Taking a shot at specific gravity measurements

I'm another one of those guys who wants to use Arduino to automate and monitor homebrewing. It's nothing too complicated, just manage a heater pad using a thermometer mostly, but I'd also like to monitor the specific gravity so I can check it out without opening it up. There's been a few threads on this but I haven't seen a single on where they came up with a good answer, mostly due to all the different types of gunk you'd get in a fermenting bin while it's doing it's business.

I figure the best way of doing it is to recreate the normal method of measuring the pressure with a flotation device. My plan is just to create a buoy with a pressure sensor on the bottom (eg. http://uk.rs-online.com/web/p/pressure-sensors/2355762/).

  1. There's a couple of issues here, the weight of the wire on the buoy is going to affect how deep it floats, it won't be super accurate but I'll just have to regulate it as much as I can.

  2. Bubbles may form under the buoy, I'm going to have to shape it so that bubbles don't get to bad, as long as they're relatively uniformed across the batches it should be fine. If I make the buoy big enough and put enough weight on it I'm hoping they won't throw it off too much.

I'll then have a wire hanging off the bottom of the buoy to drop the sensor down a bit and then attach a weight to the end of it. I'll calibrate it afterwards and I'm hoping this will give me semi-accurate readings? They don't need to be outstanding, just good enough to work with.

Can anyone spot an obvious flaw that I'm missing here? Will the pressure sensor be accurate enough?

Thanks.

You don't need a Pr Sensor to find the position.
Maybe...
Let's say you place the bouy support (tube) on the container cap. Then the bouy can slide vertically through that tube depending on the liquid voyancy.Then you place a rod with a ferrite section in the bouy shaft and windings in the tube coincident with the center of the windings at normal voyancy. Study LVDT (Linear Variable Differential Transformer). You can convert any displacement of the bouy shaft to electrical signal at the output of the LVDT. You can make your own LVDT by winding a wire on the tube (2 windings). There are specifics on how to do that of course. You need to study that. The advantage is the bouy will be a passive element with no wires and the electrical system will be on the support attached to the container cap and all that can be external if you make the bouy shaft long enough.
To make the sensor make to identical windings on the tube. One for excitation where you apply a high frecuency low voltage signal (let's say 10 KHz). The other winding is the pick up coil where you receive the signal. When the rod (ferrite position changes) the amplitude of the picked up signal changes. Then you do AM demodulation to recover the envelope with a simple diode and a low pass filter (a capacitor). The envelope is your voltage changing with the rod position and therefore voyancy. If you want you can go FM by using the inductance as a component to change the frecuency of your oscillator (study Clapp Oscillator). The ferrite movement will affect the inductance of the coil and then you can demodulate the signal by using a phase detector (XOR gate) or a PLL unit like the CD4076. You need to study that too. It will give you a voltage output dpending on the frecuency changes. You can amplify (using LM324 if required) and feed that to Arduino. The FM version will be more complicated but you will get better inmunity to noise.

The main problem I see is you will have to be Zeroing the stuff regularly as the position of the bouy will change not only with voyancy; but with the liquid level inside as it evaporates, etc... maybe others have an idea of how to solve that problem. If they don't come up with a nice solution you may end up with an evaporation meter instead. :grin:

Can you use a pressure sensor like that? The pressure sensors I'm used to using to read air pressure behave like differential pressure sensors and I don't think they'd work in this situation.

What sort of resolution do you need from the density results?

What range of specific gravity are you trying to measure? I am thinking about lowering a heavy object into the liquid (not touching the ground) and having a force gauge hold the string. Depending on specific gravity the force gauge reads different values.

Thanks for the replies guys, especially yours arduinoadrian, that's a good read. Like you said though, the problem with that solution is that it will require calibrating at the beginning of each brew and maybe constantly throughout. Admittedly this is more of a novelty than anything but I do genuinely want to make things easier for myself. After the mess and hassle of getting the fermentation process underway, and knowing the amount of cleaning I'd likely still have to do at the point in which I'm filling the fermentation bin, I'd rather not make things even more troublesome for myself. The main problem with these ideas is that they always struggle to ascertain the top of the water level.

As for the other questions, the SG range would have to be something like starting at 1.090 (in the case I decide to do a strong sugar wash) right down to 0.950 (so that it can help me spot rogue yeast infection). I haven't done the maths to work out the PSI that would convert to and I wouldn't have a clue without checking up first. I'm at work at the moment so I'll have to do this when I get home.

I have been thinking and maybe the surface level change problem can also be solved by using a differential method.

Let's say you use a second similar floating device with complete voyancy (floating at the surface of the liquid). If this second device is not affected by changes in the liquid voyancy then you can use it to measure just the surface level. Then by substracting the two signals the difference will be the difference in the voyancy of the liquid. To do that, both sensors outputs (after demodulation) can be fed to a differential amplifier ( easy to build, check the LM324 datasheet) and that will be your liquid voyancy signal to be send to Arduino analog input. That way you will not need to calibrate the system that much; but just at the beginning. I think you Project its completly doable that way.

Having two floating sensors does seem like the best way to handle the issue of changing levels, but I would have thought that all of these float based solutions would be rendered useless by bubbles forming on the submerged surfaces.

If you have a pump running at any point then one option would be to pump the fluid up a column with an open top so the column was completely full, and then measure the pressure at the base of the column. There would be some variations from surface tension, I suppose, but I would have thought that with care you could get consistent readings from it.

So what's wrong with my weighing a mass in liquid with force gauge method? Worse than having multiple floaters? You are talking about +-5%-10% change in SG so a decent force gauge can pick up the change easily. In intro labs, we use regular balances to find density of salt water.

I think the weight mass method you are proposing is good; but you will need to set a spring or something to make the thing retract when the weight goes up and calibrating that spring will be a difficult task. The floaters move freely. Floaters have been used forever to meassure liquids specific gravities and that maybe the reason why.
Anyway if the weight (which is acctually a floater too) can be attached to the weighting system without affecting the meassurement then maybe it is another solution to the same problem.
A basic rule in measurements is that the meassuring system used should not affect the system being measured.

The bubbles issue I think is not a problem as they don't form constantly in a stream or something like that. On the other hand the liquid voyancy is not a rapid changing thing. It will take hours for the yeast to do their business before it changes in a detectable way, therefore a bubble here and there will not affect the results I guess. If a bubble appears then you get a way off reading and you just need to way for a bit and take another reading. The soft can take care of that. A very slow changing parameter is being meassured therefore rapid changes will be noise and can be discarded.

arduinoadrian:
I think the weight mass method you are proposing is good; but you will need to set a spring or something to make the thing retract when the weight goes up and calibrating that spring will be a difficult task. The floaters move freely. Floaters have been used forever to meassure liquids specific gravities and that maybe the reason why.
Anyway if the weight (which is acctually a floater too) can be attached to the weighting system without affecting the meassurement then maybe it is another solution to the same problem.
A basic rule in measurements is that the meassuring system used should not affect the system being measured.

I'm not sure you understood my method. I said a heavy weight thus submerged. Something like steel or aluminum (is that alright in drinks?). You need not pull it down with springs. I don't think a floater can be easily attached to a weighing system. It's already balanced in its forces and designed to do so.

I don't get it. How wil the liquid specific gavity will affect your weight? Oh the other hand if you are weighting the liquid why do you need the weight? I understand the liquid weight changes; but the weight pounds will change with the liquid specfic gravity.
Sorry, you are right I don't understand the method.

liudr:
What range of specific gravity are you trying to measure? I am thinking about lowering a heavy object into the liquid (not touching the ground) and having a force gauge hold the string. Depending on specific gravity the force gauge reads different values.

Your weight does not touch the ground and you need to hold it with a string. A heavy object will not be affected by changes in the liquid voyancy which is what's being measured. Without the string it will sink. Once the object especific weight surpasses the espeific weight of the liquid it sinks. To do that the especific weights of the liquid and the object must be similar and that's exactly what they do with the gravity measuring bulbs when filling them with pellets of heavier metals. They are trying to match both to make the object float at a certain depth/ Soemting like subs. Please don't take me wrong we are just brainstorming here and we all learn from each other. That's all. Thanks :wink:

The heavy object is submerged but not touching the bottom of the container. That I hope is clear enough for everyone.

A heavy object will not be affected by changes in the liquid voyancy which is what's being measured

that's under the impression the object is sitting on the bottom or just not understanding what buoyancy force means? If liquid density (specific gravity) changes, buoyancy force changes. You should check some reference before "brainstorming". My understand of the word is that you think of some random stuff but still they have to be correct. Otherwise it's hard to differentiate it from bullshitting.

Bullshittinmg is posting without reading or understanding. Read my post before posting and yu'll see I mentioned exactly that.
By the way no one here voted fo your load cell anyway. No one liked it; don't take it pesonal, it happens...

arduinoadrian:
Bullshittinmg is posting without reading or understanding. Read my post before posting and yu'll see I mentioned exactly that.
By the way no one here voted fo your load cell anyway. No one liked it; don't take it pesonal, it happens...

Please stop spreading your misunderstanding about "voyancy". It's not even a word. No one needs to know that you don't know. Don't pretend to know what you don't. We see plenty that everyday, won't make you special. Who said I care about a vote? Did someone call for a vote or you just did that in your head?

Taking it personal the way you are doing when you have it wrong and nobody liked your idea, really makes you special; but not smart.

arduinoadrian:
By the way no one here voted fo your load cell anyway. No one liked it;

Didn't realise it was a poll. It's a perfectly feasible solution, and the only thing I don't much like about it is that it seems to require moving parts which I think would be problematic in a hygienic environment. And just about all the other solutions being suggested suffer from a similar problem.

You don't even know how buoyancy force changes with specific gravity and you are talking it and giving advice, so feel embarrassed and go google it. I don't even know you so how am I making it personal? Like by saying your name? I don't know that, just some @$$&9@& pretending to know some science. That is as personal as I may get. I don't even bother that much. All I want is for nonsense to stop. You happen to be that person spreading it. I can't stand it. You want to take it personal then it is your own problem. I can see that in a separate post you like "brainstorming" and tend to fight back with others unfavorable comments. Why don't you humble yourself with some wiki reading about buoyancy force and correct some spellings at the time? Too old to take unfavorable comments or too young to be polite?

PeterH:
Didn't realise it was a poll. It's a perfectly feasible solution, and the only thing I don't much like about it is that it seems to require moving parts which I think would be problematic in a hygienic environment. And just about all the other solutions being suggested suffer from a similar problem.

Peter,

Not a moving part. You lower a weight into the liquid and tie the string to a force gauge. The weight balances with gravity, buoyancy force and string. If specific gravity increases then buoyancy force increases and string force decreases. The force gauge reads it as a decrease. The weight stays immobile, no moving parts. We do this lab in intro physics labs to determine salt water density with a 0.1g accuracy balance. The result is not bad. With a floating meter you can read it but salt starts gathering on it and the reading is affected. I imagine similar effects with floaters inside the liquid.

Well, this other guy is doing a vote all by himself, apparently that is what he considers as a vote. I must have not learned my English well enough. Sounds like communist vote. One person votes :slight_smile: