160 MW Fossil Unit Feedwater Heater Drain Cooler 1

[MechEng1995]

I am retrofitting a Horizontal Low Pressure Heater Train with drain pumps. The current configuration is a pure cascading drain design from high to low and to the condenser. Our performance engineers determined that we could realize a heat rate improvement by pumping the #2 heater drains back into the consendate line between the #1 & #2 Heaters. The #2 Heater shell pressure is 14.0 psia and drain tempeature is @ 209.4F. There is no drain cooler on this heater. I am waffling on whether or not to put in a drain cooler to help prevent flashing in the pump(s). The reason I am unsure is that the pump(s) will be on the ground floor, 34 feet below the heater, which is on the turbine deck. My first blush look is that there is approx. 14.7 psi (34ft) of elevtaion head that should get the liquig entering the pump sufficiently away from the saturation line. However, if any system upset or transient load drops the pressure in the heater more than 14.7 psi, the liquid will flash, possible in the pump. As we all know, this is bad for the pump. Any suggestions, or thoughts?

 

[toothless]

Calculate your "NPSH available" in units of feet. Remember that a 34 foot column of water at saturated conditions is not the same as 34 feet of 70 F water. Then look at your pump curve to see what your "NPSH required" is.

Off the top of my head I think you'll be OK. Our BFW pumps pull saturated water out of a receiver tank at 5 psig saturated conditions. The receiver is about 20-25 feet above the pump suction and we've never had a problem.

 

[rmw]

The reason your performance engineers are recommending that this condensate be 'pumped ahead' is to capture the heat that is in it. If you put a drain cooler in, you lose that heat to whatever cools the drain cooler.

Looking at your physical conditions, you should have beau coup NPSHA for any pump that I ever worked with. Would that I had had 34 feet of static head for some of the pumps I have suffered with.

Do a search on this site for some threads that give the NPSH formula and do the math. I predict that you will have NPSHA and to spare.

rmw

 

[mauner]

I'm not sure I understand. Does the #2 feed water drains cascade into # 1 and then into the condenser via a drain cooler?

If you pump the #2 drains into the condensate immediately upstream of the #2 feed water heater then the amount of drain cooling (reheating feedwater) would be less and the drain temperature will increase (hotter than 209.4 F).

It sounds like your extraction steam system or one of the dump valves is not operating properly. The feedwater heater train is a delicate balance. If a high level dump valve is leaking back to the condenser then the loss of reheat would result in colder drain water. This may be why he is suggesting pumping the drains back into the feed water, short cycling the condenser.

Do you routinely check turbine performance and monitor extraction steam and feed water heater temperatures?
All dump valves to condenser should be checked for excessive temperatures.

I think you should step back and conduct a full cycle evaluation. This should be compared to the original design. Usually turbine performance/orifices degrade resulting in excessive extraction steam and drains that are too hot to be entering the condenser. It doesn't make sense that all heat is reclaimed before the # 1 heater.

 

[rmw]

mauner,

I haven't seen the heat balance on this, so with out the benefit of facts here is how I see ME'95's situation.

Cascading the #2 drains to the #1 heater will provide precious little flash steam for the #1 heater from the drains coming from #2. The drains do heat some feedwater coming from the condenser in the #1 heater, but the drains from the #1 heater then flash into the condenser, adding some flash steam to its condensing load, and reducing the condensate to hotwell temp where it has to be reheated in the #1 heater.

By bypassing the bottom of the cycle I see it that the performance engr's are trying to keep from losing the flash from the #1 heater to the condenser.

Now, the question in my mind is whether or not the Performance guys have run the #1 and #2 heaters based on the reduced flow to the #1 htr and have verified that the heater will or will not perform with reduced condensate flow and therefore not have a detrimental effect on the HTC's due to reduced velocity and different approach temperatures to the #2 htr.

That wasn't ME'95's question, so I had to take it at face value that the performance guys had done their homework on those issues.

rmw

 

[mauner]

rmw,

I was responding to ME'95's general question for "any suggestions or thoughts". It was your point regarding the reason for adding the drain pump that made me want to question if the performance guys had done their homework.

My experience has been that extraction steam or feed water heater shell side relief valves and heater high level dump valves can leak directly back to the condenser. This loss of reheat results in higher contact temperature readings on downstream piping. A stuck open relief can make the #2 heater drains relatively cold enough to make this plant modification desirable. Subsequent repair/reseating of the relief valve may increase the drain temperature back to near design value (what ever it may be)- this could make the pump modification unnecessary.

If it was my money paying for the pumps - I certainly would want to understand why the plant's operating cycle has changed so much that this design modification will improve the heat rate.

 

[rmw]

Mauner,

I am with you. I wondered too if they weren't trying to solve the symptom of a problem rather than the problem itself. Pump ahead systems were common a long time ago, but each and every heater in the stream was designed for the resulting flows (on both sides).

I was thinking that the reduction of the flow through the last heater's drain cooler reduced the heat transfer there enough so as to negate any perceived gain that was to be had by pumping ahead the drains from the #2 heater. Of course, if it is a true and total cascade system, that removed a lot of condensate flow from the last heater, severely reducing the flow throught the drain cooler.

rmw