PUMP MAGAZINE: Questions and Answers (71-80)
Question #71 Dear Sir,
I enjoy the technical articles and Q&A sections, - they are very informative and helpful.
Do you provide pump consulting? Is it strictly online, or can you come to the field? We have a couple of problematic pumps, and would be interested if you could help us solve these problems. How do we start?
Answer: The first step is for you to email us a description of the problem. Try to include as much information as possible so that we get a better picture. We will evaluate and, if anything obvious, may suggest a solution right away, at no charge. If the problem appears to be more involved, we may need to ask a few follow-up questions, and will give you an estimate of the time and cost it would take to fully assess your problem and solve it. We can often provide a solution and recommendations interactively including a final report if required, but if a field visit is required we can offer a field troubleshooting approach.
Dr. Lev Nelik, P.E.
Question #72 Dear Dr. Pump,
I just saw a question (#71) regarding your engineering consulting and field assistance. We are a pump distributor, and occasionally have a need for a technical support. We have our inside sales people who are good and know the lines we handle, but occasionally the questions that our customers pose are somewhat beyond our expertise, or too involved technically. We also have questions beyond the immediate product lines we handle, and having someone like yourself “on call” for such situations would be good for us and our customers.
In cases like these, we wonder if you might be interested to be our “technical backing”?
Answer: No problem. As you can see from our web site, the types of questions we get involved with are technical, and require engineering insight. Some questions can be answered right on the spot, and others may be more involved and require calculations, drawings, engineering source books, lookup tables, etc.
We get questions from direct pump users, as well as pump distributors, - from US as well as worldwide. In most cases, our response is within 24 hours.
Let us know more about your needs, your location, and some additional background. There are several arrangements we can work out to provide the technical support you are referring to, to extend and expand your technical muscle, by becoming an addition or an extension of your Tech Support.
Dr. Lev Nelik, P.E.
Question #73 Hi,
I found your site on the web during a pump search and am quite impressed at the info you have compiled.
What I am looking for is info on "The Best" pump in the World/USA for Agricultural/Nursery/Home use. The well was just drilled at 500 ft and I am looking for the most reliable pump on the market. Looking at pumping about 50-70 gallons a minute from around 300-400 ft.
Thanks for any info you might have to help me.
Answer: Ron, - thanks for your kind words in reference to our web site.
We are forwarding your question to our associate pump distributor, who is very competent with pumps, including submersibles, - they will respond to you directly shortly. If there is such “The Best Pump in the World” that you are looking for – they will find it for you!
Dr. Lev Nelik, P.E.
Question #74 Dear Sir,
What is the effect on power when discharge head of positive displacement pump/reciprocating pump is raised, while keeping the suction head constant. What is the relationship of power with head? Kindly explain.
Answer: Power is in direct proportion to the differential pressure. For example, in US units it is:
BHP = (ΔP x Q) / 1714 / EFF,
where (ΔP) is differential pressure in PSI units; (Q) is flow in gallons per minute; (EFF) is pump efficiency; and 1714 is a conversion coefficient.
Question #75 Dear Dr. Pump,
What causes a fluid end to crack around the suction valve seat deck for Triplex and Quintuplex Plunger Pumps?
Answer: The area between the suction and discharge valves is subject to the most fatigue stress. It cycles from zero psi to discharge pressure with every rotation of the crankshaft. For a conservative 300 RPM application that is 432,000 cycles per day. Most valves are seated in the fluid end on a taper, so the higher the pressure the more the suction valve is pressed down into the fluid end. This force, along with the usual stress risers caused by corners near the suction valve, cause this area to have the highest stress.
Question #76 Dear Doctor:
After reading the articles published at your publication, I would like to congratulate you for this kind of help for all who have interest in learning more about centrifugal pumps. If possible, I would like to receive more details about the term "RATED CAPACITY". What is the difference between "RATED CAPACITY” and “NORMAL CAPACITY"? This concept is not clear to me.
Agnaldo Borges, Mechanical Engineer
Answer: Dear Agnaldo, -
Use SEARCH function of our web site, and type the word “API” – several answers will come back. You should find some discussion there.
You are right – this is confusing, and API-610 (American Petroleum Institute) should change this definitions. This is why, in practice, nobody really applies this definitions literally (because nobody really knows what this means!), but instead elaborate a bit more when discussing. Usually, there is a condition at which a pump is expected to operate – and it really (should!) be called “rated” point. The word “normal” should really go away! There is little “normal” about a “typical” pump operation! In general, a pump would almost never have the rated point exactly where best efficiency point is. It is almost imposiible to select a pump perfectly at the BEP: there is infinite number of operating conditions, and only limited number of pump sizes to pick from. It would be “nice” to “hit the BEP” all the time, but practically impossible. In real world, a centrifugal pump operates all over the place, depending on a flow need – and the “hope” is that operators will not operate the pump too far outside recommended envelope of operation, which is typically between 70 to 120% of BEP point, which is what API 610 calls for.
Hydraulic Institute, which is sometimes a good source of information on mainly definitions, unfortunately does not cover this subject sufficiently. API, which traditionally has been more practical in this regard, has much more vigorous approach, with more details. The reason API deals with a subject of limited flow range is because it has a direct impact on reliability of equipment. API committee members are practicing engineers, from the pump user community, and form the pump manufacturers, while Hydraulic Institute has taken somewhat more general approach, with not as much practically useful items, - although they have published occasional useful material, such as pump sump guidelines, and also a viscosity correction chart. HI has not been known, traditionally, as having a strong focus on technical details of engineering aspects of pumps.
So, why 70% is a limit? What if a pump runs down to 69% BEP? Or 60%? Or 40%? How bad is it? - The further away from BEP a pump runs, the trouble may happen. Radial thrust increases, suction recirculation starts, shaft deflects, seal leaks, bearings get overloaded, etc. How bad is bad? There are several related articles throughout the Pump Magazine, and you can also look at those via SEARCH function and typing-in a keyword, such as “recirculation”, “thrust”, “cavitation”, etc. In fact, there are formulas and more articles written on this subject, where minimum flow relates to suction conditions, pump specific speed, pump suction specific speed, pump energy level, etc., etc., and we would be glad to help you with those if you like. However, for now, your next step is to do some more SEARCHing by clicking a SEARCH button on our entry page.
Thanks for asking! – and keep on pumping!
Question #77 Dear Dr. Nelik,
I have been trying to find out what does a "STRING TEST" mean for centrifugal pump train? Where can I find a good explanation about this subject?
Agnaldo Borges, Mechanical Engineer
Answer: Dear Agnaldo, - you are becoming our regular reader! Good for you! Keep learning!
A “String Test” refers to a rotordynamic analysis of a complete “train”, i.e. a pump, coupling, drives, flywheels, and anything else that is a connected to a rotating “string” of rotors. There is a lateral and torsional analysis involved. For example, let’s say an engineering company, such as Bechtel, Foster-Wheeler, or similar is conducting an engineering analysis of such “train”. They would obtain information on pump rotor mass and moment of inertia. They would get similar data from a coupling manufacturer, motor manufacturer, etc. They will combine these together into a “string” of computer data, and run a computer program to calculate major critical speeds – usually the first 3-4 are most important. Both lateral and torsional results are produced, and if a “train” rotating speed is close to one of the critical speeds determined by the analysis – there is a problem, and something then needs to change. A change may mean increasing a rotor shaft, or decreasing its length, or modifying a coupling design, or changing bearings with other types having different stiffness and damping characteristics, etc.
In addition to analysis, sometimes an actual field test is performed. However, the analysis should be done first. Typically, this type of engineering work is rather sophisticated, and is done for large units, with high energy density and critical applications. For these, a field testing is to find a problem, for example if vibrations are high and failures are frequent. A proper computer analysis could avoid troubleshooting of a problem on a first place.
Pumping Machinery is often involved in field troubleshooting. If you should have a specific need to fix a problem with an installed piece of machinery, and a “string” is “not behaving” – give us a call, we may be able to help.
Dr. Lev Nelik, Pumping Machinery
Question #78 Dear Doctor Pump,
I'm new in the industry and rotating machinery has become the field which I have chosen to pursue. A simple (for you I think) question I have: can or should a pump operate below its minimum impeller diameter? I ask because I have new operating point for an existing (used) pump. Unfortunately the new point is below the minimum impeller diameter as shown on the performance curve.
Hope to receive your reply very soon!
Answer: Dear Louie:
There is not much magic behind the minimum diameter. The main limitation is the impeller wear ring diameter – at certain trim you would cut below the available metal, i.e. the impeller will get worked out into a pile of machined chips! However, strictly “hydraulically speaking” – you can extrapolate below the minimum, and it will probably work fine, although obviously at poor efficiency (although even that you can extrapolate from the existing plotted curve that you have). Years ago, when I worked at Ingersoll-Rand as a pump hydraulics designer, we tested impellers trimmed below the minimum published diameter, with performance essentially following pump affinity laws (subject to a so-called Stepanoff’s correction factor for the diameter reduction below certain value, when affinity laws begin to be less accurate).
From the logistics perspective, however, a pump manufacturer usually will not guarantee a pump operation outside of their published envelope (the less guarantees the better for them), so if you cut the impeller below the published minimum, - you are on your own.
We invite other people to comment. It would be of interest to hear the input of large and reputable engineering construction firms, like Bechtel, Fluor, and others, for example, what typical practices, regarding impeller re-rates they encounter, in their interactions between the pump users, and pump suppliers? I am sure our readers would be interested to know, and a discussion could make an interesting topic and a good contribution to the pumping community.
I hope it helps,- let us know how your trimmed-below-the-minimum impeller works! By the way – don’t forget to check out a Pump School section of our web site! – the next Pump Training Event is coming up soon! – or, if you like, we can set up a group training for your technical team at your own facility site? Let us know.
Dr. Lev Nelik, P.E.
Question #79 Dear DrPump,
In one of the projects there are (2) boiler feed pumps of capacity 190 m3/hr, with head of 1220 m, operating in parallel and feeding two boilers through common suction and discharge header. The shut-off head proposed by the consultant is minimum 125 % of rated head for better control using feed control valve regulation since pumps are intended for parallel operation. However, one of the Bidders offering the pumps for the project, indicated that only 110% shut off head is sufficient for parallel operation. Is 110% shut-off head is acceptable or a minimum of 125 % shut-off head is preferable for the pumps in parallel operation?
Answer: API-610 spec (which is also often used by the boiler feed pump manufacturers and users) 8th Edition states in paragraph 2.1.11 that “Pumps that have stable head/capacity curves (continuous head rise to shutoff) are preferred for all applications and are required when parallel operation is specified. When parallel operation is specified, the head rise shall be at least 10 percent of the head at rated capacity…”
The Bidder that quoted 110% is thus technically within the acceptable range, although just at the bare limit of it. As a note, for critical equipment, which boiler feed pumps are, it may be a good idea for your engineers or whoever you use as representatives, to actually witness the pumps being tested at the factory, and not simply approve the curves after the fact. You want to make sure the testing indeed produces a “continuously rising curve”, and not an “approximated” one. You should specify in your purchase order witness test, so that the pump manufacturer knows upfront this will be watched carefully. You representative should be well qualified in pump testing, and should be standing right there, next to the test engineer, actually taking the data along (pressures, flows, power, temperature, etc.), and comparing his own plotted curve versus manufacturers test department.
Dr. Lev Nelik, P.E.
We often get inquiries from Paper Plants asking for pumps for pumping 2 T/HR, at 4% pulp consistency. How do we convert this into M3/HR for selection?
Answer: Fundamentally, a pulp mill produces pulp stock, which is essentially a pulp (cellulosic fibers) mixed (or suspended) in water. A mill’s objective is production of the fibers, not water. The water is there only because a mill uses it as a medium to transfer fibers. A mill would like, of course, to use as little water as possible, but then the consistency of fibers would be so “thick” that the pumps would not be able to pump it. Thus a compromise is found from experience. A typical pulp stock (at the Pulp Mill) is 10 to15% consistency, and a typical paper stock (which is actually used to make paper at the Paper Mill) is 0.5 to 2%. Some plants have a Pulp Mill and a Paper Mill at the same location, and others have them separately, and transport pulp to a Paper Mill by trucks or some other means.
Mills use various terms to denote pulp consistency. A common term is “OD” (over dry), which is a moisture-free fibers. If 100 kilograms of pulp stock contains 10 kilogram of fibers, then the Oven Dry consistency is 10%, i.e. OD=10% .
So, the people who produce fibers want to know how many kilograms (or pounds, if you are in a non-metric country) of fibers they produce, but the people who pump this stock want to know how many gallons per hour (gpm) or cubic meters per hour (M3/hr), etc., will be flowing through the pump, so that they can select the pump for the mill needs.
Thus the conversion becomes rather straightforward. First, think how you would convert only water from tons per day (TPD) to gallons per minute (GPM)? In US, 1 ton (called “short ton”) is 2000 pounds. 1 gallon of water weighs 8.34 pounds. You can now easily see that:
GMP(water) x 6 = TPD(water)
(conversion coefficient convenient come out as almost an exact number 6 for the US system of units)
But that is for water only. If, only 4% is fibers, then multiplying by 4/100, we get the tons of fibers:
GPM(stock) x 0.06 x C = TPD(“dried” pulp)
(where C is stock consistency)
If your mill produces 2 tons per hour (2x24 = 48 tons per day), of a 4% pulp, then back-solving for GPM we get:
Pumped stock flow = 48 / 0.06 / 4 = 200 GPM
This means a pump must pump 200 gpm of stock. You can convert to metric units by straight conversion:
200 / 4.4 = 45 m3/Hr (approximately)
Keep in mind proper units. Metric ton (I guess you would spell it as “tonne”?!) is 1000 kilograms, but the US “short ton” is 2000 lbs (not the 1000 x 2.2 = 2200 pounds).
Also, paper mills typically do not bother with the formulas, - they have charts where the horizontal axis is gallons of stock per minute, vertical axis shows tons of pulp per 24 hours, and there are lines of percent stock consistency on the chart.
Let us know if this helps,