Oct 23, 2026
Inline leak testing for automatic blow molding machines checks each empty bottle or container during production and sends a reject signal when the measured pressure loss exceeds the approved limit. I use a pressure-decay tester, product sensors, PLC logic, and a timed mechanical actuator to create a controlled sequence from bottle detection through reject confirmation. This guide explains how I set up that sequence for PET bottles, HDPE containers, and other blow-molded packaging.
!
An inline leak testing station is normally installed after the Automatic Blow Molding Machine has formed, trimmed, and transferred the containers to a conveyor. The station detects the arrival of each bottle, seals the opening with a test head, applies compressed air, and observes pressure behavior during a defined measurement period. A passing container continues toward packing, while a failed container receives a reject command from the PLC.
For empty blow-molded containers, pressure-decay testing is usually practical because the tester can pressurize the internal cavity without filling the bottle with water or another liquid. The tester first fills the container to a set test pressure, allows the pressure to stabilize, and then isolates the air supply. If the pressure drops faster than the approved limit, the container is classified as leaking.
The tester does not directly determine every possible defect type. It identifies a measurable loss of pressure caused by a hole, split, weak seam, incomplete neck, damaged closure area, or poor sealing between the test fixture and the container. For that reason, I separate leak-test failure logic from mechanical rejection logic in the PLC.
Before installing an automatic reject system for blow molding machines, I collect the container drawings, material information, production rate, cavity count, conveyor layout, and required leak acceptance limit. PET and HDPE bottles can have different wall stiffness, neck dimensions, temperature behavior, and deformation under test pressure. The same pressure-decay program should not be copied between products without validation.
The equipment list normally includes:
I also confirm that the conveyor can hold a repeatable bottle position. A leak tester may operate correctly while the line still produces false failures if bottles arrive tilted, touch one another, or enter the test head with inconsistent spacing.
The following sequence is the setup method I use when integrating an inline leak tester with a blow molding line.
I place the leak tester after bottle trimming, cooling, and transfer operations that may affect the container shape. Testing immediately after molding can produce inconsistent results if the plastic is still warm or if flash removal has not been completed.
The conveyor should present one container at a time to the sensing point and test fixture. If the production line uses multiple cavities, I identify whether bottles remain in cavity order or become mixed during transfer. This information determines whether the PLC must track individual containers by position, cavity number, or encoder distance.
The tester should have enough upstream and downstream conveyor length for test-head movement, reject delay, and reject confirmation. I also leave access space for fixture adjustment, sensor cleaning, and removal of jammed bottles.
The first sensor establishes the reference position for each container. I mount it where the bottle is stable and where the sensor beam cannot be interrupted by handles, flash, conveyor rails, or nearby bottles.
The PLC should record a clean detection edge rather than repeatedly reading the same bottle while it remains in front of the sensor. I normally use a one-shot rising-edge signal and add a product-present timeout. If the sensor remains blocked longer than the expected bottle-passage time, the PLC should generate a handling alarm instead of creating repeated test commands.
Bottle spacing must be measured rather than assumed. If the line produces 120 bottles per minute, the average pitch interval is:
[ T = \frac{60}{120} = 0.5\text{ seconds} ]
If bottle spacing changes during acceleration or line stoppage, the system should use conveyor encoder distance or a shift-register tracking method instead of relying only on fixed time.
For pressure decay testing for blow molded containers, I begin with a pressure level that gives a measurable response without deforming the bottle. The correct value depends on the container material, wall thickness, volume, neck design, and leak specification.
A typical development sequence includes:
For example, a development trial may use a 300 millisecond fill period, a 200 millisecond stabilization period, and a 500 millisecond measurement period. These figures are examples for commissioning, not universal settings. I adjust them after testing known-good bottles, controlled challenge bottles, and containers with production defects.
The leak limit should be based on the customer specification, package function, and validated test method. I record the measured results for multiple good bottles across several production cycles, then set the reject threshold with enough separation from normal process variation.
The leak tester should provide a clear result such as TEST_PASS, TEST_FAIL, TEST_BUSY, or TEST_ERROR. The PLC then decides what to do with that result. This separation prevents a communication fault, sensor fault, or test timeout from being treated as an ordinary bottle leak.
A practical logic structure is:
I recommend treating TEST_ERROR differently from TEST_FAIL. A failed test means the container exceeded the leak limit. A test error may indicate low air pressure, missing bottle, poor fixture sealing, communication loss, or an invalid cycle. The line may need to stop or divert the product for inspection rather than automatically accepting it.
PLC reject timing for inline leak testing must account for the distance between the test point and the reject point, conveyor speed, bottle spacing, and actuator response. The basic delay is:
[ t_d = \frac{D}{V} ]
where (D) is the distance from the test reference point to the reject reference point and (V) is conveyor speed in the same unit per second.
For example, if the distance is 1.2 meters and the conveyor moves at 0.6 meters per second:
[ t_d = \frac{1.2}{0.6} = 2.0\text{ seconds} ]
If the pneumatic valve and cylinder require 120 milliseconds to extend, I do not simply add 120 milliseconds to the delay without checking the actuator direction and sensor location. I measure the actual time from output energization to effective bottle displacement, then compensate for that response in the PLC.
When production speed changes, a fixed time delay can become inaccurate. For variable-speed conveyors, I use encoder pulses or a virtual shift register based on measured bottle travel. For fixed-speed lines, a timed delay can work if the conveyor speed is monitored and the line stops safely when tracking becomes invalid.
Before introducing bottles, I operate the pneumatic pusher, air blast, diverter, or servo mechanism through several dry cycles. I check the stroke length, response time, return position, guarding, and access to the reject bin.
The actuator should remove a failed bottle without touching the next passing bottle. I measure the bottle pitch and compare it with the actuator's active duration. If the actuator remains extended too long, it may push two containers into the reject bin or create a downstream jam.
For air-blast rejection, I verify nozzle alignment and air consumption. For a mechanical pusher, I check that the pusher face does not damage acceptable containers when bottle spacing changes.
Reject verification confirms that the defective bottle actually left the accepted product path. I install a sensor at the reject chute, reject bin, or downstream conveyor depending on the physical layout.
The confirmation window should be linked to the expected bottle travel time. If the reject command is issued but no bottle is detected in the reject path, the PLC should create a reject-validation alarm and stop or quarantine production according to the risk level.
I also test the opposite condition: a bottle enters the reject path when no failed result is active. This can indicate sensor bounce, tracking errors, actuator leakage, or incorrect PLC memory handling.
Calibration begins with the test fixture, not only the pressure sensor. I inspect the sealing plug, gasket, test head, air tubing, valves, and fittings for wear or leakage. A leaking fixture can produce false rejects even when every bottle is sound.
I first run an empty fixture test with no container installed. If the pressure decays during the fixture test, I repair the tester circuit before testing products. I then run a group of known-good containers and record the pressure-decay values, cycle time, and repeatability.
Challenge bottles are used to confirm detection capability. A challenge bottle can contain a controlled defect produced under a defined method, but the defect size must be verified independently and handled safely. I test the challenge bottle at multiple positions, including the body, shoulder, base, neck, and weld or pinch area when those locations are relevant to the container design.
Calibration records should include product code, test pressure, fill time, stabilization time, measurement time, pass limit, ambient conditions, fixture identification, and operator name. When changing from a small PET bottle to a larger HDPE container, I create a separate recipe rather than modifying the active recipe without approval.
False rejects often result from unstable bottles, poor neck sealing, wet or contaminated sealing surfaces, pressure fluctuations, or excessive test sensitivity. I check the air regulator, filter, test-head gasket, bottle temperature, and conveyor positioning before changing the leak limit.
Missed rejects usually involve incorrect delay values, a failed tracking register, sensor bounce, or a reject actuator that is too slow. I place a temporary indicator on the PLC output and compare the commanded reject position with the actual bottle position using a slow-speed dry run.
Bottle jams can occur when the pusher stroke is too long, the reject bin is full, the conveyor rails are too close, or two bottles arrive inside one reject window. I reduce actuator duration, correct spacing, and add a full-bin sensor where practical.
Reject-validation failures require a review of sensor placement and confirmation timing. The sensor may be watching the wrong path, detecting a neighboring bottle, or operating outside the expected travel window.
I complete commissioning in several stages rather than starting immediately at full production speed.
| Check | Verification |
|---|---|
| Dry run | Test sensor, PLC inputs, outputs, actuator, and alarms without bottles |
| Fixture test | Confirm pressure stability with the test head sealed and no container |
| Good-bottle test | Run known-good bottles and verify repeatable pass results |
| Challenge-bottle test | Confirm failed containers receive a fail result |
| Reject timing | Verify distance, conveyor speed, actuator response, and bottle position |
| Reject confirmation | Confirm every commanded reject reaches the reject path |
| Interlocks | Test guards, emergency stops, low-air alarm, and access switches |
| Line-stop recovery | Confirm tracked results are cleared or safely quarantined after a stop |
| Production validation | Run the system at normal speed and inspect accepted and rejected products |
I also review the first production batch after commissioning. The review should compare the leak tester's reject count with manual inspection results, rejected-bottle samples, and any downstream customer complaints. If the counts do not agree, I stop treating the system as validated until the tracking and test logic are corrected.
Offline leak testing is useful for laboratory approval, periodic sampling, troubleshooting, and product development. It gives an operator more time to position the container and inspect the test result, but it does not automatically evaluate every bottle leaving the blow molding line.
Inline testing evaluates containers at production speed and can connect the result to automatic rejection. Its main technical demands are accurate product tracking, stable handling, repeatable fixture sealing, and correct PLC timing. For high-volume PET bottle manufacturing or HDPE container production, inline inspection can reduce the delay between defect formation and process correction.
The best leak test method for empty plastic bottles depends on the defect type and package design. Pressure decay is suitable when the container can be sealed and pressurized consistently, while vacuum methods may be useful for certain shapes or test arrangements. I select the method only after confirming sensitivity, cycle time, container deformation, and the acceptance criteria required by the finished product.
Inline Leak Testing for Automatic Blow Molding Machines: How to Set Up Rejection requires more than connecting a leak tester to a conveyor. I first stabilize bottle handling, install a reliable presence sensor, select pressure-decay parameters, and validate the test fixture with good and challenge containers. I then separate tester results from PLC reject logic, calculate the delay from conveyor geometry and speed, test the actuator, and confirm that each failed bottle reaches the reject path.
For an Automatic Blow Molding Machine supplied or configured by Cangzhou Yi Sunda Mechanical Technology Co.,LTD, I would define the leak-test station around the actual bottle material, volume, cavity layout, transfer method, and production rate. The company describes equipment covering containers from 100 ml to 1,250 liters, including HDPE and other polymer applications, so the leak-testing recipe must be matched to each product rather than treated as a universal setting. After calibration, I would complete dry runs, interlock tests, challenge-bottle trials, reject verification, and production validation before releasing the line for routine operation.
