Homeroasters - Home Roasting Coffee Community » BUILDING A ROASTER » Building a Coffee Roaster
Who is here? 1 guest(s)
 Print Thread
How to handle convection and conduction in a home built roaster
LiamVanderwood
Top  #1 Print Post
Posted on 05/31/2020 10:25 PM
Hi Home Roasters forum! I'm brand new to the website, but I've been really diving deep into the nerdiness of coffee roasting since about early March. I started off with the aluminum Whirley Pop and a cast-iron skillet, but that wasn't enough to sate my curiosity. I moved onto a quick heat gun/flour sifter setup. It's served me well, but both are on extreme ends of the spectrum of convection and conduction roasting. For the Whirley pop, those smooth and malty notes are prominent, but a lot of acidity is masked due to the roasting deriving from contact with the metal rather than airflow, and the flip side applies to the sifter with the roasting deriving from cooking entirely through airflow
This has led me to want to build my own roaster. I'm moving to France soon to be reunited with my wonderful wife after a prolonged wait due to the current global mess. As I'll be jobless for a little while, I'll have time on my hand to hone my engineering skills while I buff up my French.

My primary concern at the moment is workin' on a way to alternate between more conduction or more convection. I figure that I could have some fans positioned behind the perforated back of the drum and flame heat underneath the drum, and hook both of those elements up to the Arduino panel I plan to rig into the machine, but I'd like to hear everyone else's input on that concept.

My goal is to be able to roast about a kilogram maximum, with the 500-gram load being the sweet spot. The drum I intend to fashion from a restaurant-quality stainless steel pot or cannibalised ss Whirley pop (or French equivalent). I'll also be getting a nice k-type in there so I can monitor ET and BT, so I can get it hooked up to Artisan.

This'll be my first foray into engineering and struggling with an Arduino. While there'll be trials and tribulations, my love for deconstructing things and coffee will make this a passion project that I wanna see through.
I know the coffee I'm buying can be better than it is now, despite it being quite nice. But something's lacking, and having the ability to fully eke out those nuanced flavors is the end goal for me. As I can't afford one of those dreamy Aillio Bullets, I'm gonna go ahead and have some fun by going full "Home engineer".
Edited by LiamVanderwood on 06/01/2020 7:30 AM
 
renatoa
Top  #2 Print Post
Posted on 06/01/2020 3:42 AM
Why would you need conduction heat transfer through a single contact point to a spherical surface, as is the bean ?
This means burning the bean in the contact point, to succeed transferring a significant quota of heat.
Assuming you succeed, what are the virtues of contact transferred heat in coffee roasting? We want uniform deep penetration of heat, not a crust as in bread...
 
LiamVanderwood
Top  #3 Print Post
Posted on 06/01/2020 7:48 AM
Okay, I should focus more on airflow throughout the drum then? Are there any benefits to having some level of conductive heat through the drum present or is it all based on airflow?
I?m not anywhere near a scientifically minded sort of person, so I?m kinda new to learning the deeply technical details of roasting. I appreciate any advice I can get!
 
renatoa
Top  #4 Print Post
Posted on 06/01/2020 9:12 AM
welcome to forum
I see solid drum roasters as a heat exchange device, kind of...
The flame (or heater) heats the drum, then the airflow extracts the heat and transfer to the beans mostly in a (auto)convective way. Auto prefix meaning that the bean-air movement is more due to the beans movement, than due to airflow movement, compared to other roasting machines, based on hot air blowing.
There is also a small radiant heat transfer in the area surrounding the bean-drum contact point. Probably this heat transfer type is mistakenly seen by many as "contact", which is not true.
These are the only two ways I see desirable during the roast. No true contact heat transfer for me, thank you.
In an ideal world, the beans should be winnow permanently by the vanes, and don't touch at all the drum, because the surface temperature, supposedly in excess of 520F/270C, should burn the beans, if longer contact is maintained. Probably short knocks/hits aren't affecting the beans in a detectable way.

Due to the above, I see drum machines understanding and operating too complicate for home/hobby usage. Building side... probably a choice for those who like more tinkering, than roasting and drinking Grin
500-700 grams are much easier attainable target using a pure convection hot air machine, with a touch of radiant heat, as a turbo oven lid based setup, at least this was my way.
Cheaper, easier and faster to build and control, you focus on a single variable: heat, airflow is no more variable.
Good luck in your endeavour !
Edited by renatoa on 06/01/2020 9:43 AM
 
allenb
Top  #5 Print Post
Posted on 06/01/2020 8:12 PM

Quote

LiamVanderwood wrote:

Okay, I should focus more on airflow throughout the drum then? Are there any benefits to having some level of conductive heat through the drum present or is it all based on airflow?
I?m not anywhere near a scientifically minded sort of person, so I?m kinda new to learning the deeply technical details of roasting. I appreciate any advice I can get!


renatoa brings up some good points about drum roasters. Although I once theorized that there was much more radiant heat transfer from drum wall to bean, it appears to be much less but is an active player. renatoa has hit on something that most fail to recognize in drum roaster heat transfer modes. I firmly believe as he is alluding to that most of the heat transfer in solid drum roasters with heat sources beneath the drum, is from the inner drum wall heating the thin layer of air existing within the bean bed and beyond the area beans occupy coupled with the combustion air being pulled through the drum's rear end and what ever the % of drum wall radiant energy transfer is taking place. As he is stating, there is little to no useful conduction taking place. With all that said, with proper vane design and proper rotation speed, drum roasters can produce stunning roasts if used properly and I've found the conventional gas fired drum roaster with solid drum and under the drum gas burner heat source to be the easiest and most consistent means of roasting requiring very little interaction throughout the roast. If one desires the so-called Rao roast curve (constantly declining rate of rise), the classic gas fired drum roaster can accomplish this with two or three burner adjustments throughout the roast and a couple of air adjustments. Nothing could be simpler to operate.
In regards to potential deleterious effects of bean to drum wall contact, I've never seen any tipping, scorching or any hint of bean surface damage and never noticed any overly roasty notes unless I made a serious error in heat input which is almost impossible to do in my 1lb gas fired drum roaster.
But, as renatoa rightly mentioned, building a drum roaster is much more of a mechanical engineering challenge than a turbo oven lid based roaster and if not carefully designed and built, can end up being a sheet metal mess and possibly never able to produce good results. The turbo oven lid based roaster is much easier to build and you'd have a much higher probability of having a highly functional and enjoyable roaster as well as one that can produce as great a roast as any other method.
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
allenb
Top  #6 Print Post
Posted on 06/04/2020 7:35 AM

Quote

LiamVanderwood wrote:

Okay, I should focus more on airflow throughout the drum then? Are there any benefits to having some level of conductive heat through the drum present or is it all based on airflow?
I?m not anywhere near a scientifically minded sort of person, so I?m kinda new to learning the deeply technical details of roasting. I appreciate any advice I can get!


I'll take a stab at the question about focusing on airflow versus drum wall heat. In classic drum roaster design, there's typically been two directions taken. One, like Probat ended up doing after the 1960's, is to limit to some degree the heat transfer via drum wall by using a double walled drum and increase the exhaust flow cfm to make it favor through-the-drum convection and the other one is to use standard single wall drum and design air flow cfm only to the point where there is enough to pull chaff and smoke out of the drum which is what most of us do. One can simulate the double wall drum by adding a good sized baffle between burner and drum that extends further up the side of the space between drum and inner chamber wall which limits heat transfer to drum wall and forces more heat through the back of the drum. It wouldn't be too hard to experiment with different sized baffles to see if changing heat transfer characteristics changed results in the cup. I've had equally great coffee from roasters with double wall drum and ones with single wall.
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
 
Jump to Forum: