Coffee Data Science
Oil Analysis for Total Dissolved Solids (TDS) Readings in Coffee
Searching for understanding
In some discussions about filtering samples before measuring Total Dissolved Solids (TDS), the topic of oils in espresso came up. TDS is used to calculate Extraction Yield (EY), and it is the primary tool to quantify coffee extraction. Oils in coffee could change how TDS is measured because TDS is measured using a refractometer, and oil has a different refractive index than water. To have a deeper understanding, I collected some data, and I was surprised.
Previously, I looked at someone else’s data to show that filtering causes a linear shift by about 0.5%.
Since then, I have advocated for not using filters for TDS samples because they are expensive, and according to available data, it is uncertain how they affect the sample. It is quite possible the centrifuge separated oil from water, and that’s why the measurement was close to syringed filtering samples.
If filtering samples changed the TDS reading, maybe the filter was getting the oil out of the sample. This was easy enough to test by adding some oil to the same sample to look the the opposite of filtering a sample.
I started by placing some oil in filtered water, and I gradually increased the amount of oil to water.
At first, the readings went down slightly. Then all of the sudden, the reading maxed out (HHH), which means the refractometer read a value over 24% TDS.
So I collected more data, but just using a tiny amount of oil. This is a single drop.
A tiny amount of oil still maxes out, and once you add a drop or two of water, the reading hits LLL which means there isn’t enough liquid to take a valid reading.
More Data
I looked to repeat, and I varied the oil and water.
Then I added oil into coffee grounds to get an idea of what happens when oil is filtered out.
I expected some oil to slightly change the reading, but I didn’t expect the reading to change so drastically or only very little. I’m not sure if a syringe filter is filtering out oil or fines or something else, but I am certain that oil causes wild measurements with a digital refractometer.
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:
Collection of Espresso Articles
A Collection of Work and School Stories
A Summary of the Staccato Lifestyle
Measuring Coffee Grind Distribution
Espresso Baskets and Related Topics
