Steam table data vs experimental : Why the difference?

1. Aug 26, 2009

Doctor_Doom

Hey guys! Was wondering if someone could help me out. I'm doing a thermodynamics topic and have an issue explaining the following :

I conducted and experimental procedure involving a boiler apparatus. (Similar to this one >> http://www.armfield.co.uk/th3_datasheet.html" [Broken]) Now the idea was to find the pressure and temperature for a set of data points. These data points are then to be compared to the "Steam Table" data points.

The red line is the experimental results and the blue is the data gained from the steam tables. There is roughly a 4 degree Celsius difference at each pressure point.

I have to explain why there is a variance in my results? Now human error or apparatus error is just not good enough for the explanation. Is there a concrete thermodynamical explanation?

Help would be greatly appreciated!

Ohh on the side note: How can I improve the experimental apparatus or conditions to get closer to the steam table results??

Last edited by a moderator: May 4, 2017
2. Aug 26, 2009

CFDFEAGURU

I design boilers and heat exchangers for a living. The steam tables are solid as a rock. I think you should think hard on how your system of measurement differs from the system used to create the steam tables.

3. Aug 26, 2009

xxChrisxx

On the plus side, the trend is bob on.

4. Aug 26, 2009

gmax137

How did you calibrate the pressure & temperature indications?

5. Aug 26, 2009

Doctor_Doom

The calibration was done by a resistance correction chart.(The temperature measurements were done via an electronic resistor).

So my question just to repeat it again was : How did "they" assemble the steam tables? Under what conditions? And why do my results differ by 4 degrees Celsius ?? There must be a logical explanation why my results are in the same trend but at slightly lower temperature values!

6. Aug 27, 2009

Q_Goest

As mentioned, the steam tables are solid. Calibrating sensors should be done by testing at a known temperature to ensure the measurement corresponds. In this case, putting the sensor in boiling water at a known ambient pressure should provide an accurate basis for the measurement. In other words, put the sensor in water that's known to be boiling at 100 C and make sure the reading is correct.

EDIT: Isn't there an 'offset' adjustment on the device you were using to read out the measurement?

Last edited: Aug 27, 2009
7. Aug 27, 2009

redargon

Thermocouples can have an error of 1.5°C or more. Pt100's are more accurate. What is the quality of the steam you are using? Pressure sensors also have accuracy issues (1-3% FS for festo parts for example). Try putting some error bars on your measured points.

8. Aug 27, 2009

Doctor_Doom

There is only calibration tables available, which in a way acts as an 'offset'(but I applied them). The funny thing is I cant find any historical conditions / equipment that was used to produce the steam table data points. Can anyone post a link of the conditions and equipment used at the time of the steam table creation.

9. Aug 27, 2009

Q_Goest

Tables such as these are generally created from a large amount of data taken from numerous papers. I have to assume the editor of the table takes the data and does some statistical analysis on it in order to get the degree of accuracy found in the tables. If you're interested in determining how the experiments were done, you might look through some of the references used to create the steam table.