Boiler feed water start up procedure or standard

[josemg]

After checking the ASME PTC4 Power Test Code for Fired Steam Boilers I could not found boiler feed water system balancing procedures.

How should the boiler feed water system be adjusted or balanced to prevent overflow pump conditions?

How to account for delivering proper minimums flows to all boiler feed water valves without overflowing the closer boiler and giving less flow to the farther boiler?

Should a balance valve be throttled before the feed water valve of each boiler to equalize all the boiler pressure drops at full load conditions?

Should this be done with a dummy steam load at start up, or could it be done by pressurizing the boiler before first firing and use cold water only and letting water to flow back to the deaerators?

For example, everybody says to install an orifice or bypass valve at pump discharge to account for dead headed situations. But what about the overflow possibility of a feed water pump?

If for some reason you have three boilers been supplied from two feed water pumps in a common header. The closer boiler will receive more available pressure and flow than the farther boiler when all the feed water valves start opening. Still during the boiler start up almost never will the load be available at the building to simulate the worst case scenario of a 100% load. So the start up is done with the boiler plant at less than full load capacity and the feed water pumps pressure is not equally balanced between boilers.

This procedure should be part of a standard or code to make sure the boiler plants are tested at full load conditions.

 

[Sharik]

Either we are doing things far too simply or you are making things far more complicated then they have to be.

At some of our plants we have similar systems, three boilers being fed by two 50% load BFW pumps. The third BFW pump is on standby.

We have never balanced the FW system to ensure there is the same pressure at each boiler FW valve and I don't see the need to. To set up these plants we take the total steaming load for each boiler, convert that to a total FW flow and ensure we had pumps with performance curves which would ensure, even if all three boiler FW valves were 100% open, the pumps could supply the demand, at the operating pressure.

Each boiler has it's own FW control valve so there is little chance of overfilling the boiler (well until operations gets too involved!!).

The more critical problem is maintaining a minimum flow through the BFW pumps in the times of low or no demand. Each BFW pump has a minimum flow requirement (set up from the manufacturer) and we have installed automatic flow bypass valves on the BFW pump discharge. This valve opens automatically to maintain the minimum flow through the BFW pump during low demand periods. The discharge from the minimum flow valve goes back to the deaerator storage tank.

So it doesn't seem to matter if the BFW pressures at each of the boiler FW control valves is the same. As long as the pressure is sufficiently greater than the boiler pressure to ensure enough water can be pumped into the boiler when required. And as long as the boiler feed pumps can maintain this pressure at full flow conditions.

It will be interesting to see what everyone else has to say.

 

[ZHUANGFEIYANG]

I do agree Sharik's opinion.

 

[TBP]

Sharik - As someone who ran powerhouses for a lot of years, I can tell you that in the vast majority of cases, the only times "operations get too involved" are:

A/ The initial design of a given system was sloppy.

B/ Construction botched (and was allowed to botch by virtue of little or no site supervision by engineering) the initial installation.

C/ Both of the above.

Spend some night shifts in a powerhouse, struggling to keep the plant on-line by fighting with the square peg and round hole combination you've been left with. I've been "too involved" more times than I care to remember.

 

[Sharik]

TBP

You are correct and I am forgetting my own roots. I too operated steam plants, both stationary and marine, for many years prior to becoming mechanical.

We are just going through a serious episode at one of our facilities. Operations did not maintain correct calibration of the pH meters testing the boiler feedwater. So instead of boiler water they were boiling a high pH caustic solution in the boilers. There is now extensive caustic embrittlement and cracking in the steam and muddrums of these high pressure Stirling boilers.

It is going to be expensive and there doesn't seem to be any options but to replace the drums on all three boilers.

This is getting off the original topic but I really don't think this would have happened years ago. Due to current circumstances, I don't think operations have the same input into operations, design or construction. Maybe as a result of this, they also don't seem to have the same attitude towards the equipment. Many plants I deal with, the roll of operations seems to be just to take readings. Any decisions concerning operations, repairs, etc. are handled by committees of mostly inexperienced personnel where the budgets are of primary concern.

Well that's my soapbox for the day. Now I have to rush and attend one of those committee meetings and try to keep things realistic.

 

[josemg]

Sharik,

Is good to know that no one has seen this problem happening in their installations.

I think that almost all boiler installations that have this same situation will not see a 100% steam load demanding a 100% feed water flow thru all the boilers at the same time.

So I am probably complicating things too much, since in almost all the applications there is more capacity of pumping than demand for flow.

But my concern is, if the boiler plant has to go thru a 100% load performance test to justify a new expansion project with the same boiler equipment, how you account for your system having a start up that will prove not allow this problem from happening if the 100% load exist at any time.

Probably with today's three element controls for boiler feed water systems and variable speed pump systems you do not need to balance the system at all and the feed water valves will equalize all the pressure drops at full load.

My question still exist whether or not you need to account for balancing all flows or just let the feed water modulation valves do the work.

 

[Sharik]

We have never balanced all flows for the BFW. The FW control valves do the work. The pumps merely have to supply the required total capacity at the maximum pressure, steam out = water in.

I actually did some investigating at one facility and sure enough the FW inlet pressures at each boiler on the FW header are slightly different. The difference is minimal and variable depending on the FW flows to each boiler.

There is no problem maintaining boiler water levels even during upset conditions.

The biggest problem is balancing the loads on the BFW pumps. Sometimes the turbine driven pumps are turning at different speeds than the electric driven pumps.

 

[TBP]

Sharik - I agree. Until the accounting dept's & MBAs stop running engineering, construction, maintenance and operations, things will continue to slide. In many plants (even entire companies), whole areas of expertise have simply ceased to exist. The people who knew what to do have retired, been downsized, or quit. Even if a few old hands are still around, they're usually spread so thin they don't even have time to think. They can't train anybody, because they have no time, and even if they did, there's nobody to train anyway. I hate to say it, but I doubt that anything will change until there are some serious accidents with multiple fatalities. EOR (end of rant)

 

[abcmex]

Josemg,
Sharik is right.
The pressure drop on the lines inside a boiler house are not very relevant and anyhow the level control valve would change this scenario continiously.

The control valve adjusts the pressure drop to keep the righ amount of BF-water going to the each boiler so you probably never balance. The Balance is done by the level control valves. :-(

To protect the BFW pumps you may: use a orifice plate, use a check/orifice (

http://www.yarway.com/control_valves.asp)

which is more economical in high flow/pressure applications or use a VFD.

It is recommended that you install a higher pumping capacity than what your boiler consumes and that you have a proper sequence control.
rgs

 

[unclesyd]

I have to agree with Sharik and TBK don't complicate things. Just remember the K.I.S.S. principle

I just talked with one of my old cohorts from our powerhouse and confirmed the operation of BFW Header that serves 5 power boilers, 650 psi, 3 waste heat steam generators operating at 650 psi and 4 waste heat recovery units.

The BFW header is operated almost as the output steam header. The BFW Pumps maintain a header pressure of 900 psig, with some by pass back to the deareators storage tank, and BFW Control Valves for each generator. There are different control schemes on each generating source. The only requirement is that if a big user or generator is going to go off line there is a notification of the outage transmitted to the powerhouse. There are both electric and steam turbine driven BFW pumps, 7 total with one always down. The only problem that requires intervention is when a large user or generator has an unscheduled outage, which usually requires operator intervention otherwise everything else is handled by the control scheme. There are no VSD pumps in the system, though there is a project to install 2 such drives. In 50 years operation the has never been a problem with flow to the boilers, we have never quenched a boiler, or over filled a boiler.