Typica

Mastering the Roasting Process with Profile Translation Analysis

Today a new article went up at Daily Coffee News where I've written about an old technique that I've used to help develop my own understanding of how changes in the roasting process affect what comes out in the cup. This is useful when trying to track down the source of unwanted differences between production batches and can also be used to more quickly develop the experience needed to understand how specific changes to a roasting plan can produce specific desired changes in the cup during product development.

Shift and zoom to spot significant roasting differences

Experiments conducted through the lens of roast profile translation are responsible for the profile translation feature in Typica that many companies now use to improve production consistency, but it still hasn't really caught on as a post-roast analysis tool. I think a big reason for that is the lack of tool support. I was doing this on paper well before I started to work on Typica, but that's time consuming and not something that a lot of roasters will try unless they have a good reason to think they might gain some useful insight from it. It's another case where people probably either underestimate just how useful this is or don't really understand why it should make any sense to do that.

One of the side projects that I've been working on as I get the time for it is a stand alone program that can be used for doing this sort of analysis more easily. It's not quite ready for other people to use, but once it is I'll probably ask people who are using programs other than Typica to send some example data files as I'd like to make this usable for as many roasters as possible. This reuses some of the code that was developed for Typica 2.0, so it's also a way to get more people using that code and finding the bugs that I might miss. I'm hoping to have this profile translation analysis tool ready to release either shortly before I leave for the UAE International Coffee and Tea Festival in Dubai (I'm teaching roasting classes) or shortly after I return from that. Don't expect a lot from the first release. There are a lot of features that I'd like this to have that won't be in version 1.0 just for the sake of getting something released, but I intend to continue developing that up until and then through the Typica 2.x release cycle. At some point the features will be integrated into Typica, but I'd still like to have a stand alone program available for people who aren't using Typica.