E-Zine April 2016
Pumps can be as simple as a waterwheel, and as complex as a multistage deep well turbine pump with 15 bowls. Applying pumps to processes is important, and doing it correctly can eliminate some extremely serious problems in the system. For example, selecting a high volume pump based on future expansion may produce problems when the pump is throttled down to current needs, and it may also waste power, which is costly. Using a centrifugal pump for chemical transfer, instead of a metering pump, may not be accurate enough to deliver the same volume or mass in a custody transfer application.
Pumps are based on trade-offs. In sizing and selecting a pump, you trade off pressure and friction and throughput to get the very best performance you can get.
You need to know a few important pieces of information to properly specify a pump. First, you need to know what the fluid is and what the fluid properties are.
If you have a process fluid that is toxic or corrosive, you need to know what materials you can use in contact with the fluid. If you have a process fluid that is abrasive or that contains solids, you need to know how large the entrained particles are and how abrasive the fluid is. If you have a process fluid that is flammable, you need to know that. If you have a process fluid that is ultrapure, you must know what materials you can use for wetted parts.
You also need to know the total head and the desired flow out of the pump.
To calculate total head, you need to know the flow rate through the pump; the size and number of fittings in the system; the length and size of the piping in the system and the number, type and location of all of the valves and process equipment, such as inline filters, in the system. The reason you need to know total head is that pump performance curves are a function of flow rate versus total head for the various impeller diameters of that model pump.
Click here to read “So Many Pumps, So Many Applications... Centrifugal, Turbine and Rotary Pumps”
From Flow Control (January 2002)
Quiz Corner: Variable Speed Drives
By David W. Spitzer
E-Zine April 2016
The news that a motor operated by a variable speed drive that I specified for a new agitator was overloading and shutting down was a prelude to such a very interesting situation. Sometimes you have to do more than your part of the project to keep yourself out of trouble and the company afloat.
While it was possible that incorrect overload settings could potentially be the root cause, the problem was likely more serious. Observation revealed that the motor rotated without incident at low speeds, but would consistently trip out on current overload at less than one-third speed. The overload settings were appropriate and the motor current did exceed its settings when the variable speed drive tripped.
What do you think is causing the problem?
A. The motor is too small.
B. The variable speed drive is too small.
C. The agitator is too big.
D. The agitator is damaged.
E. The motor is damaged.
F. The variable speed drive is damaged.
How would you approach resolving the problem?
A. Only operate the mixer at low speed.
B. Investigate the agitator for size and damage.
C. Investigate the motor for size and damage.
D. Verify that the agitator operates properly with a starter.
Surprisingly, all of the answers are potentially correct problem causes and approaches to resolving the problem, however they require further investigation. Verification of the physical and electrical size/dimensions of the agitator, motor and drive revealed that they conformed to their respective specifications. The variable speed drive was properly sized for the motor.
None of the equipment was physically damaged. Although lacking accuracy, manual (ammeter) motor current measurements confirmed that the motor current was relatively high. This implies that it is likely that the variable speed drive was tripping appropriately.
At my request, a review of the project engineer’s files was performed. It revealed that the maximum agitator speed needed to achieve adequate mixing was 50 percent, and that the agitator shaft required 2 hp of rotational energy at that speed. The 3 hp motor installed on the agitator would seem appropriate. However, the rotational energy (horsepower) that a motor can produce is proportional to motor speed, so a 3 hp motor operating at 50 percent speed is only capable of generating 1.5 hp of rotational energy. Therefore, the installed 3 hp motor is not sufficient to operate the agitator.
This explains why the motor was drawing high current and tripping the drive. To resolve the problem, the motor and drive were increased to 5 hp in order to provide 2.5 hp of rotational energy at 50 percent speed.
Sometimes doing only your part of a project is just not enough. In this case, investigating beyond my discipline allowed the problem to be quickly defined and resolved.
Additional Complicating Factors
If the agitator were required to operate at full speed, the rotational energy requirement for a centrifugal load would increase as the cube of the speed. In this application, doubling the speed would require 8 times more rotational energy (hp). This would increase the requirement to 16 hp, so a 20 hp motor/drive would be applicable.
Not only would the installation of a 20 hp motor/drive be more expensive, the motor will operate less efficiently its typical (low) operating speeds. In addition, physical constraints of the agitator motor mount might preclude its installation.
From Flow Control (February 2002)
Quiz Corner: High Turndown Sewage Flowmeter Application
By David W. Spitzer
E-Zine April 2016
In 1990, part of my birthday “present” was the privilege of inspecting flowmeters used to bill for sewage flowing from one authority into the sewer system of a second authority. A sewage treatment plant treated sewage from the first authority, sewage from the second authority, and storm water from the second authority.
The total flow delivered by the first authority was determined by adding the flows from each of the flowmeters. The total flow delivered by the second authority was determined by subtracting the first authority’s total flow from the total plant flow measurement. Half of the sewage treatment plant budget was allocated based upon the relative amounts of sewage delivered by each authority.
Recently, I received a phone call from the first authority stating that their percentage of the total flow had increased significantly in previous months and was fluctuating on a daily/weekly basis. What do you think is causing the problem?
A. A break in the first authority’s sewers had allowed flow from an underground spring to enter the sewer system.
B. A customer in the first authority’s system was pumping his basement.
C. The second authority removed some storm drains from their sewer system.
D. The four flowmeters are measuring high.
E. The total plant flowmeter is measuring low.
Surprisingly, all of the answers could have potentially caused the problem. In the mid-1990’s, there were stories of an underground spring infiltrating the sewer system (Answer A). And then there was the rumor about the homeowner with a water problem in the basement that was resolved with a sump pump that operated continuously (Answer B).
The second authority could upgrade its sewer system to reduce the inflow of storm water. However, without contemporaneous sewer improvements and a steady shift in the percentage, this would not likely be the cause of the increased percentage (Answer C).
I was contacted in 1990 because the percentage of flow delivered by the first authority was higher than would normally be expected for its population, size and location (Answer D). Examination of the flowmeters between the first and second authorities revealed that all of these flowmeters were measuring high. However, by the end of the 1990’s, most of the equipment had been replaced and appeared to measure with reasonable accuracy. In addition, the total flow delivered by the first authority had not increased dramatically, so it was unlikely that the flowmeters between the authorities were causing the problem.
After eliminating the other possibilities, the likely correct answer is that the total plant flowmeter could be measuring low (Answer E). If this were the case, the difference between the total plant flow and the total sewage flow from the first authority would be lower, with a higher percentage of flow allocated to the first authority. (On one occasion some years ago, the flowmeters between the authorities measured so high, and/or the total plant flowmeter measured so low, that the total plant flow was less than the flow from the first authority. Therefore, by calculation, the second authority actually consumed sewage on that day!)
Control of the plant was such that the three wet well pumps were cycled on/off as the wet well level changed. Single-pump operation occurred most often and was approximately 25 percent of the total plant flowmeter full-scale flow. Because the total plant flowmeter was a differential pressure device, the differential pressure generated across the flowmeter was only approximately 6 percent of the transmitter full-scale. In other words, normal flow produced a differential pressure that was low in the transmitter range where accuracy was poor.
Additional Complicating Factors
Passing conversation revealed that the sewage treatment plant was upgrading their wet well pumps to operate using variable speed drives. Control strategies that use variable speed drives typically operate equipment to match load. If this is the case, single-pump operation can occur at perhaps 10 percent (or less) of the total plant full-scale flow. This corresponds to a differential pressure of only 1 percent of full scale --- further compounding the problem.
From Flow Control (March 2002)