Coffee Data Science
DiFluid Omix Coffee Analysis: Technical Review
Another day, another moisture meter?
Green coffee beans have been evaluated primarily by moisture, density, color, and water activity. There are a few solutions on the market, but often not in one device. Additionally, these can be a few thousand dollars. More recently, Syncfo came out with the 4in1 which measures 3 of the 4 variables. DiFluid is now coming out with the Omix that can measure all 4 of those variables.
DiFluid sent me their new Omix to test to see how well it functions as a tool for coffee roasting. To evaluate it, I did four large series of tests:
- Green bean comparison
- Roasted bean comparison
- After adding moisture to roasted coffee
- Controlled moisture addition
The last two will be covered in the next article because the article length was too long
Devices
Both devices are simple to use, and they record a log of data. The DiFluid Omix is priced in below $3,000, and the Syncfo 4in1 priced in around $4,000. There is now also a Syncfo with a water activity function built-in, but that is not the one I have.
The shape of the Syncfo is unique, and I prefer the size and shape of the Omix because it is easier to store.
The Syncfo 4in1 is a qualified device by SCA. Additionally, I have done a few studies to qualify its performance. The aim of these experiments is to see how well the Difluid Omix compares, but because there is not groundtruth data, it is more difficult to state whether one device is more accurate than another.
Green Beans
I took readings across 10 coffee, and I took samples of some coffees more than once on different days. I ended up with 20 samples.
For green moisture measurements, there is not a statistical difference between the two devices.
Density measurements varied. Both devices assume a constant volume, so as long as one does the same sample preparation, these density readings will be just fine.
Similarly for color, there is an offset, but the key is that the device gives consistent results. From prior experience with the DiFluid Omni vs the Syncfo, there seems to be a calibration offset between the two devices.
Syncfo doesn’t have water activity, and I wanted to plot it in some way to compare to bean moisture. It seems pretty flat and lacking correlation.
Roast Beans
I also compared roasted beans, and I had a few more samples. The challenge is that the Syncfo doesn’t go below 1%, and it seems like the DiFluid Omix is more granular. There is an offset, but a bit more testing should be done. I think water activity might be more interesting for roasted coffee than moisture.
For density, the data seems a bit closer, but like I said about green coffee density, the key was consistency in sample preparation.
Color was far more connected, but the Omix had the advantage of only one mode. It can decide on its on whether it is looking at green or roasted coffee, but for the Syncfo, the user has to select the mode.
Water activity was interesting. For moisture above 2%, most of those data points were after 4% of water being added to the coffee which then seems to have some trend, but the moisture meter isn’t quite accurate or rather adding moisture to coffee is not well understood.
I made a table to talk about statistical significance using a two-tailed paired t-test. The difference in color is not that large but it is statistically significant.
This is a lot of data and analysis, which shows the DiFluid Omix performs close to the Syncfo. I think the water activity feature is helpful. The Omix gives a histogram for color analysis similar to the DiFluid Omni, but that is not in other color meters. I suspect that kind of data will be more important to analyzing coffee roasts.
Part 2 is coming where I dive into the moisture accuracy. I originally had one write-up, but it kept growing in length.
If you like, follow me on Twitter, YouTube, and Instagram where I post videos of espresso shots on different machines and espresso related stuff. You can also find me on LinkedIn. You can also follow me on Medium and Subscribe.
Further readings of mine:
My Second Book: Advanced Espresso
My First Book: Engineering Better Espresso
