Fabricating Custom Displays For A Commercial Coffee Roaster


Roasting the perfect coffee bean is an art form. But even the most talented of roasters can use a little feedback on what’s going on with their equipment. [Ludzinc] recently helped out a friend of his by building this set of 7-segment displays to show what’s happening with this coffee roaster.

The yellow modules hiding underneath the display panel are responsible for setting the speed of the hot air blower and the rate at which the drum turns. They’re adjustable using some trimpots, but it sounds like the stock machine doesn’t give any type of speed feedback other than direct observation.

The solution was to patch into those speed controllers using the ADC of a PIC chip. They each output 0-10V, which [Ludzinc] measures via a voltage divider. After the speed is quantified the microcontroller outputs to one of the displays. Since there’s a different chip for each readout, the firmware can be custom tuned to suit the operator’s needs.

Keep this in mind if you’re still planning to build that coffee roaster out of a washing machine.

14 thoughts on “Fabricating Custom Displays For A Commercial Coffee Roaster

    1. i thought those buttons looked familiar (I recently spec’ed something very similar (Apex I believe) for some instrumentation at work.

      A really attractive panel! Any reason you chose digital display over something analog? I imagine it might be useful to be able to see everything at a quick glance (which analog graphs are great for) rather than seeing exact numbers – keeping with the theme that roasting is more art than science – the analog vs digital would perhaps give a better “feel” to the process rather than having to parse numbers?

    2. I have designed similar switches from Schurter into products that I have worked on. They can be expensive, but there are not many substitutes for a high quality elevator button. Those things will last until the end of time and still look good the entire time.

      One way to reduce cost would be using a slightly smaller switch at a lower voltage, then use a solid state relay to drive your load.


      The MCS19 series is ~$10 and the MCS30 is ~$50.

      The majority of the cost is the IP-protection ratings (water/dust ingress), and adding illumination adds a lot of complexity to that sort of thing (additionally areas to seal.)

      Rolling your own illumination would reduce cost significantly, but it is not always easy.

      Either way, you built a fantastic looking controller. The two-year-old in me makes me want to push buttons on that all day long.

  1. I was going to ask about the switches too. I’m getting them come up as £64.70 (about US$96) in the UK! They’re not that nice.

    It makes me wonder what sort of staff he employs if he needs vandal resistant switches on the machine.

    1. I’m sure they were going more for switches that can take the wear and tear of daily usage by people who might not treat it as delicately as the the ones who paid for or made it. If they are vandal resistant, then they are pretty robust… well, in theory.

      If not that, then they were going for a specific look. In the grand scheme of things, $100 per switch isn’t that bad ( when you consider how much the whole project must have cost ) – it’s more than I would pay for something on one of my projects, but I have only really made things for my own use.

  2. Wouldn’t take too much to do a really wild retro version of this with magic eye tubes in place of the multi-seg displays. you’d have to have the switch for each function adjacent to the top of the tube doing the display, but you could also facricate a ring-switch with a hollow middle to accommodate the tube and a clear top dome to still allow for button pressing. Overkill, yes. But why the heck not… :-)

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.