Hi,

There are a couple of facts that you need to know. The rate of temperature drop depends on the mass and specific heat of the liquid, assuming that it is homogeneous throughout. The rate of heat loss depends on the surface area of the tank. The two statements above are linked by the heat transfer mechanisms whose mathematical model will need the thermal conductivity and viscosity values. If you want to loose heat, stirring the liquid usually helps. External air blast can sometimes be helpful. If you want to retain heat, then insulation is usually required. All the above items can be found usually in a good book on heat transfer, and is a bit cumbersome to answer here.

Regards Stephen.

Hello,

Thank you for the reply. Assuming I have the rate of heat loss, how would I find the temperature drop rate specifically at the top of the tank, not in the liquid?

First question: what does your sensor measure =

- Internal ambient or wall temperature ?

- Which is the fluid in the "empty" volume ?

The top temperature depends on :

- the internal free convection above the full volume and free convection of the fluid in the upper volume to external ambiance.

- conduction in the wall from the hot full half.

If the tank is not insulated the free convection at the outside of upper part will be function of the temperature of the lower part this one being hotter the temperature difference at the wall for external convection will be less since the air has been warmed by convection on the lower part.

As you see it is a VERY complex "cooling" and not very easy to quantify.

If you want a precise information put the sensor in the bottom part.

@Nick Hamburger Thank you for your reply. The sensor is a thermistor system mounted on the outside of the tank's top, which would measure wall temperature. The liquid is pressurised nitrous oxide where the tank has liquid and vapour nitrous phases. I don't necessarily need a highly precise value, all I need is roughly the time range taken for the top of the tank to drop in temperature for the sensor/thermistor to measure this change

Which material is used for the wall ?

How thick is the wall ? Which are the tank dimensions and its geometry ?

How far is the sensor system from the liquid level ?

In which range are the liquid temperatures ?

It could be possible that in such an arrangement your measures can be not far from the truth depending on the wall conductivity and the convection intensity as well inside as outside.

But if you have an external sensor why not place it at the bottom ?

Dear Nick Hamburger ,

I have the tank dimensions, it's just how to link them that I don't know. The purpose of this system is to use the temperature change at the top as a method of sensing the fill level of the tank, since the temperature would drop the more the tank is filled

I shall not be available for about a week but send answers to my questions and I will have a look to see the feasability of your approach. If the parameters are in the good range it can be quite Ok.I

Nick Hamburger

The tank wall is made of 6061 t6 aluminium, 0.1875" thick. Outer diameter is 6", height is x, where ambient temperature in the tank is T2 ( the goal is to find it in terms of x, T2 and T1). If the tank is half full with nitrous oxide, and my sensor/thermistor setup is at the top of the tank, how long would it take for the temperature at the top to drop from T2 to T1.

This would mean the sensor is (x/2) inches away from the liquid phase of the tank

as promised here is a path you can follow to solve your problem.

If you have more questions do not hesitate to ask for further explanations.

Nick Hamburger Thank you very much for your help!

Is every thing clear ?

Nick Hamburger Just to be sure, does φ represent heat transfer per unit area(flux)? And does that mean h is the gas' thermal conductivity (W/(m2K))?

Yes To jour 1st. Question.

It is the "convection coefficient " in W/(square m x °C ) for the second.

Other question ?

This is not an answer, just an observation, but can you gentlemen exchange e-mail or text addresses? If you need the privacy to use initials then it will be far more discrete.

Regards Stephen

Stephen Lloyd

Noted with thanks. The intention was to keep it there if someone had a similar question in the future

@ S. Lloyd

I see that you read with attention what do you think about my suggestions ?

I would appreciate your opinion.

Thanks