Shining a Light on Defect Detection:
How a $9 UV torch could be one of the best investments for the coffee value chain
8 Minutes Read
Tim Choi
Quality and Sales Assistant Europe
Phosphors are the organic compounds responsible for this glowing effect. They are a specific combination of compounds present in cells that absorb energy from UV wavelengths, and emit visible light wavelengths that we experience as a glowing colour in the dark. Microbes contain high quantities of these phosphors, and also affect the way coffee tastes and ages. Their presence in both washed and naturally processed coffees can be analysed with UV light to determine how they impact cup quality over time.
I spend a lot of my time lurking in the shadows of the darkest room in our office with 100g samples of green coffee and a UV flashlight ostensibly designed to detect pet urine. It might seem tedious, but this activity can reveal a great deal about the history and potential longevity of our coffees.
We will demonstrate how a basic UV torch can detect common defects in green coffee and how feedback from this analysis can diagnose some of the root problems for coffee producers.
Phil Sung (our Global Traffic Manager) developed a sampling protocol for our import offices in 2017, which includes UV analysis as a integral component of quality assessment. Here’s how we do it:
1. Weight out 100gr of green beans into a sample tray.
2. Expose the beans to UV light in a darkened room.
3. Place another sample tray on a scale and tare the weight.
4. Pick out any fluorescent green speckled beans:
from the 100gr sample and place in the sample tray on the scale.
5. Once you have weighed out all the speckled and fully phosphorescent beans, record this weight in gr over 100gr for a percentage of UV defects. (0% = Zero, 0.01-5% = Light, 5.1% – 10% = Medium, 10.01% + = Heavy)
6. Finally, record the degree of overall whitish glow or mottling – Zero, Light, Medium, Heavy
Roasting
UV light analysis can be a useful measure of quality for green coffee traders, but thus far there has been limited research in this area. UV can indicate if a coffee is old, poorly grown, or processed incorrectly. But roasters are also adopting this method to improve their roast quality. Some have even gone to the extent of investing in optical sorting machinery that will further sort their green coffee before roasting, essentially adding one extra step of processing and quality control to the value chain. We love that roasters are as passionate about quality as we are, but feel this extra step is unnecessary if the correct purchasing decisions are made first. A great deal of time, effort, and money can be saved if we focus on improving quality at the farm-level and by working with producers that understand and care about physical and cup quality as much as we (you) do. And it only really costs $9 to get started.
Mottling & Age
Mottling is caused by the oxidation of polysaccharides, acids, and lipids found in the coffee seed, which creates a whitish glow under UV light. If the drying process has not been given the correct attention and time, it can negatively impact the lot’s longevity. When coffee is very old or improperly dried (often the result of drying too fast), many of these compounds migrate from the interior to the exterior of the bean, thereby increasing their exposure to air/oxygen and resulting in a bright white glow under UV light. However, when coffee is fresh and given proper care during the drying process, you won’t see this. A coffee glowing bright white usually equates to a paper/woody taste when roasted, and this can tell you whether it’s been stored in a warehouse for a long period of time, if it’s up to scratch quality-wise for filter, if it is more suitable for espresso roasts, or if it’s simply not worth investing in at all. We limit our work to origins where we have our own purchasing/export operations, so that we can ensure that all the coffee we buy meets or exceeds our quality standards; doing so helps us minimize premature aging, maintain consistent cup quality, store coffee properly after purchase, and orchestrate regular shipments.
Phenol
Phenolic compounds impart an astringent taste and are quite unpleasant. The phenol defect occurs when the environmental conditions for growing coffee aren’t optimal in terms of humidity and temperature. It isn’t always easy to detect truly phenolic defects by sensorial analysis alone and flavor-wise it can also be easily confused for other defects, such as Rio. UV light reveals this defect as a fully glowing bean, no spots or speckles. The most effective method of detecting this defect is with fresh green beans, as aged/mottled coffee will make it difficult to see the difference between mottling and microbial activity.
Processing & Longevity
Microbial activity changes the flavor and quality characteristics of coffee over time. Speckling is the term used to describe an isolated spot or spots of microbial growth that glow bright green on the coffee seed under UV light. Two frequent culprits behind speckling in washed coffees are poor water quality, and poorly maintained machinery which can nick beans during the de-pulping process. When coffee is analyzed for speckling, it can tell us a lot about the history of the processing and provide useful feedback for producers who want to improve quality and increase the price per pound of coffee they grow. Beans with heavier speckling have a higher amount of microbial activity and a tendency to degrade more rapidly. They do not retain the initial fresh flavor characteristics you loved about the coffee for very long. Natural and honey processed coffees will inherently have more microbial activity as they are dried in the cherry or mucilage, which attracts an accumulation of bacteria. We offer predominantly washed coffees to ensure that the landed cup quality matches the first offer sample as closely as possible and the longevity of the lot is maximized.
Country comparisons of UV defect %
The data cited below was selected from 731 samples analyzed in the London office over the first half of 2019. It is important to note that we assessed a disproportionate amount of Colombian and Nicaraguan coffees of varying grades due to the harvest calendar, and that our analysis does not necessarily give the most accurate or complete comparison of quality between origins. This data merely serves as a snapshot of observations from Caravela’s coffees in 2019.
Guatemala had the highest median of UV defects and, upon further investigation, we discovered this was attributable to the recycled water used for washing coffee, which increased exposure to microbes and encouraged speckling. In Nicaragua, we were able to see how one producer’s machinery was creating heavily speckled coffee that was well above the median for the rest of the country. The pulping blades on the machine were too close together and were damaging the coffee parchment, thus increasing exposure to bacteria. Both issues appear easy to resolve by either changing water quality or recalibrating the blades of the pulper, respectively. However, many producers have limited access to education and resources, which is why our PECA program continuously shares our QA findings and educates producers in the most current quality practices. We believe that this will genuinely help them to achieve fairer prices than C-Market pricing in the long run.
How UV Defects affect Cup Scores
We have tried to analyze how different UV defect percentage readings change the final cup score of the 731 samples. Graph 1 shows how lowest cup score (83) had the highest median UV defect reading (7.75%) and a wider range of UV readings. The highest cup scores (86 & 87) had the lowest median UV defect reading (around 4%) and generally had a smaller range of UV readings. It seems that there is a correlation between the UV readings and the cup score that is worthy of further analysis.
Whilst our data is limited by the short time frame and the number of samples analyzed, there is a trend shown in Graph 2 that highlights the relationship between UV defect readings and overall cup score. It seems that we can safely assume a higher percentage of UV defects will yield a lower cup score, and vice versa.
Conclusion
This report has analyzed six months’ worth of samples, albeit primarily from two origins. We hope to learn if the trends examined in this report are still evident over the course of a year, and across a more representative mix country-wise. It would also be interesting to determine if there is any correlation between UV readings and the demographic variations within countries. For example, the difference between UV % defects in the north of Ecuador, where producers are predominantly wealthier and have better resources, and the south, where this is not the case.
For exporters of specialty coffee, UV analysis provides useful insights into green coffee quality and longevity, offering a deeper understanding of growing, processing, transporting, and storing green coffee. We can also see how important it is to reduce the percentage of UV defects in coffee to reach a higher cup score and effectively achieve better prices. However, the only way this information will become truly useful is through sharing our knowledge and techniques of green coffee analysis with producers so they can accurately diagnose any issues at the source. We are keenly aware that many do not have access to the resources or funds to invest in advanced optical sorting machinery; our aim is to provide better access to educational material and affordable technology that can improve the quality of coffee and the lives of those who grow it. As evidenced above, a $9 pet urine UV torch is a worthwhile investment for the anyone working in the coffee value chain. Just make sure you have rechargeable batteries!

How to Create a Specialty Coffee Line to Attract Commercial Coffee Drinkers Into The Specialty Coffee World? Part II

How to Create a Specialty Coffee Line to Attract Commercial Coffee Drinkers Into The Specialty Coffee World?

LOTSA Potential: The Long-Term Supply Agreement as a Tool for Financial Stability and Building Relationships

Hey guys…fantastic article and thanks so much for shining a light (I know…bad pun!) on this issue. I have been working with coffee farmers in rural Yunnan, China for the last 10 years. Over the last year, we have been trying to utilize UV lighting for quality control measures and to better enhance our processing control parameters. I haven’t been able to find a lot of research papers and so I appreciate the work and insights you have here. I do have several follow up questions, I’d love to dialogue with someone about…
(Sincerest apologies if these questions are elementary in nature!)
1. Can you tell us more about phosphors? What specifically are they? How do they exist in coffee? Are they naturally occurring or developed? Even if you have another resource on them, that would be wonderful!
2. “UV can indicate if a coffee is old, poorly grown, or processed incorrectly.” Do you have any results from just cupping the UV reflected defects? As a producer and exporter, I just want to make sure that if we are telling people reflected beans are linked to certain things that in fact they are. We’ve had some interestingly mixed results from cupping UV reflected beans.
3. “Oxidation of polysaccharides, acids, and lipids create whitish glow,” is this how phosphors are developed? Would be fun to know how oxidation process creates those phosphors.
4. Very interesting stuff about phenol. This is one of the biggest struggles for producers in our region, and it is the “silent killer!” While not detectable by merely looking at the green, it is incredibly offensive in the cup. You say that phenol under UV presents as a full white bean. Any thought on why/how? You make a link between mircobial activity and phenol. Would love to hear more about this.
5. “Beans with heavier speckling have a higher amount of microbial activity and a tendency to degrade more rapidly.” The link here between higher microbial activity and coffees degrading quicker, do you have any more information or source for this research? Very interesting and would love to dig more into this.
Again, just want to say thanks again for the article. I’m excited to see how this can continue to expand into newer innovations, research, and understanding!
Hey dear Tim,
beautiful questions!
We are from the educational sector snd recently shed light on this topic. Your questions are beautiful!
I would be interested in getting in touch and share knowledge and inspiration.
What do you think?
Best wishes, Nadine
kaffeeschule.com
Hey Tim,
Please excuse the delay in getting back to you and thanks for reading this. We really appreciate your interest in our approach to UV analysis and the questions you have asked. We will never claim to be academic/scientific experts in this area (not sure anyone really is to be completely honest) but we would like to find out the exact same answers you are seeking. Please email me – tim.choi@caravela.coffee. It would be great to discuss this further.
Cheers!
Apologies for spelling mistakes above…my keyboard was not cooperating with me today…
Thanks so much for this information.
Dear Tim,
wonderful job and fantastic work! 731 is a very good amount of inspected samples, thanks a lot for spending so much hours on it!
We are on the topic and your article is a great step ahead!
We would be interested in keeping in touch about this issue and will drop you an email the upcoming days.
Thanks a lot, great work!
Best wishes, Nadine
So a optical sorter with a UV instead of white light should in theory work.
Thanks for putting this together!
We’ve been using UV analysis to aid purchasing decisions at Kickapoo for a few years now, and we’ve found consistently that coffee that looks better under UV tends to taste better, longer. Really interesting to see how this data breaks down country to country. It would be interesting to keep up with this research as you get more data points over time.
One interesting experiment would be to measure how much UV activity in a green coffee improves on a specific farm after making some of the changes you mentioned (changing water, calibrating de-pulper, etc.) to start digging deeper into what causes this and how to improve it.
Keep up the good work!
Great Article! I haven’t read this anywhere else and will begin implementing into our buying process.
Thanks Tim and everyone for your efforts and everything you put into this work! Outstanding and interesting research.
Im a home roaster/brewer, so Im gonna invest in UV light to take a notch up in my coffee. So curious to try it out and compare the results.
Cheers!