Electronics > Projects, Designs, and Technical Stuff
Volumetric measurement of liquid in transparent container
<< < (6/6)
splin:
Shine a light/IR beam at an angle towards the surface and use a linear optical sensor to detect the angle of the reflected beam. A vertical baffle across the container base, just past the point where the beam enters the container, is needed to block reflections from the container base reaching the sensor. The beam would not need to be especially well focused/collimated as the sensor output can be heavily filtered to determine the mean, or centre (or edges) of the reflected beam position. Modulating the light beam at even a lowish frequency (but not 50/60Hz or harmonics!) would allow ambient light to be ignored.

Increasing the distance between the container bottom and the sensor increases the resolution.

I just tried this with an LED torch (flashlight) and a plastic container with between 1 and 5cms of water and it looks promising.

Problems:

1) Preventing ambient light saturating the sensor - vertical baffles either side of the sensor will prevent light entering sideways but bright overhead lighting could be a big problem especially if it is incandescent which has lots of energy acrosss the spectrum including IR - LED or fluorescent  lighting would allow you to select a suitable emitter wavelength and use a corresponding filter over the sensor.

2) There is much less choice of linear sensors than single photodiodes/transistors and may be more expensive and difficult to work with. An array of, say 16, individual sensors may be enough so long as the reflected beam is wide enough to illuminate several sensors, the relative output levels being used to determine the centre or edges of the beam.  Alternatively, if you were to relax your 'no moving parts' requirement, then you could use a fixed single sensor and oscillate the beam angle. Using a moving coil should be very reliable and bearings can be avoided by using a spring material, plastic or metal, to mount a mirror or an LED emitter. Avoiding linear sensors would give you much more flexibility in the choice of sensor wrt sensitivity, dynamic range and wavelength (including UV).

3) If the container is very tall relative to its width, the beam would have to be near vertical and the reflectance may be insuffcient.

4) It might not work well at very low water levels as some (or the majority if the beam is too 'wide') of the reflected beam may hit the vertical baffle seperating the emitter and sensors.

[EDIT] Just spotted the requirement that this has to work with cloudy liquids such as milk or orange juice. The latter is likely to be a big problem especially if it is the 'with bits' variety. It may just be possible though if the beam is powerful enough and well focused. Using multiple emitters at different wavelengths may be needed to cope with a wide range of coloured liquids.
SiliconWizard:
Since the liquid probably has to get in and out in some way, wouldn't measuring the flow (with a flow sensor), instead of estimating the overall volume at any given time, adequate for your intended accuracy?
Rerouter:
Grab a suitably sized kitchen scale without a removable measuring plate. And adhere or mount it to it. Steal the strain gauge wires and run to your own device.
speechsupply:

--- Quote from: SiliconWizard on February 13, 2019, 01:53:23 am ---Since the liquid probably has to get in and out in some way, wouldn't measuring the flow (with a flow sensor), instead of estimating the overall volume at any given time, adequate for your intended accuracy?

--- End quote ---
In theory that could work.  But the container mouth / top has a wide openingen. Meaning, that liquids can be poured from many angles.
Navigation
Message Index
Previous page
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod