8 replies to this topic

[deltaChe]

Dear Sir

I have a Postive displacement rotary pump at the bottom of the column ( the column is to remove all MEA and some DEA from the feed for recycle to the reaction system , the proccess diagram is attached).

My question is this type Postive displacement rotary pump is variable speed, and I didn't see any minimun flow line back to the column. Is there no need for min flow protetion at all, if the pump is Postive displacement rotary pump with variable speed control? And could you please kindly tell me why is the reason behind?

Thank you so much

Attached Files

•  Book1.xlsx   14.23KB   157 downloads

[Art Montemayor]

Steven:

Good basic data sketch and good communications.

The installed positive displacement pump - even though it is of the rotary type – does not require any concern for having a minimum pumped flow going through it. Unlike centrifugal pumps, this is a typical, positive characteristic of all positive displacement pumps.

A centrifugal pump can operate (rotate) and try to move fluid –even when its discharge piping is totally blocked. Since the centrifugal pump has no positive displacement ability, it merely spins and churns the fluid in the casing while the discharge pipe is blocked closed. However, should you allow this type of condition to continue and persist during the total operation of the pump, then you will cause pump problems that could lead to mechanical damage. A centrifugal pump adds energy to the liquid that the moving liquid normally carries away. If the flow is blocked, the temperature within the pump will rise steadily until the liquid boils (net positive pressure is now zero). Damage to the pump is quite likely. Seals and bearing suffer damage. For this reason some form of minimum flow is almost always employed on centrifugal pumps.

Positive displacement pumps are tigers of a different stripe - and more dangerous. They will pump liquid fluid and try to continue to pump it even though you totally block the discharge. These pumps have no “slippage” or internal bypassing of the liquid, so they will continue to pump liquid into the blocked discharge line and build up hydraulic pressure there until one of several things happen:

• A pressure relief valve on the discharge line opens to relieve at a set pressure;
• The pump fails to pump due to a bad NPSH value (the liquid vaporizes inside the pump casing;
• The pump driver stops due to an excessive power requirement caused by the high discharge pressure;
• The pump ruptures due to excessive hydraulic pressure produced.

All of the above – except the first – are BAD things that can happen. That is why ALL positive displacement pumps should have a reliable PSV installed on the discharge line UPSTREAM of any discharge line block valve. This PSV should normally discharge to the source tank or vessel. (NOTE: you do not show this required PSV in your sketch. Make sure you have one.)

There is no need for minimum flow protection on postive displacement rotary pumps – even those with variable speed control. HOWEVER, there is an important need for a discharge PSV on the same pump.

I hope this explanation helps you to understand the basic difference between both pumps and why a centrifugal is more demanding.

[deltaChe]

Dear Mr. Montemayor

I miss to put PSV on the discharge line upstream of the block valve in case to protect the downstream equipment overpuressure.

Thank you very much for your insightful comment.
I really appreciate it.

[kkala]

Quite agreeing to previous posts, it is noted that a constant speed rotary pump could have a sending line and a return line with a control valve on each of them. The two control valves are linked to each other, so that sending (i.e. required) + return flow rate equals total pump capacity. This control scheme is often applied in e.g. fuel oil feed to a boiler. The return line is not a min flow line, since all pump capacity passes from it if a valve on sending line is closed. And PSVs are still necessary in this electronic control scheme.

[manchester]

May I ask why is discharge line of PSV may not be connected to pump suction line?

[Art Montemayor]

Manchester:

It is not smart or safe engineering to connect a PSV discharge from the outlet of a positive displacement device back into the suction line of the same device. The same applies to any recycle stream from the same discharge port. The reasons are very basic and simple:

• as stated previously, a positive displacement device is very efficient about adding momentum energy to a fluid. Consequently, if the discharge is throttletd, the fluid will increase in hydraulic pressure and temperature. If you recirculate the discharge back to the pump's suction, you will start increasing the fluid's temperature in that recycle - in a very quick and efficient manner. You are advised not to do this because this will lead to flashing of the fluid, loss of pumping, thermal damage, etc., etc.
• the suction piping of a pump is not designed to handle a recycle stream; flow turbulance and upsets are inevitable at the entrance of the pump if you try to do this; this creates flow eddies and fluctuations in pressure. Remember, you are releasing pump energy within the fluid back into the suction side as a fluid expansion. Pumps are designed for steady flow - not flow like this.
• any PSV discharge will have to overcome a built-up backpressure. This changes the characteristics of the PSVs capacity and its ability to safely release the excess pressure. You are advised not to mess around with adding more resistance to the PSVs designed back pressure.

Any attempt to recycle a pump's discharge back into the suction port is a naive attempt to try to "save" money by cheapening the piping requirement to send the fluid back to its source. There is always a trade off to pay for this: more operational, safety, and maintenance problems will result.

I hope this experience is of help.

[kkala]

In addition to the previous post, it is noted that we send min flow return to the suction tank as a normal practice, even though pump is out of tank dike (if any).
There are some cases when this is not easy; then we connect to the suction line as far as possible from the pump, so that design temperature can be approached after at least half an hour (a TI had better be also placed on suction line near the pump, with alarm). In one case client had objection even to it, and finally TI/TSHH was placed on the loop to stop pump motor in case of excessive temperature (the control is electronic).
Of course each of above steps brings a reduction in safety; connection to the tank (where the pump takes suction from each time) is the first priority.

[deltaChe]

Dear Mr. Montemayor

I still have some questions.

When we start a postive displacement pump or any type pump, the pump usually required a recycle valve that can be opened during start up.

Therefore, the pump can start without a load and a soft starter of VFD so that it can slowly ramps up to full speed.

It seems that starting the pump under full load and no return line on the discharge side is very hard on the pump as start-up proceeding.

So is that means every pumps' discharge side should have a return line or pumparound loop back to its upstream tank or vessel.

Thank you so much for your kindly help.

[Art Montemayor]

Steven:

You state you have questions, but yet you make statements rather than ask questions. Nevertheless, your basic understandings of the principles behind a positive displacement pump are correct and valuable because they lead you to understand how best to start, operate, and stop such a device.

You are correct in surmising that a “soft” start is what is proper for a positive displacement pump. This is always the recommended manner of starting a positive pump: always afford an “open” (non pressurized) discharge to the positive displacement pump. This gives the pump and its valves an opportunity to initiate the passage and stoppage of liquid fluid through its pumping chamber in the first initial strokes. This action removes any possible gases or non-compressibles trapped in the system and starts the fluid lubrication of the chamber and the valves. Doing this initially without any pressure load and slowly bringing up the discharge pressure allows the pump a steady increase in load that does not give it any sudden shocks or “hammering” of fluid. The obvious way to ensure this is to allow the pump to pump initially in a “pump-around” circuit that is allowed by using a discharge check valve and an open by-pass valve, located immediately upstream of the check valve, that allows for 100% recycle back to the source. I do this startup procedure using a PLC and slowly and step-wise close the by-pass valve allowing the discharge pressure to increase in increments that ultimately reach a sufficient level of pressure to open the discharge check valve and commence pumping fluid downstream. This method ensures a steady, and increasing electrical load on the driver motor that gives it a longer life and less wear. It also ensures a safe startup of a positive displacement pump.

Centrifugal pumps don’t require this type of careful startup (AND SHUT DOWN) because they are not a threat to unrestricted discharge pressure increases. Centrifugal pumps are usually started up with their discharge totally blocked off and subsequently slowly allowed to commence pumping by throttling the discharge block valve. This, you will note, is the direct opposite of what is done with a positive displacement pump. It is desirable to have fluid flowing through a positive displacement pump both during an unloaded startup AND an unloaded shutdown. I hope it is needless to state that all positive displacement pumps should always be shut down totally unloaded.

I hope this helps you out.