ASME STP-PT-065-2013 BRANCH LEG STUDY FOR BIOPROCESSING EQUIPMENT《生物工艺设备的分支支柱研究》.pdf

1、STP-PT-065BRANCH LEG STUDY FOR BIOPROCESSING EQUIPMENTSTP-PT-065 BRANCH LEG STUDY FOR BIOPROCESSING EQUIPMENT Contributing Authors and Editors: Ethan Babcock, URI Mechanical Engineering Graduate Mallory Corbin, Stevens Institute, Applied Chemistry Graduate Randy Cotter, Sr., Cotter Brothers Corporat

2、ion Matthew Deane, URI Environmental Science Graduate Student Bo Boye Busk Jensen, Ph.D., Alfa Laval Dan Mathien, Behringer Corporation Phil Paquette, P.E. Marc Pelletier, CRB Engineering Joe Serdakowski, AutoSoft Systems James Dean Vogel, P.E. The BioProcess Institute Deborah Botham, Cotter Brother

3、s Corporation Jay Ankers, M+W U.S., Inc. Date of Issuance: December 19, 2013 This report was prepared as an account of work sponsored by ASME Pressure Technology Codes it is therefore important to define term when presenting and comparing testing information. The “L” dimension is not readily obtaina

4、ble from vendor literature as they have long provided fitting dimensions from the centerline to the edge. (Vendors and the BPE Standard refer to this centerline dimension with the letter designation A or B.) Both piping designers and detailers need these centerline dimensions to accurately develop d

5、esign drawings and models. The ASME BPE Standard has established a “benchmark” set of these centerline-based dimensions (see Part DT) for stainless steel tube and fittings to permit designers and detailers to have uniform dimensions available regardless of individual manufacturers. The “L” is a proc

6、ess dimension, calculated as: ( ) 2.2 Standard and Short Outlet Tee The introduction of short outlet (SO) tee fittings where the A and B dimensions are significantly reduced (thus reducing the process dimension L), has greatly enhanced the ability of piping designers and detailers to achieve the tar

7、get L/D of 2:1. Not all short outlet branched fittings yield an L/D ratio less than the recommended two value when used as is. The addition of a standard valve versus a cap would increase the L and resultant L/D ratio. This study will examine both the L/D ratio obtained from using a standard short o

8、utlet tee fitting and a fixed L/D ratio equal to two. (See Figure 5-3 in Section 5). STP-PT-065: Branch Leg Study for Bioprocessing Equipment 5 3 LITERATURE REVIEW Most of the prior articles covered the mixing within dead legs, with a focus on achieving effective cleaning in the branch leg. No refer

9、ence was found which focused on the removal of air, and only a few references even mentioned it. Gaerkes study 3 on the flow rate required to displace the air/flood sections of straight piping (without branched fittings) of different diameters of transparent schedule 40/80 PVC piping fabricated with

10、 socket joints in a variety of installation configurations (horizontal, sloped, and vertical) with different outlet configurations (open on the end and liquid sealed) indicated the following: The piping configuration that required the highest flow rate to displace the air was a vertical pipe with th

11、e flow directed downward that was open on the end. The addition of backpressure had no impact on the flow rate required to displace the air from the piping systems evaluated. Although it was not the primary focus of the study, Gaerke looked briefly at transparent PVC branched tees fabricated with so

12、cket joints and determined that liquid velocities as high as 7 feet/sec were insufficient to displace air from a 2 inch branched tee directed upward to an L/D ratio of 2. Youngs study 4 discusses air removal in the effective SIP of a system. Air is heavier than steam at routine SIP conditions. The p

13、rimary issue is to displace the air with the steam. They concluded: L/D values do not provide a general guideline which can be used to predict sterilization. Tubes with similar L/Ds, but different diameters, showed sterilization times varying up to 250% SIP of dead legs with saturated steam is depen

14、dent on tube diameter, length and orientation with respect to the gravitational vector Saturated steam sterilization did not occur at any location above the initial steam-air interface Young stresses that the proper sizing of the tube diameter had the greatest effect on sterilization as it increased

15、 the ratio of buoyant forces to viscous dissipation forces. They effectively showed how a small tube of 40mm ID and 88mm long, L/D of 22, exhibited little buoyant driven convective flows, and the minimal air displacement observed was due primarily to diffusion. These buoyant forces are in reverse in

16、 CIP where the process liquid is heavier than the air. Even with proper tube diameter the air requires more than gravity to remove it from a branch sloped above the horizontal centerline. Grasshoff 56 verified that flows into the dead leg, rather than away from the dead leg, provided better mixing i

17、n the dead leg, but their work did not mention the removal of air. They showed that if the L/D is large enough, a secondary recirculation zone is formed at the end of the dead leg reducing exchange of mass (liquid or air) from the dead leg to the main pipe. They also showed that the net velocities i

18、n the primary recirculation zone were as low as one-eighth of the bulk velocity. These results were confirmed in Jensens Computational Fluid Dynamic (CFD) and experimentation. Seven Helium (He) experimented with a flow diverter to improve cleaning. Sassanami 8 noted that the soil removal rate would

19、be much lower in the dead space area than the pipe because the fluid mechanics involved in this area are significantly different, reducing the ability to clean this area. This study showed that for a dead space with an L/D value of 1, there was a significant advantage to operating at Reynolds number

20、s above 70,000 since this caused a significant increase in cleaning. For dead zones with an L/D of 4, the cleaning rate becomes independent of the Reynolds number. STP-PT-065: Branch Leg Study for Bioprocessing Equipment 6 4 EXPERIMENT Experiments for estimating the removal of air were conducted usi

21、ng a translucent test spool system by recirculating liquid in the system. Removal of air was recorded on video to estimate the time for removal with various configurations. A series of experiments were conducted over a wide range of flow rates, different geometrical configurations, two different tem

22、peratures, and different back pressures. A grade system was developed for quantification of air removal. The following sections provide detail on the experimental set-up and data analysis, ending with the configurations investigated. 4.1 System Description The system (Figure 4-1) consisted of a manu

23、ally operated liquid reservoir tank, pump, recirculation piping, instrumentation, and hand valves. The system was reconfigured as needed to allow for different arrangements. Most of the system piping was comprised of two inch stainless steel and silicone hose. Experiments for estimating the removal

24、of air were done in translucent natural polypropylene (PP-R) test spool pieces while recirculating liquid in the system. The major system components are: Centrifugal Pump, maximum 155 gpm, (flow energy) Turbine Flow Meters, two, ranges 1.75-150 gpm, (flow rate) Digital Temperature Gauge, 0-100 C, (f

25、luid temperature) Analog Pressure Gauges, two, 0-100 psig, (inlet/outlet pressure) Digital Pressure Gauge, 0-100 psig, (outlet pressure) 2 inch Diaphragm Valves, three, (feed control, bypass control, backpressure control) 2 inch Ball Valves, three, (feed on/off, bypass on/off, feed outlet on/off) in

26、ch Ball Valves, two (air bleed valve, pump drain valve) 1 inch Ball Valve (spool drain valve) 250 gallon Tank, natural polypropylene (system sump) 500 liter Tank , stainless steel (system sump) Heat Exchanger (heat energy) Pipe Fittings, stainless steel and polypropylene Piping, stainless steel, mos

27、tly 2 inch Hoses, 2 inch, stainless steel end crimped silicone The central system consists of a 250 gallon feed tank that serves as the liquid reservoir, in which gravity feeds the centrifugal pump. The pump drives the water at the desired flow rate and is adjusted by varying the opening of the diap

28、hragm valve at the beginning of the feed line. Opening this valve allows a higher flow rate through the line, while closing it decreases the flow rate through the line. The bypass diaphragm valve allows the operator to divert some amount of flow from the main line so as to more easily find the corre

29、ct flow rate through the main line. The water next moves in ascending order through: the pump pressure gauge the fluid temperature gauge the flow meter the inlet pressure gauge the test spool outgoing pressure gauge It is then directed back into the tank, using flexible hosing, to be recirculated. S

30、TP-PT-065: Branch Leg Study for Bioprocessing Equipment 7 Figure 4-1: Test Schematic for Horizontal Runs Note: The rest spool is the milk white section in the top part of the picture. Schematic for the other configurations can be seen in Appendix A. The remaining valves are used to prepare the syste

31、m for tests, e.g. drain the system. An air bleed ball valve, which can be opened to allow air into the system and out of the test spool, is located above the feed control diaphragm valve. A back pressure diaphragm valve is located on the return line just before the STP-PT-065: Branch Leg Study for B

32、ioprocessing Equipment 8 water re-enters the feed tank. Closing this valve will increase the pressure on the line to test effects of increasing back pressure on the test spools. A heat exchanger is added to the return line and the tank is changed to the smaller stainless steel tank during runs which

33、 require heating. 4.2 Testing Procedure 1. Start up the pump in order to run water through the pipeline. 2. Adjust the flow rate using the diaphragm valve until the desired BPI Visual Air Grade is reached. A low flow rate is suggested to begin, so that immediate flooding of the branch leg does not o

34、ccur. 3. Once the desired visual air grade is reached (a) Photograph the result. (b) Record the flow rate, pump pressure, water temperature, incoming pressure, outgoing pressure, and visual air grade. (c) Close the test spool using the two ball valves on either side. (d) Photograph the still spool.

35、(e) Remove the cap from the branch leg, and pour water into it from a graduated cylinder. (f) Record the volume of water needed to fill the branch leg, as that is the volume of air that was left in the branch leg. 4. Drain and repeat two subsequent times for this amount of air, and the repeat the pr

36、ocedure for each individual visual air grade (from 0-5 see Section 5.6). This experiment will also be repeated for a test spool with its branch leg rotated from the vertical position to an angle of 45, (from 1-4). This, however, does require one to rotate the test spool back into the vertical positi

37、on once the photos have been taken. This allows the experimenter to safely measure the amount of air using the above method with a graduated cylinder without losing water and contaminating the results. 4.3 Experimental Technique For each configuration, four different tests were performed to collect

38、data. Each test was repeated three times to ensure redundancy of results. These four tests are as follows: 4.3.1 Flow Rate Increase (FRI) Test Increase flow incrementally (minimal back pressure). This test is a quick screening to assess the high level performance characteristics of the configuration

39、. Record temperature, flow rate, air quality, and pressures at each flow increment Take photos of branch leg at each flow increment Determine what flow rates to run Flow Rate Maintain tests 4.3.2 Flow Rate Maintain (FRM) Test Maintain a certain flow rate over time (minimal back pressure). This test

40、simulates most process conditions, including CIP. Record temperature, flow rate, air quality, and pressures at regular time increments Record video (whether real time or 2 second snapshots) of branch leg at regular time increments Determine flow rate needed to clear branch leg at 1 min and 5 minutes

41、 Anything over 5 minutes is considered unacceptable. 4.3.3 Pressure Increase (PI) Test Increase pressure incrementally while maintaining a certain flow rate. This test is a quick screening test to assess if increasing the pressure will assist in air elimination. Record temperature, flow rate, air qu

42、ality, and pressures at each pressure increment, and with pump shut off STP-PT-065: Branch Leg Study for Bioprocessing Equipment 9 Take photos of the branch leg at each pressure increment, and when pump is shut off Determine what pressures and flow rates to run Pressure Maintain tests 4.3.4 Pressure

43、 Maintain (PM) Test Maintain a certain pressure and flow rate over time. This test simulates some process conditions where pressure in maintained. Record temperature, flow rate, air quality, and pressures at regular time increments Record video (whether real time or 2 second snapshots) of branch leg

44、 at regular time increments Determine flow rate needed to clear branch leg at 1 min and 5 minutes. Anything over 5 minutes is considered unacceptable. 4.4 Explanation of Rotation/Slope A digital level that measured in tenths of degrees was used for the different slopes and rotations. For the slopes,

45、 it was calculated that 1/8 by 12 inches is about 0.6 degrees; that set point was employed to find the correct slopes. For the rotations, the level was placed on the dead leg until the correct rotation in degrees for that orientation was achieved. 4.5 Data Collected Visual observations, such as the

46、amount of air left in the branch leg, were recorded with the test spools. Flow rate, temperature, time, and pressure data were all recorded where applicable. 4.6 Evaluation Method A qualitative scale to judge the amount of air still trapped in the branch leg was used to assess the level of air elimi

47、nation in the system. This scale ranged from 0 to 5, making it possible to record how much air was still left in the dead leg for different configurations (see Figure 4-2). 0. No water in the branch 1. Less than full of water (more than full of air) 2. to full of water ( to full of air) 3. to comple

48、tely full of water (clear to full of air), there is always a standing bubble of air 4. Completely full of water, many bubbles remain throughout the branch leg (nearly clear of air) 5. Completely full of water (clear of air) STP-PT-065: Branch Leg Study for Bioprocessing Equipment 10 Figure 4-2: Exam

49、ple Images of a Visual Air Grade From left to right, the Air Grade is 1,2,3,4 and 5 respectively Figure 4-3: Example Image of Air Grade 1 Left: No rotation with the pump off and the spool closed. Right: No rotation with the pump running and the spool open. By performing this experiment, a quantitative standard has been introduced in which the amount of air can be measured in a branch leg. This is important to the BioProcess Equipment (BPE) community, as a standard will help when