Hi there

I thought you guys might find this interesting: I'm in the process of designing and building a coffee roaster from scratch as part of my studies. The project is named: "Design and Automation of a Profile Controlled Sample Coffee Roaster". The aim is to build a small scale coffee roaster that is electronically controlled and can adjust for environmental variables - such as ambient temperature and a few others. After months of research and planning, I have finally entered the construction phase of the project. The frame has already been built and all of the steel parts have been laser cut. Materials include brushed stainless steel, mirror (BA) finished stainless steel and also some mild steel. Other specs include:
3-plaque ceramic infrared burner
LPG Gas powered
Proportional solenoid valve controlled gas line
Thermocouple temperature readings
Digital signal processor controlled
Thermocouple compensation electronics
PID controlled
PWM controlled motors (to vary the drum speed)
Computer software for making the profile
Serial connection between the roaster and the electronics.
Stainless steel drum
Hopefully around 3kg per batch

On the electronics side, things are moving really fast. I have completed the PID algorithm, along with the A/D (analog-to-digital) conversions of the thermocouples, PWM (Pulse width modulation) of the motors and UART (serial) communication to and from the computer. The electronics are being programmed in C with Microchip's MPLAB C30 compiler.

I am currently busy with the software side of things. The software will feature a really cool interface that will allow roasters to easily create, save and load roast profiles. A short video of the Alpha version of the software can be seen on http://www.youtube.com/watch?v=btxx_HlibOw . Please just note that this is not yet functional and that I still have masses of functions to code before I can call it proper Roast Profiling software. The objective here is to build a prototype...

I am also a barista, not a roaster, so I don't know too much about coffee roasting yet. I know what I want, but not how to get it. I did get 150kg's of green beans sponsored, so there is enough coffee for me to perfect the PID alogorithm and hone my roasting skills.

What I'd like from you guys is any input, ideas or thoughts that you might have about the project. I am building this for home-roasters, coffee shops and commercial roasters alike and would really appreciate your inputs.

First step was to get a nice drum. I found this 219mm (diameter) 3.5mm thick stainless steel drum



Then I began with building a model of the inside of the drum by using laser cut perspex fins.



Here is a pic of the finished model, although all of the reverse fins is not shown in this pic:

The next step was to test the burner for flow rate, ramp temp etc. This was done in my University's chemical engineering labs.



Between it all I learned how to work with Solidworks 2008, a very powerful but difficult CAD modeling program. Here is the initial design:

After rendering the image, this is what I got:







An here is a taste of the final product. It is being built as we speak, but I ran out of stainless steel bolts to fasten all the panels. All that needs to be done is some serious TIK welding, a lot of drilling and then it's off to the powdercoaters for the final touch. I have my first demo of the project in 2 week's time (along with exams), so I'm building like mad!

you should check out the forums at http://homeroasters.org/ if you haven't already. many of the member's there are members here, but they have a whole section for building your own roaster.

project looks sweet. out of curiosity, what's it cost to build from scratch (assuming you're not billing your hours designing it)?

Thanks for the tip. The price is a difficult topic. Too cheap an people won't even look at it twice. Too expensive and sales go down. I'm hoping to put it on the market for around the same price as an espresso machine...maybe around 5000 USD. But again: too cheap and people will think it's a tin can with a candle. I guess it also depends on the target market. Marketing is still a looooong way to go

I have one small question...

Isn't perspex a type of lexan?

PMMA ignites at 460 °C and burns, forming carbon dioxide, water, carbon monoxide and low molecular weight hydrocarbons, including formaldehyde

I'm guessing that you plan to replace those with metal fins?...

http://en.wikipedia.org/wiki/Acrylic_glass

Any plans for managing/metering airflow through the roaster?

Perspex is a transparent plastic like thing. Works like a charm for rapid prototyping and out university has a laser cutter for perspex. The perspex was just a mockup of the drum. The final drum will have 2mm stainless fins.

As for metering of air flow: The temperature of the air will be adjustable, but not the air flow. I'll fine tune the speed of the extractor fan and maybe later add PWM control, but that's not a prime concern for now.

Hey guys.

There's loads of updates on homeroasters.org if you'd like to follow the roaster's progress

http://www.homeroasters.org/php/forum/v ... post_16347