SJ:

Scotch Marine or Fire Tube boilers don't use Overall "U"s in their design, to my knowledge.  The area is figured empirically and sometimes with a heat flux between 8,000 and 10,000 Btu/hr-ft2.

What is it that you want to do?  Design a boiler?  Rate an existing boiler?  Calculate the "efficiency" of the heat recovery?

Art Montemayor
Spring, TX

Hi Art,

Thanks for the reply.  Presently, we have a prototype firetube boiler in our plant.  I am trying to calculate the rate of heat transfer that can be achieved between the fire in the tubes to the water. I would like this to assess the maximum steam output from the current boiler configuration.

Thanks,
SJ

SJ:

On these types of boilers, the heat transfer efficiency is usually an empirical value that is obtained from experience and fabrication criteria and design.  There are some efficient Fire Tube designs (with "wet-backs" & other multiple tube passes) and then there are some very inefficient, single pass designs (the old locomotives were one example) that are built that way for a specific purpose.  If you want an accurate rating, you're going to have to use an experienced boiler engineer to do the assesment or go to the fabricator.  Otherwise, the fastest and most accurate way to do it is to fire it up and check out the steam production and firing rate.

Art Montemayor
Spring, TX

What you need to do is break the boiler down into sections for the analysis. This is usually taken as changes in flue gas directions. I have designed many fire tube boilers and this is the approach I use. In the furnace you can calculate the heat transfer on the gas side being a mixture of radiation and convection. On the water side use the values for full steaming at maximum pressure with a difference between the steel surface temp on the water side and water of approx 8 degrees C this should give you nucleate boiling. As the temp drop across the steel will be proportional to the resistances you can work out the gas sidewall temp and the corresponding radiative transfer. You may need several goes to get it to all balance with radiative and convective transfers.
Knowing the properties of the combustion gases and the above you should be able to calculate Uo and the resulting furnace exit flue gas temps and next section of the boiler inlet temp. For the fire tubes you can use simple convection analysis on the gas side and calculate Uo for each tube pass giving the gas outlet temps & inlet temps for sequential passes. The tube plates and reversal chambers will need to be calculated in the sequence of the flue gases trough the boiler giving Uo for each section. Very simply calculate Uo for each section-giving outlet from one pass to the inlet of the next pass. Tube plate calculations can be taken from texts such as BS1113 or AS1228. This is usually quite an involved process.

Bogu

SJ1-are you sure that you have a vertical fire tube boiler?That type of design went out of the window in the 1940's or even before that with the last of steam powered shovels.  As far as process fire tube boilers these were probably last manufactured in the early 1900's.
As far as determining steaming capacity use 5.00 to 7.50 lb of saturated steam per sq.ft. of heating surface for horizontal fire tube boilers. Steaming surfaces include include 2/3 of net tubesheet(front and rear) area, total inside surface area of tubes and 1/2 of shell surfaces.  Superheater and economizer surface areas contribute nothing to steaming capacities.    

SJ1 By the way,  any process vertical firetube boiler that I ever encountered were of the one-pass design--very inefficient which is the reason for the introduction of economizers on the top of such boilers.

Have a look at

http://v_ganapathy.tripod.com/surfarea.pdf

it may be helpful.

Hi Everyone,

25362's suggested article by Ganapathy for determining heat transfer rates in FT and WT boilers has been a huge help (thanks 25362!).  I have been trying to work through the model equations suggested by Ganapathy for analyzing the performance of my own FT Boiler.  However, in building the model I have encountered discrepancies between my calculated answers with the answers from examples in the article(unfortunately there are no sample calculations).

I wanted to know if it is valid to assume that the properties of the flue gas (density, thermal conductivity and viscosity) are approxiately equal to those of air at the same T and P?  

Any input would helpful.

Thanks,
SJ