Sciencemadness Discussion Board

Repairing or Constructing a Liquid in Glass Thermometer

earpain - 12-5-2020 at 05:29

So I broke my only two lab grade Liquid-in-Glass thermometers in one day. One was Hg, the other red dyed alcohol.

Although most would likely order replacements and/or explore electronic equivalents in the meantime, I'm somewhat savvy with glasswork and did some research.

Is the Hg inside a glass thermometer sealed in just inert gas, or under vacuum, or under pressure?
Quote:



When the partly made thermometer is cooled, the quicksilver fills the tube and bulb below the upper bulb. The lower bulb is then heated until the mercury expands and runs into the upper bulb, and just as the column of mercury begins to go back in cooling the glassblower hermetically seals the tube just below the upper bulb, which is then drawn off". This leaves the mercury in a vacuum in the tube, and the thermometer is ready to he graduated.

Trumbull White, 1902


The Hg thermometer broke right around the 100C mark. Experimentally, I noticed that it still seemed to work ok. I tested in boiling acetone and it read about 6C too high.

Then every modern reference to both Hg or the thus-far inescapably inferior non-toxic equivalents, made no mention of the internal chamber's pressure being above, below, or at Atmospheric.

https://themcaa.org/industry-insight/standards/
https://themcaa.org/wp-content/uploads/filledsystemthermomet...
https://themcaa.org/wp-content/uploads/ligthermometers.pdf

I ended up sealing the barely-visible opening to the capillary column of the Hg thermometer by fusing the opening with a thin borosilicate rod, heated via propane torch. Though it only goes up to 90C, and is now off by ~6C, it still seems to work.

I also have evaporated away all of the red alcohol in an almost full length lab thermometer, and i have two capillaries, with temp index paper still glued on, grinded off from candy thermometers. The metallic bulb I'm almost certain is not the actual reservoir for the blue colored indicator liquid, but just a highly thermally conductive secondary chamber.

And, i actually have some lengths of suitable boro tubing. I'm not afraid to perform the procedure outlined in the 1902 text, though it would take a month for it to equalize before I could calibrate it.


Sulaiman - 12-5-2020 at 09:16

There is some interesting reading here
Attachment: R133-e02_Thermometers.pdf (79kB)
This file has been downloaded 293 times

earpain - 12-5-2020 at 12:23

Quote: Originally posted by Sulaiman  
There is some interesting reading here


Thank you! I think that answers my question, along with all of the questions I was going to think of later.

In summary:
I infer that the organic fluid, aka alcohol thermometers hold the liquid in above atmospheric pressure. Precisely just like automobile brake fluid.

Hg thermometers, it won't make a huge difference if the chamber is in vacuum, at atmosphere, or above. The expansion will still be fairly linear.

I have changed my mind about chemical vs. electronic temperature devices. Honestly I don't see any real advantage to chemical devices in today's world, as long as the reference parameters are the same, electronic thermometers are easier to make, and less likely to break.

Mercury in glass is more classy and glamorous though.

Chemetix - 12-5-2020 at 15:50

To make a thermometer using a liquid it might help to note that the volume change with temperature of a substance usually does so at it's third decimal place. So I means that the main volume of the liquid needs to be 1000 times larger than the volume where the scale is marked.

A lab grade thermometer that is only 6mm or so in diameter has a small bulb and a correspondingly small volume in the scale, the only glass to be used in this application is precision bore capillary tubing.

The use of vacuum will prevent tiny air bubbles sticking to the surface of the glass and changing the reading. This is what I think is happening to make the thermometer lose its calibration. The other thing is that air inside the thermometer will pressurise and create a non linear expansion of the liquid. I'd think that effect would be tiny but still, a vacuum is best.

earpain - 12-5-2020 at 20:59

Quote: Originally posted by Chemetix  
To make a thermometer using a liquid it might help to note that the volume change with temperature of a substance usually does so at it's third decimal place. So I means that the main volume of the liquid needs to be 1000 times larger than the volume where the scale is marked.

A lab grade thermometer that is only 6mm or so in diameter has a small bulb and a correspondingly small volume in the scale, the only glass to be used in this application is precision bore capillary tubing.

The use of vacuum will prevent tiny air bubbles sticking to the surface of the glass and changing the reading. This is what I think is happening to make the thermometer lose its calibration. The other thing is that air inside the thermometer will pressurise and create a non linear expansion of the liquid. I'd think that effect would be tiny but still, a vacuum is best.



Yes indeed looking at a cross section of the Hg thermometer shows that the capillary column itself as about the same diameter as the hole that the propane in my torch venturi's out of to suck in ambient air and burn hotter, much skinnier than most sewing needles, barely visible.

For non Hg thermometers I thought the same things you did, however the much bigger issue is: evaporation

Buy having the bulb, the column, and the expansion/extra reservoir on the top be sealed at high pressure, this raises the boiling point of the alcohol or blake fluid like substance, allowing for it to reach much higher temperatures.

All of the current texts clearly state an inert gas such as N2 is crucial. This makes sense to me, for the same reason that vacuum pumps are used to fix refrigeration systems. O2 is corrosive, but much worse H2O vapor creates a chaotic system sealed hermetically.

If you ever have the problem of air bubbles creating a gap in the indication line, you can dip the bulb in dry ice for a moment, and then gently tap the thermometer, upside down, against a piece of rubber or something similar. Similar effect happens with a syringe.