Progress for Week 3 of Building the Roaster

2012 May 3
Comments Off on Progress for Week 3 of Building the Roaster

This week, the machining of the bowl was completed. The edge was milled to exactly 36″ in diameter on a vertical mill.

Next, the bowl was sent to the QC lab for inspection. It was tested for concentricity.

Extremely accurate measurements were taken at the area where the hub will attach. These measurements will be used to do a perfect match in machining the hub.

The main bearings for the bowl shaft arrived. They are high-temperature bearings since some heat will be transferred down the shaft.

The bowl drive shaft has been fully machined with a threaded end for the hub.

The opposite end has been machined down and keyed for the coupler to the drive motor.


My Tabletop Vertical Mill

2012 April 27

While ACME Metalworks will be doing the bulk of the machining work for the roaster, I have always wanted to make some of the parts myself. It is impractical for me to make large components, because the machines need to do so are large and expensive. However, there are lots of parts such as control panels and detailed parts that are small enough to be made in a home-based machine shop.

The most important and flexible machine tool is the vertical mill. It can drill, bore, cut, and shape parts in infinite variety. Today, my tabletop vertical mill arrived from Sherline in San Marcos, CA. It is small but powerful. This little mill can machine steel, aluminum, brass, plastic, wood, etc. to a tolerance of 0.0005 inches, which is just as good as big commercial mills. I ordered mine with digital readout, so that it is easier to use. The mill can also be converted to CNC (computer numerical control) in which a computer drives the mill and machines parts from a program written in G-code, the programming language for machine tools.

The Sherline 5400 is able to move the piece being machined in X, Y, and Z directions. The movement is controlled by a handwheel for each direction. The hand wheel has a digital encoder that measures how far the wheel has been turned.

This is indicated on the digital readout along with the speed of the tool in RPM. It is important to know the speed of the tool, as this affects cutting speed.

The tools are driven by a variable speed motor connected to the head by a drive belt. The black and white disk on top of the pulley is the disk from which the speed is measured.

The Sherline mills have many tools that can be used with them to be able to produce very intricate precision parts. People have made everything from integrated circuit testing machines to wristwatches using them. The photo below is of 1/8 scale Harley-Davidson cylinders machined from a solid piece using Sherline equipment!


Progress for Week 2 of Building the Roaster

2012 April 24
Comments Off on Progress for Week 2 of Building the Roaster

This week, ACME Metalworks continued machining on the bowl. Due to the fact that the bowl is very large and the thickness is 1/4″, it must be machined gradually. The bowl tends to flex a bit when being machined and set up harmonic vibrations. To damp the vibrations, the bowl is filled with cooling/lubricating fluid.

Slowly, the milling tool is run around the bowl. Cutting a layer at a time, the tool is moved closer and closer to the final diameter, which is 36″. It must be precisely round (concentric) for the roaster to work properly.

By the way, the machining is being done using Haas machining centers which are also American-made. It is quite a success story of American ingenuity and craftsmanship. The machines are built in Oxnard, California. Since our goal is a 100% US-made roaster, it even more satisfying that the operators are local machinists and that the machines that make components for the roaster are made in our neighboring state.

My drawings are being converted to CAD drawings for machining the other components for Module 1. We have also ordered the US-made bowl drive motor and speed controller from Bodine Electric.


Progress for Week 1 of Building the Roaster

2012 April 13

Physical work on machining the real deal started this week at ACME Metalworks and I have been designing and drawing feverishly to keep up. There are 22 modules in the build sequence for the roaster and sub-assemblies within each module. A list of the modules and current subassemblies are in this list. We are working on Module 1, the bowl and its drive system, which is perhaps the most difficult and critical to the everything else.

Last week, Tim Eckholdt from ACME designed a fixturing system to hold the bowl in the vertical mill to do the machining of the part. This is the fixture on CAD.

This week, he has fabricated the fixture and has started to use it. He welded temporary tabs on the bowl, which will be cut off later.

The bowl has been cut down from 20″ tall to about 11″ tall which results in the shape which we feel is optimal for bean dynamics.

Last week he also converted my engineering drawing for the drive hub into a CAD file for machining.

So far the hub has been partially machined and awaits a key slot for the drive shaft and mounting holes for the securing studs.

This thrust bearing will carry the weight of the bowl and the coffee while it spins at 90 RPM.

I have also completed my drawings for numerous other parts, including the aluminum carrier for the bowl, bearings and drive shaft. This is my engineering drawing of the upper plate which will me machined out of 1/2″ thick aluminum.


Machining Details for the Bowl

2012 April 9
Comments Off on Machining Details for the Bowl

We met recently at ACME Metalworks to discuss how to machine the bowl. It is certainly the most critical component in the entire roaster, so we really need to sweat the details. The bowl must be concentric (round) to within a few thousandths of an inch so that, while rotating, it does not scrape the cylinder it is spinning within. Furthermore, we must be able to move the bowl up and down about 3 inches while it is spins and stay in perfect alignment.

I rely upon Tim Eckholdt of ACME to come up with methods for machining parts that meet my design specifications. Since we do not have a giant lathe at our disposal, the bowl will be machined radially using a CNC mill. Tim designed an aluminum cradle to attach to the bowl. It is attached via welded tabs. Once attached, he will find and mark the dead center of the bowl. Then he will mill the edge radius of one half of the bowl, turn it, and machine the other half.

After this is finished and inspected, he will keep the bowl in the machining cradle and do the boring for the drive hub. This includes as large centering hole and six attachment holes. The hub mates to the driveshaft with a spline and a nut to secure it. The machining should begin this week. Afterwards, it will be heat treated and nickel plated.


Naming the Roaster

2012 April 4

As I sketch away on parts drawings for the roaster project, I have been thinking about what to name the roaster. “Roaster project” is the title of the process of designing and building the roaster, but it is not a good name for the finished product. You may think it is too early to think of naming the machine, but I would like to include the name on title blocks for each of the part drawings and quite likely inscribed in various ways on parts, even internal parts.

Naming is a very important activity. My general approach is to find the shortest name that conveys the most important characteristics. For the roaster, I could include the idea of a bowl or mechanical fluidized bed or the triple hexagonal shape. I would also like to honor the family name and our family heritage of design and building machines. I don’t want to use my full name, though. The concept of locally made and US made would be desirable, as well as the idea of permanence and craft.

If  you have some names you’d like to share, please make a comment, e-mail, or send a message on our Twitter account! I’d like to make a decision within the next few days. In my next post, I will share the design of Module 1.


Roasting Module: New Bowl Arrives

2012 March 15

We are now beginning the production of the roaster as mentioned in the post “Moving to the Building Process“. The methodology is DBTM (Design-Build-Test-Modify) and the project is broken down in to sections which are perfected before moving on to the next section. The first section is the roasting bowl and drive system.

Our prototype roasting bowl and drive worked rather well, but we felt that the semi-elliptical shape of the bowl (in cross-section) created slowing of the beans at the edge since the bowl is essentially vertical at that point. The forces tend to push the beans against the edge of the bowl instead of up the edge and into the return vanes. Our solution is to start with an oversized hemispherical bowl and cut it down to approximately half of its original depth. This results in an edge with a generous slope. The forces on the beans are then both  pushing the beans against the bowl and upward into the return vanes. This is also important for effective bowl evacuation when the roasting is completed. We plan on spinning the beans out of the bowl and edge stalling slows their departure, taking longer to complete the evacuation.

Our new “rough” bowl has arrived from a manufacturer in Ohio. It has an outside diameter of 40.5″ and a nominal 1/4″ thickness. The bowl is made of carbon steel, which has excellent thermal characteristics for roasting.

This modified photo shows the basic machining we plan to do, after heat treating the bowl to relax any existing stresses. The bowl will be cut down to 36″ in diameter with a corresponding depth of approximately 11″. We will machine a 2″ locating hole for the drive hub along with six smaller holes for the bolt down studs to secure the bowl to the hub.

We’ve also decided that instead of moving the return vane ring upward to open up a passageway for the beans to escape at the end of the roasting cycle, we will move the bowl downward. The diagram below shows the new methodology.

The bowl will move down approximately 3″ (all while spinning), uncovering ports that the beans flow into. Note that since the bowl now moves up an down within a cylinder, the edge needs to be flat vertically rather than horizontally and very concentric to keep from scraping the sides of the cylinder. This means the bowl will need to be precisely machined and the lift mechanism must be very stable and accurate.

Our next step is to begin machining the new bowl. Simultaneously, I will be finishing the design of the structural frame for the roasting module and the aluminum carrier for the bowl and its motor.


Aloof Coffee: In Defense of Aloofness

2012 March 2
Comments Off on Aloof Coffee: In Defense of Aloofness
by Joe Johnston

Aloof Coffee (see earlier post “New Coffee Concept Makes Roasting Obsolete“) has received unwarranted criticism from many potential customers and “third wave” coffee proponents regarding our aloofness. It is as though aloofness is considered a negative trait. We chose our company name carefully and consider “aloof” to be a strongly positive word, a virtuous word. Though we have ignored the criticism until now, we determined it is time to explain our point of view.

Webster’s Dictionary defines aloof as “removed or distant either physically or emotionally”. Since we know that our idea (coffee must not be roasted: it is a violent act which degrades coffee and the coffee growers) is superior to the ideas of others, we must distance ourselves from them. We must set ourselves on a pedestal, knowing that this will give us the role of teacher to instruct and guide others into the truth. Any sort of bonding with those holding opposing views only implies that their ideas are of equal weight, which ultimately is not even in their best interest. Our highest aim is to convert those who hold wrong views on coffee to our way of thinking. We understand that this may involve some arguing, berating, and condescension, but it is with the noblest of goals. We would also point out that those of the “third wave” who practice aloofness against “second wave” coffee people may be disingenuous in their criticism of our aloofness.

Furthermore, the fact that society clearly admires aloofness is demonstrated by mass advertising. Browse any fashion magazine and assess the attitude of the models. Are they wearing a welcoming, happy expression? Of course not. Madison Avenue understands that people are attracted to aloofness. The emotional distancing implies that the product the model is promoting is superior to yours and that there is distain for your wrongheadedness. This brings the viewer to the conclusion that they have an inferior fashion sense that can be remedied by purchasing the product shown. Aloofness is a positive motivator for change and action. Advertisers would not bother with it, if it did not work.

In conclusion, we are proud to be aloof. Are we smug? Perhaps. But is it wrong to be certain of the truth and uphold it all costs, including possible relational damage? We think not. We remain unapologetic about being aloof and the great positive our remaining so brings to the coffee world.


(Note: This continues the entirely fictitious Aloof Coffee saga. The story is meant to foment thought on how we think about coffee and people. You may wish to review my earlier posts on the “Religion of Coffee” for a deeper look.)


A Preview of Form and Scale

2012 February 24

I’ve been doing a lot of math and sketching the internal design of the roaster. To see what the final product might look like, I used Google Sketch-Up to do two 3-D models of likely layouts. They are obviously very basic and devoid of beauty and detail, but do give an idea of the overall design. The first is the staggered tower version.

The second is the clustered tower version.





















Any thoughts you’d like to share?


Moving to the Building Process

2012 February 17
Comments Off on Moving to the Building Process

The tests on the prototype have gone about as far as possible. The results have been promising and so it is time to move to building the real roaster. I have been sketching for days on various aspects of the design. It is coming to the point where ideas and sketches need to make their way to steel, copper, plastic, brass, glass, electronics, motors, etc.

My approach will be Design-Build-Test-Modify (DBTM). Basically, I will start out with a grand design that does not have every detail worked out — just the most important ones. Without a team of engineers to power through the whole project and 3D modeling, the idea of building the roaster in one try is ridiculous. Plus, that would take the fun out of the whole project. Edison famously had numerous failures in trying to develop the light bulb and one learns much from failures (as much as from successes, really).

Instead, I will do a detail design of a key section and then build it. That section will be tested and then redesigned and modified, if necessary. Once that section is perfected, I will move on to the next section. Going step-by-step in a logical succession, the final product will emerge.

Because I may end up drilling holes and attaching ultimately useless items to the design while going through this process, I am prepared to go back and remanufacture final components. Rolls Royce used to build their engines, run them on a test platform, then take them apart to measure and reassemble. The final roaster is going to be done in the same way. When a component is perfected, but has been cobbled up during the process, it will be discarded and a pure one made to replace it.

I will attempt to do a modular design to make things easier to assemble in the first place and to maintain later. The order of module development, at this point, is as follows:

1. Roasting bowl mount and drive system.
2. Fixed return blade assembly (roasting bowl).
3. Movable return blade assembly (roasting bowl).
4. Cooling bowl mount and drive system.
5. Fixed return blade assembly (cooling bowl).
6. Movable return blade assembly (cooling bowl).
7. Roaster ductwork.
8. Roaster blower.
9. Rough enclosure for roasting chamber.
10. Burner assembly.
11. Cooler blower.
12. Cooler ductwork.
13. Rough enclosure for cooling chamber.
14. Green coffee delivery system.
15. Cooled coffee output system.
16. Water misting system.
17. Sensors.
18. Control system.
19. Cyclone separator.
20. Cooling hex frame and cladding.
21. Cyclone hex frame and cladding.
22. Roasting hex frame and cladding.
23. Detailing.
24. Certification and listing.