Taking a shot at specific gravity measurements

I get what you mean but I plan to make sure I always use the same volume by pumping more than the container can hold so the excess overspills. By example I mean, if you get a 1 litre container and pour 2 litres into it, there's still only going to be 1 litre in the container. By taking the measurement straight away and then flushing it should that not solve the issue of having to have exactly the same volume each time and by allowing a few seconds for it to settle and then taking the measurement straight away would that not prevent evaporation from having a noticeable influence? It might take some calibration to set up but I would think it was only required once.

The way you are expalining it now might work. Looks very reasonable to me if you keep volume changes.
I've been thinking there might be a correlation between the liquid specific gravity and its conductivity... After all, changes in the liquid density are related to changes in its contents; that is its chemical composition. Even if the relationship is not linear the mathematical equation can be obtained (not hard to do I've done it several times) and linearize the thing in the soft to extrapolate the specific gravity from the measured conductivity. You can to research that. Probably someone tried that already. If you can't find nothing some experimentation may yield some measurement values we can work with. Conductivity measurements are way easier to perform and it can be monitored continuosly. I have doubts about the cleanliness of the eletrodes ,etc to be suitable for your particular case though. :grin:

What about liquid opacity or its ability to transmit light. Does it change during the brewing? I've never done any brewing; but from the experience of consuming its product, I guess so.., :smiley:
If you set a glass tube parallel to the main container similar to a level indication tube and then an optical system with an LED and a Light depending resistor using infrarred maybe you can meassure the opacity of the liquid and also correlate it with specific gravity. Again taking measurements we can find the equation correlating them and extrapolate specific gravity from the light transmission measurement. This will leave the electrodes corrosion and cleeanliness problem out that the conductivity method will create. The tube just need to be long enough for the sensor to be attached to it and have a glass window for the light to pass through . In that way you won't have ambient light affecting your brewing. It wil require maintenace (cleaning) from time to time; but that's minimal.
You will need to set up the system and run a test brewing where you log in the PIR sensor readings and meassure specific gravity with a real bulb from time to time. You post here the samples and we will find the equation. It may not matter that much if the opacity changes with other variables as the mathematical equation found will remove them out. It will correlate just the variables measured. Several trials maybe required.

Optical sensing:

I like your thinking but it wouldn't work I'm afraid. There can be any amount of gunk in there and opaqueness can vary wildly, if you make a sugar was it can be all the way up to 1.090 (strong pre-ferment) and yet be almost as clear as water, while if you were using a coffee malt it would be really dark down at 0.90 (fermentation complete).

Conductivity:

I'm not so sure about this one but I have heard it mentioned before. I saw a discussion which went into greater lengths about using capacitance as the indicator, treating the wash as the dielectric. It was ultimately determined that the ingredients were so variable, with so many different types of crap making up the mash, that it couldn't be done in any kind of predictable way. I suppose it's likely that conductivity might go the same way.

I'm sure there must be some way of just dropping a probe into the wash to determine the SG but buggered if I know what it is. I remember something like a Kickstarter project to create a beer computer and they were going to release a cheaper model and then a more expensive model later on that would be able to read the SG and it would just be this little green box. They didn't give anything away about they planned to take the reading. I've just been googling it now and I can't find it.

The light meassuring thing can be zeroed at the begining depending on the original color of the liquid. On the other hand the fact that a human can't detect differences it does not necessarily mean the meassuring system can't also. For example CO2 is transparent to humans; but there are CO2 detectors working on that principle outthere. They use a specific infrared wavelenght that's absorbed by CO2 only. I guess the CO2 content of the liquid increases with time as the yeast works. ???. maybe you can monitor CO2 content with such sensor which I belive is used in fire alarms and smoke detector. I think this is not so hard to try compared with other methods suggested like my floaters for example. I would give a simple trial and see what happens. We maybe surprised by the results...
Check densitometers and spectrophotometers.

How do I upload a picture here?

On a much larger scale, but at the refinery I worked at when they needed a SG measurement of some kind of composite fluid in a vessel, column, or tank, they would would often take a differential pressure measurement between two known points of a fixed vertical separation, that were at all times filled to least the top tap or higher. So often these measurement taps were located at the very bottom of the tank. As the separation pressure taps of the liquid being measured was a constant then the created differential 'head pressure' measurement would reflect the SG of the liquid. So if the tap separation was 12" and the liquid was pure water the pressure reading would be 12" of H20 pressure, and any deviation higher or lower in that 'head pressure' measurement reading reflected the SG of the actual liquid in service. Worked well. It did assume that the components of the composite liquid didn't naturally stratify, as in say a water oil mix where the water would sink and the oil would rise in a tank. Although in that situation there is a way to use a fixed separation differential pressure measurement that could tell you where in the tank the actual water-oil point was, called the 'interface level', that only assumed the vessel/tank was 100% liquid filled and that the fluid comprised of only two different components of known SG. It's was all done via how one calibrated the measurement setup.

Lefty

![](http://C:\Users\Adrian\Desktop\Floatres SG Sensor)
Tryibg to upload a picture here. 1,2,3 testing...

Didn't work... :relaxed:

Hang on Lefty, are you just saying that SG can be determined simply by having two pressure sensors take a differential reading from fixed points? Because that would be much more simple (though far less sexy) than my idea. I could even attach the sensors to rod or something, the way I could move them from one bin to another providing I used the same type.

Twinnie:
Hang on Lefty, are you just saying that SG can be determined simply by having two pressure sensors take a differential reading from fixed points? Because that would be much more simple (though far less sexy) than my idea. I could even attach the sensors to rod or something, the way I could move them from one bin to another providing I used the same type.

That is one method used in the process control industry, so yes. However one usually uses a single differential pressure sensor instead of two psig single port pressure sensors as the measurement range is usually small and using two different sensors would bring in calibration variations into the differential measurement. Again this method only works if the liquid level is always higher then the higher pressure tap, and there is no stratification of the composite liquid being measured. Also your sensor must be rated for liquid service and many of the inexpensive pressure sensors are for air/inert gas service only, so check the datasheet carefully.

I could even attach the sensors to rod or something,

Immersing the complete sensor(s) assembly including lead wires leaving is rather unorthodox and probably fraught with problems? Pressure taps are usually drilled into the side of the the vessel being measured. Both sensor leads should route down below the lower pressure tap level such that any trapped vapour bubbles will migrate up and back into the process being measured.

Lefty

Hang on Lefty, are you just saying that SG can be determined simply by having two pressure sensors take a differential reading from fixed points? Because that would be much more simple (though far less sexy) than my idea. I could even attach the sensors to rod or something, the way I could move them from one bin to another providing I used the same type.

To give you a feel for how that could work in a home brewing setup, let's say your two pressure taps are about 30 inches apart. The differential pressure would be about 1 psi. If your SG changed by 0.05, you would see a differential pressure change of about 0.05 psi. Could work nicely with the proper differential pressure transducer.

i'm not planning on just dunking them into the solution, what i'll probably do is have a tall curved piece of plastic made of a quartered pipe and a straight piece of plastic with holes drilled in and the pressure sensor nozzles pushed through. Then I'll have them glued together and the bottom glued up and weighed down before putting some kind of hook on the top so I can hang it off the side of a barrel. If I do it that way then I can probably calibrate it for each barrel and move it around. The only complication remaining is working out which sensors to use. I can only assume that beer isn't corrosive since you can leave it in a plastic bottle for years and nothing will happen so I'm planning on getting any that are suitable for water but I need ones that will be accurate enough and I don't want to pay loads. There's hundreds on RS Online so I'l just have to spend some time going through them.

Twinnie:
i'm not planning on just dunking them into the solution, what i'll probably do is have a tall curved piece of plastic made of a quartered pipe and a straight piece of plastic with holes drilled in and the pressure sensor nozzles pushed through. Then I'll have them glued together and the bottom glued up and weighed down before putting some kind of hook on the top so I can hang it off the side of a barrel. If I do it that way then I can probably calibrate it for each barrel and move it around. The only complication remaining is working out which sensors to use. I can only assume that beer isn't corrosive since you can leave it in a plastic bottle for years and nothing will happen so I'm planning on getting any that are suitable for water but I need ones that will be accurate enough and I don't want to pay loads. There's hundreds on RS Online so I'l just have to spend some time going through them.

That might work indeed and save you having to drill pressure taps into the barrels. As far as using a sensor rated for just water, who knows, however you also that to research the temperature range of the sensor as I assume this measurement is continuing during the 'cooking' process. You will want pretty low range differential sensor that is for sure.

Lefty

For measure the SG attach a very light plastic
In the top of the hydrometer use a ultrasonic
Distance sensor and determinate the SG using
The de aviation in the variation of the buoyancy

There have been many ideas expressed here, but I think the simplest and most robust solution is using a submerged mass suspended by a string as suggested by liudr. If I was making my own beer (I wish I was) it's the method I would use. Here are a few of the benefits:

  1. If you want to leave it there all the time, you can.
  2. It's easy to move to another tank if desired.
  3. It will be immune to foam on the surface or submerged particles.
  4. If you suspect bubbles have formed on it, just lower it until it hits bottom to knock them off.
  5. If you suspect the liquid is stratified, simply take measurements at different levels to verify.
  6. The weight and volume of the mass is not important, as long as it's large compared to the string.
  7. It requires exactly 1 calibration ever - by taking 1 measurement in plain water.

I noticed no one posted any equations to calculate specific gravity. Not even those who claimed to have an understanding... so I'll post equations for liudr's method only:

float mAir;       // mass suspended in air (load cell measurement)
float mWater;     // mass suspended in water (load cell measurement)
float mBeer;      // mass suspended in beer (load cell measuremet)
float bfWater;    // buoyant force in water
float bfBeer;     // buoyant force in beer
float sgBeer;     // specific gravity of beer

...

bfBeer = mAir - mBeer;
bfWater = mAir - mWater;

sgBeer = bfBeer / bfWater;
  1. It requires exactly 1 calibration ever - by taking 1 measurement in plain water.

I would want to make that one calibration with the water temperature at the same temperature that you will be cooking the process at.

Lefty

What do the pros do? Maybe a simplified (or cheapified) similar approach?

JavaMan:
There have been many ideas expressed here, but I think the simplest and most robust solution is using a submerged mass suspended by a string as suggested by liudr. If I was making my own beer (I wish I was) it's the method I would use. Here are a few of the benefits:

  1. If you want to leave it there all the time, you can.
  2. It's easy to move to another tank if desired.
  3. It will be immune to foam on the surface or submerged particles.
  4. If you suspect bubbles have formed on it, just lower it until it hits bottom to knock them off.
  5. If you suspect the liquid is stratified, simply take measurements at different levels to verify.
  6. The weight and volume of the mass is not important, as long as it's large compared to the string.
  7. It requires exactly 1 calibration ever - by taking 1 measurement in plain water.

I noticed no one posted any equations to calculate specific gravity. Not even those who claimed to have an understanding... so I'll post equations for liudr's method only:

float mAir;       // mass suspended in air (load cell measurement)

float mWater;     // mass suspended in water (load cell measurement)
float mBeer;      // mass suspended in beer (load cell measuremet)
float bfWater;    // buoyant force in water
float bfBeer;     // buoyant force in beer
float sgBeer;     // specific gravity of beer

...

bfBeer = mAir - mBeer;
bfWater = mAir - mWater;

sgBeer = bfBeer / bfWater;

This thread, http://arduino.cc/forum/index.php/topic,66603.0.html, from the old forum, discusses the hanging weight approach.