Addressing the Accuracy Concerns of Measuring MOV Stem ... - CB&I

Proceedings of the ASME/NRC 2014 Twelfth Symposium on Valves, Pumps, and In-service Testing for Operating and New Reactors NRC2014 June 23-25, 2014, North Bethesda, Maryland, USA NRC2014-5038

Addressing the Accuracy Concerns of Measuring MOV Stem Nut Wear From Diagnostic Testing Carter (Chuck) Reames, Jr. CB&I 4171 Essen Lane Baton Rouge, LA 70809 method that more accurately efficiently validates this wear.

Abstract: Excessive stem nut thread wear represents a potential common-cause failure mode that could impact all risingstem valves. The consequence of unexpected failure of stem nuts emphasizes the importance of improving condition monitoring and maintenance practice activities by identifying, quantifying and minimizing stem nut thread wear. In the nuclear industry, MOV (motor operated valve) Diagnostics estimates stem nut thread wear on safety-related valves using the stem-to-stem nut transition time (zero plateau). But, the stem-to-stem nut transition time could also be affected by other variables that would lead to an inaccurate calculation of wear. Using stem-to-stem nut transition time to estimate wear, coupled with generally erring on the conservative side, usually indicates higher than actual wear. This method, combined with all of the unknown variables, results in nuclear plants using valuable outage resources and dose to pull good stem nuts unnecessarily. This white paper identifies some of the variables that can be mistakenly construed as wear and offers a

and

Introduction: Diagnostics have been used for decades on MOVs and proved valuable in determining MOV set points and health condition. They measure some parameters very well: e.g., stem thrust, switch settings and motor current. Certain measured parameters have limited effectiveness in determining MOV health: e.g., motor health and stem nut wear. Motors have unique issues that require specialized equipment designed to monitor all three phases of voltage and current for online tests and that equipment is used to determine rotor and stator health. Using the motor current signature from MOV diagnostics equipment alone to determine the motor health would either be inadequate to see anomalies, or any anomalies seen would prompt the use of specialized equipment. Stem nut wear measurement is very similar: using the transition time to flag stem nuts with high wear may be a good first identifier but using a more accurate

1

Proceedings of the ASME/NRC 2014 Twelfth Symposium on Valves, Pumps, and In-service Testing for Operating and New Reactors NRC2014 June 23-25, 2014, North Bethesda, Maryland, USA

method to confirm is prudent. Otherwise many stem nuts with low wear will be replaced. Removing a MOV stem nut for visual inspection can take 4-12 manhours. Performing a post-maintenance diagnostic test can also take an additional 42 man-hours.

however, using diagnostic testing only, it is more likely to be replaced before replacement is actually needed. Discussion: Analyzing stem nut wear from a MOV diagnostic thrust trace begins by measuring the longest plateau time of the close-open (O4), open-close (C3) or calibration zero plateau. This plateau is produced from the time it takes the stem nut threads to transition from compressing to tensioning (or tensioning to compressing) the valve stem. Essential in calculating this distance (in a perfect world) is the following:

Background: A MOV stem nut is typically made of a bronze alloy material. Stem nut threads will wear from opening or closing the valve and can be affected by several factors, including the following: (1) stem nut material; (2) normal operating loads and maximum loads; (3) number of stem nut rotations during the valve stroke; (4) number and frequency of valve strokes; (5) stem nut manufacture and threaded length; (6) stem to stem nut fit; (7) valve stem thread condition; and (8) stem lubricant, lubrication method and frequency, and environmental conditions for the lubricant. Because the valve stems are made of stainless or carbon steel, wear will first occur on the stem nut threads.

Actual motor “RPM” during transition event Actuator gearbox overall gear ratio “OAR” Stem threads per inch “TPI” Number of stem thread “starts”

A rising-stem valve stem nut transfers rotational motion “torque” to axial stem movement “thrust”. MOV diagnostics to estimate stem nut wear is used to avoid stem nut failure, which in turn prevents valve operation and may cause valve position indication to be incorrectly displayed in the control room.

“

”

60

To calculate the distance between the stem and stem nut threads (wear), the plateau transition time (sec) is multiplied by the stem speed (in/sec). This number, by itself, is meaningless to determine stem nut thread wear unless compared to the calculated stem nut thread dimension being measured.

With program valves, it is always better to replace a stem nut before it fails; 2


The BWR Owners Group paper, released in 2008, identified the “Original Stem Nut Thread Thickness at Min-Major Stem Diameter”i as the dimension used to calculate thread thickness from diagnostics. Another measurement that can be used to calculate stem nut thread wear is to use the stem thread profile. Since the stem is designed not to be the wearing component, it can be used as the gauge to measure stem nut wear. This dimension can be calculated by subtracting stem thread crest “Fcs” from stem pitch. There is a minimal difference in the results of the calculation between these two methods.

Exception Motor speed (RPM) is used to calculate stem speed. Typically MOV actuators use 2, 4 and 6-pole induction motors. During the lost motion portion, and through stem nut transition, the motor is lightly loaded. An unloaded induction motor, with no broken rotor bars, will run just below synchronous speed. If the exact motor speed is not measured at the transition event, the rest of the calculation to determine stem speed will be wrong. The synchronous speed of a 2-pole motor is 3600 RPM, 4pole is 1800 RPM and 6-pole is 1200 RPM. So, for induction motors, it would be expected that an unloaded 2-pole would run around 3590 RPM, 4- pole around 1790 RPM and 6-pole around 1390.

Data Acquisition Variables There are testing situations that may indicate more wear than actually exists or could make finding the zero plateau difficult. A loose stem nut locknut will increase the plateau time. This condition would be easy to identify by inspection.

For the stem speed to be more accurate, the calculation should not use name plate RPM but rather an RPM closer to synchronous. Using the name plate RPM, is for a fully-loaded motor, will calculate a slower stem speed, increasing the apparent stem nut wear. Using an RPM closer to synchronous would cause the measured stem nut clearance to be shorter; decreasing the calculated stem nut wear. So, using the name plate speed would make this portion of the calculation more conservative.

Stem rotation is a condition that could significantly increase the plateau time. It is very difficult to detect by inspection and almost impossible to quantify its impact on increasing the plateau time. Identification of the zero plateau requires enough packing load to see a change in running loads between the open-close or close-open strokes. Valves with live-load packing can make identification of this event difficult.

3


Rotork actuators have a floating stem nut in the thrust base. Though the zero plateau may be identifiable it is impossible to use to calculate stem nut wear.

the stem nut that was replaced was a result of galled threads from actuator issues at RNP. Maintenance time can vary based on the size of the Limitorque actuator (i.e., an SMB-3 may take a shift to pull and reinstall a stem nut, while an SMB-000 may take 3 hours). However, if we consider an SMB-00 actuator for a stem nut inspection, it is estimated that approximately 50 man-hours would be needed to complete the entire scope of work at RNP, HNP, or BNP. This includes walk-downs, paperwork, removal and reinstallation of the stem nut lock nut / stem nut, and post-maintenance diagnostic testing. Since five out of six inspections above reflected no stem nut wear, the SNAP tool could have saved a large number of man-hours associated with those inspections. This doesn’t take into consideration that these activities can be performed in radiation fields. The dose and man-hour savings, in addition to the fact that the device measures actual physical stem nut wear (as opposed to reliance on zero plateau differences that are merely guessing), makes the application of the SNAP tool a potentially very beneficial practice.” Doug Coon (MOV Engineer, Robinson Nuclear Plant)

Larger actuators, especially those using a thrust base, could have significant looseness in the drive sleeve/stem nut assembly and will cause the transition time to increase. Compensator slop in SB and SBD actuators could make finding the zero plateau difficult to identify. Examples Peach Bottom - “HPCI Injection MOVs: This issue seems to primarily apply to the larger actuators; SMB-4 and 5 and especially the “T”, torque only actuators (SMB-4T, 5T).” Curt Reynolds (MOV Engineer, Peach Bottom APS) Brunswick, Harris, & Robinson – “There have been eight stem nut inspections scheduled between Jan. 2011 and Mar. 2015 between the three legacy Progress Energy plants. Half of the inspections were baseline and the other half were as a result of greater than 25% differences in diagnostic trace zeros between tests. The four baseline inspections results were all satisfactory. Two of the four zerodifference inspections have been completed. One of the stem nuts was replaced as a result of the inspections. Therefore, five out of six inspections showed no signs of stem nut wear. Also,

Fermi 2 – “In some cases there is wide variation between the transition time from a power trace and from a strain gauge. And, we have a number of SB/SBD actuators for which transition time isn't really apparent.” (Brett Gallatin,

4


1. It is designed that wear will occur on the stem nut, not the stem. 2. The stem is used as the gauge to determine the wear by detecting stem movement when stem nut thread contact has been made.

Fermi 2, Performance Engineering Lead Engineer, Programs License Renewal Project) Solution The Stem Nut Analysis Protractor “SNAP” tool was developed to provide a nonintrusive and quick means of quantifying stem nut thread wear without having to remove the stem nut from the actuator or the valve from service. This process directly converts the backlash between the stem and stem nut threads to a rotational percentage on rising stem valves. From this it is possible to determine the remaining stem nut thread metal. The area between the valve STEM thread peaks, identified as P- Fcs is being used to calculate the region of 100% stem nut wear indicated by “X” (Fig. 1). “P” is the thread pitch and Fcs is the basic flat at crest of the stem threadii. The formulae in Fig. 1 are used to calculate “X” then convert it to its rotational equivalence, represented in percent. This region “X” represents 100% of possible stem nut thread wear and is valid for all General Purpose Acme Threads with one thread start. The linear relationship between stem nut rotation and stem nut thread advancement can be found in the definition of “Stem Lead” – the distance a thread advances in one turn of the nut.

Fig. 1 SNAP Scale Calculation Х or (P- Fcs) equals 100% of the thickest part of the stem nut thread; this is where the backlash is indicated. As the stem nut thread wears, its backlash will increase allowing more free rotation. The purpose of the SNAP tool is to determine thread wear percent (backlash percent). Once the backlash percent is determined, remaining thread material can be calculated to provide the basis to determine valve operability. The data collected can be used to calculate thread structural integrity, and ultimately, to predict its functional-failure point to preempt failure by monitoring stem nut life cycle and wear rate (See Figs. 2 & 3).

P- Fcs, on the stem, is used to calculate 100% of stem nut thread wear for two reasons:

5


NOTE: The SNAP scale assumes that the valve stem threads are properly manufactured with a 29° angle to the dimensions detailed in the Machinery’s Handbook 25 for ACME Thread Design.

Fig. 3 SNAP Reading at 65% Wear Conclusion: If a baseline diagnostic test exists of a MOV with a new stem nut, it can be helpful in trending wear. It’s helpful because many of the variables present in the transition time, that aren’t associated with thread wear, will usually cause the indicated stem nut thread wear to be greater than actual. It can usually be assumed that any future increases in the transition time be attributed to thread wear; that is, assuming that none of the variables that falsely indicate thread wear would develop. The SNAP tool can be used to validate stem nuts with suspected high wear, instead of using maintenance to pull it for visual inspection. Because the test usually takes fewer than 30 minutes, at the valve, there could be a radiation dose savings as well. In a nuclear plant, the total man-

Fig. 2 SNAP Reading at 7% Wear

6


hours expected to perform a SNAP test should be about 3 man-hours. In addition to validating stem nut wear on Program

Valves, maintenance technicians can use SNAP on all rising-stem MOVs, including BOP valves with a high economic value.

i

ii

Guideline for Monitoring and Evaluating Motor Operated Valve Stem Nut Wear “BWROG-TP-07-004 (112)” Machinery’s Handbook 25 General Purpose Acme Threads pp 1716-1724)

7