Homeroasters - Home Roasting Coffee Community » BUILDING A ROASTER » Fluidbed Roaster
Who is here? 1 guest(s)
Page 4 of 4: 1234
 Print Thread
All About the Bubble Bed Roaster
Bhante
Top  #76 Print Post
Posted on 07/08/2023 8:27 AM

Quote

Bhante wrote:2) We are looking at more conduction, i.e. the beans are really in close contact.

In the video above Typhoon talks exclusively of conduction from the metal, but when the beans are in close contact the conduction between beans is also a factor, equalising local temperature variations.
Edited by Bhante on 07/08/2023 10:43 AM
 
renatoa
Top  #77 Print Post
Posted on 07/08/2023 11:05 AM
The conduction transfer in a BB machine is far from the conduction concept in drum roaster.
Yes, in both machines is about bean to metal plate contact... but... the details follows:
- in a BB machine more contact means less airflow, to reduce bubbling, thus reducing agitation, thus uniformity of roast changing. For a drum machine the agitation remains the same, whatever is the heat input in the drum.
- in a BB machine roast plate is heated by same air that heats the beans too, so they are always close temperature, there is very low heat transfer from plate to beans. In a drum roaster is hugely different, the drum metal is almost every moment at least 50 C hotter than the air inside.

So, imo, the claimed contact transfer it's just a marketing ploy of the Typhoon guys, to attract guys who learnt roasting the classic way, on a commercial drum machine.
 
allenb
Top  #78 Print Post
Posted on 07/08/2023 12:10 PM

Quote

Bhante wrote:

Quote

allenb wrote:

Let me know the post # so I can go back and take a look


#27, as reply to #25


Got it. The reason I mentioned that it would launch the beans was just to make the point that I thought the airflow would need to be dialed way back once the beans lightened up. I do have a variable speed control on that blower so was not an issue.
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
allenb
Top  #79 Print Post
Posted on 07/08/2023 12:36 PM

Quote

Bhante wrote:

Quote

allenb wrote:I think their idea of rows of holes in a repeating fashion sets up longitudinal waves that carries the effect on to the perimeter due to the wave effect carrying on to the extreme sides of the roast chamber.


I am slightly intrigued why you suggest there many be longitudinal waves in the Typhoon roaster, and yet you assert this cannot happen in a round chamber! The latter assertion is in case of fact incorrect! (Well, at least in part).

Have you seen waves in the typhoon roaster, or are there assertions from someone else that they have seen waves? Actually, one reason I suggested reducing the air speed is because I think the excessive air speed will break up the waves.

The Zoellner patent you referenced in the earlier thread relates, as far as I know, to a specific large capacity roaster built in Germany (40 tonnes, 4 tonnes ... not sure, maybe the former but I can't remember). I seem to remember the coffee tooks 24 hours or numerous days to cool. That roaster had a torroidal roasting region, and a slow-moving wave traversed around that torroid in a continuous motion. I witnessed a presentation concerning that roaster sometime around 2010 by someone who had seen it in action, including one or more videos (not just the Nepro video which I think is misleading).

Anything more specific I say about that presentation must be with that caveat that the presentation was a long time ago and my memory may not serve me correctly in some details. On 11th November 2011 I downloaded a copy of the Zoellner patent, almost certainly in response to a forum discussion on some coffee forum at that time, which I seem to remember contributing to - at that time obviously I would have remembered the details more reliably, maybe someone could find it with more success than I had yesterday (Homeroasters ... Home Barister ... Kaffee-Netz.net ...?)

I think there are two important hints in the Typhoon video we need to take stock of:

1) The hole pattern on those two plates seem to be identical, as far as I can see (see 1:45 - just one plate it thick and the other thin, and obviously for one the blower is turned up.

2) We are looking at more conduction, i.e. the beans are really in close contact.

According to my feint memory of the presentation, I saw a video of the wave movement in the actual roaster, and/or a small scale model (which was also torroidal).


Many good questions/thoughts!
On the wave theory in respect to round versus square RC, the reason I don't see a practical way to produce the rolling wave action exhibited in the square typhoon RC is this. When you place the double/single repeating rows of holes in a round chamber, you end up with the flowing wave of beans having to cascade from one arc curvature into a smaller arc or vice/versa. A wave likes to travel in a uniform flow and not have to compress or expand as it travels. At least thats my theory. Sort of like sound waves, when you try and bend it, it will keep moving but becomes distorted in shape. When I use the term "wave", I am really referring to the mass of beans being lifted by the double row of holes and falling into the lower pressure valleys of the single rows.
While it may be possible to pull it off in a round RC, I think it would require a whole lot of trial and error and may not be worth it.

On the thicker and thinner perf plates, It is as renatoa stated, clever marketing to less savy roasters who think they will be able to retain some of the supposed attributes of drums. Without a means to add additional heat to the plate, its just a heavier slab of perf plate.
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
Bhante
Top  #80 Print Post
Posted on 07/09/2023 6:41 AM
The wave flow in the original Zoellner roaster was in a radial orientation flowing in a circular motion parallel to the circumpherence, therefore there is no change in the dimensions of the wave as it goes slowly round. The Typhone patent shows circular rows of holes, which obviously would not work for producing the waves. Both the real Zoellner roaster and the mockup model I saw (which had a real wave of coffee beans) were torroidal, i.e. it seems to require an inner wall. It looks just like a water wave, and like a water wave the actual motion at the bean level is nothing like the motion of the mass of the wave as a whole, rather, there must be motions in a perpendicular direction. Yesterday I was reading a paper about formation of bubbles in a bubble bed, their properties, how the beans move, what prevents bubbles from collapsing etc. It didn't mention anything about transverse waves, but intuitively it seems to discuss some relevant parameters. In the Zoellner roaster most of the bed is obviously in a state of insipient bubbling, and the wave must represent a state change to actual bubbling (in a special phase).

How I imagine the state of the bed is this, and it is pure speculation of course, from beginning to end, although a few of the details must logically be true. The entire bed of beans is fluid at all times - that means it is in constant motion - but most of the time this is probably very small scale motion on the very local level; but as the wave passes by the whole bed gets churned over in the region of the wave. All the beans will be in very close contact all the time, with a very thin film of air passing over them the whole time - with the exception of the bubbles. How and where the bubbles are and how they develop/move is of course the big unknown, but I am assumng the wave itself is a special case of bubble behaviour.

I partially take back what I said earlier about bean-bean contact as a significant component of heat transfer, because of this thin film of air - which must necessarily exist if the method really works and there is a wave (and I know that it does work). The thin film of air is by definition highly efficient for heat transfer (but not necessarily by definition uniform, that is another matter, but allegedly it is). This heat transfer is gas-bean not bean-bean (nor metal-bean), but it is a conductive heat transfer not convective. I don't believe the Typhoon people know what is going on in detail with the conductive heat transfer, and they may well be simply assuming the thick plate is better by analogy with the drum roaster - but I think we cannot rule out that it is also in fact better for reasons that we cannot conceive; it is my assumption they have tested different plate thicknesses and might have found a difference. Probably they have designed the controls in such a way that when the thicker or thinner plate is detected the airflow is automatically changed between conductive (low airflow) and convective (high airflow).
 
Bhante
Top  #81 Print Post
Posted on 07/09/2023 6:52 AM

Quote

allenb wrote:

When I use the term "wave", I am really referring to the mass of beans being lifted by the double row of holes and falling into the lower pressure valleys of the single rows.


The wave certainly would not constitute the type of motion you describe, that would not work. Almost certainly it will involve some complex motion of bubbles inside the mass of coffee, whose complex behaviour results in the formation and motion of the wave. This is a highly complex process of aerodynamics.
 
Bhante
Top  #82 Print Post
Posted on 07/09/2023 7:06 AM
On a micro scale, the airflow through the thin plate would be much more turbulent than through the thick plate, and it is highly plausible that that might break up the delicate flow parameters required for the conductive bed to work properly. hypothetically, one might create a hollow plate that is just as lightweight as the thin plate but which has the same hole depth as the solid thick plate, and that should have the same aerodynamic properties. I suspect such a lightweight hollow plate would give much the same results as the solif plate, but different from the thin plate (of course it would be impractical to construct, it is just a thought experiment).

Therefore I don't think the thick and thin plates are just PR. Obviously it has implications for test rigs!
 
Bhante
Top  #83 Print Post
Posted on 07/09/2023 7:18 AM

Quote

renatoa wrote:

The conduction transfer in a BB machine is far from the conduction concept in drum roaster.
Yes, in both machines is about bean to metal plate contact


In the drum the conduction is metal to bean, with a very high thermal gradient and very low thermal transfer efficiency. With the supposed bubble bed conductive transfer, it is gas to bean conduction, with a very low thermal gradient and very high thermal transfer efficiency. The reason the thermal transfer efficiency is so much greater is (a) because the surface area for the thermal transfer is really MASSIVELY greater - the entire surface area of every single bean simultaneously; and (b) because the gas is constantly flowing over the entire surface area of the beans. That's why you can get a very fast change in temperature; but more importantly, the low thermal gradient means that the bean is not subjected to such extremes of temperature.
 
renatoa
Top  #84 Print Post
Posted on 07/09/2023 8:00 AM
There is no such thing as gas to bean "conduction", this is called convection.

Quote

Convection (or convective heat transfer) is the transfer of heat from one place to another due to the movement of a fluid
 
allenb
Top  #85 Print Post
Posted on 07/09/2023 8:03 AM

Quote

That's why you can get a very fast change in temperature; but more importantly, the low thermal gradient means that the bean is not subjected to such extremes of temperature.

Wow, perfectly stated! That fact (lower environment temperature) and potentially less bean surface damage, is one of the attributes claimed by earlier papers by Mr. Zoellner. He unfortunately went in the direction of “flash roasting” for his intended market for consumer grade coffee roasting.
It would be a fascinating cupping session to actually compare the lower BB ET roasted coffee next to the typical fluidbed roast of the same coffee.
Edited by allenb on 07/09/2023 8:12 AM
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
Piotrkurak
Top  #86 Print Post
Posted on 07/09/2023 9:36 AM
Only difficulty I see with your comparison idea is getting equivalent roasts (unless that is the point you actually/want to make).
 
Bhante
Top  #87 Print Post
Posted on 07/09/2023 9:54 AM

Quote

renatoa wrote:

There is no such thing as gas to bean "conduction", this is called convection.

Nonesense. Convection means the transfer of heat through mixing (strictly speaking the term convection normally refers to convective movement that is induced by density differences caused by temperature, namely hot fluids rising and cold fluids sinking). Conduction means the transfer of heat across a thermal gradient. Conduction can be across any materials in direct contact - solid, liquid or gas. The only medium incapable of thermal conduction is an absoute vacuum.

The radiator in a car works by circulating water under pressure through passages in the metal housing of the engine, where it picks up heat from the engine by conduction. The heated water is carried to the "radiator" part where it flows through a region with a large surface area where the heat is transferred primarily by conduction (but also partially by radiation and convection) to air blown over it by a fan.

Incidentally, just a brief note on the word "fluid", which usually means either a solid or a liquid. The earth that we walk on we normally think of as solid, but it can take on a liquid form, for example during an earthquake. For example the entire main populated region of Bangladesh comprises a very deep layer of silt that happens to be situated at the confluence of three major fault lines along the edges of continental plates that are converging under huge tension. It is virtually certain that sometime - eventually - a massive earthquake will occur when these three continental plates move relative to each other to release the built-up tension. During that movement the entire body of silt - i.e. the "earth" that Bangladeshis walk on, from the surface down to a considerable depth - will instantly turn into fluid - i.e. a physical form that flows like liquid (thus during an earthquake a car or even a person can simply sink into the earth). This earth consists of small particles of solid (especially sand), tightly compacted and with very thin layers of water in between. I suppose technically we should describe this mixture of sand and water as a colloid.

The coffee beans sitting in the roaster with the airflow turned off is just a pile of solid particles; but once the airflow is turned on and the beans are moving, the bed of coffee is "fluidised" - i.e. it has certain properties of a fluid - just like the earth during an earthquake. I think this analogy is useful to think about in the context of a fluidbed roaster, because it helps us to understand that the solid particles don't need to be spaced far apart, but actually can be very tightly packed, and yet in constant motion and behaving as a fluid. In the Typhoon example, we have two different fluidised beds of coffee - one where the particles are relatively close together and relatively slow moving, the other much more widely spaced and faster moving. Both are fluids.
 
Bhante
Top  #88 Print Post
Posted on 07/09/2023 10:14 AM

Quote

Piotrkurak wrote:

Only difficulty I see with your comparison idea is getting equivalent roasts (unless that is the point you actually/want to make).


Agreed it would be a difficulty, but potentially possible. In particular, based on the claims of Typhoon, combined with what I am suggesting, you should be able to take a roast well past 2nd crack with a standard fluidbed roaster with the usual high flow rate, and a second fluidbed roaster with beans in much closer contact, and adjust both roasts to give broadly similar results - weight loss, colour, sweetness, and broad flavour characteristics.

If you then do a cupping on both side by side, you should find that the standard fluidbed roast should have more burnt flavours than the second one, because the thermal gradients would be higher.
 
allenb
Top  #89 Print Post
Posted on 07/10/2023 11:38 AM
From Bhante in 2011 within Jkoll's build thread

Quote

I had a very interesting chat with my friend yesterday about the Zoellner/Nepro roasters, which he knows well and has seen several times.

What happened is that a gentle circular wave motion arose sponaneously, INSTEAD OF the gentle (and as I understand, intermittent) bulges over the high density regions that were observed with the rectangular plate. As the wave passed over a certain region the beans from underneath would be deposited on the topsurface of the trailing edge of the wave, while the rest of the bean mass appeared virtually motionless apart from the levitation. The wave would move around the roasting chamber, according to my friend about once every thirty seconds which is somewhat slower than the rate given on the Nepro site, but the difference is probably just due to the fact that this demo was seen some years ago. I asked which direction the wave moved - he reckoned probably clockwise.


I re-read your post from 2011 and am reposting part of it as it is very informative relating to the "wave" phenomenon.

In one, and the only one I noticed of the Typhoon videos of their rectangular beds, there is clearly seen a slow traveling wave from right to left. Yet, as far as I know, the perforations are not drilled on an angle but are perpendicular to the surface.
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
Bhante
Top  #90 Print Post
Posted on 07/11/2023 10:04 AM

Quote

allenb wrote:
From Bhante in 2011 within Jkoll's build thread


Ah, good that you found my old post, thanks, I was looking for it earlier without success. It was interesting to read after all that time.

Here is the link to the Jkoll thread:

http://homeroaste...rowstart=0

Allen, you mentioned the Rototerm roaster with a wave motion http://homeroaste...post_32451 Can you give me any more information about that roaster?

Also do you have a link to the Typhoon video where you saw the wave, or at least the name of the video?
Edited by Bhante on 07/11/2023 11:05 AM
 
allenb
Top  #91 Print Post
Posted on 07/11/2023 12:27 PM
Here's a couple of links to the Rototerm on Nepro's site. I never understood exactly how the thing operated.

http://www.nepro-...oterm1.htm

http://www.nepro-...oterm2.htm

I'll try and locate the Typhoon vid.
Edited by allenb on 07/11/2023 1:25 PM
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
allenb
Top  #92 Print Post
Posted on 07/11/2023 12:50 PM

Quote

Also do you have a link to the Typhoon video where you saw the wave, or at least the name of the video?


This is not the one but shows a little hint of the flow at 3:19

https://www.youtu...JI6CUidVjg
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
allenb
Top  #93 Print Post
Posted on 07/12/2023 9:12 AM
I think there's a need to define what we mean by wave in relation to the typhoon roaster. When I initially mentioned wave related to the Typhoon, I was really referring to something less than a true wave but couldn't think of a better term. What I was referring to was the bean flow rising from the higher density rows of holes and dropping into the lower density ones and since this results in a row formation, it resembles a wave action but doesn't meet the definition of a traveling wave at all. Even the movement of the beans that appear to slowly migrate from the right side of the Typhoon roaster to the left would not be considered a wave either. So, it appears that within certain Nepro designs, as you previously described, there is an actual rolling wave action and in addition, in the Typhoon designs, there is migration of the bean mass from one end of the roaster to the other. These two phenomenon are most likely not connected to each other at all.
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
Bhante
Top  #94 Print Post
Posted on 07/13/2023 10:07 AM

Quote

allenb wrote:

This is not the one but shows a little hint of the flow at 3:19

https://www.youtu...JI6CUidVjg


I am doubtful whether it is useful to call that a "wave" - (a) it has a majority random compnent, (b) the movement is too fast, (c) it certainly has nothing to do with the wave I am talking about.

I see why you talk about an apparent movement from right to left - but that may not be real. To check that you would have to extract the individual frames from the video (less than 2 seconds worth that is relevant) and see from the stills what is really happening. The apparent motion from right to left might be a visual illusion, you'd have to compare successive bean positions to see whether there is a real movement. It would be a useful exercise anyway, but I would want to see a longer segment. Also I would want to see the "conduction" phase not the "convection" phase which is unlikely to have a proper wave.
 
Bhante
Top  #95 Print Post
Posted on 07/13/2023 10:18 AM

Quote

allenb wrote:

I think there's a need to define what we mean by wave in relation to the typhoon roaster.


Agreed.

Quote

allenb wrote:

When I initially mentioned wave related to the Typhoon, I was really referring to something less than a true wave but couldn't think of a better term. What I was referring to was the bean flow rising from the higher density rows of holes and dropping into the lower density ones


Well, for the convection phase maybe. In support of that, the high and low points are clearly fixed in location on the horizontal axis. Your supposition depends on the right-left apparent motion being real rather than a visual illusion (resulting from artefacts such as the frame rate of the camera - for example, if you film a moving object relative to a fluorescent light bulb, or sometimes an LED light bulb). But in any case it is somewhat irrelevant to the question of what movement is taking place when the air movement is slower and a real wave forms, which is what interests me.

I suspect the motion you describe is probably a sort of design artefact, a byproduct of the rows of holes and the high airflow rates.

Are there any Typhoon videos clearly showing the bean movement with the "conduction" drawer? They seem to be more keen to show the "convection" drawer, perhaps because the "conduction" drawer videos tend to make drum roasters sceptical (it probably doesn't give a realistic impression of the movement if you don't understand how the movement works).
Edited by Bhante on 07/13/2023 10:24 AM
 
Page 4 of 4: 1234
Jump to Forum: