11 replies to this topic

[betty]

Explanation (the relation between Q-H) affecting pumps…..

In series: pressure will increase. But the quantity is remaining the same.

• H= H1 + H2
• Q= Remain the same.

In parallel: quantity (flow rate) will increase. But the pressure (head) is remaining the same.

• Q= Q1+ Q2
• H= Remain the same.

The series & parallel combination holds for centrifugal pumps only.

In case of positive displacement Pump, discharge head is determined by system back pressure.

Could any body clarify more this point (In case of positive displacement Pump, discharge head is determined by system back pressure.), and what about quantity, is it determined by system back pressure too?

thanks

[ARAZA]

Hi Betty,

AS the name implies that a positive displacement pump is meant to deliver fixed amount of liquid to the process. The discharge pressure is a function of the system backpressure means that, say you want to pump 50 ml of chemical into a tank operating at 100 psig, the d/s pressure of the pump would be 100 psig + static head + dynamic head. If for instance the op. pressure of tank is lowered to 50 psig, then the d/s pressure of the pump would be 50 psig + static + dynamic head. Thus the d/s pressure vary with the downstream pressure.

The volume pumped is controlled by different means, depends on the pump being selected. In case of a recipracting pump, the flowrate is controlled by stroke adjustment at the pump itself or in combination with a speed controller. In case of a gear pump, the flowrate is controlled by a speed controller either cascaded with a flow controller or a level controller. There might be many control schemes to control the rates.

You need to spend more time reading through good books on postive displacement pump.

ARAZA

[betty]

thank you

But what about series and parallel operation in P.D pumps? There is no comparation between the Q & H.

I search about this topic but I find this Info..

Attached Files

•  FLOW_CONTROL_IN_POSITIVE_DISPLACEMENT_PUMPS_IN_SERIES.doc   28KB   74 downloads

[Art Montemayor]

Betty:

The way I have always approached a pumping application or problem is the way that I was first taught by my first engineering mentor – who happened to be a great and wonderful mechanical engineer. He taught me to acquire the skill of visualization when analyzing how a hydraulic system will work.

When comparing the pumping actions of centrifugal pumps with those of positive displacement pumps (PDs), one should look at the means and parts that each pump uses to generate the discharge head that we recognize as discharge pressure.

In a centrifugal pump (also called a "kinetic displacement" pump), a centrifugal force produced by the rotating element, called an impeller, "impels" kinetic energy to the fluid, moving the fluid from pump suction to the discharge. A PD pump uses the reciprocating action of one or several pistons, or a squeezing action of meshing gears, lobes, or other moving bodies, to physically displace the media from one area into another (i.e., moving the material from suction to discharge by imposing a direct, physical force on the fluid). The amount of positive head that a centrifugal pump can develop is limited by the diameter of its impeller, the speed at which it rotates, and the horsepower applied to it. All centrifugal pumps have a maximum total developed head that they can generate. The PD pump, on the other hand, is limited in total developed head ONLY by two factors: (1) the physical strength of its metallic components, and (2) the horsepower of its driver.

Another characteristic that separates the two types of pumps is that of pumping rate. The centrifugal pump pumps less flow rate as the total developed head increases. However, the PD pump ideally pumps a constant flow rate of fluid regardless of the total developed head. In reality, a PD pump suffers a decrease in the actual flow rate it pumps as its discharge pressure increases. This is called "slippage" and its amount depends on the type of pump, its mechanical condition, and the fluid in question. However, from a practical standpoint one must respect the fact that the PD pump will raise the discharge pressure to a very high level if allowed to do so and, therefore, must have an over-pressure device to protect it and the system it operates in.

You have compared centrifugal pumps operating in parallel and in series and have stated "In parallel: quantity (flow rate) will increase. But the pressure (head) is remaining the same." Actually, this is not true. If you connect two identical centrifugal pumps to a common manifold so that their discharge flowrate is additive, the discharge head will increase if the manifold is the same one as used for only one pump. The discharge pressure of both pumps will be the same, but it will depend on the resistance found in the discharge system down stream of the pump's discharge nozzle.

The PD pump will also pump with a discharge pressure that is fixed by the resistance found in the discharge system down stream of the pump's discharge nozzle. The only difference in the PD pump's performance is that it will not "dead-head" – it will not reach a maximum discharge pressure above which it cannot perform. The PD will, if allowed to, simply continue to raise its discharge pressure until the resistance found in its discharge system is overcome. If the resistance in the discharge system cannot be overcome – such as a block valves shut off accidentally or by mistake – then one is at the mercy of a proper relief device being located in the discharge of the pump and successfully relieving the excessive pressure of the PD at the pre-determined setpoint. If such a relief device is not installed on the PD pump's discharge, then the system is in big trouble.

I think I answered your specific questions. If I failed to do so, come back with additional questions or comments.

[ARAZA]

Hi Betty,

I've never seen a P.D pump in series or in parallel.

I guess, the reason being, there is no need to do that. A single P.D pump can generate a huge amount of pressure, I've seen some plunger pumps rated up to 3000 psig.

Similarly there is no need for a P.D to be connected in parallel since a P.D pump is employed when there is a need for a fixed amount of rate such as in chemical injection, catalyst injection etc.. and usually these types of application with small flow-rates can be performed using a single pump. You can get a P.D pump specially gear pumps up to several gallons per minute.

ARAZA

[Art Montemayor]

Betty:

Araza is very correct in placing the PD pump within its logical and practical design and application When debating series flow. It simply isn't practical to arrange PD pumps in series if one is free to design for the application. As I stated previously, A PD cannot be dead-headed. If one designs the mechanical components to contain the desired discharge pressure safely, then all one has to do is ensure that enough horsepower is connected to the PD and the pump will deliver the desired discharge head in one simple stage. This is a NET advantage over a centrifugal model. I routinely have pumped liquids to 3,000 psig discharge pressure in a single-stage piston pump. And I did this pumping saturated liquid Oxygen at -350 oF, 24 hours a day, 7 days a week – without stopping. So what is commonly done is that ONE PD pump is designed and used for an application regardless of the discharge pressure.

Additionally, I recently worked on a methanol pumping application: 15 gpm of methanol from an API storage tank to a discharge pressure of 10,000 psig – in one reciprocating pump stage. Four of these pumps were connected in parallel. Therefore, I don't agree with Araza's statement that there is no need for a P.D to be connected in parallel. I also have had multiple chemical addition PD pumps connected in parallel and pumping simultaneously to the same manifold when injecting chemicals.

[Zauberberg]

I agree with Art. Depending on application, you really may need several PD pumps connected in parallel. I remember seing those in refinery (bitumen plant), or in production fields where extremely viscous fluid has to be pumped miles away from the station. In such circumstances, it is better to go for PD pump, and depending on overall pumping rates there can be a requirement for multiple PD pumps operating in parallel.

Best regards,

[ARAZA]

Dear Art,

Thanks for the wonderful analysis & comments. I think, I've learned something new today.

I guess, so far I've never seen any application with P.D pump arranged in parallel. I've always ran into appication such as chemical injection, precise monitoring of the reactants in the reactor etc. which I always seems to accomplish using a single pump.

Best regards,

ARAZA

[fallah]

If the resistance in the discharge system cannot be overcome – such as a block valves shut off accidentally or by mistake – then one is at the mercy of a proper relief device being located in the discharge of the pump and successfully relieving the excessive pressure of the PD at the pre-determined setpoint.

Dear Art

Considering PD pumps don't have classic shut-off head as centrifugal one,is it correct to say:

PRV setpoint is specified by vendor based on material strength and power of the PD pump?
Indeed,which of two above is usually governing parameter for specifying the setpoint?

Regards

[fallah]

Is there any piont?

[Art Montemayor]

Fallah:

You asked: "PRV setpoint is specified by vendor based on material strength and power of the PD pump?
Indeed,which of two above is usually governing parameter for specifying the setpoint?"

I have trouble understanding what you are asking, so I'll guess that you mean: "Is the set point fixed by the material strength of the pump or its power?"

To that question I would answer that the manufacturer is expected to set the relief point of his pump based on the maximum allowable pressure permitted by his design - and this may involve the material strength, the mechanical design, and (of course) the weakest component in that design.

The purchaser should always insist or demand that the MAWP of the pump SYSTEM supplied be identified (and stamped on the assembly) for the protection of the user. It is important to note that the pump is only ONE component within an entire pumping system and it is possible that it could wind up as the weakest component in the entire system. This is vital to know when considering any safety scenario during its design, a hazop, or normal operations.

Of course, the above only considers the maximum PRESSURE on the system. That is not the only safety concern involving rotary or moving equipment connected to mechanical drivers. As you have probably insinuated, there is yet another danger involved every time we employ a mechanical driver to a rotary piece of equipment: it is possible to apply an excessively powerful driver to a piece of equipment (a pump, for instances) that could mechanically rupture or destroy the equipment due to excessive horsepower applied. This is another hazard that certainly has to always be considered.

[djack77494]

May I make an assumption here? I believe fallah is anticipating two scenarios. Let's say the driver applies power to the pump which cannot be properly dissipated by fluid flowing through the pump. Typically the case can be considered a blocked discharge valve. In one scenario, there are mechanical components of the pump that cannot withstand the resulting forces. The result is some type of mechanical failure with accompanying loss of containment. In the second scenario, the pump and connected system CAN withstand the mechanical forces applied up to the point where the force exerted by the driver is overcome. Then the result is that the moving parts cease to move and the motor is forced into a "locked rotor" condition. I believe this is what fallah is asking, but I cannot answer this question. Help please from a knowledgable rotating equipment person.