EEVblog Electronics Community Forum

General => General Technical Chat => Topic started by: sarel.wagner on April 21, 2018, 07:19:45 am

Title: How to Determine boiling point from temp only
Post by: sarel.wagner on April 21, 2018, 07:19:45 am
There are various ways to electronically determine boiling point of water. Also there are various things that influence the boiling point of water like altitude, pressure variations, purity of water and water and other chemical (or binary) mixtures. Most of these are in general well understood.

There are many ways like acoustical detection, detecting steam, and so on. I am interisted in detecting it in the simplest possible way, (preferre by single parameter like temp) accurately and without measuring many factors like pressure etc. This is for a hobby project and because I find it stimulating as a first principle challenge.

I am not concerned by what the temp is at boiling, merely that it has started boiling. So all those other factors are not important to me and the fact that the liquid is boiling is the only parameter that is important.

As part of this, I am thinking that if we could detect boiling, we can also detect when it is about to start boiling. This may be important if we should want to prevent the onset of boiling. Absolute temp cannot be used as all those other factors will influence the boiling point temp.

My thinking is as follows:
Water or a chemical mixture with water will raise its temp at a constant rate with a constant energy input. Pure water or water and sugar (or coolant antifreeze) mix (binary mix) will raise its temp at the same rate for the same heat energy input. After it has been heating for a while it will get close to boiling point and then the rate of change of temp will have to slow down more and more as the liquid gets closer to boiling point. At boiling point the temp change will stall until the water has evaporated and only the antifreeze remains. With binary mixtures in general, evaporation of the lowest boiling point chemical in the mixture causes a change in boiling point and a slow rise in boiling point temp.

Stating as follows: Detecting pre boiling, onset of boiling and the subsequent slow change in temp as the mixture changes should be possible by measuring only temperature, or rather the rate of change of temperature.
Pre boiling is a constant rate of change when the liquid starts heating with a constant input of heat.
Slowing down of rate of change indicates a region close to boiling. Finding what this slow down rate of change is close to boiling can be determined I guess, ie. change over time.
When the rate of change for pure water stops or get very small we know water is boiling. We should now be able to keep it just at the point of boiling or at the onset by varying the heat input. We can also control the rate of boiling by increasing the heat input energy.
As water evaporates into steam from our mix with Ethylene glycol (antifreeze) the temp would change by increasing, also the rate of change of temp will now be increasing from zero or from a very small number.

Does this make any sense or is my logic flawed? (No sacred cows here  :o)
I want to be able to program a PLC with 4-20mA SCR to let me know at the onset of pre boiling (just before it boils) with an alarm and then keep the liquid at that temp for a amount of time allowing me to add other chemicals to the mix. After the addition the boiling point is now changed and the temp will again rise, when the rate of change slows down because we approach boiling again it must allow the liquid to just start boiling.
Title: Re: How to Determine boiling point from temp only
Post by: ebastler on April 21, 2018, 07:31:31 am
I think you are essentially describing Calorimetry, an established method to determine melting or boiling points.

Add a reference sample which you also heat (while measuring its temperature), and you have Differential Scanning Calorimetry. That enhaced method will also tell you about the amount of energy needed to melt or boil your sample, for further material characterization or identification. Google knows more about this technique, and quite a few vendors offer commercial instruments.
Title: Re: How to Determine boiling point from temp only
Post by: sarel.wagner on April 21, 2018, 07:39:54 am
I think you are essentially describing Calorimetry, an established method to determine melting or boiling points.

Thank you indeed, was not familiar with the term. This will however not be aplicable to my application, not looking for an instrument either. Also it will be expensive and not teach me anything.  :phew:
Title: Re: How to Determine boiling point from temp only
Post by: ebastler on April 21, 2018, 07:46:36 am
Thank you indeed, was not familiar with the term. This will however not be aplicable to my application, not looking for an instrument either. Also it will be expensive and not teach me anything.  :phew:

I didn't mean to suggest that you buy a commercial instrument. Just wanted to confirm that the general approach is known and makes sense -- and suggest that reading up on "calorimetry" may provide helpful hints for your own implementation.
Title: Re: How to Determine boiling point from temp only
Post by: Marco on April 21, 2018, 05:07:56 pm
After the addition the boiling point is now changed and the temp will again rise

You'll probably need a button to tell it to start ramping/measuring again, otherwise it will likely keep ratcheting up the temperature as it repetitively tries to see if the second derivative exceeds the trip point (it always needs a bit of run up).
Title: Re: How to Determine boiling point from temp only
Post by: GerryBags on April 21, 2018, 05:16:10 pm
Maybe you could use light transmission of the fluid to measure its density? Or wait for a certain number of interruptions of an IR beam caused by rising bubbles?
Title: Re: How to Determine boiling point from temp only
Post by: TerraHertz on April 22, 2018, 02:07:23 am
I think the problem with using calorimetry (rate of temp rise vs known energy input) to determine 'just before boiling', is that there's little or no flattening of the curve until the liquid actually starts boiling. It's the phase transition of liquid molecules to vapor that takes away the extra heat, making the remaining liquid stay right at boiling point temp.

Hmm... the only thing I can think of, is to measure the partial vapor pressure of water in the air above the heated water. That should start going up sharply just before the water really starts boiling.
Not sure how you'd measure that reliably. Humidity sensor? Dielectric constant of air?
Also it would be hard to compensate for barometric pressure, and rate of intermixing of the air in the apparatus, with outside air.
Title: Re: How to Determine boiling point from temp only
Post by: IanB on April 22, 2018, 02:31:09 am
As part of this, I am thinking that if we could detect boiling, we can also detect when it is about to start boiling.

Strictly speaking, no you can't. If you had a perfect system then there would be a sharp corner in the slope of temperature vs time at the onset of boiling.

I think you can detect boiling by making sure the liquid is well mixed (have a vigorous stirrer) so that it all is at the same temperature and it all starts boiling at about the same time. Then measure the rate of change of temperature, and when the temperature stops increasing the liquid is boiling.
Title: Re: How to Determine boiling point from temp only
Post by: GerryBags on April 22, 2018, 02:37:31 am
Erm..... I do know a device that detects the boiling point of water using one parameter. It's called a kettle. You could just use a bi-metal switch out of one.
Title: Re: How to Determine boiling point from temp only
Post by: sarel.wagner on April 22, 2018, 05:42:00 am
Erm..... I do know a device that detects the boiling point of water using one parameter. It's called a kettle. You could just use a bi-metal switch out of one.
I think that detects steam in reality, in time that would be way after boiling stated, no?

Did a quick experiment, logging every 15 sec with calibrated probe....
(https://i.imgur.com/WNHs7ZCl.png)

Fixed graph error
Title: Re: How to Determine boiling point from temp only
Post by: sarel.wagner on April 22, 2018, 06:02:47 am
And the last bit of the data around the boiling point:
There is a slowdown (small) just before temp stable at boiling point. My current calibrated probe does not log fast enough, sadly. Will upgrade soon.
 
Spent 7 Seconds at 94C, only 3 Sec at 93C.
17:12:06   88.00
17:12:08   89.00
17:12:09   90.00
17:12:12   91.00
17:12:13   92.00
17:12:16   92.00
17:12:16   93.00
17:12:19   93.00
17:12:19   94.00
17:12:23   94.00
17:12:26   94.00
17:12:26   95.00
17:12:29   95.00
17:12:32   95.00
17:12:35   95.00
17:12:38   95.00
17:12:41   95.00
17:12:44   95.00
17:12:47   95.00
Title: Re: How to Determine boiling point from temp only
Post by: tpowell1830 on April 22, 2018, 09:31:30 am
I have not done such experiments before, but, it seems to me that if you measure the temperature of the liquid, and measure the air temperature at some set distance above the liquid that you would get a 'knee' after each liquid started boiling. This would be influenced by the shape of the vessel in which you hold the liquid. Your 'air' temperature measure would have faster response if the the vessel had a smaller opening where the second temperature measurement is done.

I could be all wrong, just my 2 cents...
Title: Re: How to Determine boiling point from temp only
Post by: RoGeorge on April 22, 2018, 10:15:40 am
Temperature alone (without hidden assumption, like e.g. constant heat transfer from the heat source, or other hidden assumptions) doesn't work.

Can be done in a gazillion of ways, i.e.
The problem description is too generic. You need to give same technical specifications and application in order to get realistic solutions/answers.
Title: Re: How to Determine boiling point from temp only
Post by: ebastler on April 22, 2018, 11:09:51 am
Temperature alone (without hidden assumption, like e.g. constant heat transfer from the heat source, or other hidden assumptions) doesn't work.

Sure; but constant heat transfer from the aource IS a (not-so- hidden) assumption in calorimetry. As mentioned earlier, use differential meaurements on a second, known reference mass if you want to improve precision.

I think the discussion is veering towards unnecessarily complex (although creative) alternatives at the moment.
Title: Re: How to Determine boiling point from temp only
Post by: sarel.wagner on April 22, 2018, 11:31:43 am

The problem description is too generic. You need to give same technical specifications and application in order to get realistic solutions/answers.

Its a 10 Litre insulated sous-vid processing container with a 2000W heating element and mixer for experimental purposes, and my curiosity, (does darn good sous-vid). The main Liquid is always water and the heat up will be either water or water and one other chemical like salt or sugar or glycerin, hence the changeable boiling point. Other soluble chemicals will be added later. The heater will be controlled by 4-20mA SCR via PLC with PID available.

Ideally I want to use a single parameter to determine boiling and it needs to be reliable. I will do experimentation later on with various methods to determine what works best. I want to find out what methods may prove to be worthwhile. Standard ways unlike detection will not really work as the mix boiling point is not fixed like water, it varies depending on concentration, so a set Temp wont work.

Temperature is preferred as it is easy to use and integrate into an automated system. It is also easy as no need to modify the container. Camera could work as well, don’t know about steaming up and it needs to be water proof? The sound detection is well known, but difficult to tune and external uncontrolled events causing noises frequently ruins the detection.

I like this discussion as I find many things that I have not yet contemplated.  :-+
Title: Re: How to Determine boiling point from temp only
Post by: IanB on April 22, 2018, 11:55:23 am
As has been mentioned above, the way electric kettles work is quite reliable. They pipe steam from the boiling chamber down a long tube with a temperature sensor at the end. When the water is not boiling the temperature sensor stays cool. When the water starts to boil a large volume of steam is generated which flows down the tube and heats up the sensor. When the sensor reaches a set temperature it trips and shuts off the power to the heating element.

This method is reliable because it relies on the defining feature of boiling, namely the rapid evolution of vapor. If you were to use the same method, but replace the bimetallic strip of the kettle with a thermocouple and look for a rapid rise in temperature you would readily detect the onset of boiling. You could also get advance notice of boiling as the thermocouple started to warm up, and you could regulate the heat input to a simmer by controlling the thermocouple temperature a few degrees below the maximum temperature reached.
Title: Re: How to Determine boiling point from temp only
Post by: raptor1956 on April 22, 2018, 11:56:46 pm
If you heat water its temperature will rise until it reaches the boiling point and there it stays until all or most of the water boils away.  The boiling point is effected significantly by pressure as well as contaminants and under certain conditions water, if it is very pure, can rise, temporarily, above the boiling point.  If you have a very clean glass and fill it with very pure water then put that in a microwave oven for 15 seconds or so longer than it should take to boil water it can avoid boiling until disturbed and then it explodes.


Brian