Brainstorming theoretical fluid bed build

Discuss roast levels and profiles for espresso, equipment for roasting coffee.
Elfmaze

Postby Elfmaze » Mar 19, 2019, 11:10 am

I have been talking to and reading some of Viliam's posts about fluid bed roasters and watching his videos on youtube (http://www.youtube.com/watch?v=sBMnw6Wwzi0) I had an idea about where i have seen components like that before.

In the distillation world Still Dragon makes some beautiful work utilizing Chinese stainless components. Will had to machine his funnel at significant cost even working at the shop. what if an off the shelf component worked? According to Viliam's tests 100mm wide is good for 350-400 gram batch size.

As far as I know the silicon gaskets for the column and TC fittings should be good to 500*f, Is there potential here?

Image
https://www.stilldragon.eu/en/reducers/ ... owell.html

Image
https://www.stilldragon.eu/en/sight-gla ... tower.html

User avatar
SAS

Postby SAS » Mar 19, 2019, 5:16 pm

Go on eBay and look at Tri Clamp or Tri Clover fittings. Same stuff as Still Dragon but less expensive.

Examples on eBay:
Concentric Reducer 4" to 8"

3 Inch Flow Stainless Steel 304 Tri Clamp Sanitary Sight Glass

I don't know why you would even need the silicone gaskets; try aluminum foil or metallic tape and tri-clamp.

There are all sorts of interesting fittings you can string together. Be creative with your search.

Keep us posted.
LMWDP #280
Running on fumes.

Elfmaze

Postby Elfmaze » Mar 19, 2019, 5:26 pm

SAS wrote:Go on eBay and look at Tri Clamp or Tri Clover fittings. Same stuff as Still Dragon but less expensive.
Example: Concentric Reducer 4" to 8"


There are all sorts of interesting fittings you can string together. Be creative with your search.


Edit: I see the one you posted is MUCH bigger

the concentric reducer is actually not bad from SD $47USD
http://stilldragon.com/index.php/4-x-2- ... ducer.html

the tower column is a bit rougher at $145 but still not bad for what it is... problem is I would need a longer column and I'm still verrifying that the silicon gaskets will take 500*

Elfmaze

Postby Elfmaze » Mar 19, 2019, 5:31 pm

How tall do you think my column needs to be? The deep reducers will probably hold most of the bean load... but I think their glass is only 100mm tall. I wonder if I would have beans jumping out the top without a taller glass

Elfmaze

Postby Elfmaze » Mar 21, 2019, 12:15 am

There seems to be something wrong with my math on this. A 350 gram roaster using sivitz numbers for heat application originally had me thinking I needed a 4500 watt heating element per my napkin sketch assuming a 10% machine efficiency....

But as I'm looking into the heater there is no way people are packing a 4500 watt heater into even a 2inch pipe much less a 3/4 inch inlet. That's an absolute massive amount of energy in a small space. Basically building this machine and cramming it under a roasting chamber? I must be off on my numbers by an order of magnitude, right?

Image

https://jobcosupply.com/store/076008-sy ... gle-phase/

false1001

Postby false1001 » replying to Elfmaze » Mar 21, 2019, 1:30 pm

Be careful of supposed plug and play formulas for DIY roasters, especially for fluid bed roasters. It is nigh upon impossible to control for enough variables in a home setting for them to be useful. For example: It is quite possible to do 350 grams with a lowly 1kw heating element, however it seems most people use 1.5-2kw elements for roast capacities of around 400-500 grams. I personally have used both a 1.6kw and a 2kw heating element... the 1.6kw was perfect as far as responsiveness and heating curves went, but wasn't able to do a full pound of very dense beans. 2kw had plenty of power, but had a relatively small sweet spot and I ended up thrashing my SSR more than I'd like to maintain temps toward the end of the roast.

The devil is in two details: airflow and insulation. The big reason I couldn't do a pound of dense beans with the 1.6kw element is mostly because I needed such high airflow to loft the beans the heating element couldn't keep up. If I had a better funnel design I might have been able to loft those same beans with less airflow and still been able to use the smaller heating element. Insulation is also huge. Remember your heat source is the heated air you're pushing through, and any holes or conductive areas will dramatically reduce the heat capacity of that air and force you to oversize or overwork your heating element. There are plenty of easy to use insulation products that can help here, from flue tape to cut up fire blankets (make sure you're not cutting fiberglass insulation though) to HVAC insulation and beyond. Make sure you're using the proper metal alloys as well... i.e. use aluminum inside the roaster, but food grade steel for the outsides and structure. Also don't forget that air itself is one of the best and handiest insulators around... I double chambered my heating element (an inner ceramic cylinder that housed the element with an outer steel cylinder, just a couple mm of air separating them) and that made more of a difference than any other single modification I could have done. I reduced my power needs by roughly 200 watts.

Last but not least, remember that when you are dealing with cheaper heating elements the QC will not be great. I've ordered several heat gun heating elements and even whole heatguns that I've disassembled... the wattage is as advertised only about 50% of the time and often off by more than 100 watts. You definitely get what you pay for. If ordering from alibaba or aliexpress always get 3 of a kind for this same reason, and always ask for pre-shipping QC photos.

Good luck!

Elfmaze

Postby Elfmaze » Mar 22, 2019, 11:55 am

false1001 wrote: the 1.6kw was perfect as far as responsiveness and heating curves went, but wasn't able to do a full pound of very dense beans. 2kw had plenty of power, but had a relatively small sweet spot and I ended up thrashing my SSR more than I'd like to maintain temps toward the end of the roast.


Did the SSR actually give up the ghost? I thought that was kinda their designed purpose was to rapidly switch with PWM loads? The old better to throttle a larger element than over heat a smaller? But I am seeing more evidence of 2KW is the better option which is good because 4500 was a bit more difficult to source.

I have been eye balling my old leaf blower as a fan... It has an adjustable speed knob already built in and even if it kicks the bucket A new one is only $80. I would think it has enough flow to clear the beans right out of the chamber, But there will be restrictions in the 2" tube pushing past the heating element.

Image

false1001

Postby false1001 » replying to Elfmaze » Mar 22, 2019, 4:36 pm

No mechanical problems, just made things harder for my automation. If your SSR is switching on and off (much) faster than your main control loop can execute (I ran my PID control in a separate thread) you start to get Real Time Software Problems, which I'm happy to solve for a paycheck but are very much not what I want to do with my free time, especially with constraints imposed by Artisan. Don't get me wrong, it worked, but I had to manually end roasts which kind of defeated a large purpose of my automation. I still roasted with that model just fine for about 2 months before I shattered my roast chamber (protip: metal expands, borosilicate does not).

Blower motor sounds good, keep in mind that overspec-ing your blower might interfere with your ability to hear first/second crack.

Elfmaze

Postby Elfmaze » Mar 22, 2019, 7:11 pm

false1001 wrote: I still roasted with that model just fine for about 2 months before I shattered my roast chamber (protip: metal expands, borosilicate does not).


Ah are you cyberelectronics on youtube? I think I saw the results of that one.

false1001

Postby false1001 » replying to Elfmaze » Mar 23, 2019, 2:39 pm

Nope, just incapable of learning from other people's mistakes 8)