Liquid Density Measurement with Ultrasonic Transducers

[ILetSmokeOut]

I had an idea for a project.

I'd like to measure the density of a liquid in a glass jar. My thought was to measure the speed of sound through the liquid and calculate the density based on the known distance between transmitter and receiver. The liquid would be at STP and really close to the density of water (it's a water/sugar solution). I know that sound travels about 3 times faster in glass so I'd ignore the first ultrasound burst received and just focus on the range I'm looking for.

One problem I'm running into is, I can only find transducers that operate at 40kHz (or lower). At 40kHz, I would be able to send 8 cycles (with pure water at 1 foot distance) before the receiver starts picking up the signal. I'd like a little more room to play with but maybe it's not an issue. Maybe I can send bursts of 2 cycles.

Anyone see an issue with using 40kHz transducers or know of inexpensive 1Mhz transducers?

Any other thoughts that would help me out?

 

[jpanhalt]

I think the 40 KHz unit should be fine to demonstrate your concept.

The biggest problem I foresee is knowing which surface/interface is reflecting the signal. I did a similar experiment with a laser distance measuring device that used time of flight like your sonar does. I simply positioned the device over a bucket and took a reading of the distance to the bottom. I then filled the bucket with water and took another reading. Since the speed of light was slower in water, the apparent distance increased. Reflection from the surface from the water did not interfere. The light beam was also quite narrow.

The same set-up with sound would give a shorter distance when the bucket was filled with water. But, I think reflection from the air-water interface and dispersion of the beam will be much greater problems. Since you know what the distance to the top surface is, however, you may be able to correct for reflections from it in software. I think your biggest problem will be unwanted reflections.

I know that sound travels about 3 times faster in glass so I'd ignore the first ultrasound burst received and just focus on the range I'm looking for.

Getting your sound to penetrate rather than reflect off a glass vessel may doom that approach. Alternatively, can your sensor be placed in the water to take advantage of the reflectivity of the vessel you plan to use?

John

 

[ILetSmokeOut]

I wouldn't want to put my transducers in the water just for clean up/sterile purposes.

My thoughts were to use both a transmitter and a receiver... transmitting straight through the jar. The receiver would be about a foot away. I'm hoping that I would see distinct signals at the receiver, once from the sound through the glass around my liquid, and another from the direct line of sight through the liquid. I'm assuming the signal will travel through the glass just fine...?

I think I'll go ahead and buy a couple 40kHz transducers and build a 2 cycle burst circuit. I'll hook it up to an O-scope and see what it does. Any idea where I could find 1MHz transducers for under 10 bucks?

 

[jpanhalt]

Do you want the $10-dollar transducers to be new or will like-new be OK? I assume you are kidding.

It sounds like you are doing some biomedical/microbiological research since you mention sterility. If you look at fetal ultrasound as an example of a high frequency use, note the rigorous conditions that are used to avoid reflections at the skin surface. I suggest borrowing a ultrasonic tape measure and see whether you can penetrate a glass vessel or even a single pane of glass. I have never done that experiment, but my ultrasonic tape measure reflects well off of windows and give an accurate distance. Most important, a piece of single-pane glass in front of it prevents it from measuring the distance to a wall behind located 75 cm beyond. Maybe try some oil or similar (e.g., epoxy) between the transducer and glass wall.

John

 

[ILetSmokeOut]

You're close on guessing microbiological research... I'm making beer. Home brewing is one of my hobbies and I thought it would be cool to measure the specific gravity (density relative to water) of the wort as it ferments. I wanted to have it on the outside of the fermenter to avoid contamination/infection. I suppose, if I could encase the sensors in stainless steel, I could sanitize it and drop it in but then I would have to worry about the signal propagating through the stainless. Maybe there's a reason I'm only finding medical ultrasound probes that are 1Mhz... 300 bucks too.

The ultrasonic tape measure is looking for the first reflected signal to calculate distance. I would ignore the first received signal because it would be from the glass.

I guess the only way to find out if it'll work is to just try it. If it doesn't work, It's only 50 bucks in parts total... if that.

I appreciate your help,
Matt

 

[FvM]

I think, you should start with gathering some basic knowledge about acoustics. Reviewing state-of-the-art regarding ultrasonic density measurement won't be bad as well.

Some comments about your considerations:
- a setup with transducer-air-glass-liquid-glass-air-transducer interfaces is simply doomed to failure. There are some points about acoustic impedances against it.

- even without detailed understanding of a suitable instrument design, you can calculate some sound speed and delta t numbers to get an idea of the required time respectively phase resolution.

- I assume that you'll only achieve reasonable results with frequencies above 1 MHz. Transducers for these frequencies are basically piezo ceramic discs operated in thickness resonance - with some supplements provding impedance matching and dampening. Piezo discs aren't too expensive, but most likely not available at a DIY electronics shop around the corner.

 

[jpanhalt]

I suspected it might be something like that. While acoustics may be fun to do for the challenge, I would recommend using refractive index. Specific gravity and refractive index are actually quite closely correlated for a given system. Did you know, for example, that the urine specific gravity that gets reported to your medical record is very likely done by refractive index?

Measuring RI is quite easy and easily automated. If you want to run a correlation with specific gravity, I would measure that with a float hydrometer and generate my own calibration curve.

John

 

[FvM]

Another interesting density measurement method has been presented by Endress+Hauser. They extended the capabilities of the classical vibrating tuning fork level switch to density measurement.
Endress+Hauser: flowmeters, level switches, pressure transmitters, pH, process automation

 

[ILetSmokeOut]

Funny you should mention the refractive index John. I use a handheld refractometer to measure the gravity of my beer. It only works before the beer has alcohol in it. As soon as it starts fermenting, I need to correct for the alcohol. I use a calculator that takes the original gravity and the current reading. I only use a few drops to get a reading. I don't know how well light would travel through 1 foot of dark beer. I'm not sure that would be easy. Oh, and UV light is bad for beer, it makes it taste bad... thats why bottles are brown.

I just ordered a set of ten 300kHz peizo discs from Steminc.com for $31 shipped. STEMINC - SMD07T02R412WL - Piezo Electric Ceramic Disc Transducer

It'll be fun to play with them even if it doesn't work for my application.

Matt

 

[jpanhalt]

RI is rarely done with UV. By far the most common wavelength used is 589 nm (sodium D line). Near IR absorption is another option to consider and is gaining increased recognition as a method of choice for process control of liquids. Good luck with the piezo's. Please keep this thread updated. I like good beer.

John