By Jerome Fryer, DC
Introduction: why joint cracking needed a laboratory
Joint cracking is one of the most familiar—and most misunderstood—phenomena in musculoskeletal care. For decades, explanations focused on tissues snapping or bubbles collapsing, largely because the event happens too quickly and too deep inside the joint to observe directly in living humans.
To move beyond inference, we needed a system where joint cracking could be provoked on demand, measured precisely, and observed at high temporal resolution. That was the motivation behind a series of controlled experiments conducted at the University of British Columbia (UBC), using a sealed in-vitro joint model designed to reproduce the essential mechanical and fluid conditions of a real synovial joint.
This article explains what those experiments revealed—and why they matter for how we understand joint motion.
Why in-vivo observation alone is not enough
Real-time MRI has shown that joint cracking coincides with the sudden appearance of an intra-articular cavity. What MRI cannot resolve, however, are the millisecond-level mechanical events that lead up to that moment. The forces, accelerations, sound generation, and early fluid behavior all occur faster than most clinical imaging modalities can capture.
A laboratory joint model allows those missing pieces to be measured simultaneously.
Building a joint that behaves like a joint
The UBC experimental system was designed around one principle: a joint must be sealed and fluid-filled to crack.
The model consisted of two rigid surfaces joined by a compliant enclosure and submerged in degassed water to eliminate pre-existing gas nuclei. A motorized translation system applied controlled vertical distraction while multiple sensors recorded the event:
- Load cell to measure tensile force
- Accelerometer to capture rapid mechanical acceleration
- Hydrophone to record sound pressure
- High-speed camera operating at thousands of frames per second
All signals were synchronized through a data-acquisition system, allowing force, motion, sound, and imagery to be aligned on a single timeline.
This setup made it possible to observe not just that a joint cracked, but how and when each physical variable changed.
What happens in the milliseconds around a crack
One of the most important findings from the UBC experiments was the sequence of events leading up to the crack.
As tensile load increased, the joint surfaces did not separate smoothly. Instead, separation resisted until a critical threshold was reached. At that moment:
- Joint separation accelerated abruptly
- A sharp acoustic signal was recorded
- A fluid cavity appeared in the joint space within the next imaging frame
At first glance, the sound appeared to precede cavitation. However, careful synchronization testing revealed a small, consistent delay between the camera trigger and the sensor signals. When this delay was accounted for, the data showed that sound and cavity formation are effectively simultaneous.
This aligns with in-vivo MRI findings and supports the conclusion that the audible crack corresponds to cavity inception, not collapse.
Why some cracks are louder than others
The UBC data also clarified why joint cracks vary so widely in loudness.
Across repeated trials, two relationships were consistent:
- Higher tensile load at the moment of cracking produced higher sound amplitude
- Greater pre-compression of the joint reduced the load required to crack
In other words, louder cracks reflected higher stored mechanical energy at the moment the cavity formed. The frequency content of the sound remained relatively stable, while amplitude varied with load.
This explains a common clinical observation: the same joint can crack quietly one day and loudly another, without any implication of tissue damage.
The refractory period explained mechanically
After a joint cracks, it cannot be cracked again immediately. This refractory period has been recognized clinically for decades, but its mechanism was unclear.
The UBC experiments demonstrated that refractory behavior depends strongly on post-crack compression. When the joint was sufficiently compressed after cracking, the cavity dissolved back into solution over time, allowing the joint to crack again. Without compression, the cavity persisted and further cracking was not possible.
From a physical standpoint, this means:
A joint cannot crack again until the fluid environment returns to a pre-cavitation state.
The refractory period is therefore a fluid-mechanical reset, not a neurological or tissue-fatigue phenomenon.
How this fits with earlier joint-cracking research
The UBC findings complete a progression of evidence:
- In-vivo MRI demonstrated that cracking coincides with cavity formation
- Early in-vitro models showed that sealed, denucleated systems are required
- UBC experiments revealed how force, acceleration, sound, and cavitation are coupled in time
Together, these studies support a unified view: joint cracking is a rapid, localized fluid event that occurs when synovial joints separate under tension.
Why this matters for understanding joints
Joint cracking is not a sign of tissue failure. It is a normal consequence of how synovial fluid behaves when joints move rapidly under load. Understanding this reframes cracking as a mechanical marker of joint separation, not something inherently harmful.
More broadly, these findings reinforce the idea that joints should be understood as dynamic fluid systems, not simply as bones connected by ligaments. Appreciating the role of fluid mechanics opens new ways of thinking about joint motion, modeling, and education.
About the author
Jerome Fryer DC, is a chiropractor and clinical researcher with a longstanding focus on joint biomechanics and patient education. He is a co-author of peer-reviewed studies on joint cavitation, including real-time MRI investigations and laboratory-based joint models. His work bridges clinical practice, experimental research, and anatomical modeling to better understand how joints function under load.
This work was conducted by Johanna SUAT
06/03/2019 – 09/06/2019
Under the supervision of Dana Grecov pHD at the University of British Columbia, Mechanical Engineering Department, with the lead investigator, Jerome Fryer, DC. Funding was secured through a GoFundMe campaign.
