018 5:47 AM
 
Brewin_Bruin
I did in fact build the Arduino readout/datalogging unit. I created a new thread under "Dataloggers-Controller-Rate of Rise Meters" called "Simple Arduino-based datalogger" which gives the details. I only used two thermocouples, and really only pay attention to the BT; the ET isn't very interesting in my current grasp of things. Being able to monitor RoR (rate of rise) in real time, however, is VERY helpful. I use it primarily to tweak the heater power, helping me to pin down the right power for various size bean loads. The air needs minimal tweaking during the roast, and I don't feel a compelling need to have the Arduino try to control that, it's not likely to be very helpful at this juncture.
Edited by Brewin_Bruin on 05/29/2018 12:20 PM
 
Jump to Forum:

Similar Threads

Thread Forum Replies Last Post
Fluid Bed Roaster Build - bye bye Heat Gun and Flour Sifter Fluidbed Roaster 55 11/13/2019 3:05 PM
Pyrex Bake A Round Tube Mounting Help Fluidbed Roaster 7 11/06/2019 3:09 AM
New 1kg drum roaster build Drum Roasters 8 10/10/2019 9:53 AM
first DIY-Build 0,5-1kg WHAT SHOULD I BUY OR MAKE 12 10/01/2019 8:57 PM
Bake A Round Alternatives Fluidbed Roaster 11 09/30/2019 9:54 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