- FlowStream Blog
- The Magazine
- FlowTube Videos
- White Papers
- Vendor Directory
- Pump Guy Seminar
|Larry Bachus |
(a.k.a. Pump Guy)
To: Larry Bachus
Subject: Pumping Viscous Fluids
Larry “Pump Guy” Bachus,
I have some issues regarding pumping of a viscous fluid. The fluid is a juice concentrate with a reported viscosity of 15,000 CPS. Does a viscosity this high make sense?
The fluid is pumped with a positive-displacement pump from a holding tank through stainless steel pipe. The pipe has a number of bends, elbows, valves, and Ts. The fluid then passes through a meter and into a mixing tank. When the holding tank level drops to a certain point, the meter becomes inaccurate.
We suspect that air is being drawn through the system because of the low suction pressure. But we are not sure.
We are trying to calculate the pressure drop in the suction piping, but the numbers we have calculated using the normal method of “equivalent lengths” do not make sense because of the viscosity. For example, the Reynolds number we calculated is about 2.5, which goes off the chart for locating the friction factor. I have two questions: 1. What method can I use for calculating the pressure drop in the system? 2. Do you have any suggestions to resolve the problems we have at lower tank levels?
From: Larry Bachus
RE: Pumping Viscous Fluids
A viscosity of 15,000 CPS (Centipoise) is completely logical for juice concentrate. I tend to think of viscosity in the Centistokes Scale (CKS). Centipoise = Centistokes x sp. gr. Toothpaste is rated at about 25,000 CKS. Peanut butter is rated at about 350,000 CKS. (If I knew the sp.gr. of peanut butter, I could calculate its viscosity in CPS). So, juice concentrate at 15,000 CPS is perfectly logical. (If the juice concentrate has a sp.gr. of two, then the CKS viscosity would be 7,500). This is about right for juice concentrate.
Pipe friction tables (resistance and the Reynolds number) are based on ambient water, not juice concentrate. The tables assume that the pipe is new and free of scale. The tables also assume that all valves are completely open with no resistance. The tables assume that all filters and strainers are new and clean. The tables assume that the construction and installation instructions were followed to the letter with no deviation. None of this is reality.
|Resistance tables don’t reflect reality, and they often lead to pump problems.|
Install a pressure gauge at the beginning of a run of pipe-and-fittings. This will be the upstream gauge. Install another gauge at the end of the run of pipe and fittings. This is the downstream gauge. Do this for both the suction side and discharge side of your piping system. Note the differential between the upstream and downstream gauge readings (suction and discharge) with the system off and no movement through the pipe. If there is a differential, it will reflect the different static liquid columns on the gauges.
Next, start your pump and allow the flow to begin and stabilize. Now, note the differential pressure from the upstream to the downstream gauges (suction and discharge) with the system functioning. The further differential pressure between the two gauges will reflect the energy lost into the system from velocity and friction. Convert this pressure into head. This is the friction and velocity resistance across your piping system. The resistance formula is called the “Bachus-Custodio Formula.”
The tables are only a suggestion of possible friction. They are moderately useful when the system only exists on the drawing board. But once a plant is built and commissioned, the tables are useless. Why do I say this?
Those tables don’t represent reality. Those tables don’t show the resistance across a check valve with a weighted arm. Those tables can’t record a hole, or plug, in a strainer basket. Those tables can’t record the losses across a partially clogged filter screen. Those tables can’t see a bolt or wrench stuck inside a pipe elbow. Those tables can’t reflect a change or alteration that might occur at the construction and installation phase. Those tables can’t measure the viscosity correction factors. Those tables can’t measure any change in viscosity if the liquid temperature changes. If you assume the fluid velocity from one table and insert fictional numbers into the resistance formulas from another table, then you calculate fictional friction values.
The gauges will record everything as it is now, including: the wrench stuck inside the pipe elbow; the hole in the strainer basket; the pipe joint gasket that interferes with flow: the check valve flapper that won’t completely open; the globe valve that looks like a gate valve; and the flowmeter Venturi that was inserted into the piping last weekend that you don’t know about yet.
If the temperature, and thus viscosity, changes in the operation, run the test at both temperature extremes. Now you can determine if the suction piping arrangement is starving your pump, aspirating air, or precipitating turbulence bubbles in the juice. Measure the resistance losses with gauges. Don’t guesstimate the losses with fictional numbers from those tables. This very discussion is in Chapter 8 (“System Curves and Troubleshooting Piping Systems”) of my pump book (see page 42 for details). The exercise marches you step-by-step through a realistic system.
That’s my suggestion. I employ the formula frequently to resolve problems with pumps. I hope this helps.
Larry Bachus (a.k.a. Pump Guy)
Larry Bachus, founder of pump services firm Bachus Company Inc., is a regular contributor to Flow Control magazine. He is a pump consultant, lecturer, and inventor based in Nashville, Tenn. Mr. Bachus is a member of ASME and lectures in both English and Spanish. He can be reached at email@example.com or 615 361-7295.
Get Hands-On Training at the
PUMP GUY SEMINAR
Oct. 21-23 - Houston, TX
ATTENDEES RECEIVE: Complimentary Companion Text - Free Shuttle Service To & From Hobby Airport - Free Parking - Discounted Hotel Rooms - Food & Bev Service.
Larry Bachus (a.k.a. "Pump Guy"), a regular contributor to Flow Control magazine and a widely recognized expert on pumping technology, recently presented his Pump Guy Seminar in the Chicago area to an eager crowd of pump users. Here''s what some of the attendees had to say about this training event:
KEY SEMINAR TOPICS INCLUDE:
• Basic Pump Principles
• The Affinity Laws
• Work & Efficiency
• Pump Classification
• Pump Curves
• System Curves
• Shaft Deflection
• Pump-Motor Alignment
• Pump Packing
• Mechanical Seals
• Pump Piping
For a sampling of Larry''s latest "PUMP GUY" columns from Flow Control magazine, see:
• "Cheat Sheets: Energy, Work & Power"
• "Cheat Sheets: Unwritten Pump Rules"
• "Cheat Sheets: The Affinity Laws"
FOR MORE INFORMATION & TO REGISTER FOR THE PUMP GUY SEMINAR, CLICK HERE.
If you have any questions about the PUMP GUY SEMINAR or need help registering, please contact Matt Migliore at 610.828.1711 or Matt@GrandViewMedia.com.