Measuring/Calculating Boiler Feedwater Heater Performance at Startup

[MAPower]

At our coal fired facility we recently replaced four of six feedwater heaters. With the recent replacement I must measure/calculate the new heater performance. I will need to do this for startup conditions of which I do not have all the desired measured parameters needed to calculate TTD (terminal temperature difference) and DCA (drains cooler approach). The performance curves for TTD and DCA are given based on percent design feedwater flow and we are not measuring this. How can I calculate performance of the heater at intermediate loads without directly measuring the flow?

 

[stgrme]

I don't believe you can solve your problem without measurement of at least one flow. Even if you wanted to make a heat balance at each heater, you would need the extraction flow in or drain flow out, and any drain flows entering on the shell side of the heater to calculate the heat duty. Once you had the heat duty you could calculate the feedwater flow.

Best of luck!

 

[davefitz]

see the following link for typical monitoring and analysis to determine supplied feedwater heater DCA and TTD.

<

http://pepse.scientech.us/papers/1989section10.pdf>

For startup conditions, the conditons are likely dynamic and would need to include dynamic correction factors for thermal inertia of the tubesheeet, tubes, and shell.Not an easy task- it would seem more reliable to only determine the steady full load performance( heat exchanger efffectiveness e from the e-NTU method), work backwards to the apparent U*A for each zone ( desuperheater, condensing, subooler). Then pertube this model to determine steady state part load performance.

Every plant I've seen has a flow nozzle at the dearator outlet , so obtaining HP heater flows should be straightforward. The LP heater water flow may need to be calcualted based on a heat balance at the DA if a condensate pump flow element is not provided . Cascading drain flows would seem to be problematic- perhaps an ultrasonic flow device could be used to measure those flows.

 

[MAPower]

Thanks for the input.

I recently discovered we are adding triple redundant feedwater flow meters. So now I just need to figure out the correction factors for partial load performance. As my conditions will vary from design. I guess I'll thoroughly read through the document you posted davefitz to see if I can develop the correction factor for TTD and DCA for non-design conditions.