CVT Pulley Cylinder Pressure Test: Internal Leakage and Pressure Loss
Process step 6, explained: the dedicated test for the CVT pulley cylinder, and how to read a result.
Design diagram only, not a product photograph.
Where the test sits in the process
The company process documents describe a ten-step remanufacturing process, and step 6 is component bench testing. In that step, solenoids, sensors and valve-body assemblies are tested on the bench for oil flow, pressure stability, switching response and sealing. CVT units receive something additional in the same step: a dedicated pulley-cylinder pressure test, checking for internal leakage and pressure loss. The word dedicated matters. The cylinder is not tested only as part of a general bench test. It has a test of its own, because its pressure behaviour is central to how a CVT works, and because its failure mode, leakage, is not visible from outside.
The parts table gives the same story from the catalogue side. The CVT pulley cylinder row carries the note CVT core pressure-test part. Core, because the cylinder is central to the pulley movement. Pressure-test part, because pressure testing is how it is checked. Table and process agree, and this guide rests on that agreement: the cylinder is defined, in the materials, by its pressure test.
What the cylinder does, and why pressure is the test
A pulley cylinder moves a pulley. Hydraulic pressure applied to the cylinder changes the pulley's working diameter, and the ratio follows the diameters. The cylinder is therefore a pressure device before it is anything else: its job is to convert pressure into movement, and to hold the movement while the pressure is held. A test that checks anything other than pressure behaviour would miss the job. A test that checks pressure behaviour, and only pressure behaviour, checks exactly the job.
Two pressure failures are named in the process summary, and this guide keeps them distinct, because they are different failures with different consequences. Internal leakage is pressure escaping past a seal or a surface inside the cylinder, so the pressure that should move or hold the pulley is lost internally, without any visible external leak. Pressure loss is the broader outcome: the pressure available at the cylinder falls below what is needed, whether from leakage or from another cause upstream. Internal leakage is a cause. Pressure loss is the result that the driver or technician experiences. A test report that says only pressure loss, without saying whether internal leakage was found, has reported the result without the cause, and the reader should ask for the cause.
What the test checks, in the documents' own terms
The process summary says the test checks for internal leakage and pressure loss. It does not give a test pressure, a hold time, a leakage rate or a pass threshold, and this guide does not invent any of those numbers. Inventing a threshold would be worse than giving none, because a reader who compared a supplier's result against an invented threshold would draw a false conclusion in either direction: passing a real cylinder against a made-up bar, or failing a good cylinder against one.
What the guide can say, without inventing a number, is what a complete test report should contain. It should name the cylinder tested, by row and by identification. It should name the test, pulley-cylinder pressure test, as distinct from the general bench testing in the same step. It should report whether internal leakage was found, and whether pressure was held or lost, in the terms the tester uses. And it should be tied to the specific cylinder or unit ordered, not presented as a general claim that all cylinders pass. A report that does those four things is checkable. A report that says only tested and passed is not, however reassuring it sounds.
Why the test matters to a buyer who is not testing anything
Most readers of this guide will never run a pulley-cylinder pressure test themselves. They will buy a cylinder, a pulley set or an assembly, and the test matters to them in a different way: as a question to ask. A buyer who knows the test exists can ask whether the cylinder offered has had it, and can ask for the result in the terms above. A buyer who does not know the test exists will accept a general testing claim, and will not know that a specific, named test was never mentioned.
The question is worth asking for a second reason. A cylinder fault produces symptoms that look like other faults: slipping, ratio behaviour changes, drive hesitation. A buyer who orders a belt for a slipping unit, without asking about the cylinder test, may be ordering for the wrong row. A buyer who asks about the cylinder test, and about the pulley inspection that sits alongside it in the process, is asking the questions that separate the candidate rows. The slipping guide on this site shows that separation in full, and this guide is its companion on the pressure side.
The test in relation to the other testing layers
The pressure test is one layer in a three-layer testing story, and confusing the layers is a common source of false confidence. The first layer is defect screening in step 2, where pulleys and other core parts found worn, deformed, fatigued or out of tolerance are rejected. The second layer is the bench testing of step 6, including the cylinder pressure test. The third layer is the finished-unit testing of step 9, where the documents say every finished unit receives full-load dynamometer testing, with a quality file kept for each unit. A cylinder pressure-test result is a step 6 result. It is not a step 9 finished-unit result, and it does not replace one. A buyer who is offered a finished unit should ask about both layers, and a buyer who is offered a single cylinder should ask about the layer that applies to that cylinder.
The company materials also mention intelligent testing equipment for initial inspection and a vehicle test in the quality summary. Those descriptions sit alongside the ten-step process, and a buyer can ask which of them applies to a CVT order. This guide does not merge them into the step 6 test, because the documents do not say they are the same test, and merging them would create a test that exists in no document.
Questions to ask about a pressure-test result
Ask which cylinder the result covers, by row and identification. Ask whether the test was the dedicated pulley-cylinder pressure test or a general bench test. Ask whether internal leakage was found, separately from whether pressure was lost. Ask whether the result belongs to the cylinder being ordered, or to a sample, a batch or a different unit. Ask for the result in writing, kept with the order record, so the delivered cylinder can be tied back to the tested cylinder. Those five questions cost nothing, and they turn a reassuring sentence into a checkable record.
That is the whole guide, and it deliberately ends without a threshold. The threshold for a specific cylinder is the supplier's threshold for that cylinder, reported with the test, not a number in a guide. A guide that gave you a number would be giving you a false certainty. A guide that gives you the questions gives you something better: the ability to tell a real test result from a reassuring sentence.