boiling water reactor feedwater control during plant startup

[BWR trainee]

Can someone help me understand how feedwater flow is controlled during reactor heatup
and turbine-generator startup on a boiling water reactor. Is feedwater pump turbine speed varied or held constant? Or is there a flow control valve that's used to regulate flow
during startup? Are there recirculation lines that allow water to flow back to the main
condenser during low flow periods for both the condensate pump and main feedwater pump?

[tr]

It sort of depends on the plant design.  Some plants have motor driven pumps (or at least a motor driven startup pump), where there is a valve to control flow.  For turbine driven pumps there is usually some sort of smaller startup line with a flow control valve.  Turbine speed adjusts as necessary to maintain the desired flow rate and pressure.

Pretty much all feedwater pumps have a minimum flow line with a valve that adjusts to get keep the pump above minimum flow.

[Fermi2]

My old plant utilized two variable speed feedpumps. There were two 20 inch feed lines. During startup and to around 20 to 25 power we used a bypass line with an air operated Startup Level Control Valve. (SULCV) Now I;m going from memory here and my last start up was in 2004 so I'll try my best.

Assume the reactor is cold. We'd be lined up through the SULCV feeding with the Condensate Pumps. I say feeding but in reality the control Rod Drive Pumps were all we ever needed at that point.

1: We'd pull the reactor Critical.

2: Reach the POAH. Given you're a trainee I'll give you a GREAT hint. Long before you call the POAH on rising temperature you'll see the IRM Trace start flattening out and Rx Water Level Swell. Call the POAH there. Otherwise you're just a rookie. At this point there's rarely any feed flow. CRD Flow is enough. Fermis Gland Seal was a load right away so at that time it MIGHT be feeding via the SULCV.

3: Start your heat up. In most cases the SULCV is still closed. At 140 Pounds or so we'd start what we called the Heater Feed Pumps. They were good to about 650-700 #.

4: 650 # Roll the first Feed Pump. After it's rolled an operator would try to keep it about 100 to 150 # above Reactor Pressure

5: About 1000# Now your line up is one feedpump in Manual. Maintaining 100 or so # above RPV Pressure. The SULCV is actually controlling RPV Water Level. When the SULCV got to 70% open we'd bump Feedpump speed up to get it to about 20% open. That way the Operator wasn't constantly playing with feedpump controls.

6: Roll the turbine at about 10% Power I;m a bit shaky on this, At Fermi we actually based it on position of the Reactor Pressure Bypass Valves. When they got to about 55% open we'd roll the turbine. After it was rolled we'd Synch it. At this point your feedwater controls are still the same.

7: Raise power to 20 to 25%, Start a second Heater Feed Pump. Now it's time to get off SULCV and get on Single Element Control. Which changes the line up. We'd place the Running Feedpump in Auto, and it would change speed to control Reactor Vessel Level. Then we'd place the SULCV in Manual and fully Open it. At this point the Lineup is via the SULCV with RFP Speed controlling water level about 197 inches above the Top Of Active Fuel.

8: NOW the hairiest evolution, at least in my mind we ever did at Fermi. Lining Up the Feedpumps off SULCV. We'd lower the auto setpoint for Single Element control around 2 or 3 inches. When level stabilized we'd open the Motor Operated Isolating valve for the running feedpump. This would line up the Feedpumps for the Big feedlines. Usually you'd overfeed at first, then underfeed. Eventually level would stabilize. Then we'd open the nNon Running Feedpump Iso Valve which was pretty much a non event. Then close the SULCV and set it up for Post SCRAM Feedwater Logic. Now your line up is. Both Feedpumps lined up through the big lines. One Feedpump in Single Element Control. SULCV is Closed. Operators would LOVE to place Feedpump controls in manual when Unisolating the Big Lines. On my shift I wouldn;t allow it. Auto worked just fine.

9: As soon as we reached the lineup described in Step 8 we'd go to Three Element Control. That was as simple as turning a switch. Again the feedpump is getting an error signal which controls feedpump speed.

10: Shortly thereafter we'd warm up the second feedpump. At about 55% Power or so we'd parallel the Feedpumps and head to 100% power.


Hope this helped

Mike

[BWR trainee]

mike, when you feed with codensate pumps alone(lower pressure), does water flow through or around the big feed pumps. Is the SULSV line downstrem of the big feed pumps. What are heater feed pumps; are they electrically driven? 

[Fermi2]

Yeah both Condensate Pumps and Heater Feedpumps are Electric Motor Driven. Their nomenclature varies from plant to plant. TVA calls them Hotwell Pumps and Condenstate Booster Pumps.

The flowpath at Fermi was as follows,

1: The Hotwell, it carried 4 minutes of water IIRC about 90,000 gallons.

2: 3 Condensate Pumps in Parallel. Each rated for 7 to 8 thousand gallons.

3: 7 Heat Exchangers. 4 SJAE Condensers, 2 Off Gas Condensers, 1 Gland Seal Condenser. These had a Bypass Valve maintaininng 24# DP across Heat exchangers..

4: 8 Precoat Type Condensate Filter Demins

5: The first three stages of Feedwater Heating (2 stages of LP Heating and a Drains Cooler)

6: 3 Electric Driven Heater Feed Pumps.

7: The next three stages of Feedwater Heaers.

8: Two Reactor Feed Pumps in Parallel. Each had a suction strainer. The Pumps COULD be bypassed but NEVER were the whole time I was there.

9: The last Stage of Feedwater Heating.

10: The Reactor.

During SULCV Operations and until the Big Valves on the outlet of the Feedpumps were open the flow path would be Condensate Pumps>Filter Demins> Heater FeedPumps> Suction Strainers of the Feedpumps> Through the feedpumps> Through a bypass line on the outlet of the feedpumps that led to the SULCV. Each Feedpump was capable of feeding on its own to about 70% power but that was ragged edge. IIRC correctly if a feedpump tripped the reactor would run back to a Recirc Flow equivalent to about 65% power. It knew this by whether the vessel level had reached Level 4 (I'm THINKING 193 inches or so) with either contacts on a feedpump indicating it was tripped OR less than a certain flow through a feedpump.

Also IIRC the Condensate system provided about 67% of the water for rat6ed feed flow, the other 33% came from the Heater Drains System. The #5 Heaters drained to Flash Tanks and 2 Heater Drains Pumps pumped about 5000 GPM each into the line between the #5 Heaters and the Feedpump Suction. I CAN tell you if one of those pumps goes away you can still safely get at least 98% power and have margin to a feedpump suction trip provided all other condensate system and heater drain pumps are running.

Also, if any heater feed pump or Condensate Pump tripped the remaining pumps would support 100% Power. We ran them at a LOT less than rated. Once I did some quick number crunching and figured if any of those pumps went away we could probably safely stay at about 95% power. The next day an engineer showed me a curve that showed I was correct.

I'll give you a bit of advice. KNOW the TRUE capabilities of your equipment, especially pumps. The trip of a pump might put you in a position where you don't want to stay long term BUT if you know what your equipment is capable of you'll have time to think things through.

A quick Thumb Rule I've always used. Once the plant is stabilized if you're about at least 25 to 50 pounds above your pump suction trips you're ok FOR NOW. 25 # is ragged edge low but it gives you time to think, call the boss and get a plan together for where you really want the plant to be.

Mike

[BWR trainee]

Thanks mike and tr

[Len61]

Can someone help me understand how feedwater flow is controlled during reactor heatup
and turbine-generator startup on a boiling water reactor. Is feedwater pump turbine speed varied or held constant? Or is there a flow control valve that's used to regulate flow
during startup? Are there recirculation lines that allow water to flow back to the main
condenser during low flow periods for both the condensate pump and main feedwater pump?

During startup (Lacking electric motor driven feed pumps) we use the condensate booster pumps and a startup level control valve that bypasses the turbine driven feed pumps. The condesate booster pumps are good for up to about 425 psig.
Normal level control at power is via variable speed turbine driven feed pumps. Each feed pump (3 total) has it's own min flow valve. There is also a min flow line for the condensate system.

[RRhoads]

During Startup, we use condensate booster pumps & i think DEH in there somewhere as well.

[BWR trainee]

Mike, you provided me with some great information on managing the feedwater during startup on a BWR4. BUT DO YOU RECALL WHEN TO START THE 2nd and 3rd condensate pumps and the
3rd heater feed pump. I'm guessing you start the 2nd condesate pump when you start the 2nd heater feed pump, at about 20% power prior to transfer of level control from SULCV to the running reactor feed pump. And you start the 3rd condesate pump and 3rd heater feed pump just before you place the 2nd Reactor feed pump into service at 55% power. If I'm wrong, can you correct me? THANKS. I know it's been some time since your last startup on a BWR.

[Fermi2]

IIRC Fermi lineup at startup was 2 Condenser Pumps, 1 Heater Feedpump Though the HFP may have been started when Reactor Pressure reached 140 #. The 140# was based on the Condenser pumps not being able to pump into the reactor at that pressure.

After that it was based on Power OR the status of the Reactor Feedpumps. If we were going to start the second feedpump we started a Second Heater Feedpump. The reason was the Minimum Flow Valve for the RFP used 4,000 GPM or so and without the second HFP Minimum Flow valves throughout the Condensate system would cycle. Usually to support the Second HFP we'd start the 3rd Condenser Pump.  So call it roughly 25 to 35% Power or so.

The last HFP was based on some power level, IIRC about 85%. We didn't want to start it too early as around 55 to 65% Power we'd start pumping forward with Heater Drains and if you had too many HFP running the Heater Drains System wouldn't pump the way it should.

Around 55% was putting the Second Feedpump in Service.

Things may have changed since I left but Fermi had to get the second RFP (The South RFP, warming up early as it had a 12 hour warmup time due to moisture intrusion issues). We always came up on the North Pump.

Mike

[BWR trainee]

All 3 condesate pumps and 2 heater feed pumps were in service at 25% or so. That
sounds like a lot of pumping capacity for such a low power level.  I always thought the volumetric flow rate was held constant by the SULCV or the Running RFP (which ever was in AUTO at the time) when a new pump was put into service, so I didn't quite understand your reference to the operation of the minimum flow valves for the RFP and the codensate system. Could you explain further?

[BWR trainee]

Mike, Do you start the 3rd conedesate pump and second heater feed pump before or after you
have transfered feedwater flow from the SULCV line to the 2 Big 20inch Lines.

[Dreamza]

Around 55% was putting the Second Feedpump in Service.