Simple colorimeter for food / beverage

I’m trying to standardize the brewing of some common tea in terms of temperatures and infusion time in order to optimize yield and character of the brew. Tea is mostly a ‘transparent’ liquid (vs turbid or reflective). I figured I could easily find some sort of beverage colorimeter to quantify ‘concentration’ (think Beer’s law). But surprisingly after a couple hours of searching, I don’t find such an instrument on ebay or from other sources.

I don’t need to characterize the color spectrum, only the concentration via the absorption. Or perhaps a refractometer might work but I doubt it. I’m just trying to get a relative concentration in comparison to another brew, not any sort of mg or solids or other specific.

I could probably build something but I don’t have the time. Don’t want to spend too much. I’d hoped to find some common instrument that showed absorption with a zero or blank using water and scale factor perhaps. Seems like that’s what I remember from college chemistry.

Or maybe there is something like a chorine test meter that I can hack to give a reading with a brown liquid.

Maybe that’s a common beer brewing thing?

Your just trying to achieve consistency of color from one batch to another?

If so, take a look at Adafruit’s RGB Color Sensor with IR filter and White LED board. The board is based up on the TCS34725 from TAOS (Texas Advanced Optoelectronics Solution, A Dallas based company.

I have used several different devices from TAOS in the past. They have very good technology, devices & support.

Link to Adafruit’s board -
https://www.adafruit.com/product/1334

Below is a copy of the TCS34725 data sheet -
TAOS-color-light-sensor.pdf (237.5 KB)

1 Like

Great example to use for a Design of Experiments.

  1. Conductivity could be one of the measures.
  2. Sample the water before infusion for chemical content, use that as a zero point. Variations in the water will vary unless you always use distilled or DI water, at least during your experimental phases. Or use commercial bottled water that is made by reverse osmosis, e.g Dasani brand by companies like Coca-Cola) and have minerals added back in. At least it would be fairly constant. Any of these would be a controlled variable.
  3. pH? Have no idea of variation but would impact taste.
  4. I would think that even the same varietal of tea will have color variations that impact the flavor.
  5. Age of tea could be a factor. Country of origin even if same varietal.
  6. Not sure how temperature will effect the measurements - but is another parameter that impacts concentration levels/density. I know temperature and time impacts coffee. Hot brewed vs cold brewed (High temps impact caffeine content which changes taste). Of course there is cold brewed tea. Question does cold brewed and hot brewed with same color taste the same?
  7. Refraction: Could borrow the one we use for coolant on the Haas.
  8. Science has a Raman something or other. Measures things. Some magical device some of them were excited about. Me I keep thinking noodles.

Then when you find the right concentration of your chosen ambrosia - should/may have someway to measure it - verses just tasting it. No reason a probe that tests multiple factors can’t be made during the sampling process. Major problem is subjective tatse at the end.

Sound like a great set of experiments - get with Science Committee and develop lists of vairables, design series of experiments and measurement tools and have it.

Me, I can’t stand tea. So subjectively to me, it would be just finding one that consistently tastes the same degree of bad.

Sorry, was a Process Control dweeb when I worked - goal was to chase down sources of variation.

Who knows, could develop … pun coming … the Teaminator! You’ll be Back for More!

2 Likes

In theory, using colorimetry to measure concentration is fairly straightforward. You make a “known” concentration series and measure the transmission, or absorption, or reflectance (or even luminance in some experiments) for your knowns. So you’d have dilutions like 100%, 95%, 90%, 80%, etc., and a value for each. Then you plot them and figure out the equation for the response curve. The response curve will be s-shaped so you need to manage your dilutions so that you’re only using the relatively straight part of the curve in the middle.

Then you measure your unknowns (in this case your tea) and use the equation to figure out the concentration based on your “knowns.” You adjust the concentration of your batch to reach the desired concentration.

The part you have to figure out is, how does color/darkness correlate with taste or other subjective factors? Which wavelengths best describe the color/darkness you want? Can you measure those wavelengths appropriately in the “straight” portion of the curve.

Oh, and you have to come up with a stable “standard” to measure between batches. That could be fun too.

https://www.vernier.com/products/sensors/colorimeters/gdx-col/

or

https://www.vernier.com/products/sensors/colorimeters/col-bta/

wandrson - Thanks so much. One of those might work. I did a fair amount of searching and could not find a product like those. And they are in the ballpark of what I might spend on the item.

JK,

Thank you so much for the reference to the Adafruit item. That’s more of a reflective device but a quantitative colorimeter might be a good factor to include in the analysis.

And regardless, what a cool item to know about that would otherwise be quite a challenge to design!!!

Cheers,

DJ

1 Like

jswilson64

Thanks, That’s sort of what I’d be trying to do. Just trying to find or build a simple instrument to measure. Looks like the ones above could do the job.

DJ

1 Like

Photomancer,

Thanks for your many ideas. Great analysis. You are thinking like I am.
Conductivity might not do much because most of the compounds are organic and I suspect not ionized. pH is a good idea to evaluate anyway because there might be a correlation between pH and some quality I want in the tea. Yes, the varietal or moreso the type (green, oolong, black, pu-erh) makes a huge difference in optics. Temperature is a huge factor and that is what I’m investigating. I’m not going for speed but instead some product of yield x quality and temp / brew time will be the independent variable. Have thought about refractometry. Might be too low in TDS but might work.

BTW, I was in the same camp re tea for years. I grew up with instant tea an could more or less take it or leave it - mostly leave it. I’d rather have water. But my wife got interested in tea from an agricultural point of view. We planted some in east Tx, about an acre. Turns out that tea is the #2 drink in the world behind water. You would not believe the amazing qualities of a good tea. I am still amazed. What we get in this country is basically trash tea, the stems, the chaff and dust from the tea industry that is bagged and standardized to taste exactly the same every year x forever = terrible. It is cut from the bush or ‘tree’ by mowers and vacuumed off the ground. The average wholesale price of this product is around $2.5 per kilo.

Now, look at ‘real’ tea that I would now drink and that you might even like. Even teas from ‘Teavana’ etc are almost junk and they are still standardized for that franchise. They mostly sell herbal infusions or tisanes. Tea comes from the camelia plant C. Sinensis only. Although there are different varietals or sub species, the different ‘kinds’ like green, white, oolong, black, pu-erh are the exact same leaf, just processed differently and allowed a different fermentation process. I only learned this stuff a little over a year ago so I’m really not an expert on this, but fascinated by what I did not know.

Good teas are hand picked (incredibly labor intensive) - like my wife and I picked for over an hour for one # of tea that yielded about 100 gm after processing. But the taste was amazing. Proper tea picking of excellent tea is typically the end of the stem in the form of 1) just the bud 2) bud and one leaf 3) bud and two leaves. Only a small percent are picked so as not to disturb the plant. Some are wild and over 100 years old and grow under a forest canopy. Most are cultivated in small villages or on plantations.

The descriptions of them are insane. Like reading about a great wine or scotch. I’ll find some and post. But seriously, the good ones are amazing in their quality and nothing like the bitter stuff that I can totally do without.

OK, a long winded story to explain what I’m doing. I’m experimenting with titration of temp, leaf to water ratio, brewing time and even pressure to try and extract the best from the best of these really nice teas. I’m sure it’s been done in some form but I can’t find that info and really need to do it myself since it’s really subjective.

Real tea aficionados would say that is getting too technical and ruining the ceremonial aspect of tea brewing but being an engineering nerd and just wanting a good and consistent product without all the fuss, I thought I’d look at being scientific with the process. I have about a dozen really good Chinese teas from Yunnan province that I’m playing with. Some are from small farms that only produce 10 or fewer kg per year. It took me months to find and get them.

Prices on the high end of rare, loose tea can be $1 per gram (almost $600 per pound) but most are much less. As crazy as that sounds, an efficient brewing of these could mean a lot in terms of cost of the final product. At somewhere around 200 ml per loose leaf tea gram for the final extraction ratio, that is still under $1 per brewed cup for the most expensive one. BTW I only bought samples, like 25 gm of the ultra pricey ones.

Would be cool to do a sampling at DMS sometime of the really good ones. I’ve got about 6 I really like but it is, of course, subjective.

Here’s a simple video on tea that will put you at about the 99.9% level of tea knowledge in comparison to others. . .

2 Likes

Will have to pick your brain sometime about DOE. Its all fun in theory but in practice is a bit messy. Would love to hear about your process control experience.

riceball2015, I’ll update you when I get going. . .

2 Likes

Hi. The spectrophotometer that science is about to buy would work beautifully as a colorimeter.

I see you noticed it, but oddly you asked about infrared instead of RGB detection which you discuss in this thread. RGB is a piece of cake, infrared not so much.

Hi. Really enjoyed your post about tea. Tea has all sorts of interesting chemistry that could be used to categorize and analyze them. Here is a fun tea reaction:
http://pubs.acs.org/doi/abs/10.1021/ed2002175

Cool! Spectrophotometer has been ordered.

Kobin Caddick
Dallas Makerspace
Science Procurement Officer
[email protected]