Design minimum flow line

[evdbroek]

In my opinion you have three options for minimum flow protectin.
1. minimum flow line with orifice
2. minimum flow line with control valve
3. minimum flow line with automatic return valve.

When do you decide which type of minimum flow protection is the best?

 

[BigInch]

#3 apparently won't work in this case.

http://virtualpipeline.spaces.msn.com

 

[trashcanman]

Here is a radical idea. Ask the pump manufacturer what to use.

 

[BigInch]

Maybe they'd like to pay for it too.

http://virtualpipeline.spaces.msn.com

 

[Valvecrazy]

Asking the pump manufacturer what to use is like letting the fox guard the hen house. Most pump and motor manufacturers are promoting VSD. However, I have learned that this is not to make pumps and motors last longer and also does not save energy as they claim. It seems pump and motor manufacturers like to sell pumps and motors. Therefore, they are not in the business of showing you what makes pumps and motors last longer. A simple pump control valve is your best option. A PCV will maintain a constant pressure by varying the flow from the pump. It will do this without any of the negative side effects associated with VSD controls, that actually shorten the life of your pump system.

 

[dickon17]

There is no where near enough information provided in the question to even start discussing practical solutions. Who set the minimum flow requirements and for what reason? What type of pump is this and what are the duty conditions? Is cold water being pumped? What size electric motor is being used? Is this a single pump system?

Assuming min flow conditions are set by pump manufacturer to protect against excessive heat rise under low flow conditions, then the equation to calculate min flow requirements is relatively simple provided you have enough data (which we don’t). For pumps in this size range, actual low flow limits are typically about 5% of design flow (or only about 9m3/h in this case).

 

[ClicketyClack]

Valvecrazy, I feel the same way about people selling valves!

Once again, valves and drives have their place. Drives are not a fad nor are they a gimmick. They work very well in the right application. True, overzealous vendors can give them a bad name.

A well applied and programmed VSD will not shorten the life of a pump, and it will extend it. Excessively throttling a pump WILL shorten the life of a pump.

I would use a large pressure tank to add some capacity to the system with a high and low pressure switch to control the pump, if they make them that big.

 

[Valvecrazy]

Ouch!! I never said drives were a fad or a gimmick. Just that they are not the cure for all ills as the overzealous salespeople would like for you to believe. Many even get very defensive if you bring up a few facts.

I do not see how a drive can extend the life of a pump or motor when compared to a correctly sized pump running across the line. Considering the harmonics, voltage spikes, partial discharges, resonance frequencies, motor shaft currents, and many other characteristics of drive control, across the line will always last longer.

Excessive throttling can destroy a pump but, I have seen controlled throttling increase pump life by more than 400% many times, as compared to cycling with a pressure tank and pressure switch.

I don't think they can even make pressure tanks large enough to solve the problem.

 

[BigInch]

I just don't understand why its so hard for "engineers" to believe there is nothing more efficient than a properly sized pump!

http://virtualpipeline.spaces.msn.com

 

[ClicketyClack]

Sorry Valvecrazy, I had flash backs of my old boss when I read your post. He hated drives and fought tool and nail to keep them out of his plant, even when it was clearly needed.

If you throttle back to prevent runout conditions, yes you will extend the pump's life. Also, I have seen 5000 gallon pressure tanks in the entertainment industry but they are probably custom made and very expensive.

I have installed hundreds of VSD's on pumps and have never seen a problem with anything you suggested. Of course, they were all properly applied. I have VSD's on air compressors that have been running for 10 years.

BigInch, I agree. However when flows vary dramatically (heating plants, sanitary water, water shows, etc), VSD's are sometimes a good option.

Sorry to beat the point in. I feel like VSD's get a bad rap on this forum. These sales "engineers" are giving them a bad name.

 

[BigInch]

Variable flow is a bit of a different animal, but then there's still nothing more efficient than a properly sized pump ... with a tank.

http://virtualpipeline.spaces.msn.com

 

[evdbroek]

I was trying to find out if there were any rules of thumb on minimum flow protection.

dickon17 is right about the information I gave about my problem is not sufficient to get to a good fundamented sollution.

There is an expansion needed of the current condensate supply in the plant, because a new unit will be operated parallel of the existing ones. A new condensate tank will be placed of 55m3. This is an atmospheric tank and the condensate temperature is 90°C. If all the units are in operation the condensate demand is 140m3/h, but every 8 hour an extra demand of 40m3/h is required (so 180m3/h total demand) Several months per year not all units are running and then the demand will drop to 15-30m3/h.
Each unit require condensate supply @ about 4 bara.

1)
The main question is that I require a pump which can do all these cases. The pump vendor specified a pump with a minimum flowrate of 35m3/h, but this is less than the minimum requirement of the process.
2)
How do I keep a constant supply pressure of condensate the eacht unit?

 

[BigInch]

You mean minimum process flow is less than minimum pump flow.

On Aug 17 you said, "Maximum flow is 180m³/h and minimum is 20m³/h with almost same discharge head."

If minimum pump flow is 35 and min process is 20, then when you are running minimum process flow, you must recycle 15 m3/h to maintain a total flow through the pump of 35.

The requirement you have already stated for "minimum is 20m³/h with almost same discharge head". Will prohibit a VSD from working. As already has been explained, dropping the speed will drop the head and you will not be able to get the process minimum flow into the vessel (or wherever its going).

You must get a pump with a curve showing that minimum flowrate at vessel entry pressure (or head equivalent) is possible and the pump curve must also show capability to do maximum flow at that (apparently "almost the same") head.

You should use a recirculation line and a possibly a discharge control valve to control those flows. A VSD is not possible, given what you have said. You must look at the pump curve you have, the line sizes, the control valve and recycle valve options, the pipe sizes, the minimum pump flow and the pressures involved and design accordingly.

http://virtualpipeline.spaces.msn.com

 

[ClicketyClack]

The rule of thumb on min flow protection is: Don't operate below min flow. I wouldn't suggest running close to it either. "Good practice", according to Flowserve, is to run -30% to +15% of the BEP.

You need a PCV, as Valvecrazy and BigInch suggested. These types of applications give VSD's a bad name. Someone will try it and it won't work.

 

[dickon17]

At 90C I would also be concerned about significantly increased pump NPSH requirement at minimum flows. Check your pump curve. If min flow of 35m3/hr is too much for your system, check out another pump supplier.

 

[BigInch]

Or use two 1/2 size pumps.

http://virtualpipeline.spaces.msn.com