Troubleshooting and root-cause analysis

Troubleshooting and root-cause analysis

From observing symptoms to diagnosis, correction and prevention of recurrence

A decrease in the height of the fountain, a group of lights turning off or a pump stopping are visible signs; But their cause may be in another part of the system. A decrease in water height can be related to strainer clogging, control command change, tank level drop or equipment wear. Multiple lights turning off at the same time may also be caused by shared power, data path, or application settings.

Successful troubleshooting clarifies the relationship between the symptom, the conditions of occurrence, and the actual data. This process should lead to a correction whose result can be tested and reduce the possibility of repeating the failure.

In chapter 31, the maintenance program and the initial handling of failures were discussed. This chapter develops the methodology of expert review and analysis of evidence.

The examples and calculations in this chapter are educational. Specialized tests must be carried out with the appropriate method and equipment, by qualified people and in compliance with safe conditions.

34–1. Separation of symptom, cause and effect

Accurate recording of the problem is the first step of troubleshooting.

For example, the statement "pump weakened" is an early diagnosis. A possible description can be as follows:

“In show program number three, the height of all J-02 orbital fountains has been reduced relative to the reference state. This situation is observed after about twenty minutes of work and changes with a short-term stop."

This description provides information about the scope, time and circumstances of the occurrence.

A failure may have several causes at the same time. For example, clogging of the suction path and low water level can together make the pump working conditions unsuitable.

34–2. Preservation of evidence and determination of safe investigation conditions

Before resetting the error or changing the settings, the available information is recorded:

• Warning code and text.

• Time of occurrence and sequence of events.

• Active program.

• Displayed values.

• Condition of related equipment.

• Recent changes or repairs.

• Photo or video reference, if useful.

Toggling may temporarily remove the token or erase part of the history. Therefore, if possible and without delaying the necessary safety action, the evidence is preserved before the intervention.

It is then determined which on-going checks are allowed and which activities require shutdown and power isolation.

Opening strainer, pump, pressure connection or electrical compartment must be done after appropriate securing. A software stop command is not a substitute for power isolation.

Removing the protection or arbitrarily increasing the permissible limits is not considered a fault diagnosis method. Abnormal conditions must also not be reproduced by creating a hazard to people or equipment.

34–3. Determining the fault range using the system structure

The primary question is: which equipments have problems and what parts are common between them?

If all the nozzles of a circuit are down, the common path of that circuit becomes more important. If only one nozzle has changed, checking the branch and the nozzle itself takes priority.

These patterns suggest a path of investigation and are not definitive diagnoses by themselves.

Up-to-date drawings, equipment ID and control input and output connection table are the main tools of this stage.

Fig. 34–1 — Defect range detection path

Suggested form: a diagram with three branches "whole set", "one circuit" and "one device", together with the display of common components of each branch.

Subtitle: "The range of symptom occurrence is the starting point for choosing measurements and tests."

34–4. Check the validity of the measurement and compare it with the reference state

Before interpreting the data, the validity of the instrument and the comparability of the conditions should be checked.

For example, today's pressure can be compared to the pressure recorded at delivery when the measurement location, valve status, number of active circuits, speed command, and water level are known.

Important items include:

• Unit and instrument range.

• Place of pressure sampling or flow measurement.

• Validation of the sensor and its scaling in the controller.

• Data synchronization.

• Recording speed according to the event.

• Status of other effective circuits.

Educational example: detecting the increase in flatness resistance

Suppose the pressure difference is measured on both sides of a strainer:

In the range where the pressure drop relation is approximately proportional to the square of the flow rate, we can write for a preliminary comparison:

$R=\frac{\Delta p}{Q^2}$

Ratio of current resistance index to reference:

$\frac{R_2}{R_1}=\frac{32}{20}(\frac{18}{15}{)}^2=2.304$

Therefore, the resistance index has increased by 2.3 times.

This calculation shows the increase in resistance of the section between two measurement points, assuming the validity of the data, the same path and the appropriate approximation of the current relationship. Filter blockage is one of the main hypotheses.

For confirmation, the smoothness condition will be checked after stopping and securing, and the measurement will be repeated after correction. If the path also includes a valve or other component, its effect should be checked separately.

This indicator is not a general limit for all filters.

Fig. 34-2 location — pressure drop measurement location

Suggested figure: A view of a strainer with pressure points before and after, the location of the flow meter and the table of values before and after the service.

Caption: "Interpretation of pressure difference should be done in conjunction with flow rate and route specifications."

34–5. Troubleshooting the drop in height and change in shape of the fountain

Simultaneous drop of a group

In this case, the review begins with the common conditions:

1. Is the display command the same as the reference state? 2. Does the actual speed or condition of the pump match the command? 3. Is the water level and suction conditions suitable? 4. Have the circuit pressure and flow rate changed? 5. Has the condition of valves, strainers or the return path changed?

The combination of pressure and flow gives more information than either alone. Flow reduction with pressure change can be related to path limitation, speed change or pump performance change; The exact interpretation depends on the location of the pressure gauge and the system curve.

Dropping or deflection of a nozzle

Checkable items include clogging, angle change, displacement of the installation site, orifice damage, branch problem and related valve adjustment.

For nozzles dependent on the water level, the alignment around the nozzle is also part of the diagnosis. For example, OASE introduces the Cascade 70 T nozzle depending on the water level; Therefore, the change in the shape of the output of such a nozzle does not necessarily result from the change in the pump's performance. Specifications of Cascade 70 T nozzle ^[241]

Cleaning should not change the dimensions or shape of the opening. After the service, the output is compared with the reference condition.

34–6. Detection of abnormal sound, air ingress and cavitation

Abnormal noise and vibration can have hydraulic or mechanical causes. Hearing the sound alone is not enough to detect cavitation.

The review path includes the following:

• Tank level and water entry conditions.

• Presence of vortex or air entry.

• Smooth condition and suction path.

• Flow rate, speed and working point.

• Water temperature.

• The condition of supports and connections.

• Mechanical signs recorded by the technician.

In heater cavitation, the local pressure drop can lead to the formation of a steam bubble and its collapse in the higher pressure area. Free air entering the circuit is a different phenomenon, although it may cause similar symptoms. KSB cavitation technical reference ^[242]

To check the suction conditions, the available NPSH of the system is compared with the pump demand at the corresponding duty point. The value included in the manufacturer's curve should be interpreted along with the criteria for its definition; NPSH associated with 3% head drop is not a measure of complete absence of cavitation. The necessary margin should be determined according to the pump and the application. Reference NPSH in KSB ^[243]

Increasing speed to compensate for nozzle drop, before the cause is known, may make suction conditions more difficult.

34–7. Investigation of level drop and compensation water increase

Increased water consumption can be related to leakage, overflow, splashing, evaporation, washing or compensatory water control error.

The review begins with recording this information:

• Compensatory water volume in a certain period.

• Changing the stored volume.

• Effective precipitation.

• Recorded draining and washing.

• Hours and type of show.

• Wind and temperature conditions.

• Status of compensating water valve and level sensor.

For a given boundary of the system:

$V_{خروجینامشخص}=V_{ورودی}-\Delta V-V_{خروجیثبت‌شده}$

In this regard:

$\Delta V=V_{پایان}-V_{آغاز}$

Educational example

In one interval, the water inflow is 0.9 m3, the decrease in stored volume is 0.12 m3 and the recorded discharge is 0.08 m3. There is no other entry:

$V_{خروجینامشخص}=0.9-(-0.12)-0.08=0.94m^3$

This amount can include evaporation, splashing, leakage and measurement error; Still not considered "leak volume".

The comparison of the beginning and end volumes should be done in similar situations. Water in the pipes, return delay and level fluctuation can change the reading.

Circuit separation tests or sealing checks must be done with approved methods and structural considerations.

34–8. Pump stop troubleshooting and electrical protection function

First, it should be determined by what mechanism the equipment is stopped:

• Normal control command.

• Removal of work permit.

• Drive error.

• Overload protection function.

• Current leakage protection function.

• Stop feeding.

• Communication or equipment error.

These situations do not have the same review path.

Low or high flow should also be interpreted along with other data. A flow value alone does not prove the health or failure of a pump.

Increasing the protection setting or repeatedly resetting the error, without investigating the cause, may prevent correct diagnosis. Electrical tests and inspection inside the panel need qualified people and suitable method.

34–9. Troubleshooting lighting and DMX communication

In lamp troubleshooting, four parts must be distinguished:

1. Nutrition. 2. Data path. 3. Settings and addressing. 4. The lamp itself or the driver.

If several lights are malfunctioning at the same time, their common components are checked. If only one color is a problem in a lamp, it becomes more important to check the relevant channel and local equipment.

A consistent and known pattern can be useful for separating program error from equipment behavior. The implementation of such a test must be compatible with the permitted conditions of power supply and cooling of the equipment.

The ETC manual explains that DMX cable, connection, topology and termination errors can lead to flickering or unstable behavior. Examining the path with a suitable test connection and separation of sections is one of the diagnosis methods. Any changes must be made in accordance with the equipment documentation. Technical guide and DMX troubleshooting in ETC ^[244]

In the fountain, the health of the gland and the connection of the seal should also be checked. Opening or replacing the connection requires restoring the seal and performing the relevant test.

34–10. Separation of control error from physical delay

To analyze the representation problem, the performance chain is examined:

Program event → Output command → Receive command → Equipment movement → Visible result

The program may give the command at the right time, but the valve or water circuit will respond later. It is also possible that the equipment is healthy and the command is sent with a delay from the beginning.

The main questions are:

• Has the version of the program and music file changed?

• Is the command issued at the expected time?

• Does the equipment receive the same command?

• Has the actual response changed from the past?

• Is the pressure, flow rate or initial state of the circuit different?

• Does the error occur only in a certain combination of equipment?

Fixed delay and increasing time difference are two different patterns. The first may be related to the response time or delay compensation setting; The latter can be related to the time base or synchronization of clocks.

Sensors should also be checked in the same chain. A "low level" message may be caused by actual level, sensor contamination or failure, wiring, or scaling. Deleting the message without checking the actual condition will not solve the problem.

34–11. Troubleshoot turbidity, foam, sediment and odor

The change in the appearance of the water should be recorded and checked as an indication. Color, foam, or turbidity alone do not determine the cause.

Monitoring records, inlet water conditions and equipment activity should be reviewed along with appropriate sampling.

For decorative fountains, the CDC lists algae, biofilm, turbidity, odor, and visible particles as signs of needing attention and emphasizes a review of the water management plan. CDC Guidelines for Decorative Fountains ^[245]

In an interactive fountain, the decision to stop and reopen should be coordinated with the health plan and site requirements. Arbitrary addition of substances to fix the undesirable appearance of water is not a substitute for diagnosis and appropriate action.

34–12. Record intermittent failures

Intermittent failure may not occur when a technician is present. In these cases, the quality of data recording becomes more important.

For each event, this information is useful:

• Accurate time and time zone recording.

• Program and show stage.

• Active equipment.

• Related commands and feedback.

• Effective level, pressure or temperature.

• Pre-stop warnings.

• Last change made.

Data should share the same clock as possible. The time difference between the controller, camera and operator report can obscure the sequence of events.

Educational sample

If the level reduction is registered first, then the work permit is removed and then the pump is stopped, the sequence of events is different from the case where the pump is first stopped and then the level is changed.

However, the conclusion depends on the temporal accuracy of the recording. One-minute recordings cannot reliably represent the sequence of events lasting several seconds.

Controlled repetition of conditions is done only within the permitted limits. Deliberate overloading or dehydration to reproduce the error is not acceptable.

Place Figure 34–3 — Timeline of events

Suggested figure: simultaneous diagram of water level, pump command, work feedback and alarm, with indication of time of first change.

Caption: "Sequence of events can separate primary cause from subsequent consequences."

34–13. Root cause analysis in repeated failures

For frequent failures, symptom resolution should be accompanied by investigation of the conditions causing it.

Educational example

Suppose the pump of a circuit has been stopped several times with a low level warning.

The survey shows:

• In a certain scene, a larger volume of water is placed in the circuits.

• The tank level approaches the stop threshold.

• The return path is also limited by waste.

• The show schedule has been changed after delivery and its effect on the surface has not been checked.

In this instance, replacing the sensor or lowering the stop threshold is not a documented answer to the problem.

Actions can include clearing the return path, checking the capacity and behavior of the reservoir, modifying the program within the scope of the plan, and retesting.

Analysis must distinguish between "confirmed fact" and "hypothesis". For each hypothesis, supporting and opposing evidence is recorded.

Tools such as a chain of "why" questions or a cause-and-effect diagram are useful, but should not limit the analysis to a preselected answer.

Instead of Figure 34–4 — Cause and effect map

Suggested figure: Level drop diagram with branches "Circulating Volume", "Water Return", "Sensor", "Settings" and "Change Program", with the evidence of each branch indicated.

Caption: "Sustainable reform depends on recognizing the conditions that cause failure."

34–14. Action priority and referral to a specialist

Diagnostic priority is not determined solely by the apparent severity of the symptom. The possible outcome, scope of the effect and the possibility of continuing the work are also important.

Three operational situations can be defined:

• Termination of the relevant scope: when the continuation of work is not compatible with safety or health requirements.

• Limited exploitation: only in pre-screened and authorized alternative mode.

• Planned handling: for defects that do not have an immediate decisive effect after evaluation.

Expert referral is required when the test requires tools, competencies, or information outside the scope of the operations team.

To contact the manufacturer or support, the following information package is sent:

1. Equipment ID and model. 2. Detailed description of the symptom and conditions of occurrence. 3. Error codes and recorded data. 4. Map of the relevant section. 5. Actions taken and the result of each. 6. Useful photo or video. 7. Latest system changes.

Crystal Fountains has introduced technical support and troubleshooting of fountain equipment and performance into its services. The quality of the information sent allows for a more targeted review by the support. Crystal Fountains technical support service ^[246]

34–15. Proof of fixing the defect and returning to service

The end of the repair should be accompanied by the result test.

First, the conditions in which the defect was observed are recreated within the allowed range. The related functions are then checked to ensure that the modification has not caused any other undesirable consequences.

For example:

• After cleaning the strainer, flow rate and pressure drop should be recorded again.

• After replacing the light driver, check the addressing, color and function of the group.

• Test level control, warning, protective action and recovery after modification.

• After changing the pump speed, height, spray, power and working conditions should be controlled.

At the end, it must be confirmed that:

• Connections and protections are established.

• Test settings have been removed.

• Necessary protections are active.

• Additional tools and parts are out of scope.

• Documentation and version settings have been updated.

• The operator is aware of the changes.

For intermittent failure, one time correct operation may not be enough. The duration and conditions of monitoring after repair should be appropriate to the previous failure pattern.

34–16. Troubleshooting reports and turning experience into project knowledge

The report should make the decision path clear to someone who was not present at the review.

In the GSO product catalog, nozzles, pumps, lighting, control and circuit accessories are introduced. In the troubleshooting file, the exact identification and specifications of these components must be attached to the project documents so that the inspection and supply of the part is not done with ambiguity. GSO urban fountain equipment groups ^[247]

Similar records should be searchable. If the same type of failure is repeated in several circuits, examining them together can reveal a problem with water conditions, installation, adjustments, or maintenance procedures.

Each closed file should specify what was corrected, how the result was verified, and what action was taken to reduce the likelihood of recurrence in the project plan.