What's new
What's new

VFD on a constant pressure feedwater pump ..... How ?

DDoug

Diamond
Joined
Oct 18, 2005
Location
NW Pa
My neighbor, a drives engineer, recently put a VFD on a boiler feedwater pump, and I can't understand how it works.

He couldn't explain it, rather he set's the programming to "hold pressure", there are preloaded programs just for this application.

So these are centrifugal (multistage) pumps, NOT positive displacement
pumps. I thought that the speed of the pump determined pressure.

How doo you turn down a pump when you don't need all the flow, but still maintain outlet pressure ?
 
They just sense pressure and modulate pump speed to maintain that pressure at a setpoint. It’s not true that without flow there’s no pressure. The pump will still make pressure. There are things to consider though. If the pump sits and runs dead headed it will burn up since the heat generated has no where to go.
 
They just sense pressure and modulate pump speed to maintain that pressure at a setpoint. It’s not true that without flow there’s no pressure. The pump will still make pressure. There are things to consider though. If the pump sits and runs dead headed it will burn up since the heat generated has no where to go.

How is the VFD going to effect any savings ?
Prior to this, the motor ran at a constant "full" speed.

If you drop the speed, won't the outlet pressure drop ?
Regardless of the flow ?
 
How is the VFD going to effect any savings ?
Prior to this, the motor ran at a constant "full" speed.

If you drop the speed, won't the outlet pressure drop ?
Regardless of the flow ?

Not sure on savings. There are some complicated boiler feed systems and I think there is not enough information to answer that question.

I do have experience with another example of this type situation where it's a little more obvious. At the pharma plant I used to work for we had very large hot water heat and chilled water systems. Instead of a constant flow system as most of our smaller systems were, they designed it as a constant pressure variable flow system. A VFD was in place and varied the pump speed and number of pumps running to maintain pressure. As flow demand increased, the pressure would drop some and pumps would be sped up or more pumps brought online to keep up. This was an energy saving measure.

Yes if you drop pump speed pressure will drop all other things being equal. Flow and pressure are interrelated but for a fixed restriction this will be true.
 
There may be more going on under the covers here - probably he's using a pressure sensor to provide feedback to the drive.

Yes, there is.

My point is, how can you slow down the speed of the motor, and still make the pressure, it's a centrifugal pump.

I could plainly see if it was a positive displacement pump like a gear pump.
 
The pump runs at whatever speed required to reach the pressure setpoint. This will be flow-dependent obviously.

If the flow rate goes down the pump speed will drop accordingly.
 
Centrifugal pumps produce pressure at no flow. In fact, pressure goes DOWN with flow. (And the pump works the hardest to run full flow at low pressure)

To maintain pressure one might be able to slow the pump, depending. Most such systems have a relief to allow some flow around to the water source, to prevent stirring the same water for a long time, with will produce heating. Slowing the pump would reduce heating.
 
Ok, I guess it's time for a little hydraulics. Pumps don't create pressure, pumps create flow. Pressure happens when flow meets a resistance (orifice) and is restricted. Now, there is no such thing as a "zero flow" situation. You may observe no fluid moving through the system, but as long as the pump is operating, fluid is moving somewhere. It may be through a relief check, or just eddy currents within the pump, but it's going somewhere and has the potential for pressure. You don't say "I want a pump that puts out this much pressure", you say "I want a pump that has the power capacity to generate this much pressure at this much flow". And as long as the pumps capacity exceeds those requirements, you can dial back it's power to hit your numbers instead of dumping the extra through a relief valve. Because running at a constant power, It will build flow, and therefore pressure, until it can no longer turn hard enough to do so. You get your desired numbers by regulating the orifice to hit your flow, and allowing any extra pressure to escape through a calibrated spring tentioned relief system.
 
For centrifugal pumps (most common), yes. They can generate pressure with no *external* (eddies only) flow but I believe it's not good for them.

Fixed displacement pumps can maintain pressure with zero flow; consider a hydraulic cylinder with a load on it.
 
Fixed displacement pumps can maintain pressure with zero flow; consider a hydraulic cylinder with a load on it.

Pumps don't work like that. If it's a true fixed displacement pump, and there is zero flow, it is either leaking (or bypassing, which is flow) or stopped, in which case it's a static system where the pressure is being generated by external forces in balance. Pressure in a closed system is a dynamic of a fluid redirecting all forces acting on it to be equal and balanced in all directions, a hydraulic cylinder with a load on it and a closed valve does not have a functioning pump in the system and does not apply. A fixed displacement pump that displaces nothing is not a functional pump.
 
Pumps don't work like that. If it's a true fixed displacement pump, and there is zero flow, it is either leaking (or bypassing, which is flow) or stopped, in which case it's a static system where the pressure is being generated by external forces in balance. Pressure in a closed system is a dynamic of a fluid redirecting all forces acting on it to be equal and balanced in all directions, a hydraulic cylinder with a load on it and a closed valve does not have a functioning pump in the system and does not apply. A fixed displacement pump that displaces nothing is not a functional pump.

Agreed...a positive displacement pump when dead headed (zero flow) will do one of the following:

1) Stall the prime mover.
2) Leak/bypass internally which will quickly cause overheating and failure.
3) Break the pump.

PD pumps are almost always protected by a relief valve to prevent 1-3.
 
I suspect we may be overlooking some important points of information...

But realize,this is not a FLUID POWER application, and it is not a 'hydraulic pump'. It's a boiler feed pump.
There cannot be a pressure relief valve isn't in the water delivery line... that would risk having the boiler discharge itself and overheat.

What defines the pressure to which the feedwater pump is working, is the temperature of the boiler, and the rate at which it is developing steam. As the steam is drawn off, boiler water level falls, water boiling rate rises, and pressure in the boiler STARTS to fall.

As that pressure falls, the multistage centrifugal pump progresses from a minimal-flow circumstance, to a higher flow circumstance, and water then moves into the boiler. This CHILLS the boiler slightly (incoming feed water is absorbing heat, hence, boiler pressure slumps for a little bit... but the water expands, recovering the hump.

All this time, the feedwater pump is running, and must remain so in order to maintain the differential pressure from feedwater input, to boiler output.

To further complicate matters, we do not know if this pump is in the path immediately PRIOR, or POST a feedwater preheater, if it is BEFORE, there'll be a return passage from the after port, back to a point BEFORE the feedwater pump, as 'stalling' water in the preheater loop will turn the preheater into a BOILER.

The feedwater pump is centrifugal, and it is designed to be stalled against an intended PRESSURE. It is not positive displacement... it is variable. Remember, it is first and foremost, centrifugal FORCE which allows it to work, and the motor's RPM determines how much centrifugal force can appear at the edges of the impellers (it's multi-stage). The only thing that will PREVENT it from generating a static pressure, is CAVITATION, which won't occur if the inlet pressure (water supply source)is at an excessively LOW pressure to start with.

A PID control signal from the boiler's master controller probably commands the pump's drive in response to several factors, as most complex boilers nowdays run nifty full-authority controls that can do instantly, what a well-seasoned boiler operator needed several minutes to sort out.
 
Agreed...a positive displacement pump when dead headed (zero flow) will do one of the following:

1) Stall the prime mover.
2) Leak/bypass internally which will quickly cause overheating and failure.
3) Break the pump.

PD pumps are almost always protected by a relief valve to prevent 1-3.

A lot of old steam-powered pumps I believe did #1. Not practical with diesel or most electric, but in theory if there's not enough steam pressure or too much water pressure, the pump just slows and stops, until outlet pressure reduces.
 
This is true of many types of steam-powered domestic lift pumps, i.e. skyscrapers built in the timeframe prior to ww2... easy way to 'self regulate' lift pressure.

Feedwater pumps for boilers are challenged in this respect- they have to be able to force water into the boiler at ANY pressure (high or low) because the water level must never fall below the crownplate, and SHOULDn'T be allowed to overfill into the steam dome...

but this is reminding me... I need to go back and recert my boiler engineer's wallpaper... it's been a while since I had cinders in my bibs...
 








 
Back
Top