Pycnometer. An unusual device for measuring density, specific gravity

[taricha]

Density/specific gravity measurements of aquarium water can make people a little crazy because if you have two devices they are going to give you different results. So you get a third, or buy a calibration standard - and the "which method is most trustworthy?" game goes on. Part of the frustration is that it's literally the most fundamental parameter of our salt water - how much salt is in it?

I ran across a pycnometer for the first time a few months ago. And I eventually got one (two actually - 100mL and 10mL) to play with. In theory, a pycnometer and a good scale ought to give you a fundamental and highly trustworthy way to measure specific gravity.

It's not such an unusual an apparatus - it's just a special kind of volumetric flask made so that at a given temperature it can only possibly hold one volume of liquid. My 100mL pycnometer has a "certified volume" of 99.727mL at 20C.

So you can either work at that specific temp, and base your calculation off that certified volume, or - my preference is to just have tank water, distilled water and the pycnometer all at the same temp (water bath) and use specific gravity instead of density. For density I need exact volume - which means I need to know the temp and use the certified volume or measure it myself. For specific gravity, I just need a repeatable volume, distilled water and aquarium water at the same temp, and a scale. Since SG is fundamentally just a ratio, it cancels out a lot of measurement errors like if the flask volume is slightly off.

Here's how you fill it. You add liquid to the neck and then push the ground glass top stopper in place, it forces excess water up through a capillary and overflows, then you wipe off all excess. So the volume of liquid contained is pretty exact, and very repeatable.

Video of the filling process here

I wanted to see how good the precision was using this 100mL pycnometer, So I checked to see if it could measure the increase in salinity due to evaporation over 6 hours.

I took a few water samples over 6 hours, sealed them and put them in a tap water bath at room temp (24C) along with a sample of distilled water. after the last sample sat in the water bath for over an hour, I used the pycnometer to measure all the samples on a scale.
If the volume and temp are the same - specific gravity is simply the ratio (mass of sample / mass of pure water). So that's all I need.

I actually used two scales - the best one I have access to, goes to 0.0001g and a very not fancy centigram scale, goes to 0.01g.
Here's the measured masses in grams.



The measurements under blue header are with the fancy scale, and the ones under the red "cg" headers are the centigram 0.01g scale.

So here's what specific gravity looks like - simply taking the aquarium water mass measurements and dividing by the average of the distilled water mass measurements.

Interestingly, the cheap scale and the really good scale gave basically the same results, which means that the scale is probably not the limitation to the precision. It's how repeatable the water volume is.

So the water started around a SG of 1.0271-1.273 and finished at 1.0274-1.0276 which is a pretty nice result for a $20 piece of glassware.

Of course, If you just wanted to quantify evaporation, there are easier ways...like a ruler. Water level in the sump dropped 8mm.

 

[KrisReef]

The first thing I thought; Is that a Pinko meter or a Pinch o meter?

I don’t think I have ever heard of this measurement tool before?

 

[Dan_P]

Density/specific gravity measurements of aquarium water can make people a little crazy because if you have two devices they are going to give you different results. So you get a third, or buy a calibration standard - and the "which method is most trustworthy?" game goes on. Part of the frustration is that it's literally the most fundamental parameter of our salt water - how much salt is in it?

I ran across a pycnometer for the first time a few months ago. And I eventually got one (two actually - 100mL and 10mL) to play with. In theory, a pycnometer and a good scale ought to give you a fundamental and highly trustworthy way to measure specific gravity.

It's not such an unusual an apparatus - it's just a special kind of volumetric flask made so that at a given temperature it can only possibly hold one volume of liquid. My 100mL pycnometer has a "certified volume" of 99.727mL at 20C.

So you can either work at that specific temp, and base your calculation off that certified volume, or - my preference is to just have tank water, distilled water and the pycnometer all at the same temp (water bath) and use specific gravity instead of density. For density I need exact volume - which means I need to know the temp and use the certified volume or measure it myself. For specific gravity, I just need a repeatable volume, distilled water and aquarium water at the same temp, and a scale. Since SG is fundamentally just a ratio, it cancels out a lot of measurement errors like if the flask volume is slightly off.

Here's how you fill it. You add liquid to the neck and then push the ground glass top stopper in place, it forces excess water up through a capillary and overflows, then you wipe off all excess. So the volume of liquid contained is pretty exact, and very repeatable.

Video of the filling process here

I wanted to see how good the precision was using this 100mL pycnometer, So I checked to see if it could measure the increase in salinity due to evaporation over 6 hours.

I took a few water samples over 6 hours, sealed them and put them in a tap water bath at room temp (24C) along with a sample of distilled water. after the last sample sat in the water bath for over an hour, I used the pycnometer to measure all the samples on a scale.
If the volume and temp are the same - specific gravity is simply the ratio (mass of sample / mass of pure water). So that's all I need.

I actually used two scales - the best one I have access to, goes to 0.0001g and a very not fancy centigram scale, goes to 0.01g.
Here's the measured masses in grams.



The measurements under blue header are with the fancy scale, and the ones under the red "cg" headers are the centigram 0.01g scale.

So here's what specific gravity looks like - simply taking the aquarium water mass measurements and dividing by the average of the distilled water mass measurements.

Interestingly, the cheap scale and the really good scale gave basically the same results, which means that the scale is probably not the limitation to the precision. It's how repeatable the water volume is.

So the water started around a SG of 1.0271-1.273 and finished at 1.0274-1.0276 which is a pretty nice result for a $20 piece of glassware.

Of course, If you just wanted to quantify evaporation, there are easier ways...like a ruler. Water level in the sump dropped 8mm.

Haven’t used one since P. Chem lab. The experiment for the day was developing a standard curve for BSA concentration v density in distilled water at a constant temperature.

Thanks for the memory!

 

[Randy Holmes-Farley]

Nice, taricha!

Here’s another thread on this topic from years ago

Salinity determination by density

The subject of various alternative means of determining salinity previously came up in this thread where I dabbled (in what I now see was an amusingly over-simplified way) with the idea of using a couple of simple tools to determine salinity by calculating the density of your tank water. I have...

www.reef2reef.com

 

[Wasabiroot]

Good stuff as usual. Thx for your contributions to the hobby!

 

[taricha]

I didn't actually give the step by step for how I could use this to get an accurate specific gravity from my tank water fairly easily.

mass the pycnometer while it's dry = m_pyc_dry
Have a bottle of distilled water sitting in the sump for at least an hour to get same temp.
When I'm ready to check, I'll swirl the pycnometer in the sump water to get the glass to the same temp.
pour in distilled water, and mass the pycnometer with distilled = m_pyc_distilled
pour in the sump water, and mass the pycnometer with the sump water = m_pyc_sample

Then subtract the mass of the pycnometer
mass distilled water = m_pyc_distilled - m_pyc_dry
mass aquarium water = m_pyc_sample - m_pyc dry

and divide
specific gravity = (mass aquarium water / mass distilled water)

 

[taricha]

Nice, taricha!

Here’s another thread on this topic from years ago

Salinity determination by density

The subject of various alternative means of determining salinity previously came up in this thread where I dabbled (in what I now see was an amusingly over-simplified way) with the idea of using a couple of simple tools to determine salinity by calculating the density of your tank water. I have...

www.reef2reef.com

Yep. I love that thread by Jim.
It partially influenced me to stop thinking in density and deal in specific gravity instead.
One point of that thread (for me) is that if you really want to go to a digit past 35 to 35.X ppt or past 1.026 to 1.026X SG, and base it on fundamentals (as opposed to trusting a hydrometer, refractomer etc), then you have to take extraordinary care with measurements.

The part of this that the pycnometer helps with is the super-repeatable volume.

Jim's thread covers how to get that level of repeatable volume from a volumetric flask in post 24 (it's great).

The next technique for refining the accuracy and precision of this method involves doing everything possible to fill the flask consistently with the same volume of water, both during calibration and when measuring the tank water density. There are two aspects to this. One is really just basic good measurement practices when reading the meniscus, but I will now elaborate on the specifics of how to do this. The other addresses a small but very real variability that can be minimized when filling the flask....

 

[taricha]

I didn't actually give the step by step for how I could use this to get an accurate specific gravity from my tank water fairly easily.

mass the pycnometer while it's dry = m_pyc_dry
Have a bottle of distilled water sitting in the sump for at least an hour to get same temp.
When I'm ready to check, I'll swirl the pycnometer in the sump water to get the glass to the same temp.
pour in distilled water, and mass the pycnometer with distilled = m_pyc_distilled
pour in the sump water, and mass the pycnometer with the sump water = m_pyc_sample

Then subtract the mass of the pycnometer
mass distilled water = m_pyc_distilled - m_pyc_dry
mass aquarium water = m_pyc_sample - m_pyc dry

and divide
specific gravity = (mass aquarium water / mass distilled water)

Example:
Dry pycnometer mass: 41.1473 g
Filled from distilled water bottle from sump: 140.4799 g
filled from tank water from sump: 143.2355 g

distilled water = 140.4799 - 41.1473 = 99.3326
Tank water = 143.2355 - 41.1473 = 102.0882

specific gravity (at sump temp 26.5C) = 102.0882 / 99.3326 = 1.02774

 

[taricha]

Let's talk about this part "Certified Volume" of 99.727mL at 20C
They say this is done by massing purified water.

At 20C for me, I measure the mass of distilled water at 99.4503g and dividing by the known density at 20C (0.9982) I get 99.630mL or 0.097% too low.
My re-filling variations looked around 0.03%, and my temp probes all read within a 1/2 degree C, but if there were a 1deg C error around 20C, that changes density by 0.02%
My scale measures three different calibration 100g masses as all 99.997 +-0.0005 so not gonna find a tenth of a percent there either.


They list the equipment they use on the calibration certificate, and it's better than mine - so If I have to pick one, I guess I'd pick theirs.

Fortunately, I'm using specific gravity, so I don't really have to care about whether their certified volume or my measured one is closer, since I'm just doing a ratio of masses - the SG measurement doesn't depend on the exact volume.

 

[Randy Holmes-Farley]

Fortunately, I'm using specific gravity, so I don't really have to care about whether their certified volume or my measured one is closer, since I'm just doing a ratio of masses - the SG measurement doesn't depend on the exact volume.

That’s a good point. The accuracy of the volume mark is irrelevant.

There’s a small variation in sg with temp for a fixed salinity, which may need to be considered for very high accuracy, but as long as the seawater and ro/di are at the same temp, and that temp is in the 20-25 deg C range, this seems like a very good method.

 

[taricha]

Interestingly, the cheap scale and the really good scale gave basically the same results, which means that the scale is probably not the limitation to the precision. It's how repeatable the water volume is.

I wanted to revisit this....
Here's three days of me comparing my measurements on a higher quality scale that goes to 0.0001g vs a significantly cheaper scale that only goes to 0.01g.

The blue is the fancier scale and the red is the 0.01 scale. They read essentially the same. The grey lines indicate how much error there would be if the 0.01g scale read high in the distilled and low on the tank water by 0.01g, and vice versa.
Clearly the two scales agreed closer than that.

The point being a good scale that goes to 0.01g is plenty acceptable for this.

Also I checked these vs my swing arm hydrometer that I had marked years ago based on standards from the DIY salinity standard article. The agreement is very good.

So if you have a decent scale and want a salinity measure based on fundamentals (to check calibration of another device such as a hydrometer) this seems a very good option.

 

[Randy Holmes-Farley]

I wanted to revisit this....
Here's three days of me comparing my measurements on a higher quality scale that goes to 0.0001g vs a significantly cheaper scale that only goes to 0.01g.

The blue is the fancier scale and the red is the 0.01 scale. They read essentially the same. The grey lines indicate how much error there would be if the 0.01g scale read high in the distilled and low on the tank water by 0.01g, and vice versa.
Clearly the two scales agreed closer than that.

The point being a good scale that goes to 0.01g is plenty acceptable for this.

Also I checked these vs my swing arm hydrometer that I had marked years ago based on standards from the DIY salinity standard article. The agreement is very good.

So if you have a decent scale and want a salinity measure based on fundamentals (to check calibration of another device such as a hydrometer) this seems a very good option.

Thanks for the update!