Question: Differential Pressure sensor set-up

[mjohn87]

Hey guys. So I would like to be able to measure the amount of beer in a corny keg and I can't really use a flow meter. I was reading up on this and differential pressure sensors seem like a good alternative.

From what I understand the differential pressure sensor basically has two inputs. One tube would go to the bottom of the keg, below all of the liquid, and the other tube would be above liquid so it can sense the difference in pressure.

I figure the easiest way to set this up would be to tee off the gas in and liquid out posts. The gas in would be connected to the co2 regulator/tank as well as an input to the first diff pressure sensor. The liquid out post would be connected to the dispensing system and the second input to the diff pressure sensor.

Does this sound like it makes sense? Any issues with this possible set-up?

 

[Stealthcruiser]

So, you would be measuring the pressure differential at the bottom of a column of liquid, vs. the "air" pressure in the headspace?

Don't sound right to me, but maybe some Engineer types will sound off.

 

[JayMac]

while differential pressure could work, it won't always be the most accurate.
a) you won't get the sensor at the VERY bottom of the keg
b) the density of your beer will fluctuate, due to differences in alcohol content and residual sugars

The pressure exerted on the sensor from the liquid will be rho*g*h, where h is the height of the liquid, g is gravity, and rho is the density of the liquid. As you can see, rho and h are where we run in to problems.

The most accurate and easiest way to determine the level in the keg is to create a sightglass on the side.

I'm sure there are ways around the problems I suggested, but I'll leave that up to people with more experience than myself.

Good luck!!

 

[FuzzeWuzze]

Unless you have a digital regulator or something that is super precise i dont see how this could work, what pressure are you comparing it too? Atmosphere?

The entire keg is under pressure and the crappy analog regulators everyone uses are hardly precise...

 

[big_len]

Specific gravity is the density of the fluid, I would hope most people know this when they keg their beers ! If your specific gravity is 1.010 then the density if 1010 kg/m^3.

The pressure of the gas is also not a problem, you don't need to know it at all. This is the reason for using a differential pressure transducer. Since the pressure of the gas is acting both on one side of the sensors diaphragm and the surface of the fluid, which is in turn acting on the other side of the diaphragm, then the reading of the sensor is purely a function of the height of the fluid.

A quick calculation, assuming a max height of fluid of around 0.5 of a meter would indicate that you need a sensor with a range of approx. 5000 Pascals.

Although perfectly feasible, it seems awful awkward...

EDIT: Also, the transducer does not need to be installed directly into the base of the keg, it only needs to be at the same height as it - tubing inbetween does not matter in this case.

 

[The10mmKid]

How about a weight scale?
Seems you could find something with a remote readout pretty cheap.

'da Kid

 

[processhead]

At the water utility where I work, we use pressure transmitters to measure the amount of water in large reservoirs .

In order to get an accurate reading, the transmitter has to be physically located at, or below the lowest point in of the container and it has to be directly connected to the container at this low point, not through connection like a keg post.

Industrial pressure transmitters are sufficiently accurate for measuring liquid levels in large containers, but not a keg sized container.

10mm is closer to a practical solution. By weighing the keg and converting the level you could track the kegs contents. (after factoring in the tare weight of the keg)

For me, I would just keep drinking till the keg blows CO2, and then change it.

 

[jhoyda]

How about a weight scale? Seems you could find something with a remote readout pretty cheap. 'da Kid

My suggestion as well. Weigh it empty, weigh it full. Linear scale from there. Readout could be a simple % remaining.

 

[kenwells]

I am currently building a web interface for some kegerators at work and I have done some research on different methods for determining keg volume.

I have not found a pressure sensor, like the kind used in bathroom scales, that is anywhere near good enough to be used for this application. There seems to be too much variance in the sensor outputs to make this work. The reason bathroom scales work is because they go through a zeroing process whenever you turn them on. It takes a few seconds for them to read 0.0lbs. My colleague and I found that the output resistance/voltage can vary by hundreds of percent from cycle to cycle. When used in a fridge, where the air temperature can vary by several degrees as the fridge cycles, and where the keg is always sitting on the sensor, you will not be able to make a zero measurement and you will not get a good reading.

I have settled on the far more expensive option of Swiss Flow SF800 flow meters. They are currently $60/ea + $20 shipping to CA, USA, but they don't produce much foam and they are very accurate. My initial tests showed roughly 170.5 pulses per oz. of water poured (water for testing). So I can get down to an accuracy of roughly 170th of an ounce. Even CO2 bubbles floating past the sensor cause this thing to tick, so some filtering is necessary. I have 3 taps for this project, so three flow meters. An Arduino Uno R3 is my hardware interface to the flow meters and temperature sensor (and, hopefully soon, a CO2 tank pressure sensor). Check out https://github.com/kencodes/reducedKegbot for my codebase, and https://github.com/kencodes/reducedKegbotWeb for the web interface.

 

[day_trippr]

fwiw, a bunch of us have been doing something similar running RaspberryPints with excellent results...

Cheers!