Hello, I am trying to measure liquid height in a 20 ul capillary tube. (https://www.sigmaaldrich.com/catalog/product/sigma/p2049?lang=en®ion=US&cm_sp=Insite-_-prodRecCold_xviews-_-prodRecCold10-7) Right now my approach is to develop a capacitance level transmitter using arduino, reasberry pi, and copper sheets.
Is there a better method for measuring capillary height? Does anyone know of commercially available capillary liquid level transmitters?
Hi Takayuki Suzuki , there seems to be a problem with the link but anyhow:
Can you see the liquid height with the naked eye or by using a magnifying glass? Perhaps it would help to have a backgrund or lighting of a certain color. Or is the liquid clear as water, and virtually invisible?
If it is visible, and if the resolution is satisfactory, a camera (1D would be enough, i. e. a 1D optical sensor array) plus image processing (in the 2D case) might work.
Hi Joerg Fricke, We can see the height with a camera, but that method is harder to make automatic and scale up to simultaneously measure multiple (50+) capillaries. I would also like to keep the cost of my device as low as possible.
There are commercially available liquid height transmitters for large tanks of liquid. I was hoping if you or someone else would know of commercial capillary height transmitters?
Hi Takayuki Suzuki , I'm sorry that I don't know of commercial sensors suitable for measurement in capillary tubes.
But with regard to the cost:
Even if you would find a commercially available sensor it would not be a mass product and consequently expensive; simultaneous measurement in 50+ capillaries would mean that you needed 50+ sensors; I guess that would take > $5000.
On the other hand, if, say, 10 capillaries could be placed stationarily within the visual field of one camera, 5+ cameras plus 5+ microcontroller boards would be enough. You should get one camera plus one microcontroller board for less than $100 dollar.
Regarding the software, I think your problem does not require full blown image processing (that's why a microcontroller of medium performance will do) but just locating every tube laterally within a narrow range (could be eased by lines of a certain color between tubes) and then following the tube upwards to the border between liquid and air. Every clever and interested student of information technology who has some experience with microcontrollers should be able to write this software within some days.
So, I still feel an optical solution would be the least expensive way.
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Condensed Matter Physics
- Soft Matter
- Liquids