Who is here? 1 guest(s)
 Print Thread
5 Lb Roaster Perf Plate Design Help!
pisanoal
Super useful information! Thanks for sharing that. I have been looking at datasheets for blowers and comparing, but i haven't really had a good idea as to what i was shooting for. All i really knew was the two blowers i was using in parallel were working great. Now i just need to decide if I want to spend $200 on this blower vs 60ish to add a second blower. With your flow chart though and the info on 20" WC required pressure, I can look for some alternatives.
 
allenb
On deciding on 2 versus 1 larger blower, it's not uncommon for medium sized fluidbeds to have two blowers so don't feel that if you design around that format that you are cutting corners or being unconventional.
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
CoffeeInterest

Quote

pisanoal wrote:

Quote

CoffeeInterest wrote:

Quote

pisanoal wrote:

Quote

CoffeeInterest wrote:

Perf plate does matter, but less so with an asymmetrical chamber. You have enough clearance to accommodate the batch size you want.

You need the right blower as well. My experience has been that the Ametek 122170-18 is the best you can use.

Additionally, you probably have too much shielding around your burner to prevent a good consistency of air and flame. Brush flame torches like the Gods BP?s, the Seiverts, Exacts are absolutely perfect for this. They won?t blow out.



Really sorry, I completely missed this post. I really appreciate the feedback.

With your blower recommendation, are you suggesting I'm lacking in CFM or pressure? Just looking to compare my blower to the one you recommended.

I do have a good bit of shielding around my burner, but I'm not sure id say its too much preventing the consistency you are talking about. I could see having trouble maintaining heat that way? But I have plenty of heat getting to my air stream, and it does not seem to fluctuate wildly.


It?s all about pressure when you get into the bigger loads. CFM is related, but it?s more on how well the loft motor can handle 6-10# or bouncing beans on its air bed. My first motor struggled a lot at 6lbs. The Ametek I shared easily lofts more than 22# of green.


I tried my new angled plate design and it helped a little but was still having issues with the temperature. I also had to turn the air back up at the end of the roast to keep beans moving which i thought was really odd. I hypothesized that maybe the dlow falloff at the pressure required to loft the beans was significant enough to not be able to deliver enough BTUs.

I tried two blowers tonight and finally had a successful roast! I was able to keep my ROR where i wanted and the ET stayed under 500 degrees. I was stuck on having enough pressure because that seems to be what people think is most important, but the pressure vs flow curve definitely comes into play when pushing the limits of the blower pressure.

The blower you recommended is almost twice the CFM at close to the same pressure so i think it would act as my two blowers. Do you see any issues with this? How quiet is the blower you recommended? Looks like a low power draw too? Do you have a source that isnt $200.

Kleen-Rite sells them cheap. It's as loud as any other Ametek I've used. This blower is very special and will loft a lot of beans for you.
 
pixelsmithy

Quote

allenb wrote:

The angle on the 1+ bag Sivetz I used to roast on was somewhere near 45 degrees and the throw of the beans was up and not toward the vertical wall.


This makes sense because air will follow the path of least resistance. It isn't going to turn a corner just because you have tilted a round hole. Instead the air will continue to flow in the same direction, only the round hole has now become elliptical, thanks to the tilt you have put in the perf plate. This means you are effectively reducing the amount of hole-square-inches and so may need more holes than you would have in a flat perf plate, to achieve an equivalent air flow.
 
pisanoal

Quote

pixelsmithy wrote:

Quote

allenb wrote:

The angle on the 1+ bag Sivetz I used to roast on was somewhere near 45 degrees and the throw of the beans was up and not toward the vertical wall.


This makes sense because air will follow the path of least resistance. It isn't going to turn a corner just because you have tilted a round hole. Instead the air will continue to flow in the same direction, only the round hole has now become elliptical, thanks to the tilt you have put in the perf plate. This means you are effectively reducing the amount of hole-square-inches and so may need more holes than you would have in a flat perf plate, to achieve an equivalent air flow.


My intuition and experience disagree. The air will turn the corner and flow perpendicular to the perforations. The best illustration of this is drill a hole in a pipe carrying air (or water). Which way does the air blow out of the hole? When the bean chute was open with this design, the air was definitely blowing perpendicular to my perf plate. I think your comment about reducing the effective area tells the story, if that were the case that is not the path of least resistance. Path of least resistance by definition in this case would be where it has the most area for flow.

The overall trajectory of the bean spout is away from the back wall because the air hits the back wall and travels up along it with the bean chute closed. My concern was not which way the spout was going to travel once over the bean bed, but that the beans at the point they entered the air stream by the perf plate would be thrown to the front wall creating more resistance. Also, the air would lose energy and the air stream becomes effectively narrower. Flattening the perf plate did make a significant difference for my RC. I think I still could have been successful with the angled perf plate knowing what I know now about required CFM, but I'm still happy with the flat perf plate.
Edited by pisanoal on 12/17/2019 8:48 AM
 
pisanoal
Quick update. After thinking i had the problem solved, I have been trying to get 6 lbs to roast right. It has been back to the same issues of ROR tailing off a few minutes into the roast.

After discussing with my buddy, he suggested checking my ET thermocouple which for some reason i hadn't considered being an issue. One of my concerns with placing the TC under the perf plate from the beginning was that radiant heat from my burner would cause issues with my TC reading. So I stuck some cheapo TCs with a handheld meter in there, one below the perf plate and one right above.

The ones below agreed, the one right above was almost 200 degrees lower! I'm sure when there are beans in there, the difference isn't as dramatic, but still definitely an issue. Another reason I thought my ET was high is my door gasket kept scorching and i thought it was rated to 600. I checked the temp rating and its only 425. My RTV I'm using is what's rated for 600 +... I have since changed the door design so the gasket isnt in the direct stream of hot air anymore and havent had scorching issues since. Also, on my TC test, i had to get my ET probe to read over 700 F before I was at around 480 F above the plate. Definitely explains a lot of my problems well.
 
pisanoal
Conclusion to the high air temp issue:

Moving the probe above the perf plate helped some. I was getting faster roast times, but still not around the 10-12 minute mark that I was wanting. And in order to get under 16 minutes, I was having to push the ET to over 600 degrees. Also indicative of too hot of air temp, I was getting some bean scorching, maybe 20% of the beans. I was really stuck on two things, amount of airflow and perf plate design. Along the radiant heat issue, I added a heat shield below the angled plate, thinking maybe that was getting too hot. No change... Then we decided we should add a baffle in the furnace section to keep the direct hot air from going straight through perforations. Instant success!

With the furnace baffle I immediately noticed two things: 1) I was having to push gas flow a lot higher at the same air flow to reach roasting temps on my ET and 2) The ET trend was much, much smoother. So I concluded that I was getting some localized areas of very hot air, but overall not getting enough heat.

We did a couple of roasts this way and were able to keep air temps below 500 degrees, the roasts came out much, much more even, ROR curve was much easier to control, and roast times were between 10-14 minutes, First crack being anywhere from 8-9:30 for a reasonable roast profile. Results are very repeatable and at this time, we have done 6 or 7 successful roasts in a row.

Someone earlier on asked the question if there was proper flame and air mixing. They were right and I misunderstood their point, so thanks even though I didn't listen...

I've seen several other threads from the past where people struggle with high ET in fluid bed roasters and I think there is a general rule of thumb here. Baffle your furnace section, or build some kind of plenum to evenly mix and distribute the heated air. The perf plate alone may not take care of that.
 
allenb

Quote

pisanoal wrote:

Then we decided we should add a baffle in the furnace section to keep the direct hot air from going straight through perforations. Instant success!

With the furnace baffle I immediately noticed two things: 1) I was having to push gas flow a lot higher at the same air flow to reach roasting temps on my ET and 2) The ET trend was much, much smoother. So I concluded that I was getting some localized areas of very hot air, but overall not getting enough heat.



I'm a total believer in baffles or turbulators to break up stratification when heat sources are somewhat close to the RC inlet plate.
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
pisanoal
Yeah, I read some posts about the turbulator in your thread i think. I didnt think id need one in this roaster. I definitely wouldnt describe the heat source as close. It's 26" from the perf plate.

I think its probably a good idea even if the burner is a good distance away.
 
PhilH
Sorry Im very late on this thread
Im with Tamarian on trying the steeper angle on the ramp

My Roaster is similar and maybe some comparisons could be helpful
I have a 5kg FB using 6? cylindrical RC and Burner with 45deg ramp and flat perf plate
https://forum.hom...ad_id=5296
I measure ET about 1cm below perf plate and ET is often measuring around 150C(270F) above BT
I measure BT about 50% of the bean mass height and about 5cm from the ramp and well out of the main spouting airflow.

Its notoriously difficult to understand with certainty the fluid/thermodynamics of a specific FB roaster geometry (esp wrg to Temp measurement). I measure peak ET about 350C(662F) which would seem to theoretically break ?the rules?. However I know there is a very significant temp drop as soon as the heated airflow crosses the perf plate which is difficult to measure accurately. I am comfortable this likely puts me back inside ?the rules?. Tasting (as Tamarian points out) is a great way to assess/confirm whether changes result in improvement or deterioration of roasted beans.

I use 2x cheap VC blowers ? one was not enough) and use a long burner chamber approx 750mm (30") from burner to perf plate (probably overkill) with a baffle plate to promote turbulence / mixing of heated air prior to reaching perf plate. Leaks prior to the perf plate (even small ones) are easy to miss and will mess up bean movement and lose a lot of heat.

Sounds like you have got it largely sorted now.
Crem ONE LFPP, Compak K10F, 6kg FB LPG/PID aArtisanQ_PID, UNO/LCDapter
 
pisanoal

Quote

PhilH wrote:

Sorry Im very late on this thread
Im with Tamarian on trying the steeper angle on the ramp

My Roaster is similar and maybe some comparisons could be helpful
I have a 5kg FB using 6? cylindrical RC and Burner with 45deg ramp and flat perf plate
https://forum.hom...ad_id=5296
I measure ET about 1cm below perf plate and ET is often measuring around 150C(270F) above BT
I measure BT about 50% of the bean mass height and about 5cm from the ramp and well out of the main spouting airflow.

Its notoriously difficult to understand with certainty the fluid/thermodynamics of a specific FB roaster geometry (esp wrg to Temp measurement). I measure peak ET about 350C(662F) which would seem to theoretically break ?the rules?. However I know there is a very significant temp drop as soon as the heated airflow crosses the perf plate which is difficult to measure accurately. I am comfortable this likely puts me back inside ?the rules?. Tasting (as Tamarian points out) is a great way to assess/confirm whether changes result in improvement or deterioration of roasted beans.

I use 2x cheap VC blowers ? one was not enough) and use a long burner chamber approx 750mm (30") from burner to perf plate (probably overkill) with a baffle plate to promote turbulence / mixing of heated air prior to reaching perf plate. Leaks prior to the perf plate (even small ones) are easy to miss and will mess up bean movement and lose a lot of heat.

Sounds like you have got it largely sorted now.


Well where the heck were you! Haha. Just kidding. (Sort of) Glad to have your input! I actually looked at your thread while trying to sort out my issues and gleaned a little bit of useful info there.

The steeper angle was definitely key, I think because of the square roast chamber.

My ET was definitely causing me major issues. While I wasn't positive I needed to place a lot of emphasis on adhering to the rules, it was limiting my thinking and ultimately figuring out how to get that down was the key to my success. I've roasted some excellent coffee the lay few batches and the roaster is very controllable more. That is something I could not have said before the furnace baffling. What does your furnace baffle look like, and where is it located/how much area for flow did you leave around it? I saw a huge drop in temp across my perf plate without beans so o4 moved my ET probe to above the perf plate. But I noticed that with the roaster loaded, the temp drop was no where near as great. I was also seeing roast defects caused by high heat though. That's why we finally added the baffle. It is directly under the perf plate so air has to flow across the furnace chamber twice before going through the perf. It extends out to about 2" from the back wall.
 
PhilH
I put the baffle plate ahead of the burner. The air comes in the bottom of the burner thru the rectangular hole and has to go around the outside of the circular baffle plate there is a gap of approx 1cm between the baffle plate and the 150mm outside pipe

pics attached - I hope they come thru ok
PhilH attached the following image:
img_2234.jpg

Edited by PhilH on 12/19/2019 12:14 PM
Crem ONE LFPP, Compak K10F, 6kg FB LPG/PID aArtisanQ_PID, UNO/LCDapter
 
pisanoal
That's exactly how i used to have mine set up. If you can, I'd recommend trying a baffle above the flame. I'd bet money your ET will look better. Thats precisely what did it for me.


I attached a drawing of how mine is set up now.

I'm going add one in my little 1 lber too. The roaster was a lot more controllable after adding that baffle as well
pisanoal attached the following image:
15767841337103207610240224958851.jpg
 
PhilH
I'll try it and report back. A bit busy at the minute so wont be for a few weeks.
Crem ONE LFPP, Compak K10F, 6kg FB LPG/PID aArtisanQ_PID, UNO/LCDapter
 
renatoa
Found today this roaster... maybe the picture with lateral section help you, if still in trouble...

https://www.neuha...g-process/

Scroll down to the bottom of page
 
CharcoalRoaster
Thanks for sharing Renatoa - good find! I've never seen a curved "return" like that for the bean mass. It makes a ton of sense in reducing friction and encouraging better flow I would imagine. I wonder if production is that much more difficulty proving to be a challenge in most FB designs
 
Husamka
If you noticed, your beans are falling in the middle of the column. it should be blown to the left end so it will smoothly slide on the 45 degree plate. If you can add curved plate in the top as a trial similar to the attached design.

Edit, Oops you already have the photo.
Husamka attached the following image:
wwwneuhaus-neotecde.jpg
 
CharcoalRoaster

Quote

Husamka wrote:

If you noticed, your beans are falling in the middle of the column. it should be blown to the left end so it will smoothly slide on the 45 degree plate. If you can add curved plate in the top as a trial similar to the attached design.

Edit, Oops you already have the photo.


Are you referring to the OP roaster design or the attached image Husamka?
 
Husamka
Regarding the beans are falling in the middle, this is refer to "pisanoal" video. the curve in the photo is to correct that.
 
pisanoal
Thanks for the link renatoa. I actually saw that roaster or something similar in my search for answers during the build. I really liked the idea, unfortunately the RC was already built and the way the beans were being loaded, there was no good way to add a curved deflection plate. This is something that RoasterRob made reference to trying as well.

CharcoalRoaster - It definitely increases the difficulty of fabrication, but not too bad. If designed right, you could use a section of SS pipe, or get a piece of SS sheet bent to your desired radius. Still more complexity then flat sides. For large volumes of coffee, definitely makes sense. Also would be extremely useful for shortening the RC height.

Husamka - Thank you for taking the time to look at the thread and provide suggestions. I have actually already successfully completed the project. The design is capable of roasting upwards of at least 8 pounds of coffee in very reasonable amounts of time (drop times under 15 minutes), with good ETs (below 525F). The roaster has produced some really great coffee. The problem with adding an angled plate would have been loading the roaster. I would have had to add two angled plates stacked like a curved version of "<". The beans load from the top and would have hung on the curved plate. Ultimately, the beans falling in the middle of the pile only happened very early on in the roast. The key changes that made this successful were:
1. Increasing the angle of the angled plate (square RC requires steeper plate angle for good agitation, something around 60 degrees)
2. Baffling the furnace above the flame to ensure good air mixing and even heat distribution.
 
Jump to Forum: