5 replies to this topic

[Jiten_process]

Dear all,

I am evaluating a vendor pump performance curve for a centrifugal pump used in CO2 recovery plant. Pump is circulating lean solution pump to CO2 absorber. My required RATED flow is 308 m3/hr @ 475 mlc. The NORMAL flow is 265 m3/hr.

I have got the offer from a pump vendor (i feel, i should not disclose the name of the vendor...) who has supplied many pumps for the similar service for many clients. Upon seeing the curve i found that for the particular model, vendor has selected the impeller dia such that my rated point comes on best efficiency point. at first look it looks nice...but the problem arises when i see my normal point on the same impeller H vs Q curve. I could see that for the rated condition i.e. 308m3/hr and 475 mlc if i go towards left side of the curve upto the normal flow i.e. 265 m3/hr, the head rise is almost 40mlc. that means that for the selected impeller dia i can get my normal flow at almost 520 mlc. I understand it this way that, for the selected impeller dia and model if i want to operated my pump at normal flow (away from the rated flow) i have to drop almost 4kg/cm2 pressure at control valve.

Now while doing the pump hydraulics, i have kept approx. 0.8 kg/cm2 pressure drop for the rated flow condition for the control valve. It means that the same control valve have to drop 0.8 kg/cm2 pressure at rated condition and 4 kg/cm2 at normal condition. IS IT FEASIBLE???

problem becomes much severe when i further go towards left up to the turndown point which is 156 m3/hr. At this point head rise from rated head is almost 80 mlc (approx. 8kg/cm2). Plz note that at 156m3/hr also, i am still well away from the minimum recirculation flow which is 100 m3/hr.

the head rise from rated head to shut off head is approx. 105 mlc ( apprx 10 kg/cm2).

IN SHORT THE REGION ON CURVE WHERE THE RATED CONDITION IS ACHIEVED IS VERY STEEP.

i want to know

1) IS THERE ANY GUIDLINES OR CODE REQUIREMENT LIMITATION IN HEAD RISE FROM RATED POINT HEAD TO SHUT OFF HEAD??? i have gone through the API 610 there is a limitation of 5% head rise from rated head to shut off head but that is for PUMPS IN PARALLEL, for single pump there is no exact value?

2) what is others experience, can a control valve be designed for such a high range in pressure drop...i guess it is quite difficult.

kindly revert in case any other inputs required to comment on the problem

Awaiting responce from the experience ChE Jedis....

Edited by Jiten_process, 25 June 2010 - 01:04 PM.

[Jiten_process]

110 views... not a single comment....

is it something out of syllabus or what...???

[Art Montemayor]

Jitan:

This is not a criticism, but I find it difficult to visualize your problem without seeing the pump performance curve. I don’t care who the pump manufacturer is, but I would certainly need to see the subject performance curve for myself. Some of your terms are not readily understood:

• What is mlc? Is this the head? Does it mean meters of liquid column? I have always used feet of head or meters of head, so I get confused with your acronym.
• What do you mean to say by “the turndown point”?
• What is your design operating point on the curve? That is the most important point that you should consider, but you don’t even mention it. Why do you base yourself on the so-called “Rated” position on the curve? What do you mean by “rated”? Obviously, the pump can achieve more capacity than the “rated”, so what importance does this point have over the design point? More importantly, what capacity do you intend (or are designing) to operate at?

To answer your questions:

• There are no guidelines or code requirement limitation in head rise from rated point head to shut off head.
• You can design the appropriate control valve trim to throttle the capacity of the pump discharge. However, why do you employ (once again) a so-called “rated” flow value in selecting the pump impeller size? Why not employ the real-life value of design flow capacity?

I hope this helps you out.

[Jiten_process]

Dear art,

firstly thank you very much for your time. I beg pardon for my poor post. let me clear you some jargons which i have used in my post assuming that it was self explanatory.

u r right, mlc mean Meter Water Column.

turn down point means...flow required when plant is running on 60% turn down. It is client's requirement for our project to have plant to be able to run on 60% turn down. So turn down flow is at that condition. This flow is given by licensor. ( normal flow, 265m3/hr * 0.6 = 159 m3/hr)

Rated point means in my project, it is a overdesign flow requires to be able to run plant at 110% capacity. this flow is nothing but 16% overdesign on normal flow. (normal flow i.e. 265 m3/hr * 1.16 = 308 m3/hr.

My pump will at steady state condition run at normal flow i.e. 265 m3/hr. But since i have to design my pump at rated flow i.e. 308 m3/hr i have done hydraulic calculation taking this flow and calculated the required differential head which is 475mlc. i have kept 0.8 kg/cm2 pressure drop while doing hydraulics at this (308 m3/hr). Note that although i have done hydraulics for 308m3/hr i will run my plant at normal condition at normal flow (265 m3/hr). and for that i have to kill(means drop) some pressure across my flow control valve to reach at my normal flow i.e. 265 m3/hr.

i hope i have made myself now clear...

also find the attached pump performance curve... there i have marked my rated flow (nothing but design flow), normal flow and turndown flow.

Attached Files

•  curve.bmp   3MB   97 downloads

Edited by Jiten_process, 26 June 2010 - 01:06 AM.

[chemsac2]

Jiten_process,

Your pump curves show 3 curves each for rated, normal and minimum flows. Also, curves are flat curves against your statement that curves are steep. Are you using VFD for motor for capacity control?

I would reply based on your original post as second post has confused more.

1. SOHR (shut-off head rise) in your case is about 22% which is less than 25% considered as the limit. Generally we limit shut-off head rise to 25%. Any attempt to have higher SOHR can lead to unstable pump curve i.e. drooping curve and hence this precaution. This is not a code requirement but but general engineering guideline. Refer to phi factor in pump handbook chapter 2.1.

2. Control valve pressure drop range in your case is very steep. I have seen DP max to DPmin of about 5. Your case of 10 may be a little high. Flow range for minimum to rated of 50% is normal. SOHR of 22% is also acceptable. System resistance in your case is highly variable. Probably either your line size is small or you have very long line length. Also, ensure you have used same static and pressure heads for calculations for rated, normal and minimum cases.

Control valve designed as per your pressure drops would certainly kill pressures, but question to be asked is to if it operates in controllable region i.e. rangeability issues.

3. With high DPmax/DPmin, you may expect rangeability issues. Generally globe control valves have rangeability of 50:1 based on inherent characteristics. In your case, distortion coefficient would be significant and hence rangeability would be less (may be not more than 10:1). A solution in such a case is to use split range control valves.

Regards,

Sachin

Edited by chemsac2, 25 June 2010 - 11:38 PM.

[Propacket]

Jiten_process,

Why don’t you use other methods for capacity control like followings?
1-Recirculation control: Recalculating portion of fluid back to the suction. It has a disadvantage that if flow of recirculation fluid is large (as in your turndown case) than total flow of fluid at the suction will be extremely large causing high power consumption and high friction losses.
2-Pump speed control: Speed control is better because it reduces power consumption. This is recommended for your case due to large difference in flow rates.
If you want to use throttling for capacity control, I don’t think single port control valve can handle such high range of pressure drop. There are two solutions to your problem:
Multiple Port Control Valve: I have attached some literature about two port valves that will be helpful for you.
By Pass Control Valves: You can use multiple bypass control valves sized for required flow and pressure drop. I think this is not a feasible option due to high cost of control valves.
Hope this helps.

Attached Files

•  Two Port Valves.pdf   50.36KB   60 downloads

Edited by P.Engr, 30 June 2010 - 05:18 AM.