Estimating Hot Side Temp of Vessel

[TGHNylon]

I need some help. I have an insulated vessel for which I know the cold side skin temp to be 420F. The material is steel and it's 1 inch thick. The outside ambient temp is 80F. The vessel is insulated, but I can get a pyrometer reading on one of the exposed nozzles where it enters the vessel. I want to calculate the hot side temp.
Using Stefan Boltzman law I can calculate the radiation heat transfer from the cold side to ambient at around 1030 btu/hr assuming an area of 1 ft2. Also I can calcualte convection from the surface with q=hA(T1-T2) to be 560 btu/hr for the same one square foot. This combines to about 1590 btu/hr off of the cold surface. If I use simple conduction through the 1 inch vessel and assume the heat transfer off the cold surface is equal to the heat transfer through the vessel wall, q=kA(tT1-T2)/L I only get a temperature gradient of about 5F through the wall and I don't think that is correct. I am trying to get the temp at the top of the vessel, and the vessel level never gets anywhere near the top, so the hot side contacts only hot gases. Any help estimating hot side temperature is greatly appreciated.

 

[IRstuff]

Your radiated emission seems a bit high; I'm getting more like 880 BTU/hr, but overall, I don't see anything seriously wrong with your setup, although your convective coefficient seems a little high for natural convection.

Your radiative emission appears to have been calculated as a strictly one-way process, while, in reality, the ambient environment radiates back. This accounts for the extra 150 BTU/hr.

Don't forget that while the internal gas might be VERY hot, gaseous thermal conductivity is usually poor. If you continue the thermal transfer back to the gas, you should get a very high gas temperature.

TTFN

 

[JStephen]

It seems this might work better if you could measure temperature under the insulation and outside the insulation, then calculate heat flux from that alone (assuming you know properties of the insulation!). Then you wouldn't have to deal with convective and radiation effects. That should get you an easier analysis with more accuracy.

 

[sailoday28]

You state that the cold side skin temp is 420F. Is that the temp on the outside of the insulation or outside surface of the steel?

 

[chicopee]

In conjunction w/using the nozzle surface to determine the hot side temp, do an estimate from the insulation side whereby radiationheat loss will be marginal and probably can be dismiss in the heat transfer analysis. Also the estimate for the outside overall convective heat transfer coefficient over the entire object surface will probably be more accurate. The hot side temp result of the shell should be very close to the bulk temperature of the liquid.

 

[ko99]

TGHNylon,

Make sure you use the temperature of the outside of the insulation (not the cold side of the steel) to calculate the external convection and radiation.

If the insulation is not much hotter than the air, this method may not be very accurate. You may be getting significant heat transfer via uninsulated areas and/or conduction to other structures.

ko (

http://www.ecooling.biz)

 

[TGHNylon]

Thanks for all the feedback.
The 420 F is taken on the steel surface. The vessel is 99% insulated, but around some of the nozzles the insulation is missing. The reading was taken on the actual vessel steel very close to where a nozzle penetrates the wall. I did not take a reading on the outside of the insulation, but I can tell you it is very close to ambient temp - maybe 85 or 90 F compared to 80 F ambient.
I can say that there are several insulated lines coming out of the top of this vessel and there is probably significant heat transfered along these lines.
I suppose I could cut out some of the insulation, attach a thermocouple and put the insulation back over it and assume this is pretty close to the hot side temperature.
My goal in all of this is to determine which material seats to use in a ball valve to mount on top of the vessel. Depending on the temperature I may be able to get away with something like PEEK, or I may need to go with metal seats.

 

[ko99]

Here's how I would do it:

1. As suggested by JStephens, measure the outside of the insulation and calculate the heat flux with q=KA(T1-T2)/L K = insulation K, L = insulation thickness, T1=420F, T2=outside of insulation. Heat flux should be well below your initial calculation

2. Calculate the hot side of the steel with q=kA(T1-T2)/L K and L of steel. This number should be very close to 420F

3. As suggested by IRStuff, calculate the gas temperature with q=hA(T1-T2)
T1 = inside of steel, T2 = gas.


ko (

http://www.ecooling.biz)